From 4b536019c3ee11ad82e9815d3812a174b6fdd55f Mon Sep 17 00:00:00 2001 From: Yamagi Date: Tue, 25 May 2021 15:29:22 +0200 Subject: [PATCH] Add the vkQuake2 Vulkan renderer ported to YQ2. This was first developed in a feature branch in the main yquake2 repo. It was merged into master in early 2021, but the experiences of the following month showed that it is not ready for prime time. There're glitches with 3rd party assets, restarts are still shaky, etc. Having the code in a separate repo allows us to: * Release Vulkan independent if YQ2. * Give commit access to contributors interested in Vulkan. This code is the same as in yquake/yquake2 ecdf912713eef55d6c5d5a772259b44e3fc232c4. --- LICENSE | 396 ++ Makefile | 362 ++ README.md | 111 + shaders/basic.frag | 24 + shaders/basic.vert | 30 + shaders/basic_color_quad.frag | 10 + shaders/basic_color_quad.vert | 24 + shaders/beam.vert | 25 + shaders/d_light.vert | 21 + shaders/model.frag | 17 + shaders/model.vert | 32 + shaders/nullmodel.vert | 26 + shaders/particle.vert | 26 + shaders/point_particle.frag | 14 + shaders/point_particle.vert | 37 + shaders/polygon.vert | 30 + shaders/polygon_lmap.frag | 17 + shaders/polygon_lmap.vert | 32 + shaders/polygon_warp.vert | 34 + shaders/postprocess.frag | 33 + shaders/postprocess.vert | 17 + shaders/shaders.sh | 25 + shaders/shadows.vert | 24 + shaders/skybox.vert | 30 + shaders/sprite.vert | 26 + shaders/world_warp.frag | 55 + shaders/world_warp.vert | 17 + src/backends/hunk_unix.c | 173 + src/backends/hunk_windows.c | 97 + src/common/header/common.h | 840 +++ src/common/header/crc.h | 33 + src/common/header/files.h | 453 ++ src/common/header/ref_api.h | 283 + src/common/header/ref_shared.h | 78 + src/common/header/shared.h | 1189 ++++ src/common/header/vid.h | 69 + src/common/md4.c | 224 + src/common/shared.c | 1419 +++++ src/constants/anorms.h | 188 + src/constants/anormtab.h | 236 + src/constants/warpsin.h | 72 + src/files/pcx.c | 317 ++ src/files/pvs.c | 90 + src/files/stb.c | 425 ++ src/files/stb_image.h | 7559 ++++++++++++++++++++++++++ src/files/stb_image_resize.h | 2630 +++++++++ src/files/wal.c | 86 + src/vk/header/local.h | 331 ++ src/vk/header/model.h | 254 + src/vk/header/qvk.h | 326 ++ src/vk/header/shaders.h | 115 + src/vk/header/util.h | 71 + src/vk/spirv/basic_color_quad_frag.c | 16 + src/vk/spirv/basic_color_quad_vert.c | 48 + src/vk/spirv/basic_frag.c | 48 + src/vk/spirv/basic_vert.c | 61 + src/vk/spirv/beam_vert.c | 42 + src/vk/spirv/d_light_vert.c | 41 + src/vk/spirv/model_frag.c | 37 + src/vk/spirv/model_vert.c | 59 + src/vk/spirv/nullmodel_vert.c | 49 + src/vk/spirv/particle_vert.c | 47 + src/vk/spirv/point_particle_frag.c | 28 + src/vk/spirv/point_particle_vert.c | 82 + src/vk/spirv/polygon_lmap_frag.c | 40 + src/vk/spirv/polygon_lmap_vert.c | 59 + src/vk/spirv/polygon_vert.c | 51 + src/vk/spirv/polygon_warp_vert.c | 80 + src/vk/spirv/postprocess_frag.c | 64 + src/vk/spirv/postprocess_vert.c | 36 + src/vk/spirv/shadows_vert.c | 46 + src/vk/spirv/skybox_vert.c | 54 + src/vk/spirv/sprite_vert.c | 49 + src/vk/spirv/world_warp_frag.c | 154 + src/vk/spirv/world_warp_vert.c | 33 + src/vk/vk_buffer.c | 205 + src/vk/vk_cmd.c | 91 + src/vk/vk_common.c | 2644 +++++++++ src/vk/vk_device.c | 365 ++ src/vk/vk_draw.c | 371 ++ src/vk/vk_image.c | 1584 ++++++ src/vk/vk_light.c | 626 +++ src/vk/vk_mesh.c | 999 ++++ src/vk/vk_model.c | 1525 ++++++ src/vk/vk_pipeline.c | 222 + src/vk/vk_rmain.c | 1761 ++++++ src/vk/vk_rmisc.c | 325 ++ src/vk/vk_rsurf.c | 1324 +++++ src/vk/vk_shaders.c | 85 + src/vk/vk_swapchain.c | 250 + src/vk/vk_util.c | 620 +++ src/vk/vk_validation.c | 99 + src/vk/vk_warp.c | 732 +++ src/vk/volk/volk.c | 1905 +++++++ src/vk/volk/volk.h | 1257 +++++ 95 files changed, 37267 insertions(+) create mode 100644 LICENSE create mode 100644 Makefile create mode 100644 README.md create mode 100644 shaders/basic.frag create mode 100644 shaders/basic.vert create mode 100644 shaders/basic_color_quad.frag create mode 100644 shaders/basic_color_quad.vert create mode 100644 shaders/beam.vert create mode 100644 shaders/d_light.vert create mode 100644 shaders/model.frag create mode 100644 shaders/model.vert create mode 100644 shaders/nullmodel.vert create mode 100644 shaders/particle.vert create mode 100644 shaders/point_particle.frag create mode 100644 shaders/point_particle.vert create mode 100644 shaders/polygon.vert create mode 100644 shaders/polygon_lmap.frag create mode 100644 shaders/polygon_lmap.vert create mode 100644 shaders/polygon_warp.vert create mode 100644 shaders/postprocess.frag create mode 100644 shaders/postprocess.vert create mode 100755 shaders/shaders.sh create mode 100644 shaders/shadows.vert create mode 100644 shaders/skybox.vert create mode 100644 shaders/sprite.vert create mode 100644 shaders/world_warp.frag create mode 100644 shaders/world_warp.vert create mode 100644 src/backends/hunk_unix.c create mode 100644 src/backends/hunk_windows.c create mode 100644 src/common/header/common.h create mode 100644 src/common/header/crc.h create mode 100644 src/common/header/files.h create mode 100644 src/common/header/ref_api.h create mode 100644 src/common/header/ref_shared.h create mode 100644 src/common/header/shared.h create mode 100644 src/common/header/vid.h create mode 100644 src/common/md4.c create mode 100644 src/common/shared.c create mode 100644 src/constants/anorms.h create mode 100644 src/constants/anormtab.h create mode 100644 src/constants/warpsin.h create mode 100644 src/files/pcx.c create mode 100644 src/files/pvs.c create mode 100644 src/files/stb.c create mode 100644 src/files/stb_image.h create mode 100644 src/files/stb_image_resize.h create mode 100644 src/files/wal.c create mode 100644 src/vk/header/local.h create mode 100644 src/vk/header/model.h create mode 100644 src/vk/header/qvk.h create mode 100644 src/vk/header/shaders.h create mode 100644 src/vk/header/util.h create mode 100644 src/vk/spirv/basic_color_quad_frag.c create mode 100644 src/vk/spirv/basic_color_quad_vert.c create mode 100644 src/vk/spirv/basic_frag.c create mode 100644 src/vk/spirv/basic_vert.c create mode 100644 src/vk/spirv/beam_vert.c create mode 100644 src/vk/spirv/d_light_vert.c create mode 100644 src/vk/spirv/model_frag.c create mode 100644 src/vk/spirv/model_vert.c create mode 100644 src/vk/spirv/nullmodel_vert.c create mode 100644 src/vk/spirv/particle_vert.c create mode 100644 src/vk/spirv/point_particle_frag.c create mode 100644 src/vk/spirv/point_particle_vert.c create mode 100644 src/vk/spirv/polygon_lmap_frag.c create mode 100644 src/vk/spirv/polygon_lmap_vert.c create mode 100644 src/vk/spirv/polygon_vert.c create mode 100644 src/vk/spirv/polygon_warp_vert.c create mode 100644 src/vk/spirv/postprocess_frag.c create mode 100644 src/vk/spirv/postprocess_vert.c create mode 100644 src/vk/spirv/shadows_vert.c create mode 100644 src/vk/spirv/skybox_vert.c create mode 100644 src/vk/spirv/sprite_vert.c create mode 100644 src/vk/spirv/world_warp_frag.c create mode 100644 src/vk/spirv/world_warp_vert.c create mode 100644 src/vk/vk_buffer.c create mode 100644 src/vk/vk_cmd.c create mode 100644 src/vk/vk_common.c create mode 100644 src/vk/vk_device.c create mode 100644 src/vk/vk_draw.c create mode 100644 src/vk/vk_image.c create mode 100644 src/vk/vk_light.c create mode 100644 src/vk/vk_mesh.c create mode 100644 src/vk/vk_model.c create mode 100644 src/vk/vk_pipeline.c create mode 100644 src/vk/vk_rmain.c create mode 100644 src/vk/vk_rmisc.c create mode 100644 src/vk/vk_rsurf.c create mode 100644 src/vk/vk_shaders.c create mode 100644 src/vk/vk_swapchain.c create mode 100644 src/vk/vk_util.c create mode 100644 src/vk/vk_validation.c create mode 100644 src/vk/vk_warp.c create mode 100644 src/vk/volk/volk.c create mode 100644 src/vk/volk/volk.h diff --git a/LICENSE b/LICENSE new file mode 100644 index 0000000..7960ca3 --- /dev/null +++ b/LICENSE @@ -0,0 +1,396 @@ +The Yamagi Quake II Vulkan renderer lib is based upon vkQuake2 and +ontains software developed by multiple individuals, projects and +organisations. Following is a list of this software and copys of each +license: + +- Quake II +- stb_image.h, stb_image_write.h + +=============================================================================== + + GNU GENERAL PUBLIC LICENSE + Version 2, June 1991 + + Copyright (C) 1989, 1991 Free Software Foundation, Inc., + 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + Everyone is permitted to copy and distribute verbatim copies + of this license document, but changing it is not allowed. + + Preamble + + The licenses for most software are designed to take away your +freedom to share and change it. By contrast, the GNU General Public +License is intended to guarantee your freedom to share and change free +software--to make sure the software is free for all its users. This +General Public License applies to most of the Free Software +Foundation's software and to any other program whose authors commit to +using it. (Some other Free Software Foundation software is covered by +the GNU Lesser General Public License instead.) You can apply it to +your programs, too. + + When we speak of free software, we are referring to freedom, not +price. Our General Public Licenses are designed to make sure that you +have the freedom to distribute copies of free software (and charge for +this service if you wish), that you receive source code or can get it +if you want it, that you can change the software or use pieces of it +in new free programs; and that you know you can do these things. + + To protect your rights, we need to make restrictions that forbid +anyone to deny you these rights or to ask you to surrender the rights. +These restrictions translate to certain responsibilities for you if you +distribute copies of the software, or if you modify it. + + For example, if you distribute copies of such a program, whether +gratis or for a fee, you must give the recipients all the rights that +you have. You must make sure that they, too, receive or can get the +source code. And you must show them these terms so they know their +rights. + + We protect your rights with two steps: (1) copyright the software, and +(2) offer you this license which gives you legal permission to copy, +distribute and/or modify the software. + + Also, for each author's protection and ours, we want to make certain +that everyone understands that there is no warranty for this free +software. If the software is modified by someone else and passed on, we +want its recipients to know that what they have is not the original, so +that any problems introduced by others will not reflect on the original +authors' reputations. + + Finally, any free program is threatened constantly by software +patents. We wish to avoid the danger that redistributors of a free +program will individually obtain patent licenses, in effect making the +program proprietary. To prevent this, we have made it clear that any +patent must be licensed for everyone's free use or not licensed at all. + + The precise terms and conditions for copying, distribution and +modification follow. + + GNU GENERAL PUBLIC LICENSE + TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION + + 0. This License applies to any program or other work which contains +a notice placed by the copyright holder saying it may be distributed +under the terms of this General Public License. The "Program", below, +refers to any such program or work, and a "work based on the Program" +means either the Program or any derivative work under copyright law: +that is to say, a work containing the Program or a portion of it, +either verbatim or with modifications and/or translated into another +language. (Hereinafter, translation is included without limitation in +the term "modification".) Each licensee is addressed as "you". + +Activities other than copying, distribution and modification are not +covered by this License; they are outside its scope. The act of +running the Program is not restricted, and the output from the Program +is covered only if its contents constitute a work based on the +Program (independent of having been made by running the Program). +Whether that is true depends on what the Program does. + + 1. You may copy and distribute verbatim copies of the Program's +source code as you receive it, in any medium, provided that you +conspicuously and appropriately publish on each copy an appropriate +copyright notice and disclaimer of warranty; keep intact all the +notices that refer to this License and to the absence of any warranty; +and give any other recipients of the Program a copy of this License +along with the Program. + +You may charge a fee for the physical act of transferring a copy, and +you may at your option offer warranty protection in exchange for a fee. + + 2. You may modify your copy or copies of the Program or any portion +of it, thus forming a work based on the Program, and copy and +distribute such modifications or work under the terms of Section 1 +above, provided that you also meet all of these conditions: + + a) You must cause the modified files to carry prominent notices + stating that you changed the files and the date of any change. + + b) You must cause any work that you distribute or publish, that in + whole or in part contains or is derived from the Program or any + part thereof, to be licensed as a whole at no charge to all third + parties under the terms of this License. + + c) If the modified program normally reads commands interactively + when run, you must cause it, when started running for such + interactive use in the most ordinary way, to print or display an + announcement including an appropriate copyright notice and a + notice that there is no warranty (or else, saying that you provide + a warranty) and that users may redistribute the program under + these conditions, and telling the user how to view a copy of this + License. (Exception: if the Program itself is interactive but + does not normally print such an announcement, your work based on + the Program is not required to print an announcement.) + +These requirements apply to the modified work as a whole. If +identifiable sections of that work are not derived from the Program, +and can be reasonably considered independent and separate works in +themselves, then this License, and its terms, do not apply to those +sections when you distribute them as separate works. But when you +distribute the same sections as part of a whole which is a work based +on the Program, the distribution of the whole must be on the terms of +this License, whose permissions for other licensees extend to the +entire whole, and thus to each and every part regardless of who wrote it. + +Thus, it is not the intent of this section to claim rights or contest +your rights to work written entirely by you; rather, the intent is to +exercise the right to control the distribution of derivative or +collective works based on the Program. + +In addition, mere aggregation of another work not based on the Program +with the Program (or with a work based on the Program) on a volume of +a storage or distribution medium does not bring the other work under +the scope of this License. + + 3. You may copy and distribute the Program (or a work based on it, +under Section 2) in object code or executable form under the terms of +Sections 1 and 2 above provided that you also do one of the following: + + a) Accompany it with the complete corresponding machine-readable + source code, which must be distributed under the terms of Sections + 1 and 2 above on a medium customarily used for software interchange; or, + + b) Accompany it with a written offer, valid for at least three + years, to give any third party, for a charge no more than your + cost of physically performing source distribution, a complete + machine-readable copy of the corresponding source code, to be + distributed under the terms of Sections 1 and 2 above on a medium + customarily used for software interchange; or, + + c) Accompany it with the information you received as to the offer + to distribute corresponding source code. (This alternative is + allowed only for noncommercial distribution and only if you + received the program in object code or executable form with such + an offer, in accord with Subsection b above.) + +The source code for a work means the preferred form of the work for +making modifications to it. For an executable work, complete source +code means all the source code for all modules it contains, plus any +associated interface definition files, plus the scripts used to +control compilation and installation of the executable. However, as a +special exception, the source code distributed need not include +anything that is normally distributed (in either source or binary +form) with the major components (compiler, kernel, and so on) of the +operating system on which the executable runs, unless that component +itself accompanies the executable. + +If distribution of executable or object code is made by offering +access to copy from a designated place, then offering equivalent +access to copy the source code from the same place counts as +distribution of the source code, even though third parties are not +compelled to copy the source along with the object code. + + 4. You may not copy, modify, sublicense, or distribute the Program +except as expressly provided under this License. Any attempt +otherwise to copy, modify, sublicense or distribute the Program is +void, and will automatically terminate your rights under this License. +However, parties who have received copies, or rights, from you under +this License will not have their licenses terminated so long as such +parties remain in full compliance. + + 5. You are not required to accept this License, since you have not +signed it. However, nothing else grants you permission to modify or +distribute the Program or its derivative works. These actions are +prohibited by law if you do not accept this License. Therefore, by +modifying or distributing the Program (or any work based on the +Program), you indicate your acceptance of this License to do so, and +all its terms and conditions for copying, distributing or modifying +the Program or works based on it. + + 6. Each time you redistribute the Program (or any work based on the +Program), the recipient automatically receives a license from the +original licensor to copy, distribute or modify the Program subject to +these terms and conditions. You may not impose any further +restrictions on the recipients' exercise of the rights granted herein. +You are not responsible for enforcing compliance by third parties to +this License. + + 7. If, as a consequence of a court judgment or allegation of patent +infringement or for any other reason (not limited to patent issues), +conditions are imposed on you (whether by court order, agreement or +otherwise) that contradict the conditions of this License, they do not +excuse you from the conditions of this License. If you cannot +distribute so as to satisfy simultaneously your obligations under this +License and any other pertinent obligations, then as a consequence you +may not distribute the Program at all. For example, if a patent +license would not permit royalty-free redistribution of the Program by +all those who receive copies directly or indirectly through you, then +the only way you could satisfy both it and this License would be to +refrain entirely from distribution of the Program. + +If any portion of this section is held invalid or unenforceable under +any particular circumstance, the balance of the section is intended to +apply and the section as a whole is intended to apply in other +circumstances. + +It is not the purpose of this section to induce you to infringe any +patents or other property right claims or to contest validity of any +such claims; this section has the sole purpose of protecting the +integrity of the free software distribution system, which is +implemented by public license practices. Many people have made +generous contributions to the wide range of software distributed +through that system in reliance on consistent application of that +system; it is up to the author/donor to decide if he or she is willing +to distribute software through any other system and a licensee cannot +impose that choice. + +This section is intended to make thoroughly clear what is believed to +be a consequence of the rest of this License. + + 8. If the distribution and/or use of the Program is restricted in +certain countries either by patents or by copyrighted interfaces, the +original copyright holder who places the Program under this License +may add an explicit geographical distribution limitation excluding +those countries, so that distribution is permitted only in or among +countries not thus excluded. In such case, this License incorporates +the limitation as if written in the body of this License. + + 9. The Free Software Foundation may publish revised and/or new versions +of the General Public License from time to time. Such new versions will +be similar in spirit to the present version, but may differ in detail to +address new problems or concerns. + +Each version is given a distinguishing version number. If the Program +specifies a version number of this License which applies to it and "any +later version", you have the option of following the terms and conditions +either of that version or of any later version published by the Free +Software Foundation. If the Program does not specify a version number of +this License, you may choose any version ever published by the Free Software +Foundation. + + 10. If you wish to incorporate parts of the Program into other free +programs whose distribution conditions are different, write to the author +to ask for permission. For software which is copyrighted by the Free +Software Foundation, write to the Free Software Foundation; we sometimes +make exceptions for this. Our decision will be guided by the two goals +of preserving the free status of all derivatives of our free software and +of promoting the sharing and reuse of software generally. + + NO WARRANTY + + 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY +FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN +OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES +PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED +OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF +MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS +TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE +PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, +REPAIR OR CORRECTION. + + 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING +WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR +REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, +INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING +OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED +TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY +YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER +PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE +POSSIBILITY OF SUCH DAMAGES. + + END OF TERMS AND CONDITIONS + + How to Apply These Terms to Your New Programs + + If you develop a new program, and you want it to be of the greatest +possible use to the public, the best way to achieve this is to make it +free software which everyone can redistribute and change under these terms. + + To do so, attach the following notices to the program. It is safest +to attach them to the start of each source file to most effectively +convey the exclusion of warranty; and each file should have at least +the "copyright" line and a pointer to where the full notice is found. + + + Copyright (C) + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, Inc., + 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. + +Also add information on how to contact you by electronic and paper mail. + +If the program is interactive, make it output a short notice like this +when it starts in an interactive mode: + + Gnomovision version 69, Copyright (C) year name of author + Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'. + This is free software, and you are welcome to redistribute it + under certain conditions; type `show c' for details. + +The hypothetical commands `show w' and `show c' should show the appropriate +parts of the General Public License. Of course, the commands you use may +be called something other than `show w' and `show c'; they could even be +mouse-clicks or menu items--whatever suits your program. + +You should also get your employer (if you work as a programmer) or your +school, if any, to sign a "copyright disclaimer" for the program, if +necessary. Here is a sample; alter the names: + + Yoyodyne, Inc., hereby disclaims all copyright interest in the program + `Gnomovision' (which makes passes at compilers) written by James Hacker. + + , 1 April 1989 + Ty Coon, President of Vice + +This General Public License does not permit incorporating your program into +proprietary programs. If your program is a subroutine library, you may +consider it more useful to permit linking proprietary applications with the +library. If this is what you want to do, use the GNU Lesser General +Public License instead of this License. + +=============================================================================== + + stb single-file public domain libraries + (stb_image.h, stb_image_resize.h) + https://github.com/nothings/stb/ + Dual-Licensed: Public Domain (Unlicense) or MIT License: + +This software is available under 2 licenses -- choose whichever you prefer. +------------------------------------------------------------------------------ +ALTERNATIVE A - MIT License +Copyright (c) 2017 Sean Barrett +Permission is hereby granted, free of charge, to any person obtaining a copy of +this software and associated documentation files (the "Software"), to deal in +the Software without restriction, including without limitation the rights to +use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies +of the Software, and to permit persons to whom the Software is furnished to do +so, subject to the following conditions: +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. +------------------------------------------------------------------------------ +ALTERNATIVE B - Public Domain (www.unlicense.org) +This is free and unencumbered software released into the public domain. +Anyone is free to copy, modify, publish, use, compile, sell, or distribute this +software, either in source code form or as a compiled binary, for any purpose, +commercial or non-commercial, and by any means. +In jurisdictions that recognize copyright laws, the author or authors of this +software dedicate any and all copyright interest in the software to the public +domain. We make this dedication for the benefit of the public at large and to +the detriment of our heirs and successors. We intend this dedication to be an +overt act of relinquishment in perpetuity of all present and future rights to +this software under copyright law. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN +ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION +WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + +=============================================================================== diff --git a/Makefile b/Makefile new file mode 100644 index 0000000..d528a5a --- /dev/null +++ b/Makefile @@ -0,0 +1,362 @@ +# ----------------------------------------------------- # +# Makefile for the Vulkan renderer lib for Quake II # +# # +# Just type "make" to compile the # +# - Vulkan renderer lib (ref_vk.so / rev_vk.dll) # +# # +# Dependencies: # +# - SDL2 # +# - Vulkan headers # +# # +# Platforms: # +# - FreeBSD # +# - Linux # +# - Mac OS X # +# - OpenBSD # +# - Windows # +# ----------------------------------------------------- # + +# Detect the OS +ifdef SystemRoot +YQ2_OSTYPE ?= Windows +else +YQ2_OSTYPE ?= $(shell uname -s) +endif + +# Special case for MinGW +ifneq (,$(findstring MINGW,$(YQ2_OSTYPE))) +YQ2_OSTYPE := Windows +endif + +# Detect the architecture +ifeq ($(YQ2_OSTYPE), Windows) +ifdef MINGW_CHOST +ifeq ($(MINGW_CHOST), x86_64-w64-mingw32) +YQ2_ARCH ?= x86_64 +else # i686-w64-mingw32 +YQ2_ARCH ?= i386 +endif +else # windows, but MINGW_CHOST not defined +ifdef PROCESSOR_ARCHITEW6432 +# 64 bit Windows +YQ2_ARCH ?= $(PROCESSOR_ARCHITEW6432) +else +# 32 bit Windows +YQ2_ARCH ?= $(PROCESSOR_ARCHITECTURE) +endif +endif # windows but MINGW_CHOST not defined +else +# Normalize some abiguous YQ2_ARCH strings +YQ2_ARCH ?= $(shell uname -m | sed -e 's/i.86/i386/' -e 's/amd64/x86_64/' -e 's/^arm.*/arm/') +endif + +# On Windows / MinGW $(CC) is undefined by default. +ifeq ($(YQ2_OSTYPE),Windows) +CC ?= gcc +endif + +# Detect the compiler +ifeq ($(shell $(CC) -v 2>&1 | grep -c "clang version"), 1) +COMPILER := clang +COMPILERVER := $(shell $(CC) -dumpversion | sed -e 's/\.\([0-9][0-9]\)/\1/g' -e 's/\.\([0-9]\)/0\1/g' -e 's/^[0-9]\{3,4\}$$/&00/') +else ifeq ($(shell $(CC) -v 2>&1 | grep -c -E "(gcc version|gcc-Version)"), 1) +COMPILER := gcc +COMPILERVER := $(shell $(CC) -dumpversion | sed -e 's/\.\([0-9][0-9]\)/\1/g' -e 's/\.\([0-9]\)/0\1/g' -e 's/^[0-9]\{3,4\}$$/&00/') +else +COMPILER := unknown +endif + +# ASAN includes DEBUG +ifdef ASAN +DEBUG=1 +endif + +# UBSAN includes DEBUG +ifdef UBSAN +DEBUG=1 +endif + +# ---------- + +# Base CFLAGS. These may be overridden by the environment. +# Highest supported optimizations are -O2, higher levels +# will likely break this crappy code. +ifdef DEBUG +CFLAGS ?= -O0 -g -Wall -pipe +ifdef ASAN +CFLAGS += -fsanitize=address +endif +ifdef UBSAN +CFLAGS += -fsanitize=undefined +endif +else +CFLAGS ?= -O2 -Wall -pipe -fomit-frame-pointer +endif + +# Always needed are: +# -fno-strict-aliasing since the source doesn't comply +# with strict aliasing rules and it's next to impossible +# to get it there... +# -fwrapv for defined integer wrapping. MSVC6 did this +# and the game code requires it. +# -fvisibility=hidden to keep symbols hidden. This is +# mostly best practice and not really necessary. +override CFLAGS += -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden + +# -MMD to generate header dependencies. Unsupported by +# the Clang shipped with OS X. +ifneq ($(YQ2_OSTYPE), Darwin) +override CFLAGS += -MMD +endif + +# OS X architecture. +ifeq ($(YQ2_OSTYPE), Darwin) +override CFLAGS += -arch $(YQ2_ARCH) +endif + +# ---------- + +# Switch of some annoying warnings. +ifeq ($(COMPILER), clang) + # -Wno-missing-braces because otherwise clang complains + # about totally valid 'vec3_t bla = {0}' constructs. + CFLAGS += -Wno-missing-braces +else ifeq ($(COMPILER), gcc) + # GCC 8.0 or higher. + ifeq ($(shell test $(COMPILERVER) -ge 80000; echo $$?),0) + # -Wno-format-truncation and -Wno-format-overflow + # because GCC spams about 50 false positives. + CFLAGS += -Wno-format-truncation -Wno-format-overflow + endif +endif + +# ---------- + +# Defines the operating system and architecture +override CFLAGS += -DYQ2OSTYPE=\"$(YQ2_OSTYPE)\" -DYQ2ARCH=\"$(YQ2_ARCH)\" + +# ---------- + +# For reproduceable builds, look here for details: +# https://reproducible-builds.org/specs/source-date-epoch/ +ifdef SOURCE_DATE_EPOCH +CFLAGS += -DBUILD_DATE=\"$(shell date --utc --date="@${SOURCE_DATE_EPOCH}" +"%b %_d %Y" | sed -e 's/ /\\ /g')\" +endif + +# ---------- + +# Using the default x87 float math on 32bit x86 causes rounding trouble +# -ffloat-store could work around that, but the better solution is to +# just enforce SSE - every x86 CPU since Pentium3 supports that +# and this should even improve the performance on old CPUs +ifeq ($(YQ2_ARCH), i386) +override CFLAGS += -msse -mfpmath=sse +endif + +# Force SSE math on x86_64. All sane compilers should do this +# anyway, just to protect us from broken Linux distros. +ifeq ($(YQ2_ARCH), x86_64) +override CFLAGS += -mfpmath=sse +endif + +# ---------- + +# Extra CFLAGS for SDL. +SDLCFLAGS := $(shell sdl2-config --cflags) + +# ---------- + +# Base include path. +ifeq ($(YQ2_OSTYPE),Linux) +INCLUDE ?= -I/usr/include +else ifeq ($(YQ2_OSTYPE),FreeBSD) +INCLUDE ?= -I/usr/local/include +else ifeq ($(YQ2_OSTYPE),NetBSD) +INCLUDE ?= -I/usr/X11R7/include -I/usr/pkg/include +else ifeq ($(YQ2_OSTYPE),OpenBSD) +INCLUDE ?= -I/usr/local/include +else ifeq ($(YQ2_OSTYPE),Windows) +INCLUDE ?= -I/usr/include +endif + +# ---------- + +# Base LDFLAGS. This is just the library path. +ifeq ($(YQ2_OSTYPE),Linux) +LDFLAGS ?= -L/usr/lib +else ifeq ($(YQ2_OSTYPE),FreeBSD) +LDFLAGS ?= -L/usr/local/lib +else ifeq ($(YQ2_OSTYPE),NetBSD) +LDFLAGS ?= -L/usr/X11R7/lib -Wl,-R/usr/X11R7/lib -L/usr/pkg/lib -Wl,-R/usr/pkg/lib +else ifeq ($(YQ2_OSTYPE),OpenBSD) +LDFLAGS ?= -L/usr/local/lib +else ifeq ($(YQ2_OSTYPE),Windows) +LDFLAGS ?= -L/usr/lib +endif + +# Link address sanitizer if requested. +ifdef ASAN +LDFLAGS += -fsanitize=address +endif + +# Link undefined behavior sanitizer if requested. +ifdef UBSAN +LDFLAGS += -fsanitize=undefined +endif + +# Required libraries. +ifeq ($(YQ2_OSTYPE),Linux) +override LDFLAGS += -lm -ldl -rdynamic +else ifeq ($(YQ2_OSTYPE),FreeBSD) +override LDFLAGS += -lm +else ifeq ($(YQ2_OSTYPE),NetBSD) +override LDFLAGS += -lm +else ifeq ($(YQ2_OSTYPE),OpenBSD) +override LDFLAGS += -lm +else ifeq ($(YQ2_OSTYPE),Windows) +override LDFLAGS += -lws2_32 -lwinmm -static-libgcc +else ifeq ($(YQ2_OSTYPE), Darwin) +override LDFLAGS += -arch $(YQ2_ARCH) +else ifeq ($(YQ2_OSTYPE), Haiku) +override LDFLAGS += -lm +endif + +ifneq ($(YQ2_OSTYPE), Darwin) +ifneq ($(YQ2_OSTYPE), OpenBSD) +# For some reason the OSX & OpenBSD +# linker doesn't support this... +override LDFLAGS += -Wl,--no-undefined +endif +endif + +# It's a shared library. +override LDFLAGS += -shared + +# ---------- + +# Extra LDFLAGS for SDL +ifeq ($(YQ2_OSTYPE), Darwin) +SDLLDFLAGS := -lSDL2 +else # not Darwin +SDLLDFLAGS := $(shell sdl2-config --libs) +endif # Darwin + +# The renderer libs don't need libSDL2main, libmingw32 or -mwindows. +ifeq ($(YQ2_OSTYPE), Windows) +DLL_SDLLDFLAGS = $(subst -mwindows,,$(subst -lmingw32,,$(subst -lSDL2main,,$(SDLLDFLAGS)))) +endif + +# ---------- +# When make is invoked by "make VERBOSE=1" print +# the compiler and linker commands. + +ifdef VERBOSE +Q := +else +Q := @ +endif + +# ---------- + +# Phony targets +.PHONY : all clean xatrix + +# ---------- + +# Builds everything +all: ref_vk + +# ---------- + +# Cleanup +clean: + @echo "===> CLEAN" + ${Q}rm -Rf build release + +# ---------- + +ifeq ($(YQ2_OSTYPE), Windows) +ref_vk: + @echo "===> Building ref_vk.dll" + ${Q}mkdir -p release + $(MAKE) release/ref_vk.dll +else +ref_vk: + @echo "===> Building ref_vk.so" + ${Q}mkdir -p release + $(MAKE) release/ref_vk.so + +release/ref_vk.so : CFLAGS += -fPIC +endif + +build/%.o: %.c + @echo "===> CC $<" + ${Q}mkdir -p $(@D) + ${Q}$(CC) -c $(CFLAGS) $(SDLCFLAGS) $(INCLUDE) -o $@ $< + +# ---------- + +REFVK_OBJS_ := \ + src/vk/vk_buffer.o \ + src/vk/vk_cmd.o \ + src/vk/vk_common.o \ + src/vk/vk_device.o \ + src/vk/vk_draw.o \ + src/vk/vk_image.o \ + src/vk/vk_light.o \ + src/vk/vk_mesh.o \ + src/vk/vk_model.o \ + src/vk/vk_pipeline.o \ + src/vk/vk_rmain.o \ + src/vk/vk_rmisc.o \ + src/vk/vk_rsurf.o \ + src/vk/vk_shaders.o \ + src/vk/vk_swapchain.o \ + src/vk/vk_validation.o \ + src/vk/vk_warp.o \ + src/vk/vk_util.o \ + src/vk/volk/volk.o \ + src/files/pcx.o \ + src/files/stb.o \ + src/files/wal.o \ + src/files/pvs.o \ + src/common/shared.o \ + src/common/md4.o + +ifeq ($(YQ2_OSTYPE), Windows) +REFVK_OBJS_ += \ + src/backends/hunk_windows.o +else # not Windows +REFVK_OBJS_ += \ + src/backends/hunk_unix.o +endif + +# ---------- + +# Rewrite pathes to our object directory +REFVK_OBJS = $(patsubst %,build/%,$(REFVK_OBJS_)) + +# ---------- + +# Generate header dependencies +REFVK_DEPS= $(REFVK_OBJS:.o=.d) + +# ---------- + +# Suck header dependencies in +-include $(REFVK_DEPS) + +# ---------- + +# release/ref_vk.so +ifeq ($(YQ2_OSTYPE), Windows) +release/ref_vk.dll : $(REFVK_OBJS) + @echo "===> LD $@" + ${Q}$(CC) $(REFVK_OBJS) $(LDFLAGS) $(DLL_SDLLDFLAGS) -o $@ +else +release/ref_vk.so : $(REFVK_OBJS) + @echo "===> LD $@" + ${Q}$(CC) $(REFVK_OBJS) $(LDFLAGS) $(SDLLDFLAGS) -o $@ +endif + +# ---------- diff --git a/README.md b/README.md new file mode 100644 index 0000000..d989a09 --- /dev/null +++ b/README.md @@ -0,0 +1,111 @@ +# Vulkan Renderer Library for Yamagi Quake II + +This is the vkQuake2 vulkan renderer library ported to Yamagi Quake II. + + +## Compilation + +You'll need: +* clang or gcc +* GNU Make +* SDL2 with `sdl2-config`. +* vulkan-headers. + +Type `make` to compile the library. If the compilation is successfull, +the library can be found under *release/ref_vk.dll* (Windows) or +*release/ref_vk.so* (unixoid systems). + + +## Usage + +Copy the library next to your Yamagi Quake II executable. You can select +Vulkan through the video menu or by cvar with `vid_renderer vk` followed +by a `vid_restart`. + + +## Console Variables + +* **vk_validation**: Toggle validation layers: + * `0` - disabled (default in Release) + * `1` - only errors and warnings + * `2` - best-practices validation + +* **vk_strings**: Print some basic Vulkan/GPU information. + +* **vk_mem**: Print dynamic vertex/index/uniform/triangle fan buffer + memory usage statistics. + +* **vk_device**: Specify index of the preferred Vulkan device on systems + with multiple GPUs: + * `-1` - prefer first DISCRETE\_GPU (default) + * `0..n` - use device #n (full list of devices is returned by + `vk_strings` command) + +* **vk_sampleshading**: Toggle sample shading for MSAA. (default: `1`) + +* **vk_flashblend**: Toggle the blending of lights onto the environment. + (default: `0`) + +* **vk_polyblend**: Blend fullscreen effects: blood, powerups etc. + (default: `1`) + +* **vk_skymip**: Toggle the usage of mipmap information for the sky + graphics. (default: `0`) + +* **vk_finish**: Inserts a `vkDeviceWaitIdle()` call on frame render + start (default: `0`). Don't use this, it's there just for the sake of + having a `gl_finish` equivalent! + +* **vk_custom_particles**: Toggle particles type: + * `0` - textured triangles for particle rendering + * `1` - between using POINT\_LIST (default) + * `2` - textured square for particle rendering + +* **vk_particle_size**: Rendered particle size. (default: `40`) + +* **vk_particle_att_a**: Intensity of the particle A attribute. + (default: `0.01`) + +* **vk_particle_att_b**: Intensity of the particle B attribute. + (default: `0`) + +* **vk_particle_att_c**: Intensity of the particle C attribute. + (default: `0.01`) + +* **vk_particle_min_size**: The minimum size of a rendered particle. + (default: `2`) + +* **vk_particle_max_size**: The maximum size of a rendered particle. + (default: `40`) + +* **vk_picmip**: Shrink factor for the textures. (default: `0`) + +* **vk_pixel_size**: Pixel size when rendering the world, used to simulate + lower screen resolutions. The value represents the length, in pixels, of the + side of each pixel block. For example, with size 2 pixels are 2x2 squares, + and at 1600x1200 the image is effectively an upscaled 800x600 image. + (default: `1`) + +* **vk_dynamic**: Use dynamic lighting. (default: `1`) + +* **vk_showtris**: Display mesh triangles. (default: `0`) + +* **vk_lightmap**: Display lightmaps. (default: `0`) + +* **vk_postprocess**: Toggle additional color/gamma correction. + (default: `1`) + +* **vk_mip_nearfilter**: Use nearest-neighbor filtering for mipmaps. + (default: `0`) + +* **vk_texturemode**: Change current texture filtering mode: + * `VK_NEAREST` - nearest-neighbor interpolation, no mipmaps + * `VK_LINEAR` - linear interpolation, no mipmaps + * `VK_MIPMAP_NEAREST` - nearest-neighbor interpolation with mipmaps + * `VK_MIPMAP_LINEAR` - linear interpolation with mipmaps (default) + +* **vk_lmaptexturemode**: Same as `vk_texturemode` but applied to + lightmap textures. + +* **vk_underwater**: Warp the scene if underwater. Set to `0` to disable + the effect. Defaults to `1`. diff --git a/shaders/basic.frag b/shaders/basic.frag new file mode 100644 index 0000000..0c73c4e --- /dev/null +++ b/shaders/basic.frag @@ -0,0 +1,24 @@ +#version 450 + +layout(push_constant) uniform PushConstant +{ + // vertex shader has 'mat4 vpMatrix;' at begin. + layout(offset = 68) float gamma; +} pc; + +layout(set = 0, binding = 0) uniform sampler2D sTexture; + +layout(location = 0) in vec2 texCoord; +layout(location = 1) in vec4 color; +layout(location = 2) in float aTreshold; + +layout(location = 0) out vec4 fragmentColor; + +void main() +{ + fragmentColor = texture(sTexture, texCoord) * color; + if(fragmentColor.a < aTreshold) + discard; + + fragmentColor = vec4(pow(fragmentColor.rgb, vec3(pc.gamma)), fragmentColor.a); +} diff --git a/shaders/basic.vert b/shaders/basic.vert new file mode 100644 index 0000000..4208399 --- /dev/null +++ b/shaders/basic.vert @@ -0,0 +1,30 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +// normalized offset and scale +layout(set = 1, binding = 0) uniform imageTransform +{ + vec2 offset; + vec2 scale; + vec2 uvOffset; + vec2 uvScale; +} it; + +layout(location = 0) in vec2 inVertex; +layout(location = 1) in vec2 inTexCoord; + +layout(location = 0) out vec2 texCoord; +layout(location = 1) out vec4 color; +layout(location = 2) out float aTreshold; + +out gl_PerVertex { + vec4 gl_Position; +}; + +void main() { + vec2 vPos = inVertex.xy * it.scale - (vec2(1.0) - it.scale); + gl_Position = vec4(vPos + it.offset * 2.0, 0.0, 1.0); + texCoord = inTexCoord.xy * it.uvScale + it.uvOffset; + color = vec4(1.0, 1.0, 1.0, 1.0); + aTreshold = 0.666; +} diff --git a/shaders/basic_color_quad.frag b/shaders/basic_color_quad.frag new file mode 100644 index 0000000..aec28c4 --- /dev/null +++ b/shaders/basic_color_quad.frag @@ -0,0 +1,10 @@ +#version 450 + +layout(location = 0) in vec4 color; + +layout(location = 0) out vec4 fragmentColor; + +void main() +{ + fragmentColor = color; +} diff --git a/shaders/basic_color_quad.vert b/shaders/basic_color_quad.vert new file mode 100644 index 0000000..dd4b67b --- /dev/null +++ b/shaders/basic_color_quad.vert @@ -0,0 +1,24 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +// normalized offset and scale +layout(set = 0, binding = 0) uniform imageTransform +{ + vec2 offset; + vec2 scale; + vec4 color; +} it; + +layout(location = 0) in vec2 inVertex; + +layout(location = 0) out vec4 color; + +out gl_PerVertex { + vec4 gl_Position; +}; + +void main() { + vec2 vPos = inVertex.xy * it.scale - (vec2(1.0) - it.scale); + gl_Position = vec4(vPos + it.offset * 2.0, 0.0, 1.0); + color = it.color; +} diff --git a/shaders/beam.vert b/shaders/beam.vert new file mode 100644 index 0000000..5d107bf --- /dev/null +++ b/shaders/beam.vert @@ -0,0 +1,25 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +layout(location = 0) in vec3 inVertex; + +layout(push_constant) uniform PushConstant +{ + mat4 mvpMatrix; +} pc; + +layout(binding = 0) uniform UniformBufferObject +{ + vec4 color; +} ubo; + +layout(location = 0) out vec4 color; + +out gl_PerVertex { + vec4 gl_Position; +}; + +void main() { + gl_Position = pc.mvpMatrix * vec4(inVertex, 1.0); + color = ubo.color; +} diff --git a/shaders/d_light.vert b/shaders/d_light.vert new file mode 100644 index 0000000..cda70b4 --- /dev/null +++ b/shaders/d_light.vert @@ -0,0 +1,21 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +layout(location = 0) in vec3 inVertex; +layout(location = 1) in vec3 inColor; + +layout(binding = 0) uniform UniformBufferObject +{ + mat4 mvpMatrix; +} ubo; + +layout(location = 0) out vec4 color; + +out gl_PerVertex { + vec4 gl_Position; +}; + +void main() { + gl_Position = ubo.mvpMatrix * vec4(inVertex, 1.0); + color = vec4(inColor, 1.0); +} diff --git a/shaders/model.frag b/shaders/model.frag new file mode 100644 index 0000000..1133244 --- /dev/null +++ b/shaders/model.frag @@ -0,0 +1,17 @@ +#version 450 + +layout(set = 0, binding = 0) uniform sampler2D sTexture; + +layout(location = 0) in vec4 color; +layout(location = 1) in vec2 texCoord; +layout(location = 2) in flat int textured; + +layout(location = 0) out vec4 fragmentColor; + +void main() +{ + if(textured != 0) + fragmentColor = texture(sTexture, texCoord) * clamp(color, 0.0, 1.0); + else + fragmentColor = color; +} diff --git a/shaders/model.vert b/shaders/model.vert new file mode 100644 index 0000000..f728e25 --- /dev/null +++ b/shaders/model.vert @@ -0,0 +1,32 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +layout(location = 0) in vec3 inVertex; +layout(location = 1) in vec4 inColor; +layout(location = 2) in vec2 inTexCoord; + +layout(push_constant) uniform PushConstant +{ + mat4 vpMatrix; +} pc; + +layout(set = 1, binding = 0) uniform UniformBufferObject +{ + mat4 model; + int textured; +} ubo; + +layout(location = 0) out vec4 color; +layout(location = 1) out vec2 texCoord; +layout(location = 2) out int textured; + +out gl_PerVertex { + vec4 gl_Position; +}; + +void main() { + gl_Position = pc.vpMatrix * ubo.model * vec4(inVertex, 1.0); + color = inColor; + texCoord = inTexCoord; + textured = ubo.textured; +} diff --git a/shaders/nullmodel.vert b/shaders/nullmodel.vert new file mode 100644 index 0000000..ca69677 --- /dev/null +++ b/shaders/nullmodel.vert @@ -0,0 +1,26 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +layout(location = 0) in vec3 inVertex; +layout(location = 1) in vec3 inColor; + +layout(push_constant) uniform PushConstant +{ + mat4 vpMatrix; +} pc; + +layout(binding = 0) uniform UniformBufferObject +{ + mat4 model; +} ubo; + +layout(location = 0) out vec4 color; + +out gl_PerVertex { + vec4 gl_Position; +}; + +void main() { + gl_Position = pc.vpMatrix * ubo.model * vec4(inVertex, 1.0); + color = vec4(inColor, 1.0); +} diff --git a/shaders/particle.vert b/shaders/particle.vert new file mode 100644 index 0000000..40fd674 --- /dev/null +++ b/shaders/particle.vert @@ -0,0 +1,26 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +layout(location = 0) in vec3 inVertex; +layout(location = 1) in vec4 inColor; +layout(location = 2) in vec2 inTexCoord; + +layout(push_constant) uniform PushConstant +{ + mat4 mvpMatrix; +} pc; + +layout(location = 0) out vec2 texCoord; +layout(location = 1) out vec4 color; +layout(location = 2) out float aTreshold; + +out gl_PerVertex { + vec4 gl_Position; +}; + +void main() { + gl_Position = pc.mvpMatrix * vec4(inVertex, 1.0); + texCoord = inTexCoord; + color = inColor; + aTreshold = 0.5; +} diff --git a/shaders/point_particle.frag b/shaders/point_particle.frag new file mode 100644 index 0000000..f47c881 --- /dev/null +++ b/shaders/point_particle.frag @@ -0,0 +1,14 @@ +#version 450 + +layout(location = 0) in vec4 color; + +layout(location = 0) out vec4 fragmentColor; + +void main() +{ + vec2 cxy = 2.0 * gl_PointCoord - 1.0; + if(dot(cxy, cxy) > 1.0) + discard; + + fragmentColor = color; +} diff --git a/shaders/point_particle.vert b/shaders/point_particle.vert new file mode 100644 index 0000000..ab40744 --- /dev/null +++ b/shaders/point_particle.vert @@ -0,0 +1,37 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +layout(location = 0) in vec3 inVertex; +layout(location = 1) in vec4 inColor; + +layout(push_constant) uniform PushConstant +{ + mat4 mvpMatrix; +} pc; + +layout(binding = 0) uniform UniformBufferObject +{ + float pointSize; + float pointScale; + float minPointSize; + float maxPointSize; + float att_a; + float att_b; + float att_c; +} ubo; + +layout(location = 0) out vec4 color; + +out gl_PerVertex { + vec4 gl_Position; + float gl_PointSize; +}; + +void main() { + gl_Position = pc.mvpMatrix * vec4(inVertex, 1.0); + + float dist_atten = ubo.pointScale / (ubo.att_a + ubo.att_b * gl_Position.w + ubo.att_c * gl_Position.w * gl_Position.w); + gl_PointSize = clamp(ubo.pointScale * ubo.pointSize * sqrt(dist_atten), ubo.minPointSize, ubo.maxPointSize); + + color = inColor; +} diff --git a/shaders/polygon.vert b/shaders/polygon.vert new file mode 100644 index 0000000..120a52f --- /dev/null +++ b/shaders/polygon.vert @@ -0,0 +1,30 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +layout(location = 0) in vec3 inVertex; +layout(location = 1) in vec2 inTexCoord; + +layout(push_constant) uniform PushConstant +{ + mat4 mvpMatrix; +} pc; + +layout(set = 1, binding = 0) uniform UniformBufferObject +{ + vec4 color; +} ubo; + +layout(location = 0) out vec2 texCoord; +layout(location = 1) out vec4 color; +layout(location = 2) out float aTreshold; + +out gl_PerVertex { + vec4 gl_Position; +}; + +void main() { + gl_Position = pc.mvpMatrix * vec4(inVertex, 1.0); + texCoord = inTexCoord; + color = ubo.color; + aTreshold = 0.0; +} diff --git a/shaders/polygon_lmap.frag b/shaders/polygon_lmap.frag new file mode 100644 index 0000000..cd43691 --- /dev/null +++ b/shaders/polygon_lmap.frag @@ -0,0 +1,17 @@ +#version 450 + +layout(set = 0, binding = 0) uniform sampler2D sTexture; +layout(set = 2, binding = 0) uniform sampler2D sLightmap; + +layout(location = 0) in vec2 texCoord; +layout(location = 1) in vec2 texCoordLmap; +layout(location = 2) in float viewLightmaps; + +layout(location = 0) out vec4 fragmentColor; + +void main() +{ + vec4 color = texture(sTexture, texCoord); + vec4 light = texture(sLightmap, texCoordLmap); + fragmentColor = (1.0 - viewLightmaps) * color * light + viewLightmaps * light; +} diff --git a/shaders/polygon_lmap.vert b/shaders/polygon_lmap.vert new file mode 100644 index 0000000..88882e5 --- /dev/null +++ b/shaders/polygon_lmap.vert @@ -0,0 +1,32 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +layout(location = 0) in vec3 inVertex; +layout(location = 1) in vec2 inTexCoord; +layout(location = 2) in vec2 inTexCoordLmap; + +layout(push_constant) uniform PushConstant +{ + mat4 vpMatrix; +} pc; + +layout(set = 1, binding = 0) uniform UniformBufferObject +{ + mat4 model; + float viewLightmaps; +} ubo; + +layout(location = 0) out vec2 texCoord; +layout(location = 1) out vec2 texCoordLmap; +layout(location = 2) out float viewLightmaps; + +out gl_PerVertex { + vec4 gl_Position; +}; + +void main() { + gl_Position = pc.vpMatrix * ubo.model * vec4(inVertex, 1.0); + texCoord = inTexCoord; + texCoordLmap = inTexCoordLmap; + viewLightmaps = ubo.viewLightmaps; +} diff --git a/shaders/polygon_warp.vert b/shaders/polygon_warp.vert new file mode 100644 index 0000000..609a2c0 --- /dev/null +++ b/shaders/polygon_warp.vert @@ -0,0 +1,34 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +layout(location = 0) in vec3 inVertex; +layout(location = 1) in vec2 inTexCoord; + +layout(push_constant) uniform PushConstant +{ + mat4 vpMatrix; +} pc; + +layout(set = 1, binding = 0) uniform UniformBufferObject +{ + mat4 model; + vec4 color; + float time; + float scroll; +} ubo; + +layout(location = 0) out vec2 texCoord; +layout(location = 1) out vec4 color; +layout(location = 2) out float aTreshold; + +out gl_PerVertex { + vec4 gl_Position; +}; + +void main() { + gl_Position = pc.vpMatrix * ubo.model * vec4(inVertex, 1.0); + texCoord = inTexCoord + vec2(sin(2.0 * ubo.time + inTexCoord.y * 3.28), sin(2.0 * ubo.time + inTexCoord.x * 3.28)) * 0.05; + texCoord.x += ubo.scroll; + color = ubo.color; + aTreshold = 0.0; +} diff --git a/shaders/postprocess.frag b/shaders/postprocess.frag new file mode 100644 index 0000000..879b555 --- /dev/null +++ b/shaders/postprocess.frag @@ -0,0 +1,33 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +layout(push_constant) uniform PushConstant +{ + layout(offset = 68) float postprocess; + layout(offset = 72) float gamma; + layout(offset = 76) float scrWidth; + layout(offset = 80) float scrHeight; + layout(offset = 84) float offsetX; + layout(offset = 88) float offsetY; +} pc; + +layout(set = 0, binding = 0) uniform sampler2D sTexture; + +layout(location = 0) in vec2 texCoord; +layout(location = 0) out vec4 fragmentColor; + +void main() +{ + vec2 unnormTexCoord = texCoord * vec2(pc.scrWidth, pc.scrHeight) + vec2(pc.offsetX, pc.offsetY); + + // apply any additional world-only postprocessing effects here (if enabled) + if (pc.postprocess > 0.0) + { + //gamma + color intensity bump + fragmentColor = vec4(pow(textureLod(sTexture, unnormTexCoord, 0.0).rgb * 1.5, vec3(pc.gamma)), 1.0); + } + else + { + fragmentColor = textureLod(sTexture, unnormTexCoord, 0.0); + } +} diff --git a/shaders/postprocess.vert b/shaders/postprocess.vert new file mode 100644 index 0000000..438e62e --- /dev/null +++ b/shaders/postprocess.vert @@ -0,0 +1,17 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +// rendering a fullscreen quad (which is actually just a huge triangle) +// source: https://www.saschawillems.de/blog/2016/08/13/vulkan-tutorial-on-rendering-a-fullscreen-quad-without-buffers/ + +out gl_PerVertex { + vec4 gl_Position; +}; + +layout(location = 0) out vec2 texCoord; + +void main() +{ + texCoord = vec2((gl_VertexIndex << 1) & 2, gl_VertexIndex & 2); + gl_Position = vec4(texCoord * 2.0f + -1.0f, 0.0f, 1.0f); +} diff --git a/shaders/shaders.sh b/shaders/shaders.sh new file mode 100755 index 0000000..cebfced --- /dev/null +++ b/shaders/shaders.sh @@ -0,0 +1,25 @@ +#!/bin/sh + +glslangValidator --variable-name basic_vert_spv -V basic.vert -o ../src/vk/spirv/basic_vert.c +glslangValidator --variable-name basic_frag_spv -V basic.frag -o ../src/vk/spirv/basic_frag.c +glslangValidator --variable-name basic_color_quad_vert_spv -V basic_color_quad.vert -o ../src/vk/spirv/basic_color_quad_vert.c +glslangValidator --variable-name basic_color_quad_frag_spv -V basic_color_quad.frag -o ../src/vk/spirv/basic_color_quad_frag.c +glslangValidator --variable-name model_vert_spv -V model.vert -o ../src/vk/spirv/model_vert.c +glslangValidator --variable-name model_frag_spv -V model.frag -o ../src/vk/spirv/model_frag.c +glslangValidator --variable-name nullmodel_vert_spv -V nullmodel.vert -o ../src/vk/spirv/nullmodel_vert.c +glslangValidator --variable-name particle_vert_spv -V particle.vert -o ../src/vk/spirv/particle_vert.c +glslangValidator --variable-name point_particle_vert_spv -V point_particle.vert -o ../src/vk/spirv/point_particle_vert.c +glslangValidator --variable-name point_particle_frag_spv -V point_particle.frag -o ../src/vk/spirv/point_particle_frag.c +glslangValidator --variable-name sprite_vert_spv -V sprite.vert -o ../src/vk/spirv/sprite_vert.c +glslangValidator --variable-name beam_vert_spv -V beam.vert -o ../src/vk/spirv/beam_vert.c +glslangValidator --variable-name skybox_vert_spv -V skybox.vert -o ../src/vk/spirv/skybox_vert.c +glslangValidator --variable-name d_light_vert_spv -V d_light.vert -o ../src/vk/spirv/d_light_vert.c +glslangValidator --variable-name polygon_vert_spv -V polygon.vert -o ../src/vk/spirv/polygon_vert.c +glslangValidator --variable-name polygon_lmap_vert_spv -V polygon_lmap.vert -o ../src/vk/spirv/polygon_lmap_vert.c +glslangValidator --variable-name polygon_lmap_frag_spv -V polygon_lmap.frag -o ../src/vk/spirv/polygon_lmap_frag.c +glslangValidator --variable-name polygon_warp_vert_spv -V polygon_warp.vert -o ../src/vk/spirv/polygon_warp_vert.c +glslangValidator --variable-name shadows_vert_spv -V shadows.vert -o ../src/vk/spirv/shadows_vert.c +glslangValidator --variable-name postprocess_vert_spv -V postprocess.vert -o ../src/vk/spirv/postprocess_vert.c +glslangValidator --variable-name postprocess_frag_spv -V postprocess.frag -o ../src/vk/spirv/postprocess_frag.c +glslangValidator --variable-name world_warp_vert_spv -V world_warp.vert -o ../src/vk/spirv/world_warp_vert.c +glslangValidator --variable-name world_warp_frag_spv -V world_warp.frag -o ../src/vk/spirv/world_warp_frag.c diff --git a/shaders/shadows.vert b/shaders/shadows.vert new file mode 100644 index 0000000..24fbae9 --- /dev/null +++ b/shaders/shadows.vert @@ -0,0 +1,24 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +layout(location = 0) in vec3 inVertex; +layout(location = 0) out vec4 color; + +layout(push_constant) uniform PushConstant +{ + mat4 vpMatrix; +} pc; + +layout(binding = 0) uniform UniformBufferObject +{ + mat4 model; +} ubo; + +out gl_PerVertex { + vec4 gl_Position; +}; + +void main() { + gl_Position = pc.vpMatrix * ubo.model * vec4(inVertex, 1.0); + color = vec4(0.0, 0.0, 0.0, 0.5); +} diff --git a/shaders/skybox.vert b/shaders/skybox.vert new file mode 100644 index 0000000..348777c --- /dev/null +++ b/shaders/skybox.vert @@ -0,0 +1,30 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +layout(location = 0) in vec3 inVertex; +layout(location = 1) in vec2 inTexCoord; + +layout(push_constant) uniform PushConstant +{ + mat4 vpMatrix; +} pc; + +layout(set = 1, binding = 0) uniform UniformBufferObject +{ + mat4 model; +} ubo; + +layout(location = 0) out vec2 texCoord; +layout(location = 1) out vec4 color; +layout(location = 2) out float aTreshold; + +out gl_PerVertex { + vec4 gl_Position; +}; + +void main() { + gl_Position = pc.vpMatrix * ubo.model * vec4(inVertex, 1.0); + texCoord = inTexCoord; + color = vec4(1.0, 1.0, 1.0, 1.0); + aTreshold = 0.0; +} diff --git a/shaders/sprite.vert b/shaders/sprite.vert new file mode 100644 index 0000000..f42a66c --- /dev/null +++ b/shaders/sprite.vert @@ -0,0 +1,26 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +layout(location = 0) in vec3 inVertex; +layout(location = 1) in vec2 inTexCoord; + +layout(push_constant) uniform PushConstant +{ + mat4 mvpMatrix; + float alpha; +} pc; + +layout(location = 0) out vec2 texCoord; +layout(location = 1) out vec4 color; +layout(location = 2) out float aTreshold; + +out gl_PerVertex { + vec4 gl_Position; +}; + +void main() { + gl_Position = pc.mvpMatrix * vec4(inVertex, 1.0); + texCoord = inTexCoord; + color = vec4(1.0, 1.0, 1.0, pc.alpha); + aTreshold = 0.0666; +} diff --git a/shaders/world_warp.frag b/shaders/world_warp.frag new file mode 100644 index 0000000..cbbeea1 --- /dev/null +++ b/shaders/world_warp.frag @@ -0,0 +1,55 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +// Underwater screen warp effect similar to what software renderer provides. +// Pixel size to simulate lower screen resolutions is used to restore world to full screen size. + +layout(push_constant) uniform PushConstant +{ + layout(offset = 68) float time; + layout(offset = 72) float scale; + layout(offset = 76) float scrWidth; + layout(offset = 80) float scrHeight; + layout(offset = 84) float offsetX; + layout(offset = 88) float offsetY; + layout(offset = 92) float pixelSize; + layout(offset = 96) float refdefX; + layout(offset = 100) float refdefY; + layout(offset = 104) float refdefWidth; + layout(offset = 108) float refdefHeight; +} pc; + +layout(set = 0, binding = 0) uniform sampler2D sTexture; + +layout(location = 0) out vec4 fragmentColor; + +#define PI 3.1415 + +void main() +{ + vec2 scrSize = vec2(pc.scrWidth, pc.scrHeight); + vec2 fragCoord = (gl_FragCoord.xy - vec2(pc.offsetX, pc.offsetY)); + vec2 uv = fragCoord / scrSize; + + float xMin = pc.refdefX; + float xMax = pc.refdefX + pc.refdefWidth; + float yMin = pc.refdefY; + float yMax = pc.refdefY + pc.refdefHeight; + + if (pc.time > 0 && fragCoord.x > xMin && fragCoord.x < xMax && fragCoord.y > yMin && fragCoord.y < yMax) + { + float sx = pc.scale - abs(pc.scrWidth / 2.0 - fragCoord.x) * 2.0 / pc.scrWidth; + float sy = pc.scale - abs(pc.scrHeight / 2.0 - fragCoord.y) * 2.0 / pc.scrHeight; + float xShift = 2.0 * pc.time + uv.y * PI * 10; + float yShift = 2.0 * pc.time + uv.x * PI * 10; + vec2 distortion = vec2(sin(xShift) * sx, sin(yShift) * sy) * 0.00666; + + uv += distortion; + } + + uv /= pc.pixelSize; + + uv = clamp(uv * scrSize, vec2(0.0, 0.0), scrSize - vec2(0.5, 0.5)); + + fragmentColor = textureLod(sTexture, uv, 0.0); +} diff --git a/shaders/world_warp.vert b/shaders/world_warp.vert new file mode 100644 index 0000000..e09723b --- /dev/null +++ b/shaders/world_warp.vert @@ -0,0 +1,17 @@ +#version 450 +#extension GL_ARB_separate_shader_objects : enable + +out gl_PerVertex { + vec4 gl_Position; +}; + +void main() +{ + vec4 positions[3] = { + vec4(-1.0f, -1.0f, 0.0f, 1.0f), + vec4(3.0f, -1.0f, 0.0f, 1.0f), + vec4(-1.0f, 3.0f, 0.0f, 1.0f) + }; + + gl_Position = positions[gl_VertexIndex % 3]; +} diff --git a/src/backends/hunk_unix.c b/src/backends/hunk_unix.c new file mode 100644 index 0000000..2485fe7 --- /dev/null +++ b/src/backends/hunk_unix.c @@ -0,0 +1,173 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * This file implements the low level part of the Hunk_* memory system + * + * ======================================================================= + */ + +/* For mremap() - must be before sys/mman.h include! */ +#if defined(__linux__) && !defined(_GNU_SOURCE) + #define _GNU_SOURCE +#endif + +#include +#include +#include +#include + +#include "../common/header/common.h" + +#if defined(__FreeBSD__) || defined(__OpenBSD__) + #include + #define MAP_ANONYMOUS MAP_ANON +#endif + +#if defined(__APPLE__) + #include + #define MAP_ANONYMOUS MAP_ANON +#endif + +byte *membase; +size_t maxhunksize; +size_t curhunksize; + +void * +Hunk_Begin(int maxsize) +{ + + /* reserve a huge chunk of memory, but don't commit any yet */ + /* plus 32 bytes for cacheline */ + maxhunksize = maxsize + sizeof(size_t) + 32; + curhunksize = 0; + int flags = MAP_PRIVATE | MAP_ANONYMOUS; + int prot = PROT_READ | PROT_WRITE; + +#if defined(MAP_ALIGNED_SUPER) + const size_t hgpagesize = 1UL<<21; + size_t page_size = sysconf(_SC_PAGESIZE); + + /* Archs supported has 2MB for super pages size */ + if (maxhunksize >= hgpagesize) + { + maxhunksize = (maxhunksize & ~(page_size-1)) + page_size; + flags |= MAP_ALIGNED_SUPER; + } +#endif + +#if defined(PROT_MAX) + /* For now it is FreeBSD exclusif but could possibly be extended + to other like DFBSD for example */ + prot |= PROT_MAX(prot); +#endif + + membase = mmap(0, maxhunksize, prot, + flags, -1, 0); + + if ((membase == NULL) || (membase == (byte *)-1)) + { + Sys_Error("unable to virtual allocate %d bytes", maxsize); + } + + *((size_t *)membase) = curhunksize; + + return membase + sizeof(size_t); +} + +void * +Hunk_Alloc(int size) +{ + byte *buf; + + /* round to cacheline */ + size = (size + 31) & ~31; + + if (curhunksize + size > maxhunksize) + { + Sys_Error("Hunk_Alloc overflow"); + } + + buf = membase + sizeof(size_t) + curhunksize; + curhunksize += size; + return buf; +} + +int +Hunk_End(void) +{ + byte *n = NULL; + +#if defined(__linux__) + n = (byte *)mremap(membase, maxhunksize, curhunksize + sizeof(size_t), 0); +#elif defined(__NetBSD__) + n = (byte *)mremap(membase, maxhunksize, NULL, curhunksize + sizeof(size_t), 0); +#else + #ifndef round_page + size_t page_size = sysconf(_SC_PAGESIZE); + #define round_page(x) ((((size_t)(x)) + page_size-1) & ~(page_size-1)) + #endif + + size_t old_size = round_page(maxhunksize); + size_t new_size = round_page(curhunksize + sizeof(size_t)); + + if (new_size > old_size) + { + /* Can never happen. If it happens something's very wrong. */ + n = 0; + } + else if (new_size < old_size) + { + /* Hunk is to big, we need to shrink it. */ + n = munmap(membase + new_size, old_size - new_size) + membase; + } + else + { + /* No change necessary. */ + n = membase; + } +#endif + + if (n != membase) + { + Sys_Error("Hunk_End: Could not remap virtual block (%d)", errno); + } + + *((size_t *)membase) = curhunksize + sizeof(size_t); + + return curhunksize; +} + +void +Hunk_Free(void *base) +{ + if (base) + { + byte *m; + + m = ((byte *)base) - sizeof(size_t); + + if (munmap(m, *((size_t *)m))) + { + Sys_Error("Hunk_Free: munmap failed (%d)", errno); + } + } +} + diff --git a/src/backends/hunk_windows.c b/src/backends/hunk_windows.c new file mode 100644 index 0000000..0c5dd0a --- /dev/null +++ b/src/backends/hunk_windows.c @@ -0,0 +1,97 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * Memory handling functions. + * + * ======================================================================= + */ + +#include + +#include "../common/header/common.h" + +byte *membase; +int hunkcount; +size_t hunkmaxsize; +size_t cursize; + +void * +Hunk_Begin(int maxsize) +{ + /* reserve a huge chunk of memory, but don't commit any yet */ + /* plus 32 bytes for cacheline */ + hunkmaxsize = maxsize + sizeof(size_t) + 32; + cursize = 0; + + membase = VirtualAlloc(NULL, hunkmaxsize, MEM_RESERVE, PAGE_NOACCESS); + + if (!membase) + { + Sys_Error("VirtualAlloc reserve failed"); + } + + return (void *)membase; +} + +void * +Hunk_Alloc(int size) +{ + void *buf; + + /* round to cacheline */ + size = (size + 31) & ~31; + + /* commit pages as needed */ + buf = VirtualAlloc(membase, cursize + size, MEM_COMMIT, PAGE_READWRITE); + + if (!buf) + { + Sys_Error("VirtualAlloc commit failed.\n"); + } + + cursize += size; + + if (cursize > hunkmaxsize) + { + Sys_Error("Hunk_Alloc overflow"); + } + + return (void *)(membase + cursize - size); +} + +int +Hunk_End(void) +{ + hunkcount++; + + return cursize; +} + +void +Hunk_Free(void *base) +{ + if (base) + { + VirtualFree(base, 0, MEM_RELEASE); + } + + hunkcount--; +} diff --git a/src/common/header/common.h b/src/common/header/common.h new file mode 100644 index 0000000..ac53365 --- /dev/null +++ b/src/common/header/common.h @@ -0,0 +1,840 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * Prototypes witch are shared between the client, the server and the + * game. This is the main game API, changes here will most likely + * requiere changes to the game ddl. + * + * ======================================================================= + */ + +#ifndef CO_COMMON_H +#define CO_COMMON_H + +#include "shared.h" +#include "crc.h" + +#define YQ2VERSION "8.00pre" +#define BASEDIRNAME "baseq2" + +#ifndef YQ2OSTYPE +#error YQ2OSTYPE should be defined by the build system +#endif + +#ifndef YQ2ARCH +#error YQ2ARCH should be defined by the build system +#endif + +#ifndef BUILD_DATE +#define BUILD_DATE __DATE__ +#endif + +#ifdef _WIN32 + #define CFGDIR "YamagiQ2" +#else + #define CFGDIR ".yq2" +#endif + +/* ================================================================== */ + +typedef struct sizebuf_s +{ + qboolean allowoverflow; /* if false, do a Com_Error */ + qboolean overflowed; /* set to true if the buffer size failed */ + byte *data; + int maxsize; + int cursize; + int readcount; +} sizebuf_t; + +void SZ_Init(sizebuf_t *buf, byte *data, int length); +void SZ_Clear(sizebuf_t *buf); +void *SZ_GetSpace(sizebuf_t *buf, int length); +void SZ_Write(sizebuf_t *buf, void *data, int length); +void SZ_Print(sizebuf_t *buf, char *data); /* strcats onto the sizebuf */ + +/* ================================================================== */ + +struct usercmd_s; +struct entity_state_s; + +void MSG_WriteChar(sizebuf_t *sb, int c); +void MSG_WriteByte(sizebuf_t *sb, int c); +void MSG_WriteShort(sizebuf_t *sb, int c); +void MSG_WriteLong(sizebuf_t *sb, int c); +void MSG_WriteFloat(sizebuf_t *sb, float f); +void MSG_WriteString(sizebuf_t *sb, char *s); +void MSG_WriteCoord(sizebuf_t *sb, float f); +void MSG_WritePos(sizebuf_t *sb, vec3_t pos); +void MSG_WriteAngle(sizebuf_t *sb, float f); +void MSG_WriteAngle16(sizebuf_t *sb, float f); +void MSG_WriteDeltaUsercmd(sizebuf_t *sb, struct usercmd_s *from, + struct usercmd_s *cmd); +void MSG_WriteDeltaEntity(struct entity_state_s *from, + struct entity_state_s *to, sizebuf_t *msg, + qboolean force, qboolean newentity); +void MSG_WriteDir(sizebuf_t *sb, vec3_t vector); + +void MSG_BeginReading(sizebuf_t *sb); + +int MSG_ReadChar(sizebuf_t *sb); +int MSG_ReadByte(sizebuf_t *sb); +int MSG_ReadShort(sizebuf_t *sb); +int MSG_ReadLong(sizebuf_t *sb); +float MSG_ReadFloat(sizebuf_t *sb); +char *MSG_ReadString(sizebuf_t *sb); +char *MSG_ReadStringLine(sizebuf_t *sb); + +float MSG_ReadCoord(sizebuf_t *sb); +void MSG_ReadPos(sizebuf_t *sb, vec3_t pos); +float MSG_ReadAngle(sizebuf_t *sb); +float MSG_ReadAngle16(sizebuf_t *sb); +void MSG_ReadDeltaUsercmd(sizebuf_t *sb, + struct usercmd_s *from, + struct usercmd_s *cmd); + +void MSG_ReadDir(sizebuf_t *sb, vec3_t vector); + +void MSG_ReadData(sizebuf_t *sb, void *buffer, int size); + +/* ================================================================== */ + +extern qboolean bigendien; + +extern short BigShort(short l); +extern short LittleShort(short l); +extern int BigLong(int l); +extern int LittleLong(int l); +extern float BigFloat(float l); +extern float LittleFloat(float l); + +/* ================================================================== */ + +int COM_Argc(void); +char *COM_Argv(int arg); /* range and null checked */ +void COM_ClearArgv(int arg); +int COM_CheckParm(char *parm); +void COM_AddParm(char *parm); + +void COM_Init(void); +void COM_InitArgv(int argc, char **argv); + +char *CopyString(char *in); + +/* ================================================================== */ + +void Info_Print(char *s); + +/* PROTOCOL */ + +#define PROTOCOL_VERSION 34 + +/* ========================================= */ + +#define PORT_MASTER 27900 +#define PORT_CLIENT 27901 +#define PORT_SERVER 27910 + +/* ========================================= */ + +#define UPDATE_BACKUP 16 /* copies of entity_state_t to keep buffered */ +#define UPDATE_MASK (UPDATE_BACKUP - 1) + +/* server to client */ +enum svc_ops_e +{ + svc_bad, + + /* these ops are known to the game dll */ + svc_muzzleflash, + svc_muzzleflash2, + svc_temp_entity, + svc_layout, + svc_inventory, + + /* the rest are private to the client and server */ + svc_nop, + svc_disconnect, + svc_reconnect, + svc_sound, /* */ + svc_print, /* [byte] id [string] null terminated string */ + svc_stufftext, /* [string] stuffed into client's console buffer, should be \n terminated */ + svc_serverdata, /* [long] protocol ... */ + svc_configstring, /* [short] [string] */ + svc_spawnbaseline, + svc_centerprint, /* [string] to put in center of the screen */ + svc_download, /* [short] size [size bytes] */ + svc_playerinfo, /* variable */ + svc_packetentities, /* [...] */ + svc_deltapacketentities, /* [...] */ + svc_frame +}; + +/* ============================================== */ + +/* client to server */ +enum clc_ops_e +{ + clc_bad, + clc_nop, + clc_move, /* [[usercmd_t] */ + clc_userinfo, /* [[userinfo string] */ + clc_stringcmd /* [string] message */ +}; + +/* ============================================== */ + +/* plyer_state_t communication */ +#define PS_M_TYPE (1 << 0) +#define PS_M_ORIGIN (1 << 1) +#define PS_M_VELOCITY (1 << 2) +#define PS_M_TIME (1 << 3) +#define PS_M_FLAGS (1 << 4) +#define PS_M_GRAVITY (1 << 5) +#define PS_M_DELTA_ANGLES (1 << 6) + +#define PS_VIEWOFFSET (1 << 7) +#define PS_VIEWANGLES (1 << 8) +#define PS_KICKANGLES (1 << 9) +#define PS_BLEND (1 << 10) +#define PS_FOV (1 << 11) +#define PS_WEAPONINDEX (1 << 12) +#define PS_WEAPONFRAME (1 << 13) +#define PS_RDFLAGS (1 << 14) + +/*============================================== */ + +/* user_cmd_t communication */ + +/* ms and light always sent, the others are optional */ +#define CM_ANGLE1 (1 << 0) +#define CM_ANGLE2 (1 << 1) +#define CM_ANGLE3 (1 << 2) +#define CM_FORWARD (1 << 3) +#define CM_SIDE (1 << 4) +#define CM_UP (1 << 5) +#define CM_BUTTONS (1 << 6) +#define CM_IMPULSE (1 << 7) + +/*============================================== */ + +/* a sound without an ent or pos will be a local only sound */ +#define SND_VOLUME (1 << 0) /* a byte */ +#define SND_ATTENUATION (1 << 1) /* a byte */ +#define SND_POS (1 << 2) /* three coordinates */ +#define SND_ENT (1 << 3) /* a short 0-2: channel, 3-12: entity */ +#define SND_OFFSET (1 << 4) /* a byte, msec offset from frame start */ + +#define DEFAULT_SOUND_PACKET_VOLUME 1.0 +#define DEFAULT_SOUND_PACKET_ATTENUATION 1.0 + +/*============================================== */ + +/* entity_state_t communication */ + +/* try to pack the common update flags into the first byte */ +#define U_ORIGIN1 (1 << 0) +#define U_ORIGIN2 (1 << 1) +#define U_ANGLE2 (1 << 2) +#define U_ANGLE3 (1 << 3) +#define U_FRAME8 (1 << 4) /* frame is a byte */ +#define U_EVENT (1 << 5) +#define U_REMOVE (1 << 6) /* REMOVE this entity, don't add it */ +#define U_MOREBITS1 (1 << 7) /* read one additional byte */ + +/* second byte */ +#define U_NUMBER16 (1 << 8) /* NUMBER8 is implicit if not set */ +#define U_ORIGIN3 (1 << 9) +#define U_ANGLE1 (1 << 10) +#define U_MODEL (1 << 11) +#define U_RENDERFX8 (1 << 12) /* fullbright, etc */ +#define U_EFFECTS8 (1 << 14) /* autorotate, trails, etc */ +#define U_MOREBITS2 (1 << 15) /* read one additional byte */ + +/* third byte */ +#define U_SKIN8 (1 << 16) +#define U_FRAME16 (1 << 17) /* frame is a short */ +#define U_RENDERFX16 (1 << 18) /* 8 + 16 = 32 */ +#define U_EFFECTS16 (1 << 19) /* 8 + 16 = 32 */ +#define U_MODEL2 (1 << 20) /* weapons, flags, etc */ +#define U_MODEL3 (1 << 21) +#define U_MODEL4 (1 << 22) +#define U_MOREBITS3 (1 << 23) /* read one additional byte */ + +/* fourth byte */ +#define U_OLDORIGIN (1 << 24) +#define U_SKIN16 (1 << 25) +#define U_SOUND (1 << 26) +#define U_SOLID (1 << 27) + +/* CMD - Command text buffering and command execution */ + +/* + * Any number of commands can be added in a frame, from several different + * sources. Most commands come from either keybindings or console line + * input, but remote servers can also send across commands and entire text + * files can be execed. + * + * The + command line options are also added to the command buffer. + * + * The game starts with a Cbuf_AddText ("exec quake.rc\n"); Cbuf_Execute + * (); + */ + +#define EXEC_NOW 0 /* don't return until completed */ +#define EXEC_INSERT 1 /* insert at current position, but don't run yet */ +#define EXEC_APPEND 2 /* add to end of the command buffer */ + +void Cbuf_Init(void); + +/* allocates an initial text buffer that will grow as needed */ + +void Cbuf_AddText(char *text); + +/* as new commands are generated from the console or keybindings, */ +/* the text is added to the end of the command buffer. */ + +void Cbuf_InsertText(char *text); + +/* when a command wants to issue other commands immediately, the text is */ +/* inserted at the beginning of the buffer, before any remaining unexecuted */ +/* commands. */ + +void Cbuf_ExecuteText(int exec_when, char *text); + +/* this can be used in place of either Cbuf_AddText or Cbuf_InsertText */ + +void Cbuf_AddEarlyCommands(qboolean clear); + +/* adds all the +set commands from the command line */ + +qboolean Cbuf_AddLateCommands(void); + +/* adds all the remaining + commands from the command line */ +/* Returns true if any late commands were added, which */ +/* will keep the demoloop from immediately starting */ + +void Cbuf_Execute(void); + +/* Pulls off \n terminated lines of text from the command buffer and sends */ +/* them through Cmd_ExecuteString. Stops when the buffer is empty. */ +/* Normally called once per frame, but may be explicitly invoked. */ +/* Do not call inside a command function! */ + +void Cbuf_CopyToDefer(void); +void Cbuf_InsertFromDefer(void); + +/* These two functions are used to defer any pending commands while a map */ +/* is being loaded */ + +/*=================================================================== */ + +/* + * Command execution takes a null terminated string, breaks it into tokens, + * then searches for a command or variable that matches the first token. + */ + +typedef void (*xcommand_t)(void); + +void Cmd_Init(void); + +void Cmd_AddCommand(char *cmd_name, xcommand_t function); + +/* called by the init functions of other parts of the program to */ +/* register commands and functions to call for them. */ +/* The cmd_name is referenced later, so it should not be in temp memory */ +/* if function is NULL, the command will be forwarded to the server */ +/* as a clc_stringcmd instead of executed locally */ +void Cmd_RemoveCommand(char *cmd_name); + +qboolean Cmd_Exists(char *cmd_name); + +/* used by the cvar code to check for cvar / command name overlap */ + +char *Cmd_CompleteCommand(char *partial); + +char *Cmd_CompleteMapCommand(char *partial); + +/* attempts to match a partial command for automatic command line completion */ +/* returns NULL if nothing fits */ + +int Cmd_Argc(void); +char *Cmd_Argv(int arg); +char *Cmd_Args(void); + +/* The functions that execute commands get their parameters with these */ +/* functions. Cmd_Argv () will return an empty string, not a NULL */ +/* if arg > argc, so string operations are always safe. */ + +void Cmd_TokenizeString(char *text, qboolean macroExpand); + +/* Takes a null terminated string. Does not need to be /n terminated. */ +/* breaks the string up into arg tokens. */ + +void Cmd_ExecuteString(char *text); + +/* Parses a single line of text into arguments and tries to execute it */ +/* as if it was typed at the console */ + +void Cmd_ForwardToServer(void); + +/* adds the current command line as a clc_stringcmd to the client message. */ +/* things like godmode, noclip, etc, are commands directed to the server, */ +/* so when they are typed in at the console, they will need to be forwarded. */ + +/* CVAR */ + +/* + * cvar_t variables are used to hold scalar or string variables that can be + * changed or displayed at the console or prog code as well as accessed + * directly in C code. + * + * The user can access cvars from the console in three ways: + * r_draworder prints the current value + * r_draworder 0 sets the current value to 0 + * set r_draworder 0 as above, but creates the cvar if not present + * Cvars are restricted from having the same names as commands to keep this + * interface from being ambiguous. + */ + +extern cvar_t *cvar_vars; + +cvar_t *Cvar_Get(char *var_name, char *value, int flags); + +/* creates the variable if it doesn't exist, or returns the existing one */ +/* if it exists, the value will not be changed, but flags will be ORed in */ +/* that allows variables to be unarchived without needing bitflags */ + +cvar_t *Cvar_Set(char *var_name, char *value); + +/* will create the variable if it doesn't exist */ + +cvar_t *Cvar_ForceSet(char *var_name, char *value); + +/* will set the variable even if NOSET or LATCH */ + +cvar_t *Cvar_FullSet(char *var_name, char *value, int flags); + +void Cvar_SetValue(char *var_name, float value); + +/* expands value to a string and calls Cvar_Set */ + +float Cvar_VariableValue(char *var_name); + +/* returns 0 if not defined or non numeric */ + +const char *Cvar_VariableString(const char *var_name); + +/* returns an empty string if not defined */ + +char *Cvar_CompleteVariable(char *partial); + +/* attempts to match a partial variable name for command line completion */ +/* returns NULL if nothing fits */ + +void Cvar_GetLatchedVars(void); + +/* any CVAR_LATCHED variables that have been set will now take effect */ + +qboolean Cvar_Command(void); + +/* called by Cmd_ExecuteString when Cmd_Argv(0) doesn't match a known */ +/* command. Returns true if the command was a variable reference that */ +/* was handled. (print or change) */ + +void Cvar_WriteVariables(char *path); + +/* appends lines containing "set variable value" for all variables */ +/* with the archive flag set to true. */ + +void Cvar_Init(void); + +void Cvar_Fini(void); + +char *Cvar_Userinfo(void); + +/* returns an info string containing all the CVAR_USERINFO cvars */ + +char *Cvar_Serverinfo(void); + +/* returns an info string containing all the CVAR_SERVERINFO cvars */ + +extern qboolean userinfo_modified; +/* this is set each time a CVAR_USERINFO variable is changed */ +/* so that the client knows to send it to the server */ + +/* NET */ + +#define PORT_ANY -1 +#define MAX_MSGLEN 1400 /* max length of a message */ +#define PACKET_HEADER 10 /* two ints and a short */ + +typedef enum +{ + NA_LOOPBACK, + NA_BROADCAST, + NA_IP, + NA_IPX, + NA_BROADCAST_IPX, + NA_IP6, + NA_MULTICAST6 +} netadrtype_t; + +typedef enum {NS_CLIENT, NS_SERVER} netsrc_t; + +typedef struct +{ + netadrtype_t type; + byte ip[16]; + unsigned int scope_id; + byte ipx[10]; + + unsigned short port; +} netadr_t; + +void NET_Init(void); +void NET_Shutdown(void); + +void NET_Config(qboolean multiplayer); + +qboolean NET_GetPacket(netsrc_t sock, netadr_t *net_from, + sizebuf_t *net_message); +void NET_SendPacket(netsrc_t sock, int length, void *data, netadr_t to); + +qboolean NET_CompareAdr(netadr_t a, netadr_t b); +qboolean NET_CompareBaseAdr(netadr_t a, netadr_t b); +qboolean NET_IsLocalAddress(netadr_t adr); +char *NET_AdrToString(netadr_t a); +qboolean NET_StringToAdr(const char *s, netadr_t *a); +void NET_Sleep(int msec); + +/*=================================================================== */ + +#define OLD_AVG 0.99 +#define MAX_LATENT 32 + +typedef struct +{ + qboolean fatal_error; + + netsrc_t sock; + + int dropped; /* between last packet and previous */ + + int last_received; /* for timeouts */ + int last_sent; /* for retransmits */ + + netadr_t remote_address; + int qport; /* qport value to write when transmitting */ + + /* sequencing variables */ + int incoming_sequence; + int incoming_acknowledged; + int incoming_reliable_acknowledged; /* single bit */ + + int incoming_reliable_sequence; /* single bit, maintained local */ + + int outgoing_sequence; + int reliable_sequence; /* single bit */ + int last_reliable_sequence; /* sequence number of last send */ + + /* reliable staging and holding areas */ + sizebuf_t message; /* writing buffer to send to server */ + byte message_buf[MAX_MSGLEN - 16]; /* leave space for header */ + + /* message is copied to this buffer when it is first transfered */ + int reliable_length; + byte reliable_buf[MAX_MSGLEN - 16]; /* unacked reliable message */ +} netchan_t; + +extern netadr_t net_from; +extern sizebuf_t net_message; +extern byte net_message_buffer[MAX_MSGLEN]; + +void Netchan_Init(void); +void Netchan_Setup(netsrc_t sock, netchan_t *chan, netadr_t adr, int qport); + +qboolean Netchan_NeedReliable(netchan_t *chan); +void Netchan_Transmit(netchan_t *chan, int length, byte *data); +void Netchan_OutOfBand(int net_socket, netadr_t adr, int length, byte *data); +void Netchan_OutOfBandPrint(int net_socket, netadr_t adr, char *format, ...); +qboolean Netchan_Process(netchan_t *chan, sizebuf_t *msg); + +qboolean Netchan_CanReliable(netchan_t *chan); + +/* CMODEL */ + +#include "files.h" + +cmodel_t *CM_LoadMap(char *name, qboolean clientload, unsigned *checksum); +cmodel_t *CM_InlineModel(char *name); /* *1, *2, etc */ + +int CM_NumClusters(void); +int CM_NumInlineModels(void); +char *CM_EntityString(void); + +/* creates a clipping hull for an arbitrary box */ +int CM_HeadnodeForBox(vec3_t mins, vec3_t maxs); + +/* returns an ORed contents mask */ +int CM_PointContents(vec3_t p, int headnode); +int CM_TransformedPointContents(vec3_t p, int headnode, + vec3_t origin, vec3_t angles); + +trace_t CM_BoxTrace(vec3_t start, vec3_t end, vec3_t mins, + vec3_t maxs, int headnode, int brushmask); +trace_t CM_TransformedBoxTrace(vec3_t start, vec3_t end, + vec3_t mins, vec3_t maxs, int headnode, + int brushmask, vec3_t origin, vec3_t angles); + +byte *CM_ClusterPVS(int cluster); +byte *CM_ClusterPHS(int cluster); + +int CM_PointLeafnum(vec3_t p); + +/* call with topnode set to the headnode, returns with topnode */ +/* set to the first node that splits the box */ +int CM_BoxLeafnums(vec3_t mins, vec3_t maxs, int *list, + int listsize, int *topnode); + +int CM_LeafContents(int leafnum); +int CM_LeafCluster(int leafnum); +int CM_LeafArea(int leafnum); + +void CM_SetAreaPortalState(int portalnum, qboolean open); +qboolean CM_AreasConnected(int area1, int area2); + +int CM_WriteAreaBits(byte *buffer, int area); +qboolean CM_HeadnodeVisible(int headnode, byte *visbits); + +void CM_WritePortalState(FILE *f); + +/* PLAYER MOVEMENT CODE */ + +extern float pm_airaccelerate; + +void Pmove(pmove_t *pmove); + +/* FILESYSTEM */ + +#define SFF_INPACK 0x20 + +typedef int fileHandle_t; + +typedef enum +{ + FS_READ, + FS_WRITE, + FS_APPEND +} fsMode_t; + +typedef enum +{ + FS_SEEK_CUR, + FS_SEEK_SET, + FS_SEEK_END +} fsOrigin_t; + +typedef enum +{ + FS_SEARCH_PATH_EXTENSION, + FS_SEARCH_BY_FILTER, + FS_SEARCH_FULL_PATH +} fsSearchType_t; + +void FS_DPrintf(const char *format, ...); +int FS_FOpenFile(const char *name, fileHandle_t *f, qboolean gamedir_only); +void FS_FCloseFile(fileHandle_t f); +int FS_Read(void *buffer, int size, fileHandle_t f); +int FS_FRead(void *buffer, int size, int count, fileHandle_t f); + +// returns the filename used to open f, but (if opened from pack) in correct case +// returns NULL if f is no valid handle +const char* FS_GetFilenameForHandle(fileHandle_t f); + +char **FS_ListFiles(char *findname, int *numfiles, + unsigned musthave, unsigned canthave); +char **FS_ListFiles2(char *findname, int *numfiles, + unsigned musthave, unsigned canthave); +void FS_FreeList(char **list, int nfiles); + +void FS_InitFilesystem(void); +void FS_BuildGameSpecificSearchPath(char *dir); +char *FS_Gamedir(void); +char *FS_NextPath(char *prevpath); +int FS_LoadFile(char *path, void **buffer); +qboolean FS_FileInGamedir(const char *file); +qboolean FS_AddPAKFromGamedir(const char *pak); +const char* FS_GetNextRawPath(const char* lastRawPath); +char **FS_ListMods(int *nummods); + +/* a null buffer will just return the file length without loading */ +/* a -1 length is not present */ + +/* properly handles partial reads */ + +void FS_FreeFile(void *buffer); +void FS_CreatePath(char *path); + +/* MISC */ + +#define ERR_FATAL 0 /* exit the entire game with a popup window */ +#define ERR_DROP 1 /* print to console and disconnect from game */ +#define ERR_QUIT 2 /* not an error, just a normal exit */ + +#define EXEC_NOW 0 /* don't return until completed */ +#define EXEC_INSERT 1 /* insert at current position, but don't run yet */ +#define EXEC_APPEND 2 /* add to end of the command buffer */ + +#define PRINT_ALL 0 +#define PRINT_DEVELOPER 1 /* only print when "developer 1" */ + +void Com_BeginRedirect(int target, char *buffer, int buffersize, void (*flush)(int, char *)); +void Com_EndRedirect(void); +void Com_Printf(char *fmt, ...) __attribute__ ((format (printf, 1, 2))); +void Com_DPrintf(char *fmt, ...) __attribute__ ((format (printf, 1, 2))); +void Com_VPrintf(int print_level, const char *fmt, va_list argptr); /* print_level is PRINT_ALL or PRINT_DEVELOPER */ +void Com_MDPrintf(char *fmt, ...) __attribute__ ((format (printf, 1, 2))); +YQ2_ATTR_NORETURN void Com_Error(int code, char *fmt, ...) __attribute__ ((format (printf, 2, 3))); +YQ2_ATTR_NORETURN void Com_Quit(void); + +/* Ugly work around for unsupported + * format specifiers unter mingw. */ +#ifdef WIN32 +#define YQ2_COM_PRId64 "%I64d" +#define YQ2_COM_PRIdS "%Id" +#else +#define YQ2_COM_PRId64 "%ld" +#define YQ2_COM_PRIdS "%zd" +#endif + +// terminate yq2 (with Com_Error()) if VAR is NULL (after malloc() or similar) +// and print message about it +#define YQ2_COM_CHECK_OOM(VAR, ALLOC_FN_NAME, ALLOC_SIZE) \ + if(VAR == NULL) { \ + Com_Error(ERR_FATAL, "%s for " YQ2_COM_PRIdS " bytes failed in %s() (%s == NULL)! Out of Memory?!\n", \ + ALLOC_FN_NAME, (size_t)ALLOC_SIZE, __func__, #VAR); } + +int Com_ServerState(void); /* this should have just been a cvar... */ +void Com_SetServerState(int state); + +unsigned Com_BlockChecksum(void *buffer, int length); +byte COM_BlockSequenceCRCByte(byte *base, int length, int sequence); + +extern cvar_t *developer; +extern cvar_t *modder; +extern cvar_t *dedicated; +extern cvar_t *host_speeds; +extern cvar_t *log_stats; + +/* External entity files. */ +extern cvar_t *sv_entfile; + +/* Hack for portable client */ +extern qboolean is_portable; + +/* Hack for external datadir */ +extern char datadir[MAX_OSPATH]; + +/* Hack for external datadir */ +extern char cfgdir[MAX_OSPATH]; + +/* Hack for working 'game' cmd */ +extern char userGivenGame[MAX_QPATH]; +extern char **mapnames; +extern int nummaps; + +extern FILE *log_stats_file; + +/* host_speeds times */ +extern int time_before_game; +extern int time_after_game; +extern int time_before_ref; +extern int time_after_ref; + +void Z_Free(void *ptr); +void *Z_Malloc(int size); /* returns 0 filled memory */ +void *Z_TagMalloc(int size, int tag); +void Z_FreeTags(int tag); + +void Qcommon_Init(int argc, char **argv); +void Qcommon_ExecConfigs(qboolean addEarlyCmds); +const char* Qcommon_GetInitialGame(void); +void Qcommon_Frame(int msec); +void Qcommon_Shutdown(void); + +#define NUMVERTEXNORMALS 162 +extern vec3_t bytedirs[NUMVERTEXNORMALS]; + +/* this is in the client code, but can be used for debugging from server */ +void SCR_DebugGraph(float value, int color); + +/* CLIENT / SERVER SYSTEMS */ + +void CL_Init(void); +void CL_Drop(void); +void CL_Shutdown(void); +void CL_Frame(int packetdelta, int renderdelta, int timedelta, qboolean packetframe, qboolean renderframe); +void Con_Print(char *text); +void SCR_BeginLoadingPlaque(void); + +void SV_Init(void); +void SV_Shutdown(char *finalmsg, qboolean reconnect); +void SV_Frame(int usec); + +/* ======================================================================= */ + +// Platform specific functions. + +// system.c +char *Sys_ConsoleInput(void); +void Sys_ConsoleOutput(char *string); +YQ2_ATTR_NORETURN void Sys_Error(char *error, ...); +YQ2_ATTR_NORETURN void Sys_Quit(void); +void Sys_Init(void); +char *Sys_GetHomeDir(void); +void Sys_Remove(const char *path); +int Sys_Rename(const char *from, const char *to); +void Sys_RemoveDir(const char *path); +long long Sys_Microseconds(void); +void Sys_Nanosleep(int); +void *Sys_GetProcAddress(void *handle, const char *sym); +void Sys_FreeLibrary(void *handle); +void *Sys_LoadLibrary(const char *path, const char *sym, void **handle); +void *Sys_GetGameAPI(void *parms); +void Sys_UnloadGame(void); +void Sys_GetWorkDir(char *buffer, size_t len); +qboolean Sys_SetWorkDir(char *path); +void Sys_Realpath(const char *in, char *out, size_t size); + +// Windows only (system.c) +#ifdef _WIN32 +void Sys_RedirectStdout(void); +void Sys_SetHighDPIMode(void); +#endif + +// misc.c +const char *Sys_GetBinaryDir(void); +void Sys_SetupFPU(void); + +/* ======================================================================= */ + +#endif diff --git a/src/common/header/crc.h b/src/common/header/crc.h new file mode 100644 index 0000000..267a789 --- /dev/null +++ b/src/common/header/crc.h @@ -0,0 +1,33 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * Corresponding header for crc.c + * + * ======================================================================= + */ + +#ifndef CO_CRC_H +#define CO_CRC_H + +void CRC_Init(unsigned short *crcvalue); +unsigned short CRC_Block(byte *start, int count); + +#endif diff --git a/src/common/header/files.h b/src/common/header/files.h new file mode 100644 index 0000000..2d00a19 --- /dev/null +++ b/src/common/header/files.h @@ -0,0 +1,453 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * The prototypes for most file formats used by Quake II + * + * ======================================================================= + */ + +#ifndef CO_FILES_H +#define CO_FILES_H + +/* The .pak files are just a linear collapse of a directory tree */ + +#define IDPAKHEADER (('K' << 24) + ('C' << 16) + ('A' << 8) + 'P') + +typedef struct +{ + char name[56]; + int filepos, filelen; +} dpackfile_t; + +typedef struct +{ + int ident; /* == IDPAKHEADER */ + int dirofs; + int dirlen; +} dpackheader_t; + +#define MAX_FILES_IN_PACK 4096 + +/* PCX files are used for as many images as possible */ + +typedef struct +{ + char manufacturer; + char version; + char encoding; + char bits_per_pixel; + unsigned short xmin, ymin, xmax, ymax; + unsigned short hres, vres; + unsigned char palette[48]; + char reserved; + char color_planes; + unsigned short bytes_per_line; + unsigned short palette_type; + char filler[58]; + unsigned char data; /* unbounded */ +} pcx_t; + +/* .MD2 triangle model file format */ + +#define IDALIASHEADER (('2' << 24) + ('P' << 16) + ('D' << 8) + 'I') +#define ALIAS_VERSION 8 + +#define MAX_TRIANGLES 4096 +#define MAX_VERTS 2048 +#define MAX_FRAMES 512 +#define MAX_MD2SKINS 32 +#define MAX_SKINNAME 64 + +typedef struct +{ + short s; + short t; +} dstvert_t; + +typedef struct +{ + short index_xyz[3]; + short index_st[3]; +} dtriangle_t; + +typedef struct +{ + byte v[3]; /* scaled byte to fit in frame mins/maxs */ + byte lightnormalindex; +} dtrivertx_t; + +#define DTRIVERTX_V0 0 +#define DTRIVERTX_V1 1 +#define DTRIVERTX_V2 2 +#define DTRIVERTX_LNI 3 +#define DTRIVERTX_SIZE 4 + +typedef struct +{ + float scale[3]; /* multiply byte verts by this */ + float translate[3]; /* then add this */ + char name[16]; /* frame name from grabbing */ + dtrivertx_t verts[1]; /* variable sized */ +} daliasframe_t; + +/* the glcmd format: + * - a positive integer starts a tristrip command, followed by that many + * vertex structures. + * - a negative integer starts a trifan command, followed by -x vertexes + * a zero indicates the end of the command list. + * - a vertex consists of a floating point s, a floating point t, + * and an integer vertex index. */ + +typedef struct +{ + int ident; + int version; + + int skinwidth; + int skinheight; + int framesize; /* byte size of each frame */ + + int num_skins; + int num_xyz; + int num_st; /* greater than num_xyz for seams */ + int num_tris; + int num_glcmds; /* dwords in strip/fan command list */ + int num_frames; + + int ofs_skins; /* each skin is a MAX_SKINNAME string */ + int ofs_st; /* byte offset from start for stverts */ + int ofs_tris; /* offset for dtriangles */ + int ofs_frames; /* offset for first frame */ + int ofs_glcmds; + int ofs_end; /* end of file */ +} dmdl_t; + +/* .SP2 sprite file format */ + +#define IDSPRITEHEADER (('2' << 24) + ('S' << 16) + ('D' << 8) + 'I') /* little-endian "IDS2" */ +#define SPRITE_VERSION 2 + +typedef struct +{ + int width, height; + int origin_x, origin_y; /* raster coordinates inside pic */ + char name[MAX_SKINNAME]; /* name of pcx file */ +} dsprframe_t; + +typedef struct +{ + int ident; + int version; + int numframes; + dsprframe_t frames[1]; /* variable sized */ +} dsprite_t; + +/* .WAL texture file format */ + +#define MIPLEVELS 4 +typedef struct miptex_s +{ + char name[32]; + unsigned width, height; + unsigned offsets[MIPLEVELS]; /* four mip maps stored */ + char animname[32]; /* next frame in animation chain */ + int flags; + int contents; + int value; +} miptex_t; + +/* .M8 texture file format */ + +#define M8_MIP_LEVELS 16 +#define M8_VERSION 0x2 + +typedef struct { + unsigned char r; + unsigned char g; + unsigned char b; +} rgb_t; + +typedef struct m8tex_s +{ + unsigned version; + char name[32]; + unsigned width[M8_MIP_LEVELS]; + unsigned height[M8_MIP_LEVELS]; + unsigned offsets[M8_MIP_LEVELS]; /* 16 mip maps stored */ + char animname[32]; /* next frame in animation chain */ + rgb_t palette[256]; + int flags; + int contents; + int value; +} m8tex_t; + +/* .BSP file format */ + +#define IDBSPHEADER (('P' << 24) + ('S' << 16) + ('B' << 8) + 'I') /* little-endian "IBSP" */ +#define BSPVERSION 38 + +/* upper design bounds: leaffaces, leafbrushes, planes, and + * verts are still bounded by 16 bit short limits */ +#define MAX_MAP_MODELS 1024 +#define MAX_MAP_BRUSHES 8192 +#define MAX_MAP_ENTITIES 2048 +#define MAX_MAP_ENTSTRING 0x40000 +#define MAX_MAP_TEXINFO 8192 + +#define MAX_MAP_AREAS 256 +#define MAX_MAP_AREAPORTALS 1024 +#define MAX_MAP_PLANES 65536 +#define MAX_MAP_NODES 65536 +#define MAX_MAP_BRUSHSIDES 65536 +#define MAX_MAP_LEAFS 65536 +#define MAX_MAP_VERTS 65536 +#define MAX_MAP_FACES 65536 +#define MAX_MAP_LEAFFACES 65536 +#define MAX_MAP_LEAFBRUSHES 65536 +#define MAX_MAP_PORTALS 65536 +#define MAX_MAP_EDGES 128000 +#define MAX_MAP_SURFEDGES 256000 +#define MAX_MAP_LIGHTING 0x200000 +#define MAX_MAP_VISIBILITY 0x100000 + +/* key / value pair sizes */ + +#define MAX_KEY 32 +#define MAX_VALUE 1024 + +/* ================================================================== */ + +typedef struct +{ + int fileofs, filelen; +} lump_t; + +#define LUMP_ENTITIES 0 +#define LUMP_PLANES 1 +#define LUMP_VERTEXES 2 +#define LUMP_VISIBILITY 3 +#define LUMP_NODES 4 +#define LUMP_TEXINFO 5 +#define LUMP_FACES 6 +#define LUMP_LIGHTING 7 +#define LUMP_LEAFS 8 +#define LUMP_LEAFFACES 9 +#define LUMP_LEAFBRUSHES 10 +#define LUMP_EDGES 11 +#define LUMP_SURFEDGES 12 +#define LUMP_MODELS 13 +#define LUMP_BRUSHES 14 +#define LUMP_BRUSHSIDES 15 +#define LUMP_POP 16 +#define LUMP_AREAS 17 +#define LUMP_AREAPORTALS 18 +#define HEADER_LUMPS 19 + +typedef struct +{ + int ident; + int version; + lump_t lumps[HEADER_LUMPS]; +} dheader_t; + +typedef struct +{ + float mins[3], maxs[3]; + float origin[3]; /* for sounds or lights */ + int headnode; + int firstface, numfaces; /* submodels just draw faces without + walking the bsp tree */ +} dmodel_t; + +typedef struct +{ + float point[3]; +} dvertex_t; + +/* 0-2 are axial planes */ +#define PLANE_X 0 +#define PLANE_Y 1 +#define PLANE_Z 2 + +/* 3-5 are non-axial planes snapped to the nearest */ +#define PLANE_ANYX 3 +#define PLANE_ANYY 4 +#define PLANE_ANYZ 5 + +/* planes (x&~1) and (x&~1)+1 are always opposites */ + +typedef struct +{ + float normal[3]; + float dist; + int type; /* PLANE_X - PLANE_ANYZ */ +} dplane_t; + +/* contents flags are seperate bits + * - given brush can contribute multiple content bits + * - multiple brushes can be in a single leaf */ + +/* lower bits are stronger, and will eat weaker brushes completely */ +#define CONTENTS_SOLID 1 /* an eye is never valid in a solid */ +#define CONTENTS_WINDOW 2 /* translucent, but not watery */ +#define CONTENTS_AUX 4 +#define CONTENTS_LAVA 8 +#define CONTENTS_SLIME 16 +#define CONTENTS_WATER 32 +#define CONTENTS_MIST 64 +#define LAST_VISIBLE_CONTENTS 64 + +/* remaining contents are non-visible, and don't eat brushes */ +#define CONTENTS_AREAPORTAL 0x8000 + +#define CONTENTS_PLAYERCLIP 0x10000 +#define CONTENTS_MONSTERCLIP 0x20000 + +/* currents can be added to any other contents, and may be mixed */ +#define CONTENTS_CURRENT_0 0x40000 +#define CONTENTS_CURRENT_90 0x80000 +#define CONTENTS_CURRENT_180 0x100000 +#define CONTENTS_CURRENT_270 0x200000 +#define CONTENTS_CURRENT_UP 0x400000 +#define CONTENTS_CURRENT_DOWN 0x800000 + +#define CONTENTS_ORIGIN 0x1000000 /* removed before bsping an entity */ + +#define CONTENTS_MONSTER 0x2000000 /* should never be on a brush, only in game */ +#define CONTENTS_DEADMONSTER 0x4000000 +#define CONTENTS_DETAIL 0x8000000 /* brushes to be added after vis leafs */ +#define CONTENTS_TRANSLUCENT 0x10000000 /* auto set if any surface has trans */ +#define CONTENTS_LADDER 0x20000000 + +#define SURF_LIGHT 0x1 /* value will hold the light strength */ + +#define SURF_SLICK 0x2 /* effects game physics */ + +#define SURF_SKY 0x4 /* don't draw, but add to skybox */ +#define SURF_WARP 0x8 /* turbulent water warp */ +#define SURF_TRANS33 0x10 +#define SURF_TRANS66 0x20 +#define SURF_FLOWING 0x40 /* scroll towards angle */ +#define SURF_NODRAW 0x80 /* don't bother referencing the texture */ + +typedef struct +{ + int planenum; + int children[2]; /* negative numbers are -(leafs+1), not nodes */ + short mins[3]; /* for frustom culling */ + short maxs[3]; + unsigned short firstface; + unsigned short numfaces; /* counting both sides */ +} dnode_t; + +typedef struct texinfo_s +{ + float vecs[2][4]; /* [s/t][xyz offset] */ + int flags; /* miptex flags + overrides light emission, etc */ + int value; + char texture[32]; /* texture name (textures*.wal) */ + int nexttexinfo; /* for animations, -1 = end of chain */ +} texinfo_t; + +/* note that edge 0 is never used, because negative edge + nums are used for counterclockwise use of the edge in + a face */ +typedef struct +{ + unsigned short v[2]; /* vertex numbers */ +} dedge_t; + +#define MAXLIGHTMAPS 4 +typedef struct +{ + unsigned short planenum; + short side; + + int firstedge; /* we must support > 64k edges */ + short numedges; + short texinfo; + + /* lighting info */ + byte styles[MAXLIGHTMAPS]; + int lightofs; /* start of [numstyles*surfsize] samples */ +} dface_t; + +typedef struct +{ + int contents; /* OR of all brushes (not needed?) */ + + short cluster; + short area; + + short mins[3]; /* for frustum culling */ + short maxs[3]; + + unsigned short firstleafface; + unsigned short numleaffaces; + + unsigned short firstleafbrush; + unsigned short numleafbrushes; +} dleaf_t; + +typedef struct +{ + unsigned short planenum; /* facing out of the leaf */ + short texinfo; +} dbrushside_t; + +typedef struct +{ + int firstside; + int numsides; + int contents; +} dbrush_t; + +#define ANGLE_UP -1 +#define ANGLE_DOWN -2 + +/* the visibility lump consists of a header with a count, then + * byte offsets for the PVS and PHS of each cluster, then the raw + * compressed bit vectors */ +#define DVIS_PVS 0 +#define DVIS_PHS 1 +typedef struct +{ + int numclusters; + int bitofs[8][2]; /* bitofs[numclusters][2] */ +} dvis_t; + +/* each area has a list of portals that lead into other areas + * when portals are closed, other areas may not be visible or + * hearable even if the vis info says that it should be */ +typedef struct +{ + int portalnum; + int otherarea; +} dareaportal_t; + +typedef struct +{ + int numareaportals; + int firstareaportal; +} darea_t; + +#endif + diff --git a/src/common/header/ref_api.h b/src/common/header/ref_api.h new file mode 100644 index 0000000..a788e3e --- /dev/null +++ b/src/common/header/ref_api.h @@ -0,0 +1,283 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, + * USA. + * + * ======================================================================= + * + * ABI between client and refresher + * + * ======================================================================= + */ + +#ifndef CL_REF_H +#define CL_REF_H + +#include "common.h" +#include "vid.h" + +#define MAX_DLIGHTS 32 +#define MAX_ENTITIES 128 +#define MAX_PARTICLES 4096 +#define MAX_LIGHTSTYLES 256 + +#define POWERSUIT_SCALE 4.0F + +#define SHELL_RED_COLOR 0xF2 +#define SHELL_GREEN_COLOR 0xD0 +#define SHELL_BLUE_COLOR 0xF3 + +#define SHELL_RG_COLOR 0xDC +#define SHELL_RB_COLOR 0x68 +#define SHELL_BG_COLOR 0x78 + +#define SHELL_DOUBLE_COLOR 0xDF +#define SHELL_HALF_DAM_COLOR 0x90 +#define SHELL_CYAN_COLOR 0x72 + +#define SHELL_WHITE_COLOR 0xD7 + +#define ENTITY_FLAGS 68 + +typedef struct entity_s { + struct model_s *model; /* opaque type outside refresh */ + float angles[3]; + + /* most recent data */ + float origin[3]; /* also used as RF_BEAM's "from" */ + int frame; /* also used as RF_BEAM's diameter */ + + /* previous data for lerping */ + float oldorigin[3]; /* also used as RF_BEAM's "to" */ + int oldframe; + + /* misc */ + float backlerp; /* 0.0 = current, 1.0 = old */ + int skinnum; /* also used as RF_BEAM's palette index */ + + int lightstyle; /* for flashing entities */ + float alpha; /* ignore if RF_TRANSLUCENT isn't set */ + + struct image_s *skin; /* NULL for inline skin */ + int flags; +} entity_t; + +typedef struct { + vec3_t origin; + vec3_t color; + float intensity; +} dlight_t; + +typedef struct { + vec3_t origin; + int color; + float alpha; +} particle_t; + +typedef struct { + float rgb[3]; /* 0.0 - 2.0 */ + float white; /* r+g+b */ +} lightstyle_t; + +typedef struct { + int x, y, width, height; /* in virtual screen coordinates */ + float fov_x, fov_y; + float vieworg[3]; + float viewangles[3]; + float blend[4]; /* rgba 0-1 full screen blend */ + float time; /* time is used to auto animate */ + int rdflags; /* RDF_UNDERWATER, etc */ + + byte *areabits; /* if not NULL, only areas with set bits will be drawn */ + + lightstyle_t *lightstyles; /* [MAX_LIGHTSTYLES] */ + + int num_entities; + entity_t *entities; + + int num_dlights; // <= 32 (MAX_DLIGHTS) + dlight_t *dlights; + + int num_particles; + particle_t *particles; +} refdef_t; + +// Renderer restart type. +typedef enum { + RESTART_UNDEF, + RESTART_NO, + RESTART_FULL, + RESTART_PARTIAL +} ref_restart_t; + +// FIXME: bump API_VERSION? +#define API_VERSION 5 +#define EXPORT +#define IMPORT + +// +// these are the functions exported by the refresh module +// +typedef struct +{ + // if api_version is different, the dll cannot be used + int api_version; + + // called when the library is loaded + qboolean (EXPORT *Init) (void); + + // called before the library is unloaded + void (EXPORT *Shutdown) (void); + + // called by GLimp_InitGraphics() before creating window, + // returns flags for SDL window creation, returns -1 on error + int (EXPORT *PrepareForWindow)(void); + + // called by GLimp_InitGraphics() *after* creating window, + // passing the SDL_Window* (void* so we don't spill SDL.h here) + // (or SDL_Surface* for SDL1.2, another reason to use void*) + // returns true (1) on success + int (EXPORT *InitContext)(void* sdl_window); + + // shuts down rendering (OpenGL) context. + void (EXPORT *ShutdownContext)(void); + + // returns true if vsync is active, else false + qboolean (EXPORT *IsVSyncActive)(void); + + // All data that will be used in a level should be + // registered before rendering any frames to prevent disk hits, + // but they can still be registered at a later time + // if necessary. + // + // EndRegistration will free any remaining data that wasn't registered. + // Any model_s or skin_s pointers from before the BeginRegistration + // are no longer valid after EndRegistration. + // + // Skins and images need to be differentiated, because skins + // are flood filled to eliminate mip map edge errors, and pics have + // an implicit "pics/" prepended to the name. (a pic name that starts with a + // slash will not use the "pics/" prefix or the ".pcx" postfix) + void (EXPORT *BeginRegistration) (char *map); + struct model_s * (EXPORT *RegisterModel) (char *name); + struct image_s * (EXPORT *RegisterSkin) (char *name); + + void (EXPORT *SetSky) (char *name, float rotate, vec3_t axis); + void (EXPORT *EndRegistration) (void); + + void (EXPORT *RenderFrame) (refdef_t *fd); + + struct image_s * (EXPORT *DrawFindPic)(char *name); + + void (EXPORT *DrawGetPicSize) (int *w, int *h, char *name); // will return 0 0 if not found + void (EXPORT *DrawPicScaled) (int x, int y, char *pic, float factor); + void (EXPORT *DrawStretchPic) (int x, int y, int w, int h, char *name); + void (EXPORT *DrawCharScaled)(int x, int y, int num, float scale); + void (EXPORT *DrawTileClear) (int x, int y, int w, int h, char *name); + void (EXPORT *DrawFill) (int x, int y, int w, int h, int c); + void (EXPORT *DrawFadeScreen) (void); + + // Draw images for cinematic rendering (which can have a different palette). Note that calls + void (EXPORT *DrawStretchRaw) (int x, int y, int w, int h, int cols, int rows, byte *data); + + /* + ** video mode and refresh state management entry points + */ + void (EXPORT *SetPalette)( const unsigned char *palette); // NULL = game palette + void (EXPORT *BeginFrame)( float camera_separation ); + void (EXPORT *EndFrame) (void); + qboolean (EXPORT *EndWorldRenderpass) (void); // finish world rendering, apply postprocess and switch to UI render pass + + //void (EXPORT *AppActivate)( qboolean activate ); +} refexport_t; + +typedef struct +{ + YQ2_ATTR_NORETURN_FUNCPTR void (IMPORT *Sys_Error) (int err_level, char *str, ...) __attribute__ ((format (printf, 2, 3))); + + void (IMPORT *Cmd_AddCommand) (char *name, void(*cmd)(void)); + void (IMPORT *Cmd_RemoveCommand) (char *name); + int (IMPORT *Cmd_Argc) (void); + char *(IMPORT *Cmd_Argv) (int i); + void (IMPORT *Cmd_ExecuteText) (int exec_when, char *text); + + void (IMPORT *Com_VPrintf) (int print_level, const char *fmt, va_list argptr); + + // files will be memory mapped read only + // the returned buffer may be part of a larger pak file, + // or a discrete file from anywhere in the quake search path + // a -1 return means the file does not exist + // NULL can be passed for buf to just determine existance + int (IMPORT *FS_LoadFile) (char *name, void **buf); + void (IMPORT *FS_FreeFile) (void *buf); + + // gamedir will be the current directory that generated + // files should be stored to, ie: "f:\quake\id1" + char *(IMPORT *FS_Gamedir) (void); + + cvar_t *(IMPORT *Cvar_Get) (char *name, char *value, int flags); + cvar_t *(IMPORT *Cvar_Set) (char *name, char *value); + void (IMPORT *Cvar_SetValue) (char *name, float value); + + qboolean (IMPORT *Vid_GetModeInfo)(int *width, int *height, int mode); + void (IMPORT *Vid_MenuInit)( void ); + // called with image data of width*height pixel which comp bytes per pixel (must be 3 or 4 for RGB or RGBA) + // expects the pixels data to be row-wise, starting at top left + void (IMPORT *Vid_WriteScreenshot)( int width, int height, int comp, const void* data ); + + qboolean (IMPORT *GLimp_InitGraphics)(int fullscreen, int *pwidth, int *pheight); + qboolean (IMPORT *GLimp_GetDesktopMode)(int *pwidth, int *pheight); + + void (IMPORT *Vid_RequestRestart)(ref_restart_t rs); +} refimport_t; + +// this is the only function actually exported at the linker level +typedef refexport_t (EXPORT *GetRefAPI_t) (refimport_t); + +// FIXME: #ifdef client/ref around this +extern refexport_t re; +extern refimport_t ri; + +/* + * Refresh API + */ +void R_BeginRegistration(char *map); +void R_Clear(void); +struct model_s *R_RegisterModel(char *name); +struct image_s *R_RegisterSkin(char *name); +void R_SetSky(char *name, float rotate, vec3_t axis); +void R_EndRegistration(void); +struct image_s *Draw_FindPic(char *name); +void R_RenderFrame(refdef_t *fd); +void Draw_GetPicSize(int *w, int *h, char *name); + +void Draw_StretchPic(int x, int y, int w, int h, char *name); +void Draw_PicScaled(int x, int y, char *pic, float factor); + +void Draw_CharScaled(int x, int y, int num, float scale); +void Draw_TileClear(int x, int y, int w, int h, char *name); +void Draw_Fill(int x, int y, int w, int h, int c); +void Draw_FadeScreen(void); +void Draw_StretchRaw(int x, int y, int w, int h, int cols, int rows, byte *data); +//int R_Init(void *hinstance, void *hWnd); +//void R_Shutdown(void); +void R_SetPalette(const unsigned char *palette); +void R_BeginFrame(float camera_separation); +qboolean R_EndWorldRenderpass(void); +void R_EndFrame(void); + +#endif diff --git a/src/common/header/ref_shared.h b/src/common/header/ref_shared.h new file mode 100644 index 0000000..b6a1bec --- /dev/null +++ b/src/common/header/ref_shared.h @@ -0,0 +1,78 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * Header shared between different refreshers (but not with client) + * + * ======================================================================= + */ + +#ifndef SRC_CLIENT_REFRESH_REF_SHARED_H_ +#define SRC_CLIENT_REFRESH_REF_SHARED_H_ + +#include "ref_api.h" + +/* + * skins will be outline flood filled and mip mapped + * pics and sprites with alpha will be outline flood filled + * pic won't be mip mapped + * + * model skin + * sprite frame + * wall texture + * pic + */ +typedef enum +{ + it_skin, + it_sprite, + it_wall, + it_pic, + it_sky +} imagetype_t; + +typedef enum +{ + mod_bad, + mod_brush, + mod_sprite, + mod_alias +} modtype_t; + +#define MAX_LBM_HEIGHT 480 + +extern void R_Printf(int level, const char* msg, ...) __attribute__ ((format (printf, 2, 3))); + +extern void LoadPCX(char *origname, byte **pic, byte **palette, int *width, int *height); +extern void GetPCXInfo(char *filename, int *width, int *height); + +extern qboolean LoadSTB(const char *origname, const char* type, byte **pic, int *width, int *height); +extern qboolean ResizeSTB(byte *input_pixels, int input_width, int input_height, + byte *output_pixels, int output_width, int output_height); +extern void SmoothColorImage(unsigned *dst, size_t size, size_t rstep); +extern void scale2x(byte *src, byte *dst, int width, int height); +extern void scale3x(byte *src, byte *dst, int width, int height); + +extern void GetWalInfo(char *name, int *width, int *height); +extern void GetM8Info(char *name, int *width, int *height); + +extern float Mod_RadiusFromBounds(const vec3_t mins, const vec3_t maxs); +extern byte* Mod_DecompressVis(byte *in, int row); +#endif /* SRC_CLIENT_REFRESH_REF_SHARED_H_ */ diff --git a/src/common/header/shared.h b/src/common/header/shared.h new file mode 100644 index 0000000..74b5088 --- /dev/null +++ b/src/common/header/shared.h @@ -0,0 +1,1189 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * This is the main header file shared between client, renderer, server + * and the game. Do NOT edit this file unless you know what you're + * doing. Changes here may break the client <-> renderer <-> server + * <-> game API, leading to problems with mods! + * + * ======================================================================= + */ + +#ifndef COMMON_SHARED_H +#define COMMON_SHARED_H + +#include +#include +#include +#include +#include +#include +#include + +#ifdef true + #undef true +#endif + +#ifdef false + #undef false +#endif + +typedef enum {false, true} qboolean; +typedef unsigned char byte; + +#ifndef NULL + #define NULL ((void *)0) +#endif + +// stuff to align variables/arrays and for noreturn +#if __STDC_VERSION__ >= 201112L // C11 or newer + #define YQ2_ALIGNAS_SIZE(SIZE) _Alignas(SIZE) + #define YQ2_ALIGNAS_TYPE(TYPE) _Alignas(TYPE) + // must be used as prefix (YQ2_ATTR_NORETURN void bla();)! + #define YQ2_ATTR_NORETURN _Noreturn +#elif defined(__GNUC__) // GCC and clang should support this attribute + #define YQ2_ALIGNAS_SIZE(SIZE) __attribute__(( __aligned__(SIZE) )) + #define YQ2_ALIGNAS_TYPE(TYPE) __attribute__(( __aligned__(__alignof__(TYPE)) )) + // must be used as prefix (YQ2_ATTR_NORETURN void bla();)! + #define YQ2_ATTR_NORETURN __attribute__ ((noreturn)) +#elif defined(_MSC_VER) + #define YQ2_ALIGNAS_SIZE(SIZE) __declspec( align(SIZE) ) + #define YQ2_ALIGNAS_TYPE(TYPE) __declspec( align( __alignof(TYPE) ) ) + // must be used as prefix (YQ2_ATTR_NORETURN void bla();)! + #define YQ2_ATTR_NORETURN __declspec(noreturn) +#else + #warning "Please add a case for your compiler here to align correctly" + #define YQ2_ALIGNAS_TYPE(TYPE) + #define YQ2_ATTR_NORETURN +#endif + +#if defined(__GNUC__) + /* ISO C11 _Noreturn can't be attached to function pointers, so + * use the gcc/clang-specific version for function pointers, even + * in C11 mode. MSVC __declspec(noreturn) seems to have the same + * restriction as _Noreturn so can't be used here either. */ + #define YQ2_ATTR_NORETURN_FUNCPTR __attribute__ ((noreturn)) +#else + #define YQ2_ATTR_NORETURN_FUNCPTR /* nothing */ +#endif + +/* angle indexes */ +#define PITCH 0 /* up / down */ +#define YAW 1 /* left / right */ +#define ROLL 2 /* fall over */ + +#ifndef min +#define min(a, b) (((a) < (b)) ? (a) : (b)) +#endif +#ifndef max +#define max(a, b) (((a) > (b)) ? (a) : (b)) +#endif + +#define MAX_STRING_CHARS 2048 /* max length of a string passed to Cmd_TokenizeString */ +#define MAX_STRING_TOKENS 80 /* max tokens resulting from Cmd_TokenizeString */ +#define MAX_TOKEN_CHARS 1024 /* max length of an individual token */ + +#define MAX_QPATH 64 /* max length of a quake game pathname */ + +/* + * DG: For some stupid reason, SV_WriteServerFile() and SV_ReadeServerFile() used + * MAX_OSPATH as buffer length for CVAR_LATCH CVARS and saved the whole buffer + * into $game/save/current/server.ssv, so changing MAX_OSPATH breaks savegames... + * Unfortunately, for some other fucking reason MAX_OSPATH was 128 for non-Windows + * which is just horrible.. so I introduced LATCH_CVAR_SAVELENGTH with the stupid + * values so I could bump MAX_OSPATH. + * TODO: whenever you break savegame compatibility next, make + * LATCH_CVAR_SAVELENGTH system-independent (or remove it and hardcode a + * sensible value in the two functions) + */ + +#ifdef _WIN32 + #define MAX_OSPATH 256 /* max length of a filesystem pathname (same as MAX_PATH) */ + #define LATCH_CVAR_SAVELENGTH 256 + + // by default dlls don't export any functions, use this to + // make a function visible (for GetGameAPI(), GetRefAPI() and similar) + #define Q2_DLL_EXPORTED __declspec(dllexport) + +#else // not Win32 (Linux, BSD, Mac, ..) + + #define MAX_OSPATH 4096 /* max length of a filesystem pathname */ + #define LATCH_CVAR_SAVELENGTH 128 + + // by default our .so/.dylibs don't export any functions, use this to + // make a function visible (for GetGameAPI(), GetRefAPI() and similar) + #define Q2_DLL_EXPORTED __attribute__((__visibility__("default"))) +#endif + +/* per-level limits */ +#define MAX_CLIENTS 256 /* absolute limit */ +#define MAX_EDICTS 1024 /* must change protocol to increase more */ +#define MAX_LIGHTSTYLES 256 +#define MAX_MODELS 256 /* these are sent over the net as bytes */ +#define MAX_SOUNDS 256 /* so they cannot be blindly increased */ +#define MAX_IMAGES 256 +#define MAX_ITEMS 256 +#define MAX_GENERAL (MAX_CLIENTS * 2) /* general config strings */ + +/* game print flags */ +#define PRINT_LOW 0 /* pickup messages */ +#define PRINT_MEDIUM 1 /* death messages */ +#define PRINT_HIGH 2 /* critical messages */ +#define PRINT_CHAT 3 /* chat messages */ + +#define ERR_FATAL 0 /* exit the entire game with a popup window */ +#define ERR_DROP 1 /* print to console and disconnect from game */ +#define ERR_DISCONNECT 2 /* don't kill server */ + +#define PRINT_ALL 0 +#define PRINT_DEVELOPER 1 /* only print when "developer 1" */ +#define PRINT_ALERT 2 + +/* destination class for gi.multicast() */ +typedef enum +{ + MULTICAST_ALL, + MULTICAST_PHS, + MULTICAST_PVS, + MULTICAST_ALL_R, + MULTICAST_PHS_R, + MULTICAST_PVS_R +} multicast_t; + +/* + * ============================================================== + * + * MATHLIB + * + * ============================================================== + */ + +typedef float vec_t; +typedef vec_t vec3_t[3]; +typedef vec_t vec5_t[5]; + +typedef int fixed4_t; +typedef int fixed8_t; +typedef int fixed16_t; + +#ifndef M_PI + #define M_PI 3.14159265358979323846 /* matches value in gcc v2 math.h */ +#endif + +struct cplane_s; + +extern vec3_t vec3_origin; + +#define nanmask (255 << 23) + +#define IS_NAN(x) (((*(int *)&x) & nanmask) == nanmask) + +// FIXME: use int instead of long, it's only used with int anyway? +#define Q_ftol(f) (long)(f) + +#define DotProduct(x, y) (x[0] * y[0] + x[1] * y[1] + x[2] * y[2]) +#define VectorSubtract(a, b, c) \ + (c[0] = a[0] - b[0], c[1] = a[1] - b[1], c[2] = \ + a[2] - b[2]) +#define VectorAdd(a, b, c) \ + (c[0] = a[0] + b[0], c[1] = a[1] + b[1], c[2] = \ + a[2] + b[2]) +#define VectorCopy(a, b) (b[0] = a[0], b[1] = a[1], b[2] = a[2]) +#define VectorClear(a) (a[0] = a[1] = a[2] = 0) +#define VectorNegate(a, b) (b[0] = -a[0], b[1] = -a[1], b[2] = -a[2]) +#define VectorSet(v, x, y, z) (v[0] = (x), v[1] = (y), v[2] = (z)) + +void VectorMA(vec3_t veca, float scale, vec3_t vecb, vec3_t vecc); + +/* just in case you do't want to use the macros */ +vec_t _DotProduct(vec3_t v1, vec3_t v2); +void _VectorSubtract(vec3_t veca, vec3_t vecb, vec3_t out); +void _VectorAdd(vec3_t veca, vec3_t vecb, vec3_t out); +void _VectorCopy(vec3_t in, vec3_t out); + +void ClearBounds(vec3_t mins, vec3_t maxs); +void AddPointToBounds(vec3_t v, vec3_t mins, vec3_t maxs); +int VectorCompare(vec3_t v1, vec3_t v2); +vec_t VectorLength(vec3_t v); +void CrossProduct(vec3_t v1, vec3_t v2, vec3_t cross); +vec_t VectorNormalize(vec3_t v); /* returns vector length */ +vec_t VectorNormalize2(vec3_t v, vec3_t out); +void VectorInverse(vec3_t v); +void VectorScale(vec3_t in, vec_t scale, vec3_t out); +int Q_log2(int val); + +void R_ConcatRotations(float in1[3][3], float in2[3][3], float out[3][3]); +void R_ConcatTransforms(float in1[3][4], float in2[3][4], float out[3][4]); + +void AngleVectors(vec3_t angles, vec3_t forward, vec3_t right, vec3_t up); +void AngleVectors2(vec3_t value1, vec3_t angles); +int BoxOnPlaneSide(vec3_t emins, vec3_t emaxs, struct cplane_s *plane); +float anglemod(float a); +float LerpAngle(float a1, float a2, float frac); + +#define BOX_ON_PLANE_SIDE(emins, emaxs, p) \ + (((p)->type < 3) ? \ + ( \ + ((p)->dist <= (emins)[(p)->type]) ? \ + 1 \ + : \ + ( \ + ((p)->dist >= (emaxs)[(p)->type]) ? \ + 2 \ + : \ + 3 \ + ) \ + ) \ + : \ + BoxOnPlaneSide((emins), (emaxs), (p))) + +void ProjectPointOnPlane(vec3_t dst, const vec3_t p, const vec3_t normal); +void PerpendicularVector(vec3_t dst, const vec3_t src); +void RotatePointAroundVector(vec3_t dst, + const vec3_t dir, + const vec3_t point, + float degrees); + +/* ============================================= */ + +char *COM_SkipPath(char *pathname); +void COM_StripExtension(char *in, char *out); +const char *COM_FileExtension(const char *in); +void COM_FileBase(char *in, char *out); +void COM_FilePath(const char *in, char *out); +void COM_DefaultExtension(char *path, const char *extension); + +char *COM_Parse(char **data_p); + +/* data is an in/out parm, returns a parsed out token */ +void Com_sprintf(char *dest, int size, char *fmt, ...); + +void Com_PageInMemory(byte *buffer, int size); + +/* ============================================= */ + +/* portable case insensitive compare */ +int Q_stricmp(const char *s1, const char *s2); +int Q_strcasecmp(char *s1, char *s2); +int Q_strncasecmp(char *s1, char *s2, int n); + +/* portable string lowercase */ +char *Q_strlwr(char *s); + +/* portable safe string copy/concatenate */ +int Q_strlcpy(char *dst, const char *src, int size); +int Q_strlcat(char *dst, const char *src, int size); + +/* ============================================= */ + +/* Unicode wrappers that also make sure it's a regular file around fopen(). */ +FILE *Q_fopen(const char *file, const char *mode); + +/* Comparator function for qsort(), compares strings. */ +int Q_sort_strcomp(const void *s1, const void *s2); + +/* ============================================= */ + +short BigShort(short l); +short LittleShort(short l); +int BigLong(int l); +int LittleLong(int l); +float BigFloat(float l); +float LittleFloat(float l); + +void Swap_Init(void); +char *va(char *format, ...) __attribute__ ((format (printf, 1, 2))); + +/* ============================================= */ + +/* key / value info strings */ +#define MAX_INFO_KEY 64 +#define MAX_INFO_VALUE 64 +#define MAX_INFO_STRING 512 + +char *Info_ValueForKey(char *s, char *key); +void Info_RemoveKey(char *s, char *key); +void Info_SetValueForKey(char *s, char *key, char *value); +qboolean Info_Validate(char *s); + +/* ============================================= */ + +/* Random number generator */ +int randk(void); +float frandk(void); +float crandk(void); +void randk_seed(void); + +/* + * ============================================================== + * + * SYSTEM SPECIFIC + * + * ============================================================== + */ + +extern int curtime; /* time returned by last Sys_Milliseconds */ + +int Sys_Milliseconds(void); +void Sys_Mkdir(const char *path); +qboolean Sys_IsDir(const char *path); +qboolean Sys_IsFile(const char *path); + +/* large block stack allocation routines */ +void *Hunk_Begin(int maxsize); +void *Hunk_Alloc(int size); +void Hunk_Free(void *buf); +int Hunk_End(void); + +/* directory searching */ +#define SFF_ARCH 0x01 +#define SFF_HIDDEN 0x02 +#define SFF_RDONLY 0x04 +#define SFF_SUBDIR 0x08 +#define SFF_SYSTEM 0x10 + +/* pass in an attribute mask of things you wish to REJECT */ +char *Sys_FindFirst(char *path, unsigned musthave, unsigned canthave); +char *Sys_FindNext(unsigned musthave, unsigned canthave); +void Sys_FindClose(void); + +/* this is only here so the functions in shared source files can link */ +YQ2_ATTR_NORETURN void Sys_Error(char *error, ...); +void Com_Printf(char *msg, ...); + +/* + * ========================================================== + * + * CVARS (console variables) + * + * ========================================================== + */ + +#ifndef CVAR + #define CVAR + + #define CVAR_ARCHIVE 1 /* set to cause it to be saved to vars.rc */ + #define CVAR_USERINFO 2 /* added to userinfo when changed */ + #define CVAR_SERVERINFO 4 /* added to serverinfo when changed */ + #define CVAR_NOSET 8 /* don't allow change from console at all, */ + /* but can be set from the command line */ + #define CVAR_LATCH 16 /* save changes until server restart */ + +/* nothing outside the Cvar_*() functions should modify these fields! */ +typedef struct cvar_s +{ + char *name; + char *string; + char *latched_string; /* for CVAR_LATCH vars */ + int flags; + qboolean modified; /* set each time the cvar is changed */ + float value; + struct cvar_s *next; + + /* Added by YQ2. Must be at the end to preserve ABI. */ + char *default_string; +} cvar_t; + +#endif /* CVAR */ + +/* + * ============================================================== + * + * COLLISION DETECTION + * + * ============================================================== + */ + +/* lower bits are stronger, and will eat weaker brushes completely */ +#define CONTENTS_SOLID 1 /* an eye is never valid in a solid */ +#define CONTENTS_WINDOW 2 /* translucent, but not watery */ +#define CONTENTS_AUX 4 +#define CONTENTS_LAVA 8 +#define CONTENTS_SLIME 16 +#define CONTENTS_WATER 32 +#define CONTENTS_MIST 64 +#define LAST_VISIBLE_CONTENTS 64 + +/* remaining contents are non-visible, and don't eat brushes */ +#define CONTENTS_AREAPORTAL 0x8000 + +#define CONTENTS_PLAYERCLIP 0x10000 +#define CONTENTS_MONSTERCLIP 0x20000 + +/* currents can be added to any other contents, and may be mixed */ +#define CONTENTS_CURRENT_0 0x40000 +#define CONTENTS_CURRENT_90 0x80000 +#define CONTENTS_CURRENT_180 0x100000 +#define CONTENTS_CURRENT_270 0x200000 +#define CONTENTS_CURRENT_UP 0x400000 +#define CONTENTS_CURRENT_DOWN 0x800000 + +#define CONTENTS_ORIGIN 0x1000000 /* removed before bsping an entity */ + +#define CONTENTS_MONSTER 0x2000000 /* should never be on a brush, only in game */ +#define CONTENTS_DEADMONSTER 0x4000000 +#define CONTENTS_DETAIL 0x8000000 /* brushes to be added after vis leafs */ +#define CONTENTS_TRANSLUCENT 0x10000000 /* auto set if any surface has trans */ +#define CONTENTS_LADDER 0x20000000 + +#define SURF_LIGHT 0x1 /* value will hold the light strength */ + +#define SURF_SLICK 0x2 /* effects game physics */ + +#define SURF_SKY 0x4 /* don't draw, but add to skybox */ +#define SURF_WARP 0x8 /* turbulent water warp */ +#define SURF_TRANS33 0x10 +#define SURF_TRANS66 0x20 +#define SURF_FLOWING 0x40 /* scroll towards angle */ +#define SURF_NODRAW 0x80 /* don't bother referencing the texture */ + +/* content masks */ +#define MASK_ALL (-1) +#define MASK_SOLID (CONTENTS_SOLID | CONTENTS_WINDOW) +#define MASK_PLAYERSOLID \ + (CONTENTS_SOLID | CONTENTS_PLAYERCLIP | \ + CONTENTS_WINDOW | CONTENTS_MONSTER) +#define MASK_DEADSOLID (CONTENTS_SOLID | CONTENTS_PLAYERCLIP | CONTENTS_WINDOW) +#define MASK_MONSTERSOLID \ + (CONTENTS_SOLID | CONTENTS_MONSTERCLIP | \ + CONTENTS_WINDOW | CONTENTS_MONSTER) +#define MASK_WATER (CONTENTS_WATER | CONTENTS_LAVA | CONTENTS_SLIME) +#define MASK_OPAQUE (CONTENTS_SOLID | CONTENTS_SLIME | CONTENTS_LAVA) +#define MASK_SHOT \ + (CONTENTS_SOLID | CONTENTS_MONSTER | CONTENTS_WINDOW | \ + CONTENTS_DEADMONSTER) +#define MASK_CURRENT \ + (CONTENTS_CURRENT_0 | CONTENTS_CURRENT_90 | \ + CONTENTS_CURRENT_180 | CONTENTS_CURRENT_270 | \ + CONTENTS_CURRENT_UP | \ + CONTENTS_CURRENT_DOWN) + +/* gi.BoxEdicts() can return a list of either solid or trigger entities */ +#define AREA_SOLID 1 +#define AREA_TRIGGERS 2 + +/* plane_t structure */ +typedef struct cplane_s +{ + vec3_t normal; + float dist; + byte type; /* for fast side tests */ + byte signbits; /* signx + (signy<<1) + (signz<<2) */ + byte pad[2]; +} cplane_t; + +/* structure offset for asm code */ +#define CPLANE_NORMAL_X 0 +#define CPLANE_NORMAL_Y 4 +#define CPLANE_NORMAL_Z 8 +#define CPLANE_DIST 12 +#define CPLANE_TYPE 16 +#define CPLANE_SIGNBITS 17 +#define CPLANE_PAD0 18 +#define CPLANE_PAD1 19 + +typedef struct cmodel_s +{ + vec3_t mins, maxs; + vec3_t origin; /* for sounds or lights */ + int headnode; +} cmodel_t; + +typedef struct csurface_s +{ + char name[16]; + int flags; /* SURF_* */ + int value; /* unused */ +} csurface_t; + +typedef struct mapsurface_s /* used internally due to name len probs */ +{ + csurface_t c; + char rname[32]; +} mapsurface_t; + +/* a trace is returned when a box is swept through the world */ +typedef struct +{ + qboolean allsolid; /* if true, plane is not valid */ + qboolean startsolid; /* if true, the initial point was in a solid area */ + float fraction; /* time completed, 1.0 = didn't hit anything */ + vec3_t endpos; /* final position */ + cplane_t plane; /* surface normal at impact */ + csurface_t *surface; /* surface hit */ + int contents; /* contents on other side of surface hit */ + struct edict_s *ent; /* not set by CM_*() functions */ +} trace_t; + +/* pmove_state_t is the information necessary for client side movement */ +/* prediction */ +typedef enum +{ + /* can accelerate and turn */ + PM_NORMAL, + PM_SPECTATOR, + /* no acceleration or turning */ + PM_DEAD, + PM_GIB, /* different bounding box */ + PM_FREEZE +} pmtype_t; + +/* pmove->pm_flags */ +#define PMF_DUCKED 1 +#define PMF_JUMP_HELD 2 +#define PMF_ON_GROUND 4 +#define PMF_TIME_WATERJUMP 8 /* pm_time is waterjump */ +#define PMF_TIME_LAND 16 /* pm_time is time before rejump */ +#define PMF_TIME_TELEPORT 32 /* pm_time is non-moving time */ +#define PMF_NO_PREDICTION 64 /* temporarily disables prediction (used for grappling hook) */ + +/* this structure needs to be communicated bit-accurate/ + * from the server to the client to guarantee that + * prediction stays in sync, so no floats are used. + * if any part of the game code modifies this struct, it + * will result in a prediction error of some degree. */ +typedef struct +{ + pmtype_t pm_type; + + short origin[3]; /* 12.3 */ + short velocity[3]; /* 12.3 */ + byte pm_flags; /* ducked, jump_held, etc */ + byte pm_time; /* each unit = 8 ms */ + short gravity; + short delta_angles[3]; /* add to command angles to get view direction + * changed by spawns, rotating objects, and teleporters */ +} pmove_state_t; + +/* button bits */ +#define BUTTON_ATTACK 1 +#define BUTTON_USE 2 +#define BUTTON_ANY 128 /* any key whatsoever */ + +/* usercmd_t is sent to the server each client frame */ +typedef struct usercmd_s +{ + byte msec; + byte buttons; + short angles[3]; + short forwardmove, sidemove, upmove; + byte impulse; /* remove? */ + byte lightlevel; /* light level the player is standing on */ +} usercmd_t; + +#define MAXTOUCH 32 +typedef struct +{ + /* state (in / out) */ + pmove_state_t s; + + /* command (in) */ + usercmd_t cmd; + qboolean snapinitial; /* if s has been changed outside pmove */ + + /* results (out) */ + int numtouch; + struct edict_s *touchents[MAXTOUCH]; + + vec3_t viewangles; /* clamped */ + float viewheight; + + vec3_t mins, maxs; /* bounding box size */ + + struct edict_s *groundentity; + int watertype; + int waterlevel; + + /* callbacks to test the world */ + trace_t (*trace)(vec3_t start, vec3_t mins, vec3_t maxs, vec3_t end); + int (*pointcontents)(vec3_t point); +} pmove_t; + +/* entity_state_t->effects + * Effects are things handled on the client side (lights, particles, + * frame animations) that happen constantly on the given entity. + * An entity that has effects will be sent to the client even if + * it has a zero index model. */ +#define EF_ROTATE 0x00000001 /* rotate (bonus items) */ +#define EF_GIB 0x00000002 /* leave a trail */ +#define EF_BLASTER 0x00000008 /* redlight + trail */ +#define EF_ROCKET 0x00000010 /* redlight + trail */ +#define EF_GRENADE 0x00000020 +#define EF_HYPERBLASTER 0x00000040 +#define EF_BFG 0x00000080 +#define EF_COLOR_SHELL 0x00000100 +#define EF_POWERSCREEN 0x00000200 +#define EF_ANIM01 0x00000400 /* automatically cycle between frames 0 and 1 at 2 hz */ +#define EF_ANIM23 0x00000800 /* automatically cycle between frames 2 and 3 at 2 hz */ +#define EF_ANIM_ALL 0x00001000 /* automatically cycle through all frames at 2hz */ +#define EF_ANIM_ALLFAST 0x00002000 /* automatically cycle through all frames at 10hz */ +#define EF_FLIES 0x00004000 +#define EF_QUAD 0x00008000 +#define EF_PENT 0x00010000 +#define EF_TELEPORTER 0x00020000 /* particle fountain */ +#define EF_FLAG1 0x00040000 +#define EF_FLAG2 0x00080000 +#define EF_IONRIPPER 0x00100000 +#define EF_GREENGIB 0x00200000 +#define EF_BLUEHYPERBLASTER 0x00400000 +#define EF_SPINNINGLIGHTS 0x00800000 +#define EF_PLASMA 0x01000000 +#define EF_TRAP 0x02000000 +#define EF_TRACKER 0x04000000 +#define EF_DOUBLE 0x08000000 +#define EF_SPHERETRANS 0x10000000 +#define EF_TAGTRAIL 0x20000000 +#define EF_HALF_DAMAGE 0x40000000 +#define EF_TRACKERTRAIL 0x80000000 + +/* entity_state_t->renderfx flags */ +#define RF_MINLIGHT 1 /* allways have some light (viewmodel) */ +#define RF_VIEWERMODEL 2 /* don't draw through eyes, only mirrors */ +#define RF_WEAPONMODEL 4 /* only draw through eyes */ +#define RF_FULLBRIGHT 8 /* allways draw full intensity */ +#define RF_DEPTHHACK 16 /* for view weapon Z crunching */ +#define RF_TRANSLUCENT 32 +#define RF_FRAMELERP 64 +#define RF_BEAM 128 +#define RF_CUSTOMSKIN 256 /* skin is an index in image_precache */ +#define RF_GLOW 512 /* pulse lighting for bonus items */ +#define RF_SHELL_RED 1024 +#define RF_SHELL_GREEN 2048 +#define RF_SHELL_BLUE 4096 +#define RF_NOSHADOW 8192 /* don't draw a shadow */ +#define RF_IR_VISIBLE 0x00008000 /* 32768 */ +#define RF_SHELL_DOUBLE 0x00010000 /* 65536 */ +#define RF_SHELL_HALF_DAM 0x00020000 +#define RF_USE_DISGUISE 0x00040000 + +/* player_state_t->refdef flags */ +#define RDF_UNDERWATER 1 /* warp the screen as apropriate */ +#define RDF_NOWORLDMODEL 2 /* used for player configuration screen */ +#define RDF_IRGOGGLES 4 +#define RDF_UVGOGGLES 8 + +/* muzzle flashes / player effects */ +#define MZ_BLASTER 0 +#define MZ_MACHINEGUN 1 +#define MZ_SHOTGUN 2 +#define MZ_CHAINGUN1 3 +#define MZ_CHAINGUN2 4 +#define MZ_CHAINGUN3 5 +#define MZ_RAILGUN 6 +#define MZ_ROCKET 7 +#define MZ_GRENADE 8 +#define MZ_LOGIN 9 +#define MZ_LOGOUT 10 +#define MZ_RESPAWN 11 +#define MZ_BFG 12 +#define MZ_SSHOTGUN 13 +#define MZ_HYPERBLASTER 14 +#define MZ_ITEMRESPAWN 15 +#define MZ_IONRIPPER 16 +#define MZ_BLUEHYPERBLASTER 17 +#define MZ_PHALANX 18 +#define MZ_SILENCED 128 /* bit flag ORed with one of the above numbers */ +#define MZ_ETF_RIFLE 30 +#define MZ_UNUSED 31 +#define MZ_SHOTGUN2 32 +#define MZ_HEATBEAM 33 +#define MZ_BLASTER2 34 +#define MZ_TRACKER 35 +#define MZ_NUKE1 36 +#define MZ_NUKE2 37 +#define MZ_NUKE4 38 +#define MZ_NUKE8 39 + +/* monster muzzle flashes */ +#define MZ2_TANK_BLASTER_1 1 +#define MZ2_TANK_BLASTER_2 2 +#define MZ2_TANK_BLASTER_3 3 +#define MZ2_TANK_MACHINEGUN_1 4 +#define MZ2_TANK_MACHINEGUN_2 5 +#define MZ2_TANK_MACHINEGUN_3 6 +#define MZ2_TANK_MACHINEGUN_4 7 +#define MZ2_TANK_MACHINEGUN_5 8 +#define MZ2_TANK_MACHINEGUN_6 9 +#define MZ2_TANK_MACHINEGUN_7 10 +#define MZ2_TANK_MACHINEGUN_8 11 +#define MZ2_TANK_MACHINEGUN_9 12 +#define MZ2_TANK_MACHINEGUN_10 13 +#define MZ2_TANK_MACHINEGUN_11 14 +#define MZ2_TANK_MACHINEGUN_12 15 +#define MZ2_TANK_MACHINEGUN_13 16 +#define MZ2_TANK_MACHINEGUN_14 17 +#define MZ2_TANK_MACHINEGUN_15 18 +#define MZ2_TANK_MACHINEGUN_16 19 +#define MZ2_TANK_MACHINEGUN_17 20 +#define MZ2_TANK_MACHINEGUN_18 21 +#define MZ2_TANK_MACHINEGUN_19 22 +#define MZ2_TANK_ROCKET_1 23 +#define MZ2_TANK_ROCKET_2 24 +#define MZ2_TANK_ROCKET_3 25 + +#define MZ2_INFANTRY_MACHINEGUN_1 26 +#define MZ2_INFANTRY_MACHINEGUN_2 27 +#define MZ2_INFANTRY_MACHINEGUN_3 28 +#define MZ2_INFANTRY_MACHINEGUN_4 29 +#define MZ2_INFANTRY_MACHINEGUN_5 30 +#define MZ2_INFANTRY_MACHINEGUN_6 31 +#define MZ2_INFANTRY_MACHINEGUN_7 32 +#define MZ2_INFANTRY_MACHINEGUN_8 33 +#define MZ2_INFANTRY_MACHINEGUN_9 34 +#define MZ2_INFANTRY_MACHINEGUN_10 35 +#define MZ2_INFANTRY_MACHINEGUN_11 36 +#define MZ2_INFANTRY_MACHINEGUN_12 37 +#define MZ2_INFANTRY_MACHINEGUN_13 38 + +#define MZ2_SOLDIER_BLASTER_1 39 +#define MZ2_SOLDIER_BLASTER_2 40 +#define MZ2_SOLDIER_SHOTGUN_1 41 +#define MZ2_SOLDIER_SHOTGUN_2 42 +#define MZ2_SOLDIER_MACHINEGUN_1 43 +#define MZ2_SOLDIER_MACHINEGUN_2 44 + +#define MZ2_GUNNER_MACHINEGUN_1 45 +#define MZ2_GUNNER_MACHINEGUN_2 46 +#define MZ2_GUNNER_MACHINEGUN_3 47 +#define MZ2_GUNNER_MACHINEGUN_4 48 +#define MZ2_GUNNER_MACHINEGUN_5 49 +#define MZ2_GUNNER_MACHINEGUN_6 50 +#define MZ2_GUNNER_MACHINEGUN_7 51 +#define MZ2_GUNNER_MACHINEGUN_8 52 +#define MZ2_GUNNER_GRENADE_1 53 +#define MZ2_GUNNER_GRENADE_2 54 +#define MZ2_GUNNER_GRENADE_3 55 +#define MZ2_GUNNER_GRENADE_4 56 + +#define MZ2_CHICK_ROCKET_1 57 + +#define MZ2_FLYER_BLASTER_1 58 +#define MZ2_FLYER_BLASTER_2 59 + +#define MZ2_MEDIC_BLASTER_1 60 + +#define MZ2_GLADIATOR_RAILGUN_1 61 + +#define MZ2_HOVER_BLASTER_1 62 + +#define MZ2_ACTOR_MACHINEGUN_1 63 + +#define MZ2_SUPERTANK_MACHINEGUN_1 64 +#define MZ2_SUPERTANK_MACHINEGUN_2 65 +#define MZ2_SUPERTANK_MACHINEGUN_3 66 +#define MZ2_SUPERTANK_MACHINEGUN_4 67 +#define MZ2_SUPERTANK_MACHINEGUN_5 68 +#define MZ2_SUPERTANK_MACHINEGUN_6 69 +#define MZ2_SUPERTANK_ROCKET_1 70 +#define MZ2_SUPERTANK_ROCKET_2 71 +#define MZ2_SUPERTANK_ROCKET_3 72 + +#define MZ2_BOSS2_MACHINEGUN_L1 73 +#define MZ2_BOSS2_MACHINEGUN_L2 74 +#define MZ2_BOSS2_MACHINEGUN_L3 75 +#define MZ2_BOSS2_MACHINEGUN_L4 76 +#define MZ2_BOSS2_MACHINEGUN_L5 77 +#define MZ2_BOSS2_ROCKET_1 78 +#define MZ2_BOSS2_ROCKET_2 79 +#define MZ2_BOSS2_ROCKET_3 80 +#define MZ2_BOSS2_ROCKET_4 81 + +#define MZ2_FLOAT_BLASTER_1 82 + +#define MZ2_SOLDIER_BLASTER_3 83 +#define MZ2_SOLDIER_SHOTGUN_3 84 +#define MZ2_SOLDIER_MACHINEGUN_3 85 +#define MZ2_SOLDIER_BLASTER_4 86 +#define MZ2_SOLDIER_SHOTGUN_4 87 +#define MZ2_SOLDIER_MACHINEGUN_4 88 +#define MZ2_SOLDIER_BLASTER_5 89 +#define MZ2_SOLDIER_SHOTGUN_5 90 +#define MZ2_SOLDIER_MACHINEGUN_5 91 +#define MZ2_SOLDIER_BLASTER_6 92 +#define MZ2_SOLDIER_SHOTGUN_6 93 +#define MZ2_SOLDIER_MACHINEGUN_6 94 +#define MZ2_SOLDIER_BLASTER_7 95 +#define MZ2_SOLDIER_SHOTGUN_7 96 +#define MZ2_SOLDIER_MACHINEGUN_7 97 +#define MZ2_SOLDIER_BLASTER_8 98 +#define MZ2_SOLDIER_SHOTGUN_8 99 +#define MZ2_SOLDIER_MACHINEGUN_8 100 + +#define MZ2_MAKRON_BFG 101 +#define MZ2_MAKRON_BLASTER_1 102 +#define MZ2_MAKRON_BLASTER_2 103 +#define MZ2_MAKRON_BLASTER_3 104 +#define MZ2_MAKRON_BLASTER_4 105 +#define MZ2_MAKRON_BLASTER_5 106 +#define MZ2_MAKRON_BLASTER_6 107 +#define MZ2_MAKRON_BLASTER_7 108 +#define MZ2_MAKRON_BLASTER_8 109 +#define MZ2_MAKRON_BLASTER_9 110 +#define MZ2_MAKRON_BLASTER_10 111 +#define MZ2_MAKRON_BLASTER_11 112 +#define MZ2_MAKRON_BLASTER_12 113 +#define MZ2_MAKRON_BLASTER_13 114 +#define MZ2_MAKRON_BLASTER_14 115 +#define MZ2_MAKRON_BLASTER_15 116 +#define MZ2_MAKRON_BLASTER_16 117 +#define MZ2_MAKRON_BLASTER_17 118 +#define MZ2_MAKRON_RAILGUN_1 119 +#define MZ2_JORG_MACHINEGUN_L1 120 +#define MZ2_JORG_MACHINEGUN_L2 121 +#define MZ2_JORG_MACHINEGUN_L3 122 +#define MZ2_JORG_MACHINEGUN_L4 123 +#define MZ2_JORG_MACHINEGUN_L5 124 +#define MZ2_JORG_MACHINEGUN_L6 125 +#define MZ2_JORG_MACHINEGUN_R1 126 +#define MZ2_JORG_MACHINEGUN_R2 127 +#define MZ2_JORG_MACHINEGUN_R3 128 +#define MZ2_JORG_MACHINEGUN_R4 129 +#define MZ2_JORG_MACHINEGUN_R5 130 +#define MZ2_JORG_MACHINEGUN_R6 131 +#define MZ2_JORG_BFG_1 132 +#define MZ2_BOSS2_MACHINEGUN_R1 133 +#define MZ2_BOSS2_MACHINEGUN_R2 134 +#define MZ2_BOSS2_MACHINEGUN_R3 135 +#define MZ2_BOSS2_MACHINEGUN_R4 136 +#define MZ2_BOSS2_MACHINEGUN_R5 137 + +#define MZ2_CARRIER_MACHINEGUN_L1 138 +#define MZ2_CARRIER_MACHINEGUN_R1 139 +#define MZ2_CARRIER_GRENADE 140 +#define MZ2_TURRET_MACHINEGUN 141 +#define MZ2_TURRET_ROCKET 142 +#define MZ2_TURRET_BLASTER 143 +#define MZ2_STALKER_BLASTER 144 +#define MZ2_DAEDALUS_BLASTER 145 +#define MZ2_MEDIC_BLASTER_2 146 +#define MZ2_CARRIER_RAILGUN 147 +#define MZ2_WIDOW_DISRUPTOR 148 +#define MZ2_WIDOW_BLASTER 149 +#define MZ2_WIDOW_RAIL 150 +#define MZ2_WIDOW_PLASMABEAM 151 +#define MZ2_CARRIER_MACHINEGUN_L2 152 +#define MZ2_CARRIER_MACHINEGUN_R2 153 +#define MZ2_WIDOW_RAIL_LEFT 154 +#define MZ2_WIDOW_RAIL_RIGHT 155 +#define MZ2_WIDOW_BLASTER_SWEEP1 156 +#define MZ2_WIDOW_BLASTER_SWEEP2 157 +#define MZ2_WIDOW_BLASTER_SWEEP3 158 +#define MZ2_WIDOW_BLASTER_SWEEP4 159 +#define MZ2_WIDOW_BLASTER_SWEEP5 160 +#define MZ2_WIDOW_BLASTER_SWEEP6 161 +#define MZ2_WIDOW_BLASTER_SWEEP7 162 +#define MZ2_WIDOW_BLASTER_SWEEP8 163 +#define MZ2_WIDOW_BLASTER_SWEEP9 164 +#define MZ2_WIDOW_BLASTER_100 165 +#define MZ2_WIDOW_BLASTER_90 166 +#define MZ2_WIDOW_BLASTER_80 167 +#define MZ2_WIDOW_BLASTER_70 168 +#define MZ2_WIDOW_BLASTER_60 169 +#define MZ2_WIDOW_BLASTER_50 170 +#define MZ2_WIDOW_BLASTER_40 171 +#define MZ2_WIDOW_BLASTER_30 172 +#define MZ2_WIDOW_BLASTER_20 173 +#define MZ2_WIDOW_BLASTER_10 174 +#define MZ2_WIDOW_BLASTER_0 175 +#define MZ2_WIDOW_BLASTER_10L 176 +#define MZ2_WIDOW_BLASTER_20L 177 +#define MZ2_WIDOW_BLASTER_30L 178 +#define MZ2_WIDOW_BLASTER_40L 179 +#define MZ2_WIDOW_BLASTER_50L 180 +#define MZ2_WIDOW_BLASTER_60L 181 +#define MZ2_WIDOW_BLASTER_70L 182 +#define MZ2_WIDOW_RUN_1 183 +#define MZ2_WIDOW_RUN_2 184 +#define MZ2_WIDOW_RUN_3 185 +#define MZ2_WIDOW_RUN_4 186 +#define MZ2_WIDOW_RUN_5 187 +#define MZ2_WIDOW_RUN_6 188 +#define MZ2_WIDOW_RUN_7 189 +#define MZ2_WIDOW_RUN_8 190 +#define MZ2_CARRIER_ROCKET_1 191 +#define MZ2_CARRIER_ROCKET_2 192 +#define MZ2_CARRIER_ROCKET_3 193 +#define MZ2_CARRIER_ROCKET_4 194 +#define MZ2_WIDOW2_BEAMER_1 195 +#define MZ2_WIDOW2_BEAMER_2 196 +#define MZ2_WIDOW2_BEAMER_3 197 +#define MZ2_WIDOW2_BEAMER_4 198 +#define MZ2_WIDOW2_BEAMER_5 199 +#define MZ2_WIDOW2_BEAM_SWEEP_1 200 +#define MZ2_WIDOW2_BEAM_SWEEP_2 201 +#define MZ2_WIDOW2_BEAM_SWEEP_3 202 +#define MZ2_WIDOW2_BEAM_SWEEP_4 203 +#define MZ2_WIDOW2_BEAM_SWEEP_5 204 +#define MZ2_WIDOW2_BEAM_SWEEP_6 205 +#define MZ2_WIDOW2_BEAM_SWEEP_7 206 +#define MZ2_WIDOW2_BEAM_SWEEP_8 207 +#define MZ2_WIDOW2_BEAM_SWEEP_9 208 +#define MZ2_WIDOW2_BEAM_SWEEP_10 209 +#define MZ2_WIDOW2_BEAM_SWEEP_11 210 + +extern vec3_t monster_flash_offset[]; + +/* Temp entity events are for things that happen + * at a location seperate from any existing entity. + * Temporary entity messages are explicitly constructed + * and broadcast. */ +typedef enum +{ + TE_GUNSHOT, + TE_BLOOD, + TE_BLASTER, + TE_RAILTRAIL, + TE_SHOTGUN, + TE_EXPLOSION1, + TE_EXPLOSION2, + TE_ROCKET_EXPLOSION, + TE_GRENADE_EXPLOSION, + TE_SPARKS, + TE_SPLASH, + TE_BUBBLETRAIL, + TE_SCREEN_SPARKS, + TE_SHIELD_SPARKS, + TE_BULLET_SPARKS, + TE_LASER_SPARKS, + TE_PARASITE_ATTACK, + TE_ROCKET_EXPLOSION_WATER, + TE_GRENADE_EXPLOSION_WATER, + TE_MEDIC_CABLE_ATTACK, + TE_BFG_EXPLOSION, + TE_BFG_BIGEXPLOSION, + TE_BOSSTPORT, /* used as '22' in a map, so DON'T RENUMBER!!! */ + TE_BFG_LASER, + TE_GRAPPLE_CABLE, + TE_WELDING_SPARKS, + TE_GREENBLOOD, + TE_BLUEHYPERBLASTER, + TE_PLASMA_EXPLOSION, + TE_TUNNEL_SPARKS, + TE_BLASTER2, + TE_RAILTRAIL2, + TE_FLAME, + TE_LIGHTNING, + TE_DEBUGTRAIL, + TE_PLAIN_EXPLOSION, + TE_FLASHLIGHT, + TE_FORCEWALL, + TE_HEATBEAM, + TE_MONSTER_HEATBEAM, + TE_STEAM, + TE_BUBBLETRAIL2, + TE_MOREBLOOD, + TE_HEATBEAM_SPARKS, + TE_HEATBEAM_STEAM, + TE_CHAINFIST_SMOKE, + TE_ELECTRIC_SPARKS, + TE_TRACKER_EXPLOSION, + TE_TELEPORT_EFFECT, + TE_DBALL_GOAL, + TE_WIDOWBEAMOUT, + TE_NUKEBLAST, + TE_WIDOWSPLASH, + TE_EXPLOSION1_BIG, + TE_EXPLOSION1_NP, + TE_FLECHETTE +} temp_event_t; + +#define SPLASH_UNKNOWN 0 +#define SPLASH_SPARKS 1 +#define SPLASH_BLUE_WATER 2 +#define SPLASH_BROWN_WATER 3 +#define SPLASH_SLIME 4 +#define SPLASH_LAVA 5 +#define SPLASH_BLOOD 6 + +/* sound channels: + * channel 0 never willingly overrides + * other channels (1-7) allways override + * a playing sound on that channel */ +#define CHAN_AUTO 0 +#define CHAN_WEAPON 1 +#define CHAN_VOICE 2 +#define CHAN_ITEM 3 +#define CHAN_BODY 4 +/* modifier flags */ +#define CHAN_NO_PHS_ADD 8 /* send to all clients, not just ones in PHS (ATTN 0 will also do this) */ +#define CHAN_RELIABLE 16 /* send by reliable message, not datagram */ + +/* sound attenuation values */ +#define ATTN_NONE 0 /* full volume the entire level */ +#define ATTN_NORM 1 +#define ATTN_IDLE 2 +#define ATTN_STATIC 3 /* diminish very rapidly with distance */ + +/* player_state->stats[] indexes */ +#define STAT_HEALTH_ICON 0 +#define STAT_HEALTH 1 +#define STAT_AMMO_ICON 2 +#define STAT_AMMO 3 +#define STAT_ARMOR_ICON 4 +#define STAT_ARMOR 5 +#define STAT_SELECTED_ICON 6 +#define STAT_PICKUP_ICON 7 +#define STAT_PICKUP_STRING 8 +#define STAT_TIMER_ICON 9 +#define STAT_TIMER 10 +#define STAT_HELPICON 11 +#define STAT_SELECTED_ITEM 12 +#define STAT_LAYOUTS 13 +#define STAT_FRAGS 14 +#define STAT_FLASHES 15 /* cleared each frame, 1 = health, 2 = armor */ +#define STAT_CHASE 16 +#define STAT_SPECTATOR 17 + +#define MAX_STATS 32 + +/* dmflags->value flags */ +#define DF_NO_HEALTH 0x00000001 /* 1 */ +#define DF_NO_ITEMS 0x00000002 /* 2 */ +#define DF_WEAPONS_STAY 0x00000004 /* 4 */ +#define DF_NO_FALLING 0x00000008 /* 8 */ +#define DF_INSTANT_ITEMS 0x00000010 /* 16 */ +#define DF_SAME_LEVEL 0x00000020 /* 32 */ +#define DF_SKINTEAMS 0x00000040 /* 64 */ +#define DF_MODELTEAMS 0x00000080 /* 128 */ +#define DF_NO_FRIENDLY_FIRE 0x00000100 /* 256 */ +#define DF_SPAWN_FARTHEST 0x00000200 /* 512 */ +#define DF_FORCE_RESPAWN 0x00000400 /* 1024 */ +#define DF_NO_ARMOR 0x00000800 /* 2048 */ +#define DF_ALLOW_EXIT 0x00001000 /* 4096 */ +#define DF_INFINITE_AMMO 0x00002000 /* 8192 */ +#define DF_QUAD_DROP 0x00004000 /* 16384 */ +#define DF_FIXED_FOV 0x00008000 /* 32768 */ +#define DF_QUADFIRE_DROP 0x00010000 /* 65536 */ +#define DF_NO_MINES 0x00020000 +#define DF_NO_STACK_DOUBLE 0x00040000 +#define DF_NO_NUKES 0x00080000 +#define DF_NO_SPHERES 0x00100000 + +#define ROGUE_VERSION_STRING "08/21/1998 Beta 2 for Ensemble" + +/* + * ========================================================== + * + * ELEMENTS COMMUNICATED ACROSS THE NET + * + * ========================================================== + */ + +#define ANGLE2SHORT(x) ((int)((x) * 65536 / 360) & 65535) +#define SHORT2ANGLE(x) ((x) * (360.0 / 65536)) + +/* config strings are a general means of communication from + * the server to all connected clients. Each config string + * can be at most MAX_QPATH characters. */ +#define CS_NAME 0 +#define CS_CDTRACK 1 +#define CS_SKY 2 +#define CS_SKYAXIS 3 /* %f %f %f format */ +#define CS_SKYROTATE 4 +#define CS_STATUSBAR 5 /* display program string */ + +#define CS_AIRACCEL 29 /* air acceleration control */ +#define CS_MAXCLIENTS 30 +#define CS_MAPCHECKSUM 31 /* for catching cheater maps */ + +#define CS_MODELS 32 +#define CS_SOUNDS (CS_MODELS + MAX_MODELS) +#define CS_IMAGES (CS_SOUNDS + MAX_SOUNDS) +#define CS_LIGHTS (CS_IMAGES + MAX_IMAGES) +#define CS_ITEMS (CS_LIGHTS + MAX_LIGHTSTYLES) +#define CS_PLAYERSKINS (CS_ITEMS + MAX_ITEMS) +#define CS_GENERAL (CS_PLAYERSKINS + MAX_CLIENTS) +#define MAX_CONFIGSTRINGS (CS_GENERAL + MAX_GENERAL) + +/* ============================================== */ + +/* entity_state_t->event values + * entity events are for effects that take place reletive + * to an existing entities origin. Very network efficient. + * All muzzle flashes really should be converted to events... */ +typedef enum +{ + EV_NONE, + EV_ITEM_RESPAWN, + EV_FOOTSTEP, + EV_FALLSHORT, + EV_FALL, + EV_FALLFAR, + EV_PLAYER_TELEPORT, + EV_OTHER_TELEPORT +} entity_event_t; + +/* entity_state_t is the information conveyed from the server + * in an update message about entities that the client will + * need to render in some way */ +typedef struct entity_state_s +{ + int number; /* edict index */ + + vec3_t origin; + vec3_t angles; + vec3_t old_origin; /* for lerping */ + int modelindex; + int modelindex2, modelindex3, modelindex4; /* weapons, CTF flags, etc */ + int frame; + int skinnum; + unsigned int effects; + int renderfx; + int solid; /* for client side prediction, 8*(bits 0-4) is x/y radius */ + /* 8*(bits 5-9) is z down distance, 8(bits10-15) is z up */ + /* gi.linkentity sets this properly */ + int sound; /* for looping sounds, to guarantee shutoff */ + int event; /* impulse events -- muzzle flashes, footsteps, etc */ + /* events only go out for a single frame, they */ + /* are automatically cleared each frame */ +} entity_state_t; + +/* ============================================== */ + +/* player_state_t is the information needed in addition to pmove_state_t + * to rendered a view. There will only be 10 player_state_t sent each second, + * but the number of pmove_state_t changes will be reletive to client + * frame rates */ +typedef struct +{ + pmove_state_t pmove; /* for prediction */ + + vec3_t viewangles; /* for fixed views */ + vec3_t viewoffset; /* add to pmovestate->origin */ + vec3_t kick_angles; /* add to view direction to get render angles */ + /* set by weapon kicks, pain effects, etc */ + + vec3_t gunangles; + vec3_t gunoffset; + int gunindex; + int gunframe; + + float blend[4]; /* rgba full screen effect */ + float fov; /* horizontal field of view */ + int rdflags; /* refdef flags */ + + short stats[MAX_STATS]; /* fast status bar updates */ +} player_state_t; + +size_t verify_fread(void *, size_t, size_t, FILE *); +size_t verify_fwrite(void *, size_t, size_t, FILE *); + +#endif /* COMMON_SHARED_H */ diff --git a/src/common/header/vid.h b/src/common/header/vid.h new file mode 100644 index 0000000..c7c9c57 --- /dev/null +++ b/src/common/header/vid.h @@ -0,0 +1,69 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, + * USA. + * + * ======================================================================= + * + * API between client and renderer. + * + * ======================================================================= + */ + +#ifndef CL_VID_H +#define CL_VID_H + +#include "common.h" + +// FIXME: Remove it, it's unused. +typedef struct vrect_s { + int x,y,width,height; +} vrect_t; + +// Hold the video state. +typedef struct { + int height; + int width; +} viddef_t; + +// Global video state. +extern viddef_t viddef; + +// Generic stuff. +void VID_Init(void); +void VID_Shutdown(void); +void VID_CheckChanges(void); + +void VID_MenuInit(void); +void VID_MenuDraw(void); +const char *VID_MenuKey(int); + +// Stuff provided by platform backend. +extern int glimp_refreshRate; + +const char **GLimp_GetDisplayIndices(void); +int GLimp_GetWindowDisplayIndex(void); +int GLimp_GetNumVideoDisplays(void); +qboolean GLimp_Init(void); +void GLimp_Shutdown(void); +qboolean GLimp_InitGraphics(int fullscreen, int *pwidth, int *pheight); +void GLimp_ShutdownGraphics(void); +void GLimp_GrabInput(qboolean grab); +int GLimp_GetRefreshRate(void); +qboolean GLimp_GetDesktopMode(int *pwidth, int *pheight); + +#endif diff --git a/src/common/md4.c b/src/common/md4.c new file mode 100644 index 0000000..86f358c --- /dev/null +++ b/src/common/md4.c @@ -0,0 +1,224 @@ +/* + * Public Domain C source implementation of RFC 1320 + * - The MD4 Message-Digest Algorithm - + * + * http://www.faqs.org/rfcs/rfc1320.html + * by Steven Fuller + */ + +#include + +#define ROTATELEFT32(x, s) (((x) << (s)) | ((x) >> (32 - (s)))) + +#define F(X, Y, Z) (((X)&(Y)) | ((~X) & (Z))) +#define G(X, Y, Z) (((X)&(Y)) | ((X)&(Z)) | ((Y)&(Z))) +#define H(X, Y, Z) ((X) ^ (Y) ^ (Z)) + +#define S(a, b, c, d, k, s) \ + { \ + a += (F((b), (c), (d)) + X[(k)]); \ + a = ROTATELEFT32(a, s); \ + } +#define T(a, b, c, d, k, s) \ + { \ + a += (G((b), (c), (d)) + X[(k)] + 0x5A827999); \ + a = ROTATELEFT32(a, s); \ + } +#define U(a, b, c, d, k, s) \ + { \ + a += (H((b), (c), (d)) + X[(k)] + 0x6ED9EBA1); \ + a = ROTATELEFT32(a, s); \ + } + +static uint32_t X[16]; +static uint32_t A, AA; +static uint32_t B, BB; +static uint32_t C, CC; +static uint32_t D, DD; + +static void +DoMD4() +{ + AA = A; + BB = B; + CC = C; + DD = D; + + S(A, B, C, D, 0, 3); + S(D, A, B, C, 1, 7); + S(C, D, A, B, 2, 11); + S(B, C, D, A, 3, 19); + S(A, B, C, D, 4, 3); + S(D, A, B, C, 5, 7); + S(C, D, A, B, 6, 11); + S(B, C, D, A, 7, 19); + S(A, B, C, D, 8, 3); + S(D, A, B, C, 9, 7); + S(C, D, A, B, 10, 11); + S(B, C, D, A, 11, 19); + S(A, B, C, D, 12, 3); + S(D, A, B, C, 13, 7); + S(C, D, A, B, 14, 11); + S(B, C, D, A, 15, 19); + + T(A, B, C, D, 0, 3); + T(D, A, B, C, 4, 5); + T(C, D, A, B, 8, 9); + T(B, C, D, A, 12, 13); + T(A, B, C, D, 1, 3); + T(D, A, B, C, 5, 5); + T(C, D, A, B, 9, 9); + T(B, C, D, A, 13, 13); + T(A, B, C, D, 2, 3); + T(D, A, B, C, 6, 5); + T(C, D, A, B, 10, 9); + T(B, C, D, A, 14, 13); + T(A, B, C, D, 3, 3); + T(D, A, B, C, 7, 5); + T(C, D, A, B, 11, 9); + T(B, C, D, A, 15, 13); + + U(A, B, C, D, 0, 3); + U(D, A, B, C, 8, 9); + U(C, D, A, B, 4, 11); + U(B, C, D, A, 12, 15); + U(A, B, C, D, 2, 3); + U(D, A, B, C, 10, 9); + U(C, D, A, B, 6, 11); + U(B, C, D, A, 14, 15); + U(A, B, C, D, 1, 3); + U(D, A, B, C, 9, 9); + U(C, D, A, B, 5, 11); + U(B, C, D, A, 13, 15); + U(A, B, C, D, 3, 3); + U(D, A, B, C, 11, 9); + U(C, D, A, B, 7, 11); + U(B, C, D, A, 15, 15); + + A += AA; + B += BB; + C += CC; + D += DD; +} + +static void +PerformMD4(const unsigned char *buf, int length, unsigned char *digest) +{ + int len = length / 64; /* number of full blocks */ + int rem = length % 64; /* number of left over bytes */ + + int i, j; + const unsigned char *ptr = buf; + + /* initialize the MD buffer */ + A = 0x67452301; + B = 0xEFCDAB89; + C = 0x98BADCFE; + D = 0x10325476; + + for (i = 0; i < len; i++) + { + for (j = 0; j < 16; j++) + { + X[j] = ((ptr[0] << 0) | (ptr[1] << 8) | + (ptr[2] << 16) | (ptr[3] << 24)); + + ptr += 4; + } + + DoMD4(); + } + + i = rem / 4; + + for (j = 0; j < i; j++) + { + X[j] = ((ptr[0] << 0) | (ptr[1] << 8) | + (ptr[2] << 16) | (ptr[3] << 24)); + + ptr += 4; + } + + switch (rem % 4) + { + case 0: + X[j] = 0x80U; + break; + case 1: + X[j] = ((ptr[0] << 0) | ((0x80U) << 8)); + break; + case 2: + X[j] = ((ptr[0] << 0) | (ptr[1] << 8) | ((0x80U) << 16)); + break; + case 3: + X[j] = + ((ptr[0] << + 0) | (ptr[1] << 8) | (ptr[2] << 16) | ((0x80U) << 24)); + break; + } + + j++; + + if (j > 14) + { + for ( ; j < 16; j++) + { + X[j] = 0; + } + + DoMD4(); + + j = 0; + } + + for ( ; j < 14; j++) + { + X[j] = 0; + } + + X[14] = (length & 0x1FFFFFFF) << 3; + X[15] = (length & ~0x1FFFFFFF) >> 29; + + DoMD4(); + + digest[0] = (A & 0x000000FF) >> 0; + digest[1] = (A & 0x0000FF00) >> 8; + digest[2] = (A & 0x00FF0000) >> 16; + digest[3] = (A & 0xFF000000) >> 24; + digest[4] = (B & 0x000000FF) >> 0; + digest[5] = (B & 0x0000FF00) >> 8; + digest[6] = (B & 0x00FF0000) >> 16; + digest[7] = (B & 0xFF000000) >> 24; + digest[8] = (C & 0x000000FF) >> 0; + digest[9] = (C & 0x0000FF00) >> 8; + digest[10] = (C & 0x00FF0000) >> 16; + digest[11] = (C & 0xFF000000) >> 24; + digest[12] = (D & 0x000000FF) >> 0; + digest[13] = (D & 0x0000FF00) >> 8; + digest[14] = (D & 0x00FF0000) >> 16; + digest[15] = (D & 0xFF000000) >> 24; + + A = AA = 0; + B = BB = 0; + C = CC = 0; + D = DD = 0; + + for (j = 0; j < 16; j++) + { + X[j] = 0; + } +} + +unsigned +Com_BlockChecksum(void *buffer, int length) +{ + uint32_t digest[4]; + unsigned val; + + PerformMD4((unsigned char *)buffer, length, (unsigned char *)digest); + + val = digest[0] ^ digest[1] ^ digest[2] ^ digest[3]; + + return val; +} + diff --git a/src/common/shared.c b/src/common/shared.c new file mode 100644 index 0000000..e4e2c78 --- /dev/null +++ b/src/common/shared.c @@ -0,0 +1,1419 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * Support functions, linked into client, server, renderer and game. + * + * ======================================================================= + */ + +#include + +#include "header/shared.h" + +#define DEG2RAD(a) (a * M_PI) / 180.0F + +vec3_t vec3_origin = {0, 0, 0}; + +/* ============================================================================ */ + +void +RotatePointAroundVector(vec3_t dst, const vec3_t dir, + const vec3_t point, float degrees) +{ + float m[3][3]; + float im[3][3]; + float zrot[3][3]; + float tmpmat[3][3]; + float rot[3][3]; + int i; + vec3_t vr, vup, vf; + + vf[0] = dir[0]; + vf[1] = dir[1]; + vf[2] = dir[2]; + + PerpendicularVector(vr, dir); + CrossProduct(vr, vf, vup); + + m[0][0] = vr[0]; + m[1][0] = vr[1]; + m[2][0] = vr[2]; + + m[0][1] = vup[0]; + m[1][1] = vup[1]; + m[2][1] = vup[2]; + + m[0][2] = vf[0]; + m[1][2] = vf[1]; + m[2][2] = vf[2]; + + memcpy(im, m, sizeof(im)); + + im[0][1] = m[1][0]; + im[0][2] = m[2][0]; + im[1][0] = m[0][1]; + im[1][2] = m[2][1]; + im[2][0] = m[0][2]; + im[2][1] = m[1][2]; + + memset(zrot, 0, sizeof(zrot)); + zrot[2][2] = 1.0F; + + zrot[0][0] = (float)cos(DEG2RAD(degrees)); + zrot[0][1] = (float)sin(DEG2RAD(degrees)); + zrot[1][0] = (float)-sin(DEG2RAD(degrees)); + zrot[1][1] = (float)cos(DEG2RAD(degrees)); + + R_ConcatRotations(m, zrot, tmpmat); + R_ConcatRotations(tmpmat, im, rot); + + for (i = 0; i < 3; i++) + { + dst[i] = rot[i][0] * point[0] + rot[i][1] * point[1] + rot[i][2] * + point[2]; + } +} + +void +AngleVectors(vec3_t angles, vec3_t forward, vec3_t right, vec3_t up) +{ + float angle; + static float sr, sp, sy, cr, cp, cy; + + angle = angles[YAW] * (M_PI * 2 / 360); + sy = (float)sin(angle); + cy = (float)cos(angle); + angle = angles[PITCH] * (M_PI * 2 / 360); + sp = (float)sin(angle); + cp = (float)cos(angle); + angle = angles[ROLL] * (M_PI * 2 / 360); + sr = (float)sin(angle); + cr = (float)cos(angle); + + if (forward) + { + forward[0] = cp * cy; + forward[1] = cp * sy; + forward[2] = -sp; + } + + if (right) + { + right[0] = (-1 * sr * sp * cy + - 1 * cr * -sy); + right[1] = (-1 * sr * sp * sy + - 1 * cr * cy); + right[2] = -1 * sr * cp; + } + + if (up) + { + up[0] = (cr * sp * cy + - sr * -sy); + up[1] = (cr * sp * sy + - sr * cy); + up[2] = cr * cp; + } +} + +void +AngleVectors2(vec3_t value1, vec3_t angles) +{ + float forward; + float yaw, pitch; + + if ((value1[1] == 0) && (value1[0] == 0)) + { + yaw = 0; + + if (value1[2] > 0) + { + pitch = 90; + } + + else + { + pitch = 270; + } + } + else + { + if (value1[0]) + { + yaw = ((float)atan2(value1[1], value1[0]) * 180 / M_PI); + } + + else if (value1[1] > 0) + { + yaw = 90; + } + + else + { + yaw = 270; + } + + if (yaw < 0) + { + yaw += 360; + } + + forward = (float)sqrt(value1[0] * value1[0] + value1[1] * value1[1]); + pitch = ((float)atan2(value1[2], forward) * 180 / M_PI); + + if (pitch < 0) + { + pitch += 360; + } + } + + angles[PITCH] = -pitch; + angles[YAW] = yaw; + angles[ROLL] = 0; +} + +void +ProjectPointOnPlane(vec3_t dst, const vec3_t p, const vec3_t normal) +{ + float d; + vec3_t n; + float inv_denom; + + inv_denom = 1.0F / DotProduct(normal, normal); + + d = DotProduct(normal, p) * inv_denom; + + n[0] = normal[0] * inv_denom; + n[1] = normal[1] * inv_denom; + n[2] = normal[2] * inv_denom; + + dst[0] = p[0] - d * n[0]; + dst[1] = p[1] - d * n[1]; + dst[2] = p[2] - d * n[2]; +} + +/* assumes "src" is normalized */ +void +PerpendicularVector(vec3_t dst, const vec3_t src) +{ + int pos; + int i; + float minelem = 1.0F; + vec3_t tempvec; + + /* find the smallest magnitude axially aligned vector */ + for (pos = 0, i = 0; i < 3; i++) + { + if (fabs(src[i]) < minelem) + { + pos = i; + minelem = (float)fabs(src[i]); + } + } + + tempvec[0] = tempvec[1] = tempvec[2] = 0.0F; + tempvec[pos] = 1.0F; + + /* project the point onto the plane defined by src */ + ProjectPointOnPlane(dst, tempvec, src); + + /* normalize the result */ + VectorNormalize(dst); +} + +void +R_ConcatRotations(float in1[3][3], float in2[3][3], float out[3][3]) +{ + out[0][0] = in1[0][0] * in2[0][0] + in1[0][1] * in2[1][0] + + in1[0][2] * in2[2][0]; + out[0][1] = in1[0][0] * in2[0][1] + in1[0][1] * in2[1][1] + + in1[0][2] * in2[2][1]; + out[0][2] = in1[0][0] * in2[0][2] + in1[0][1] * in2[1][2] + + in1[0][2] * in2[2][2]; + out[1][0] = in1[1][0] * in2[0][0] + in1[1][1] * in2[1][0] + + in1[1][2] * in2[2][0]; + out[1][1] = in1[1][0] * in2[0][1] + in1[1][1] * in2[1][1] + + in1[1][2] * in2[2][1]; + out[1][2] = in1[1][0] * in2[0][2] + in1[1][1] * in2[1][2] + + in1[1][2] * in2[2][2]; + out[2][0] = in1[2][0] * in2[0][0] + in1[2][1] * in2[1][0] + + in1[2][2] * in2[2][0]; + out[2][1] = in1[2][0] * in2[0][1] + in1[2][1] * in2[1][1] + + in1[2][2] * in2[2][1]; + out[2][2] = in1[2][0] * in2[0][2] + in1[2][1] * in2[1][2] + + in1[2][2] * in2[2][2]; +} + +void +R_ConcatTransforms(float in1[3][4], float in2[3][4], float out[3][4]) +{ + out[0][0] = in1[0][0] * in2[0][0] + in1[0][1] * in2[1][0] + + in1[0][2] * in2[2][0]; + out[0][1] = in1[0][0] * in2[0][1] + in1[0][1] * in2[1][1] + + in1[0][2] * in2[2][1]; + out[0][2] = in1[0][0] * in2[0][2] + in1[0][1] * in2[1][2] + + in1[0][2] * in2[2][2]; + out[0][3] = in1[0][0] * in2[0][3] + in1[0][1] * in2[1][3] + + in1[0][2] * in2[2][3] + in1[0][3]; + out[1][0] = in1[1][0] * in2[0][0] + in1[1][1] * in2[1][0] + + in1[1][2] * in2[2][0]; + out[1][1] = in1[1][0] * in2[0][1] + in1[1][1] * in2[1][1] + + in1[1][2] * in2[2][1]; + out[1][2] = in1[1][0] * in2[0][2] + in1[1][1] * in2[1][2] + + in1[1][2] * in2[2][2]; + out[1][3] = in1[1][0] * in2[0][3] + in1[1][1] * in2[1][3] + + in1[1][2] * in2[2][3] + in1[1][3]; + out[2][0] = in1[2][0] * in2[0][0] + in1[2][1] * in2[1][0] + + in1[2][2] * in2[2][0]; + out[2][1] = in1[2][0] * in2[0][1] + in1[2][1] * in2[1][1] + + in1[2][2] * in2[2][1]; + out[2][2] = in1[2][0] * in2[0][2] + in1[2][1] * in2[1][2] + + in1[2][2] * in2[2][2]; + out[2][3] = in1[2][0] * in2[0][3] + in1[2][1] * in2[1][3] + + in1[2][2] * in2[2][3] + in1[2][3]; +} + +/* ============================================================================ */ + +float +Q_fabs(float f) +{ + int tmp = *(int *)&f; + + tmp &= 0x7FFFFFFF; + return *(float *)&tmp; +} + +float +LerpAngle(float a2, float a1, float frac) +{ + if (a1 - a2 > 180) + { + a1 -= 360; + } + + if (a1 - a2 < -180) + { + a1 += 360; + } + + return a2 + frac * (a1 - a2); +} + +float +anglemod(float a) +{ + a = (360.0 / 65536) * ((int)(a * (65536 / 360.0)) & 65535); + return a; +} + +/* + * This is the slow, general version + */ +int +BoxOnPlaneSide2(vec3_t emins, vec3_t emaxs, struct cplane_s *p) +{ + int i; + float dist1, dist2; + int sides; + vec3_t corners[2]; + + for (i = 0; i < 3; i++) + { + if (p->normal[i] < 0) + { + corners[0][i] = emins[i]; + corners[1][i] = emaxs[i]; + } + else + { + corners[1][i] = emins[i]; + corners[0][i] = emaxs[i]; + } + } + + dist1 = DotProduct(p->normal, corners[0]) - p->dist; + dist2 = DotProduct(p->normal, corners[1]) - p->dist; + sides = 0; + + if (dist1 >= 0) + { + sides = 1; + } + + if (dist2 < 0) + { + sides |= 2; + } + + return sides; +} + +/* + * Returns 1, 2, or 1 + 2 + */ +int +BoxOnPlaneSide(vec3_t emins, vec3_t emaxs, struct cplane_s *p) +{ + float dist1, dist2; + int sides; + + /* fast axial cases */ + if (p->type < 3) + { + if (p->dist <= emins[p->type]) + { + return 1; + } + + if (p->dist >= emaxs[p->type]) + { + return 2; + } + + return 3; + } + + /* general case */ + switch (p->signbits) + { + case 0: + dist1 = p->normal[0] * emaxs[0] + p->normal[1] * emaxs[1] + + p->normal[2] * emaxs[2]; + dist2 = p->normal[0] * emins[0] + p->normal[1] * emins[1] + + p->normal[2] * emins[2]; + break; + case 1: + dist1 = p->normal[0] * emins[0] + p->normal[1] * emaxs[1] + + p->normal[2] * emaxs[2]; + dist2 = p->normal[0] * emaxs[0] + p->normal[1] * emins[1] + + p->normal[2] * emins[2]; + break; + case 2: + dist1 = p->normal[0] * emaxs[0] + p->normal[1] * emins[1] + + p->normal[2] * emaxs[2]; + dist2 = p->normal[0] * emins[0] + p->normal[1] * emaxs[1] + + p->normal[2] * emins[2]; + break; + case 3: + dist1 = p->normal[0] * emins[0] + p->normal[1] * emins[1] + + p->normal[2] * emaxs[2]; + dist2 = p->normal[0] * emaxs[0] + p->normal[1] * emaxs[1] + + p->normal[2] * emins[2]; + break; + case 4: + dist1 = p->normal[0] * emaxs[0] + p->normal[1] * emaxs[1] + + p->normal[2] * emins[2]; + dist2 = p->normal[0] * emins[0] + p->normal[1] * emins[1] + + p->normal[2] * emaxs[2]; + break; + case 5: + dist1 = p->normal[0] * emins[0] + p->normal[1] * emaxs[1] + + p->normal[2] * emins[2]; + dist2 = p->normal[0] * emaxs[0] + p->normal[1] * emins[1] + + p->normal[2] * emaxs[2]; + break; + case 6: + dist1 = p->normal[0] * emaxs[0] + p->normal[1] * emins[1] + + p->normal[2] * emins[2]; + dist2 = p->normal[0] * emins[0] + p->normal[1] * emaxs[1] + + p->normal[2] * emaxs[2]; + break; + case 7: + dist1 = p->normal[0] * emins[0] + p->normal[1] * emins[1] + + p->normal[2] * emins[2]; + dist2 = p->normal[0] * emaxs[0] + p->normal[1] * emaxs[1] + + p->normal[2] * emaxs[2]; + break; + default: + dist1 = dist2 = 0; + break; + } + + sides = 0; + + if (dist1 >= p->dist) + { + sides = 1; + } + + if (dist2 < p->dist) + { + sides |= 2; + } + + return sides; +} + +void +ClearBounds(vec3_t mins, vec3_t maxs) +{ + mins[0] = mins[1] = mins[2] = 99999; + maxs[0] = maxs[1] = maxs[2] = -99999; +} + +void +AddPointToBounds(vec3_t v, vec3_t mins, vec3_t maxs) +{ + int i; + vec_t val; + + for (i = 0; i < 3; i++) + { + val = v[i]; + + if (val < mins[i]) + { + mins[i] = val; + } + + if (val > maxs[i]) + { + maxs[i] = val; + } + } +} + +int +VectorCompare(vec3_t v1, vec3_t v2) +{ + if ((v1[0] != v2[0]) || (v1[1] != v2[1]) || (v1[2] != v2[2])) + { + return 0; + } + + return 1; +} + +vec_t +VectorNormalize(vec3_t v) +{ + float length, ilength; + + length = v[0] * v[0] + v[1] * v[1] + v[2] * v[2]; + length = (float)sqrt(length); + + if (length) + { + ilength = 1 / length; + v[0] *= ilength; + v[1] *= ilength; + v[2] *= ilength; + } + + return length; +} + +vec_t +VectorNormalize2(vec3_t v, vec3_t out) +{ + float length, ilength; + + length = v[0] * v[0] + v[1] * v[1] + v[2] * v[2]; + length = (float)sqrt(length); + + if (length) + { + ilength = 1 / length; + out[0] = v[0] * ilength; + out[1] = v[1] * ilength; + out[2] = v[2] * ilength; + } + + return length; +} + +void +VectorMA(vec3_t veca, float scale, vec3_t vecb, vec3_t vecc) +{ + vecc[0] = veca[0] + scale * vecb[0]; + vecc[1] = veca[1] + scale * vecb[1]; + vecc[2] = veca[2] + scale * vecb[2]; +} + +vec_t +_DotProduct(vec3_t v1, vec3_t v2) +{ + return v1[0] * v2[0] + v1[1] * v2[1] + v1[2] * v2[2]; +} + +void +_VectorSubtract(vec3_t veca, vec3_t vecb, vec3_t out) +{ + out[0] = veca[0] - vecb[0]; + out[1] = veca[1] - vecb[1]; + out[2] = veca[2] - vecb[2]; +} + +void +_VectorAdd(vec3_t veca, vec3_t vecb, vec3_t out) +{ + out[0] = veca[0] + vecb[0]; + out[1] = veca[1] + vecb[1]; + out[2] = veca[2] + vecb[2]; +} + +void +_VectorCopy(vec3_t in, vec3_t out) +{ + out[0] = in[0]; + out[1] = in[1]; + out[2] = in[2]; +} + +void +CrossProduct(vec3_t v1, vec3_t v2, vec3_t cross) +{ + cross[0] = v1[1] * v2[2] - v1[2] * v2[1]; + cross[1] = v1[2] * v2[0] - v1[0] * v2[2]; + cross[2] = v1[0] * v2[1] - v1[1] * v2[0]; +} + +double sqrt(double x); + +vec_t +VectorLength(vec3_t v) +{ + int i; + float length; + + length = 0; + + for (i = 0; i < 3; i++) + { + length += v[i] * v[i]; + } + + length = (float)sqrt(length); + + return length; +} + +void +VectorInverse(vec3_t v) +{ + v[0] = -v[0]; + v[1] = -v[1]; + v[2] = -v[2]; +} + +void +VectorScale(vec3_t in, vec_t scale, vec3_t out) +{ + out[0] = in[0] * scale; + out[1] = in[1] * scale; + out[2] = in[2] * scale; +} + +int +Q_log2(int val) +{ + int answer = 0; + + while (val >>= 1) + { + answer++; + } + + return answer; +} + +/* ==================================================================================== */ + +char * +COM_SkipPath(char *pathname) +{ + char *last; + + last = pathname; + + while (*pathname) + { + if (*pathname == '/') + { + last = pathname + 1; + } + + pathname++; + } + + return last; +} + +void +COM_StripExtension(char *in, char *out) +{ + while (*in && *in != '.') + { + *out++ = *in++; + } + + *out = 0; +} + +const char * +COM_FileExtension(const char *in) +{ + const char *ext = strrchr(in, '.'); + + if (!ext || ext == in) + { + return ""; + } + + return ext + 1; +} + +void +COM_FileBase(char *in, char *out) +{ + char *s, *s2; + + s = in + strlen(in) - 1; + + while (s != in && *s != '.') + { + s--; + } + + for (s2 = s; s2 != in && *s2 != '/'; s2--) + { + } + + if (s - s2 < 2) + { + out[0] = 0; + } + else + { + s--; + memcpy(out, s2 + 1, s - s2); + out[s - s2] = 0; + } +} + +/* + * Returns the path up to, but not including the last / + */ +void +COM_FilePath(const char *in, char *out) +{ + const char *s; + + s = in + strlen(in) - 1; + + while (s != in && *s != '/') + { + s--; + } + + memcpy(out, in, s - in); + out[s - in] = 0; +} + +void +COM_DefaultExtension(char *path, const char *extension) +{ + char *src; + + /* */ + /* if path doesn't have a .EXT, append extension */ + /* (extension should include the .) */ + /* */ + src = path + strlen(path) - 1; + + while (*src != '/' && src != path) + { + if (*src == '.') + { + return; /* it has an extension */ + } + + src--; + } + + strcat(path, extension); +} + +/* + * ============================================================================ + * + * BYTE ORDER FUNCTIONS + * + * ============================================================================ + */ + +qboolean bigendien; + +/* can't just use function pointers, or dll linkage can + mess up when qcommon is included in multiple places */ +short (*_BigShort)(short l); +short (*_LittleShort)(short l); +int (*_BigLong)(int l); +int (*_LittleLong)(int l); +float (*_BigFloat)(float l); +float (*_LittleFloat)(float l); + +short +BigShort(short l) +{ + return _BigShort(l); +} + +short +LittleShort(short l) +{ + return _LittleShort(l); +} + +int +BigLong(int l) +{ + return _BigLong(l); +} + +int +LittleLong(int l) +{ + return _LittleLong(l); +} + +float +BigFloat(float l) +{ + return _BigFloat(l); +} + +float +LittleFloat(float l) +{ + return _LittleFloat(l); +} + +short +ShortSwap(short l) +{ + byte b1, b2; + + b1 = l & 255; + b2 = (l >> 8) & 255; + + return (b1 << 8) + b2; +} + +short +ShortNoSwap(short l) +{ + return l; +} + +int +LongSwap(int l) +{ + byte b1, b2, b3, b4; + + b1 = l & 255; + b2 = (l >> 8) & 255; + b3 = (l >> 16) & 255; + b4 = (l >> 24) & 255; + + return ((int)b1 << 24) + ((int)b2 << 16) + ((int)b3 << 8) + b4; +} + +int +LongNoSwap(int l) +{ + return l; +} + +float +FloatSwap(float f) +{ + union + { + float f; + byte b[4]; + } dat1, dat2; + + dat1.f = f; + dat2.b[0] = dat1.b[3]; + dat2.b[1] = dat1.b[2]; + dat2.b[2] = dat1.b[1]; + dat2.b[3] = dat1.b[0]; + return dat2.f; +} + +float +FloatNoSwap(float f) +{ + return f; +} + +void +Swap_Init(void) +{ + byte swaptest[2] = {1, 0}; + short swapTestShort; + assert(sizeof(short) == 2); + memcpy(&swapTestShort, swaptest, 2); + + /* set the byte swapping variables in a portable manner */ + if (swapTestShort == 1) + { + bigendien = false; + _BigShort = ShortSwap; + _LittleShort = ShortNoSwap; + _BigLong = LongSwap; + _LittleLong = LongNoSwap; + _BigFloat = FloatSwap; + _LittleFloat = FloatNoSwap; + Com_Printf("Byte ordering: little endian\n\n"); + } + else + { + bigendien = true; + _BigShort = ShortNoSwap; + _LittleShort = ShortSwap; + _BigLong = LongNoSwap; + _LittleLong = LongSwap; + _BigFloat = FloatNoSwap; + _LittleFloat = FloatSwap; + Com_Printf("Byte ordering: big endian\n\n"); + } + + if (LittleShort(swapTestShort) != 1) + assert("Error in the endian conversion!"); +} + +/* + * does a varargs printf into a temp buffer, so I don't + * need to have varargs versions of all text functions. + */ +char * +va(char *format, ...) +{ + va_list argptr; + static char string[1024]; + + va_start(argptr, format); + vsnprintf(string, 1024, format, argptr); + va_end(argptr); + + return string; +} + +char com_token[MAX_TOKEN_CHARS]; + +/* + * Parse a token out of a string + */ +char * +COM_Parse(char **data_p) +{ + int c; + int len; + char *data; + + data = *data_p; + len = 0; + com_token[0] = 0; + + if (!data) + { + *data_p = NULL; + return ""; + } + +skipwhite: + + while ((c = *data) <= ' ') + { + if (c == 0) + { + *data_p = NULL; + return ""; + } + + data++; + } + + /* skip // comments */ + if ((c == '/') && (data[1] == '/')) + { + while (*data && *data != '\n') + { + data++; + } + + goto skipwhite; + } + + /* handle quoted strings specially */ + if (c == '\"') + { + data++; + + while (1) + { + c = *data++; + + if ((c == '\"') || !c) + { + goto done; + } + + if (len < MAX_TOKEN_CHARS) + { + com_token[len] = c; + len++; + } + } + } + + /* parse a regular word */ + do + { + if (len < MAX_TOKEN_CHARS) + { + com_token[len] = c; + len++; + } + + data++; + c = *data; + } + while (c > 32); + +done: + if (len == MAX_TOKEN_CHARS) + { + len = 0; + } + + com_token[len] = 0; + + *data_p = data; + return com_token; +} + +static int paged_total = 0; + +void +Com_PageInMemory(byte *buffer, int size) +{ + int i; + + for (i = size - 1; i > 0; i -= 4096) + { + paged_total += buffer[i]; + } +} + +/* + * ============================================================================ + * + * LIBRARY REPLACEMENT FUNCTIONS + * + * ============================================================================ + */ + +int +Q_stricmp(const char *s1, const char *s2) +{ + return strcasecmp(s1, s2); +} + +int +Q_strncasecmp(char *s1, char *s2, int n) +{ + int c1, c2; + + do + { + c1 = *s1++; + c2 = *s2++; + + if (!n--) + { + return 0; /* strings are equal until end point */ + } + + if (c1 != c2) + { + if ((c1 >= 'a') && (c1 <= 'z')) + { + c1 -= ('a' - 'A'); + } + + if ((c2 >= 'a') && (c2 <= 'z')) + { + c2 -= ('a' - 'A'); + } + + if (c1 != c2) + { + return -1; /* strings not equal */ + } + } + } + while (c1); + + return 0; /* strings are equal */ +} + +int +Q_strcasecmp(char *s1, char *s2) +{ + return Q_strncasecmp(s1, s2, 99999); +} + +void +Com_sprintf(char *dest, int size, char *fmt, ...) +{ + int len; + va_list argptr; + + va_start(argptr, fmt); + len = vsnprintf(dest, size, fmt, argptr); + va_end(argptr); + + if (len >= size) + { + Com_Printf("Com_sprintf: overflow\n"); + } +} + +char * +Q_strlwr ( char *s ) +{ + char *p = s; + + while ( *s ) + { + *s = tolower( (unsigned char)*s ); + s++; + } + + return ( p ); +} + +int +Q_strlcpy(char *dst, const char *src, int size) +{ + const char *s = src; + + while (*s) + { + if (size > 1) + { + *dst++ = *s; + size--; + } + s++; + } + if (size > 0) + { + *dst = '\0'; + } + + return s - src; +} + +int +Q_strlcat(char *dst, const char *src, int size) +{ + char *d = dst; + + while (size > 0 && *d) + { + size--; + d++; + } + + return (d - dst) + Q_strlcpy(d, src, size); +} + +/* + * An unicode compatible fopen() Wrapper for Windows. + */ +#ifdef _WIN32 +#include +#include +#include +#include +FILE *Q_fopen(const char *file, const char *mode) +{ + WCHAR wfile[MAX_OSPATH]; + WCHAR wmode[16]; + + int len = MultiByteToWideChar(CP_UTF8, 0, file, -1, wfile, MAX_OSPATH); + + if (len > 0) + { + if (MultiByteToWideChar(CP_UTF8, 0, mode, -1, wmode, 16) > 0) + { + // make sure it's a regular file and not a directory or sth, see #394 + struct _stat buf; + int statret = _wstat(wfile, &buf); + if((statret == 0 && (buf.st_mode & _S_IFREG) != 0) || (statret == -1 && errno == ENOENT)) + { + return _wfopen(wfile, wmode); + } + } + } + + return NULL; +} +#else +#include +#include +FILE *Q_fopen(const char *file, const char *mode) +{ + // make sure it's a regular file and not a directory or sth, see #394 + struct stat statbuf; + int statret = stat(file, &statbuf); + // (it's ok if it doesn't exist though, maybe we wanna write/create) + if((statret == -1 && errno != ENOENT) || (statret == 0 && (statbuf.st_mode & S_IFREG) == 0)) + { + return NULL; + } + return fopen(file, mode); +} +#endif + +int +Q_sort_strcomp(const void *s1, const void *s2) +{ + return strcmp(*(char **)s1, *(char **)s2); +} + +/* + * ===================================================================== + * + * INFO STRINGS + * + * ===================================================================== + */ + +/* + * Searches the string for the given + * key and returns the associated value, + * or an empty string. + */ +char * +Info_ValueForKey(char *s, char *key) +{ + char pkey[512]; + static char value[2][512]; /* use two buffers so compares + work without stomping on each other */ + static int valueindex; + char *o; + + valueindex ^= 1; + + if (*s == '\\') + { + s++; + } + + while (1) + { + o = pkey; + + while (*s != '\\') + { + if (!*s) + { + return ""; + } + + *o++ = *s++; + } + + *o = 0; + s++; + + o = value[valueindex]; + + while (*s != '\\' && *s) + { + *o++ = *s++; + } + + *o = 0; + + if (!strcmp(key, pkey)) + { + return value[valueindex]; + } + + if (!*s) + { + return ""; + } + + s++; + } +} + +void +Info_RemoveKey(char *s, char *key) +{ + char *start; + char pkey[512]; + char value[512]; + char *o; + + if (strstr(key, "\\")) + { + return; + } + + while (1) + { + start = s; + + if (*s == '\\') + { + s++; + } + + o = pkey; + + while (*s != '\\') + { + if (!*s) + { + return; + } + + *o++ = *s++; + } + + *o = 0; + s++; + + o = value; + + while (*s != '\\' && *s) + { + *o++ = *s++; + } + + *o = 0; + + if (!strcmp(key, pkey)) + { + memmove(start, s, strlen(s) + 1); /* remove this part */ + return; + } + + if (!*s) + { + return; + } + } +} + +/* + * Some characters are illegal in info strings + * because they can mess up the server's parsing + */ +qboolean +Info_Validate(char *s) +{ + if (strstr(s, "\"")) + { + return false; + } + + if (strstr(s, ";")) + { + return false; + } + + return true; +} + +void +Info_SetValueForKey(char *s, char *key, char *value) +{ + char newi[MAX_INFO_STRING], *v; + int c; + int maxsize = MAX_INFO_STRING; + + if (!key) + { + return; + } + + if (strstr(key, "\\") || (value && strstr(value, "\\"))) + { + Com_Printf("Can't use keys or values with a \\\n"); + return; + } + + if (strstr(key, ";")) + { + Com_Printf("Can't use keys with a semicolon\n"); + return; + } + + if (strstr(key, "\"") || (value && strstr(value, "\""))) + { + Com_Printf("Can't use keys or values with a \"\n"); + return; + } + + if ((strlen(key) > MAX_INFO_KEY - 1) || (value && (strlen(value) > MAX_INFO_KEY - 1))) + { + Com_Printf("Keys and values must be < 64 characters.\n"); + return; + } + + Info_RemoveKey(s, key); + + if (!value || !strlen(value)) + { + return; + } + + Com_sprintf(newi, sizeof(newi), "\\%s\\%s", key, value); + + if (strlen(newi) + strlen(s) > maxsize) + { + Com_Printf("Info string length exceeded\n"); + return; + } + + /* only copy ascii values */ + s += strlen(s); + v = newi; + + while (*v) + { + c = *v++; + c &= 127; /* strip high bits */ + + if ((c >= 32) && (c < 127)) + { + *s++ = c; + } + } + + *s = 0; +} diff --git a/src/constants/anorms.h b/src/constants/anorms.h new file mode 100644 index 0000000..1fa8ade --- /dev/null +++ b/src/constants/anorms.h @@ -0,0 +1,188 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * Precalculates anormal values + * + * ======================================================================= + */ + +{ -0.525731, 0.000000, 0.850651 }, +{ -0.442863, 0.238856, 0.864188 }, +{ -0.295242, 0.000000, 0.955423 }, +{ -0.309017, 0.500000, 0.809017 }, +{ -0.162460, 0.262866, 0.951056 }, +{ 0.000000, 0.000000, 1.000000 }, +{ 0.000000, 0.850651, 0.525731 }, +{ -0.147621, 0.716567, 0.681718 }, +{ 0.147621, 0.716567, 0.681718 }, +{ 0.000000, 0.525731, 0.850651 }, +{ 0.309017, 0.500000, 0.809017 }, +{ 0.525731, 0.000000, 0.850651 }, +{ 0.295242, 0.000000, 0.955423 }, +{ 0.442863, 0.238856, 0.864188 }, +{ 0.162460, 0.262866, 0.951056 }, +{ -0.681718, 0.147621, 0.716567 }, +{ -0.809017, 0.309017, 0.500000 }, +{ -0.587785, 0.425325, 0.688191 }, +{ -0.850651, 0.525731, 0.000000 }, +{ -0.864188, 0.442863, 0.238856 }, +{ -0.716567, 0.681718, 0.147621 }, +{ -0.688191, 0.587785, 0.425325 }, +{ -0.500000, 0.809017, 0.309017 }, +{ -0.238856, 0.864188, 0.442863 }, +{ -0.425325, 0.688191, 0.587785 }, +{ -0.716567, 0.681718, -0.147621 }, +{ -0.500000, 0.809017, -0.309017 }, +{ -0.525731, 0.850651, 0.000000 }, +{ 0.000000, 0.850651, -0.525731 }, +{ -0.238856, 0.864188, -0.442863 }, +{ 0.000000, 0.955423, -0.295242 }, +{ -0.262866, 0.951056, -0.162460 }, +{ 0.000000, 1.000000, 0.000000 }, +{ 0.000000, 0.955423, 0.295242 }, +{ -0.262866, 0.951056, 0.162460 }, +{ 0.238856, 0.864188, 0.442863 }, +{ 0.262866, 0.951056, 0.162460 }, +{ 0.500000, 0.809017, 0.309017 }, +{ 0.238856, 0.864188, -0.442863 }, +{ 0.262866, 0.951056, -0.162460 }, +{ 0.500000, 0.809017, -0.309017 }, +{ 0.850651, 0.525731, 0.000000 }, +{ 0.716567, 0.681718, 0.147621 }, +{ 0.716567, 0.681718, -0.147621 }, +{ 0.525731, 0.850651, 0.000000 }, +{ 0.425325, 0.688191, 0.587785 }, +{ 0.864188, 0.442863, 0.238856 }, +{ 0.688191, 0.587785, 0.425325 }, +{ 0.809017, 0.309017, 0.500000 }, +{ 0.681718, 0.147621, 0.716567 }, +{ 0.587785, 0.425325, 0.688191 }, +{ 0.955423, 0.295242, 0.000000 }, +{ 1.000000, 0.000000, 0.000000 }, +{ 0.951056, 0.162460, 0.262866 }, +{ 0.850651, -0.525731, 0.000000 }, +{ 0.955423, -0.295242, 0.000000 }, +{ 0.864188, -0.442863, 0.238856 }, +{ 0.951056, -0.162460, 0.262866 }, +{ 0.809017, -0.309017, 0.500000 }, +{ 0.681718, -0.147621, 0.716567 }, +{ 0.850651, 0.000000, 0.525731 }, +{ 0.864188, 0.442863, -0.238856 }, +{ 0.809017, 0.309017, -0.500000 }, +{ 0.951056, 0.162460, -0.262866 }, +{ 0.525731, 0.000000, -0.850651 }, +{ 0.681718, 0.147621, -0.716567 }, +{ 0.681718, -0.147621, -0.716567 }, +{ 0.850651, 0.000000, -0.525731 }, +{ 0.809017, -0.309017, -0.500000 }, +{ 0.864188, -0.442863, -0.238856 }, +{ 0.951056, -0.162460, -0.262866 }, +{ 0.147621, 0.716567, -0.681718 }, +{ 0.309017, 0.500000, -0.809017 }, +{ 0.425325, 0.688191, -0.587785 }, +{ 0.442863, 0.238856, -0.864188 }, +{ 0.587785, 0.425325, -0.688191 }, +{ 0.688191, 0.587785, -0.425325 }, +{ -0.147621, 0.716567, -0.681718 }, +{ -0.309017, 0.500000, -0.809017 }, +{ 0.000000, 0.525731, -0.850651 }, +{ -0.525731, 0.000000, -0.850651 }, +{ -0.442863, 0.238856, -0.864188 }, +{ -0.295242, 0.000000, -0.955423 }, +{ -0.162460, 0.262866, -0.951056 }, +{ 0.000000, 0.000000, -1.000000 }, +{ 0.295242, 0.000000, -0.955423 }, +{ 0.162460, 0.262866, -0.951056 }, +{ -0.442863, -0.238856, -0.864188 }, +{ -0.309017, -0.500000, -0.809017 }, +{ -0.162460, -0.262866, -0.951056 }, +{ 0.000000, -0.850651, -0.525731 }, +{ -0.147621, -0.716567, -0.681718 }, +{ 0.147621, -0.716567, -0.681718 }, +{ 0.000000, -0.525731, -0.850651 }, +{ 0.309017, -0.500000, -0.809017 }, +{ 0.442863, -0.238856, -0.864188 }, +{ 0.162460, -0.262866, -0.951056 }, +{ 0.238856, -0.864188, -0.442863 }, +{ 0.500000, -0.809017, -0.309017 }, +{ 0.425325, -0.688191, -0.587785 }, +{ 0.716567, -0.681718, -0.147621 }, +{ 0.688191, -0.587785, -0.425325 }, +{ 0.587785, -0.425325, -0.688191 }, +{ 0.000000, -0.955423, -0.295242 }, +{ 0.000000, -1.000000, 0.000000 }, +{ 0.262866, -0.951056, -0.162460 }, +{ 0.000000, -0.850651, 0.525731 }, +{ 0.000000, -0.955423, 0.295242 }, +{ 0.238856, -0.864188, 0.442863 }, +{ 0.262866, -0.951056, 0.162460 }, +{ 0.500000, -0.809017, 0.309017 }, +{ 0.716567, -0.681718, 0.147621 }, +{ 0.525731, -0.850651, 0.000000 }, +{ -0.238856, -0.864188, -0.442863 }, +{ -0.500000, -0.809017, -0.309017 }, +{ -0.262866, -0.951056, -0.162460 }, +{ -0.850651, -0.525731, 0.000000 }, +{ -0.716567, -0.681718, -0.147621 }, +{ -0.716567, -0.681718, 0.147621 }, +{ -0.525731, -0.850651, 0.000000 }, +{ -0.500000, -0.809017, 0.309017 }, +{ -0.238856, -0.864188, 0.442863 }, +{ -0.262866, -0.951056, 0.162460 }, +{ -0.864188, -0.442863, 0.238856 }, +{ -0.809017, -0.309017, 0.500000 }, +{ -0.688191, -0.587785, 0.425325 }, +{ -0.681718, -0.147621, 0.716567 }, +{ -0.442863, -0.238856, 0.864188 }, +{ -0.587785, -0.425325, 0.688191 }, +{ -0.309017, -0.500000, 0.809017 }, +{ -0.147621, -0.716567, 0.681718 }, +{ -0.425325, -0.688191, 0.587785 }, +{ -0.162460, -0.262866, 0.951056 }, +{ 0.442863, -0.238856, 0.864188 }, +{ 0.162460, -0.262866, 0.951056 }, +{ 0.309017, -0.500000, 0.809017 }, +{ 0.147621, -0.716567, 0.681718 }, +{ 0.000000, -0.525731, 0.850651 }, +{ 0.425325, -0.688191, 0.587785 }, +{ 0.587785, -0.425325, 0.688191 }, +{ 0.688191, -0.587785, 0.425325 }, +{ -0.955423, 0.295242, 0.000000 }, +{ -0.951056, 0.162460, 0.262866 }, +{ -1.000000, 0.000000, 0.000000 }, +{ -0.850651, 0.000000, 0.525731 }, +{ -0.955423, -0.295242, 0.000000 }, +{ -0.951056, -0.162460, 0.262866 }, +{ -0.864188, 0.442863, -0.238856 }, +{ -0.951056, 0.162460, -0.262866 }, +{ -0.809017, 0.309017, -0.500000 }, +{ -0.864188, -0.442863, -0.238856 }, +{ -0.951056, -0.162460, -0.262866 }, +{ -0.809017, -0.309017, -0.500000 }, +{ -0.681718, 0.147621, -0.716567 }, +{ -0.681718, -0.147621, -0.716567 }, +{ -0.850651, 0.000000, -0.525731 }, +{ -0.688191, 0.587785, -0.425325 }, +{ -0.587785, 0.425325, -0.688191 }, +{ -0.425325, 0.688191, -0.587785 }, +{ -0.425325, -0.688191, -0.587785 }, +{ -0.587785, -0.425325, -0.688191 }, +{ -0.688191, -0.587785, -0.425325 }, diff --git a/src/constants/anormtab.h b/src/constants/anormtab.h new file mode 100644 index 0000000..429ffd5 --- /dev/null +++ b/src/constants/anormtab.h @@ -0,0 +1,236 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * Precalculated anormal tabulations + * + * ======================================================================= + */ + + + { 1.23, 1.30, 1.47, 1.35, 1.56, 1.71, 1.37, 1.38, 1.59, 1.60, 1.79, 1.97, 1.88, 1.92, 1.79, 1.02, 0.93, 1.07, 0.82, 0.87, + 0.88, 0.94, 0.96, 1.14, 1.11, 0.82, 0.83, 0.89, 0.89, 0.86, 0.94, 0.91, 1.00, 1.21, 0.98, 1.48, 1.30, 1.57, 0.96, 1.07, + 1.14, 1.60, 1.61, 1.40, 1.37, 1.72, 1.78, 1.79, 1.93, 1.99, 1.90, 1.68, 1.71, 1.86, 1.60, 1.68, 1.78, 1.86, 1.93, 1.99, + 1.97, 1.44, 1.22, 1.49, 0.93, 0.99, 0.99, 1.23, 1.22, 1.44, 1.49, 0.89, 0.89, 0.97, 0.91, 0.98, 1.19, 0.82, 0.76, 0.82, + 0.71, 0.72, 0.73, 0.76, 0.79, 0.86, 0.83, 0.72, 0.76, 0.76, 0.89, 0.82, 0.89, 0.82, 0.89, 0.91, 0.83, 0.96, 1.14, 0.97, + 1.40, 1.19, 0.98, 0.94, 1.00, 1.07, 1.37, 1.21, 1.48, 1.30, 1.57, 1.61, 1.37, 0.86, 0.83, 0.91, 0.82, 0.82, 0.88, 0.89, + 0.96, 1.14, 0.98, 0.87, 0.93, 0.94, 1.02, 1.30, 1.07, 1.35, 1.38, 1.11, 1.56, 1.92, 1.79, 1.79, 1.59, 1.60, 1.72, 1.90, + 1.79, 0.80, 0.85, 0.79, 0.93, 0.80, 0.85, 0.77, 0.74, 0.72, 0.77, 0.74, 0.72, 0.70, 0.70, 0.71, 0.76, 0.73, 0.79, 0.79, + 0.73, 0.76, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.26, 1.26, 1.48, 1.23, 1.50, 1.71, 1.14, 1.19, 1.38, 1.46, 1.64, 1.94, 1.87, 1.84, 1.71, 1.02, 0.92, 1.00, 0.79, 0.85, + 0.84, 0.91, 0.90, 0.98, 0.99, 0.77, 0.77, 0.83, 0.82, 0.79, 0.86, 0.84, 0.92, 0.99, 0.91, 1.24, 1.03, 1.33, 0.88, 0.94, + 0.97, 1.41, 1.39, 1.18, 1.11, 1.51, 1.61, 1.59, 1.80, 1.91, 1.76, 1.54, 1.65, 1.76, 1.70, 1.70, 1.85, 1.85, 1.97, 1.99, + 1.93, 1.28, 1.09, 1.39, 0.92, 0.97, 0.99, 1.18, 1.26, 1.52, 1.48, 0.83, 0.85, 0.90, 0.88, 0.93, 1.00, 0.77, 0.73, 0.78, + 0.72, 0.71, 0.74, 0.75, 0.79, 0.86, 0.81, 0.75, 0.81, 0.79, 0.96, 0.88, 0.94, 0.86, 0.93, 0.92, 0.85, 1.08, 1.33, 1.05, + 1.55, 1.31, 1.01, 1.05, 1.27, 1.31, 1.60, 1.47, 1.70, 1.54, 1.76, 1.76, 1.57, 0.93, 0.90, 0.99, 0.88, 0.88, 0.95, 0.97, + 1.11, 1.39, 1.20, 0.92, 0.97, 1.01, 1.10, 1.39, 1.22, 1.51, 1.58, 1.32, 1.64, 1.97, 1.85, 1.91, 1.77, 1.74, 1.88, 1.99, + 1.91, 0.79, 0.86, 0.80, 0.94, 0.84, 0.88, 0.74, 0.74, 0.71, 0.82, 0.77, 0.76, 0.70, 0.73, 0.72, 0.73, 0.70, 0.74, 0.85, + 0.77, 0.82, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.34, 1.27, 1.53, 1.17, 1.46, 1.71, 0.98, 1.05, 1.20, 1.34, 1.48, 1.86, 1.82, 1.71, 1.62, 1.09, 0.94, 0.99, 0.79, 0.85, + 0.82, 0.90, 0.87, 0.93, 0.96, 0.76, 0.74, 0.79, 0.76, 0.74, 0.79, 0.78, 0.85, 0.92, 0.85, 1.00, 0.93, 1.06, 0.81, 0.86, + 0.89, 1.16, 1.12, 0.97, 0.95, 1.28, 1.38, 1.35, 1.60, 1.77, 1.57, 1.33, 1.50, 1.58, 1.69, 1.63, 1.82, 1.74, 1.91, 1.92, + 1.80, 1.04, 0.97, 1.21, 0.90, 0.93, 0.97, 1.05, 1.21, 1.48, 1.37, 0.77, 0.80, 0.84, 0.85, 0.88, 0.92, 0.73, 0.71, 0.74, + 0.74, 0.71, 0.75, 0.73, 0.79, 0.84, 0.78, 0.79, 0.86, 0.81, 1.05, 0.94, 0.99, 0.90, 0.95, 0.92, 0.86, 1.24, 1.44, 1.14, + 1.59, 1.34, 1.02, 1.27, 1.50, 1.49, 1.80, 1.69, 1.86, 1.72, 1.87, 1.80, 1.69, 1.00, 0.98, 1.23, 0.95, 0.96, 1.09, 1.16, + 1.37, 1.63, 1.46, 0.99, 1.10, 1.25, 1.24, 1.51, 1.41, 1.67, 1.77, 1.55, 1.72, 1.95, 1.89, 1.98, 1.91, 1.86, 1.97, 1.99, + 1.94, 0.81, 0.89, 0.85, 0.98, 0.90, 0.94, 0.75, 0.78, 0.73, 0.89, 0.83, 0.82, 0.72, 0.77, 0.76, 0.72, 0.70, 0.71, 0.91, + 0.83, 0.89, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.46, 1.34, 1.60, 1.16, 1.46, 1.71, 0.94, 0.99, 1.05, 1.26, 1.33, 1.74, 1.76, 1.57, 1.54, 1.23, 0.98, 1.05, 0.83, 0.89, + 0.84, 0.92, 0.87, 0.91, 0.96, 0.78, 0.74, 0.79, 0.72, 0.72, 0.75, 0.76, 0.80, 0.88, 0.83, 0.94, 0.87, 0.95, 0.76, 0.80, + 0.82, 0.97, 0.96, 0.89, 0.88, 1.08, 1.11, 1.10, 1.37, 1.59, 1.37, 1.07, 1.27, 1.34, 1.57, 1.45, 1.69, 1.55, 1.77, 1.79, + 1.60, 0.93, 0.90, 0.99, 0.86, 0.87, 0.93, 0.96, 1.07, 1.35, 1.18, 0.73, 0.76, 0.77, 0.81, 0.82, 0.85, 0.70, 0.71, 0.72, + 0.78, 0.73, 0.77, 0.73, 0.79, 0.82, 0.76, 0.83, 0.90, 0.84, 1.18, 0.98, 1.03, 0.92, 0.95, 0.90, 0.86, 1.32, 1.45, 1.15, + 1.53, 1.27, 0.99, 1.42, 1.65, 1.58, 1.93, 1.83, 1.94, 1.81, 1.88, 1.74, 1.70, 1.19, 1.17, 1.44, 1.11, 1.15, 1.36, 1.41, + 1.61, 1.81, 1.67, 1.22, 1.34, 1.50, 1.42, 1.65, 1.61, 1.82, 1.91, 1.75, 1.80, 1.89, 1.89, 1.98, 1.99, 1.94, 1.98, 1.92, + 1.87, 0.86, 0.95, 0.92, 1.14, 0.98, 1.03, 0.79, 0.84, 0.77, 0.97, 0.90, 0.89, 0.76, 0.82, 0.82, 0.74, 0.72, 0.71, 0.98, + 0.89, 0.97, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.60, 1.44, 1.68, 1.22, 1.49, 1.71, 0.93, 0.99, 0.99, 1.23, 1.22, 1.60, 1.68, 1.44, 1.49, 1.40, 1.14, 1.19, 0.89, 0.96, + 0.89, 0.97, 0.89, 0.91, 0.98, 0.82, 0.76, 0.82, 0.71, 0.72, 0.73, 0.76, 0.79, 0.86, 0.83, 0.91, 0.83, 0.89, 0.72, 0.76, + 0.76, 0.89, 0.89, 0.82, 0.82, 0.98, 0.96, 0.97, 1.14, 1.40, 1.19, 0.94, 1.00, 1.07, 1.37, 1.21, 1.48, 1.30, 1.57, 1.61, + 1.37, 0.86, 0.83, 0.91, 0.82, 0.82, 0.88, 0.89, 0.96, 1.14, 0.98, 0.70, 0.72, 0.73, 0.77, 0.76, 0.79, 0.70, 0.72, 0.71, + 0.82, 0.77, 0.80, 0.74, 0.79, 0.80, 0.74, 0.87, 0.93, 0.85, 1.23, 1.02, 1.02, 0.93, 0.93, 0.87, 0.85, 1.30, 1.35, 1.07, + 1.38, 1.11, 0.94, 1.47, 1.71, 1.56, 1.97, 1.88, 1.92, 1.79, 1.79, 1.59, 1.60, 1.30, 1.35, 1.56, 1.37, 1.38, 1.59, 1.60, + 1.79, 1.92, 1.79, 1.48, 1.57, 1.72, 1.61, 1.78, 1.79, 1.93, 1.99, 1.90, 1.86, 1.78, 1.86, 1.93, 1.99, 1.97, 1.90, 1.79, + 1.72, 0.94, 1.07, 1.00, 1.37, 1.21, 1.30, 0.86, 0.91, 0.83, 1.14, 0.98, 0.96, 0.82, 0.88, 0.89, 0.79, 0.76, 0.73, 1.07, + 0.94, 1.11, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.74, 1.57, 1.76, 1.33, 1.54, 1.71, 0.94, 1.05, 0.99, 1.26, 1.16, 1.46, 1.60, 1.34, 1.46, 1.59, 1.37, 1.37, 0.97, 1.11, + 0.96, 1.10, 0.95, 0.94, 1.08, 0.89, 0.82, 0.88, 0.72, 0.76, 0.75, 0.80, 0.80, 0.88, 0.87, 0.91, 0.83, 0.87, 0.72, 0.76, + 0.74, 0.83, 0.84, 0.78, 0.79, 0.96, 0.89, 0.92, 0.98, 1.23, 1.05, 0.86, 0.92, 0.95, 1.11, 0.98, 1.22, 1.03, 1.34, 1.42, + 1.14, 0.79, 0.77, 0.84, 0.78, 0.76, 0.82, 0.82, 0.89, 0.97, 0.90, 0.70, 0.71, 0.71, 0.73, 0.72, 0.74, 0.73, 0.76, 0.72, + 0.86, 0.81, 0.82, 0.76, 0.79, 0.77, 0.73, 0.90, 0.95, 0.86, 1.18, 1.03, 0.98, 0.92, 0.90, 0.83, 0.84, 1.19, 1.17, 0.98, + 1.15, 0.97, 0.89, 1.42, 1.65, 1.44, 1.93, 1.83, 1.81, 1.67, 1.61, 1.36, 1.41, 1.32, 1.45, 1.58, 1.57, 1.53, 1.74, 1.70, + 1.88, 1.94, 1.81, 1.69, 1.77, 1.87, 1.79, 1.89, 1.92, 1.98, 1.99, 1.98, 1.89, 1.65, 1.80, 1.82, 1.91, 1.94, 1.75, 1.61, + 1.50, 1.07, 1.34, 1.27, 1.60, 1.45, 1.55, 0.93, 0.99, 0.90, 1.35, 1.18, 1.07, 0.87, 0.93, 0.96, 0.85, 0.82, 0.77, 1.15, + 0.99, 1.27, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.86, 1.71, 1.82, 1.48, 1.62, 1.71, 0.98, 1.20, 1.05, 1.34, 1.17, 1.34, 1.53, 1.27, 1.46, 1.77, 1.60, 1.57, 1.16, 1.38, + 1.12, 1.35, 1.06, 1.00, 1.28, 0.97, 0.89, 0.95, 0.76, 0.81, 0.79, 0.86, 0.85, 0.92, 0.93, 0.93, 0.85, 0.87, 0.74, 0.78, + 0.74, 0.79, 0.82, 0.76, 0.79, 0.96, 0.85, 0.90, 0.94, 1.09, 0.99, 0.81, 0.85, 0.89, 0.95, 0.90, 0.99, 0.94, 1.10, 1.24, + 0.98, 0.75, 0.73, 0.78, 0.74, 0.72, 0.77, 0.76, 0.82, 0.89, 0.83, 0.73, 0.71, 0.71, 0.71, 0.70, 0.72, 0.77, 0.80, 0.74, + 0.90, 0.85, 0.84, 0.78, 0.79, 0.75, 0.73, 0.92, 0.95, 0.86, 1.05, 0.99, 0.94, 0.90, 0.86, 0.79, 0.81, 1.00, 0.98, 0.91, + 0.96, 0.89, 0.83, 1.27, 1.50, 1.23, 1.80, 1.69, 1.63, 1.46, 1.37, 1.09, 1.16, 1.24, 1.44, 1.49, 1.69, 1.59, 1.80, 1.69, + 1.87, 1.86, 1.72, 1.82, 1.91, 1.94, 1.92, 1.95, 1.99, 1.98, 1.91, 1.97, 1.89, 1.51, 1.72, 1.67, 1.77, 1.86, 1.55, 1.41, + 1.25, 1.33, 1.58, 1.50, 1.80, 1.63, 1.74, 1.04, 1.21, 0.97, 1.48, 1.37, 1.21, 0.93, 0.97, 1.05, 0.92, 0.88, 0.84, 1.14, + 1.02, 1.34, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.94, 1.84, 1.87, 1.64, 1.71, 1.71, 1.14, 1.38, 1.19, 1.46, 1.23, 1.26, 1.48, 1.26, 1.50, 1.91, 1.80, 1.76, 1.41, 1.61, + 1.39, 1.59, 1.33, 1.24, 1.51, 1.18, 0.97, 1.11, 0.82, 0.88, 0.86, 0.94, 0.92, 0.99, 1.03, 0.98, 0.91, 0.90, 0.79, 0.84, + 0.77, 0.79, 0.84, 0.77, 0.83, 0.99, 0.85, 0.91, 0.92, 1.02, 1.00, 0.79, 0.80, 0.86, 0.88, 0.84, 0.92, 0.88, 0.97, 1.10, + 0.94, 0.74, 0.71, 0.74, 0.72, 0.70, 0.73, 0.72, 0.76, 0.82, 0.77, 0.77, 0.73, 0.74, 0.71, 0.70, 0.73, 0.83, 0.85, 0.78, + 0.92, 0.88, 0.86, 0.81, 0.79, 0.74, 0.75, 0.92, 0.93, 0.85, 0.96, 0.94, 0.88, 0.86, 0.81, 0.75, 0.79, 0.93, 0.90, 0.85, + 0.88, 0.82, 0.77, 1.05, 1.27, 0.99, 1.60, 1.47, 1.39, 1.20, 1.11, 0.95, 0.97, 1.08, 1.33, 1.31, 1.70, 1.55, 1.76, 1.57, + 1.76, 1.70, 1.54, 1.85, 1.97, 1.91, 1.99, 1.97, 1.99, 1.91, 1.77, 1.88, 1.85, 1.39, 1.64, 1.51, 1.58, 1.74, 1.32, 1.22, + 1.01, 1.54, 1.76, 1.65, 1.93, 1.70, 1.85, 1.28, 1.39, 1.09, 1.52, 1.48, 1.26, 0.97, 0.99, 1.18, 1.00, 0.93, 0.90, 1.05, + 1.01, 1.31, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.97, 1.92, 1.88, 1.79, 1.79, 1.71, 1.37, 1.59, 1.38, 1.60, 1.35, 1.23, 1.47, 1.30, 1.56, 1.99, 1.93, 1.90, 1.60, 1.78, + 1.61, 1.79, 1.57, 1.48, 1.72, 1.40, 1.14, 1.37, 0.89, 0.96, 0.94, 1.07, 1.00, 1.21, 1.30, 1.14, 0.98, 0.96, 0.86, 0.91, + 0.83, 0.82, 0.88, 0.82, 0.89, 1.11, 0.87, 0.94, 0.93, 1.02, 1.07, 0.80, 0.79, 0.85, 0.82, 0.80, 0.87, 0.85, 0.93, 1.02, + 0.93, 0.77, 0.72, 0.74, 0.71, 0.70, 0.70, 0.71, 0.72, 0.77, 0.74, 0.82, 0.76, 0.79, 0.72, 0.73, 0.76, 0.89, 0.89, 0.82, + 0.93, 0.91, 0.86, 0.83, 0.79, 0.73, 0.76, 0.91, 0.89, 0.83, 0.89, 0.89, 0.82, 0.82, 0.76, 0.72, 0.76, 0.86, 0.83, 0.79, + 0.82, 0.76, 0.73, 0.94, 1.00, 0.91, 1.37, 1.21, 1.14, 0.98, 0.96, 0.88, 0.89, 0.96, 1.14, 1.07, 1.60, 1.40, 1.61, 1.37, + 1.57, 1.48, 1.30, 1.78, 1.93, 1.79, 1.99, 1.92, 1.90, 1.79, 1.59, 1.72, 1.79, 1.30, 1.56, 1.35, 1.38, 1.60, 1.11, 1.07, + 0.94, 1.68, 1.86, 1.71, 1.97, 1.68, 1.86, 1.44, 1.49, 1.22, 1.44, 1.49, 1.22, 0.99, 0.99, 1.23, 1.19, 0.98, 0.97, 0.97, + 0.98, 1.19, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.94, 1.97, 1.87, 1.91, 1.85, 1.71, 1.60, 1.77, 1.58, 1.74, 1.51, 1.26, 1.48, 1.39, 1.64, 1.99, 1.97, 1.99, 1.70, 1.85, + 1.76, 1.91, 1.76, 1.70, 1.88, 1.55, 1.33, 1.57, 0.96, 1.08, 1.05, 1.31, 1.27, 1.47, 1.54, 1.39, 1.20, 1.11, 0.93, 0.99, + 0.90, 0.88, 0.95, 0.88, 0.97, 1.32, 0.92, 1.01, 0.97, 1.10, 1.22, 0.84, 0.80, 0.88, 0.79, 0.79, 0.85, 0.86, 0.92, 1.02, + 0.94, 0.82, 0.76, 0.77, 0.72, 0.73, 0.70, 0.72, 0.71, 0.74, 0.74, 0.88, 0.81, 0.85, 0.75, 0.77, 0.82, 0.94, 0.93, 0.86, + 0.92, 0.92, 0.86, 0.85, 0.79, 0.74, 0.79, 0.88, 0.85, 0.81, 0.82, 0.83, 0.77, 0.78, 0.73, 0.71, 0.75, 0.79, 0.77, 0.74, + 0.77, 0.73, 0.70, 0.86, 0.92, 0.84, 1.14, 0.99, 0.98, 0.91, 0.90, 0.84, 0.83, 0.88, 0.97, 0.94, 1.41, 1.18, 1.39, 1.11, + 1.33, 1.24, 1.03, 1.61, 1.80, 1.59, 1.91, 1.84, 1.76, 1.64, 1.38, 1.51, 1.71, 1.26, 1.50, 1.23, 1.19, 1.46, 0.99, 1.00, + 0.91, 1.70, 1.85, 1.65, 1.93, 1.54, 1.76, 1.52, 1.48, 1.26, 1.28, 1.39, 1.09, 0.99, 0.97, 1.18, 1.31, 1.01, 1.05, 0.90, + 0.93, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.86, 1.95, 1.82, 1.98, 1.89, 1.71, 1.80, 1.91, 1.77, 1.86, 1.67, 1.34, 1.53, 1.51, 1.72, 1.92, 1.91, 1.99, 1.69, 1.82, + 1.80, 1.94, 1.87, 1.86, 1.97, 1.59, 1.44, 1.69, 1.05, 1.24, 1.27, 1.49, 1.50, 1.69, 1.72, 1.63, 1.46, 1.37, 1.00, 1.23, + 0.98, 0.95, 1.09, 0.96, 1.16, 1.55, 0.99, 1.25, 1.10, 1.24, 1.41, 0.90, 0.85, 0.94, 0.79, 0.81, 0.85, 0.89, 0.94, 1.09, + 0.98, 0.89, 0.82, 0.83, 0.74, 0.77, 0.72, 0.76, 0.73, 0.75, 0.78, 0.94, 0.86, 0.91, 0.79, 0.83, 0.89, 0.99, 0.95, 0.90, + 0.90, 0.92, 0.84, 0.86, 0.79, 0.75, 0.81, 0.85, 0.80, 0.78, 0.76, 0.77, 0.73, 0.74, 0.71, 0.71, 0.73, 0.74, 0.74, 0.71, + 0.76, 0.72, 0.70, 0.79, 0.85, 0.78, 0.98, 0.92, 0.93, 0.85, 0.87, 0.82, 0.79, 0.81, 0.89, 0.86, 1.16, 0.97, 1.12, 0.95, + 1.06, 1.00, 0.93, 1.38, 1.60, 1.35, 1.77, 1.71, 1.57, 1.48, 1.20, 1.28, 1.62, 1.27, 1.46, 1.17, 1.05, 1.34, 0.96, 0.99, + 0.90, 1.63, 1.74, 1.50, 1.80, 1.33, 1.58, 1.48, 1.37, 1.21, 1.04, 1.21, 0.97, 0.97, 0.93, 1.05, 1.34, 1.02, 1.14, 0.84, + 0.88, 0.92, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.74, 1.89, 1.76, 1.98, 1.89, 1.71, 1.93, 1.99, 1.91, 1.94, 1.82, 1.46, 1.60, 1.65, 1.80, 1.79, 1.77, 1.92, 1.57, 1.69, + 1.74, 1.87, 1.88, 1.94, 1.98, 1.53, 1.45, 1.70, 1.18, 1.32, 1.42, 1.58, 1.65, 1.83, 1.81, 1.81, 1.67, 1.61, 1.19, 1.44, + 1.17, 1.11, 1.36, 1.15, 1.41, 1.75, 1.22, 1.50, 1.34, 1.42, 1.61, 0.98, 0.92, 1.03, 0.83, 0.86, 0.89, 0.95, 0.98, 1.23, + 1.14, 0.97, 0.89, 0.90, 0.78, 0.82, 0.76, 0.82, 0.77, 0.79, 0.84, 0.98, 0.90, 0.98, 0.83, 0.89, 0.97, 1.03, 0.95, 0.92, + 0.86, 0.90, 0.82, 0.86, 0.79, 0.77, 0.84, 0.81, 0.76, 0.76, 0.72, 0.73, 0.70, 0.72, 0.71, 0.73, 0.73, 0.72, 0.74, 0.71, + 0.78, 0.74, 0.72, 0.75, 0.80, 0.76, 0.94, 0.88, 0.91, 0.83, 0.87, 0.84, 0.79, 0.76, 0.82, 0.80, 0.97, 0.89, 0.96, 0.88, + 0.95, 0.94, 0.87, 1.11, 1.37, 1.10, 1.59, 1.57, 1.37, 1.33, 1.05, 1.08, 1.54, 1.34, 1.46, 1.16, 0.99, 1.26, 0.96, 1.05, + 0.92, 1.45, 1.55, 1.27, 1.60, 1.07, 1.34, 1.35, 1.18, 1.07, 0.93, 0.99, 0.90, 0.93, 0.87, 0.96, 1.27, 0.99, 1.15, 0.77, + 0.82, 0.85, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.60, 1.78, 1.68, 1.93, 1.86, 1.71, 1.97, 1.99, 1.99, 1.97, 1.93, 1.60, 1.68, 1.78, 1.86, 1.61, 1.57, 1.79, 1.37, 1.48, + 1.59, 1.72, 1.79, 1.92, 1.90, 1.38, 1.35, 1.60, 1.23, 1.30, 1.47, 1.56, 1.71, 1.88, 1.79, 1.92, 1.79, 1.79, 1.30, 1.56, + 1.35, 1.37, 1.59, 1.38, 1.60, 1.90, 1.48, 1.72, 1.57, 1.61, 1.79, 1.21, 1.00, 1.30, 0.89, 0.94, 0.96, 1.07, 1.14, 1.40, + 1.37, 1.14, 0.96, 0.98, 0.82, 0.88, 0.82, 0.89, 0.83, 0.86, 0.91, 1.02, 0.93, 1.07, 0.87, 0.94, 1.11, 1.02, 0.93, 0.93, + 0.82, 0.87, 0.80, 0.85, 0.79, 0.80, 0.85, 0.77, 0.72, 0.74, 0.71, 0.70, 0.70, 0.71, 0.72, 0.77, 0.74, 0.72, 0.76, 0.73, + 0.82, 0.79, 0.76, 0.73, 0.79, 0.76, 0.93, 0.86, 0.91, 0.83, 0.89, 0.89, 0.82, 0.72, 0.76, 0.76, 0.89, 0.82, 0.89, 0.82, + 0.89, 0.91, 0.83, 0.96, 1.14, 0.97, 1.40, 1.44, 1.19, 1.22, 0.99, 0.98, 1.49, 1.44, 1.49, 1.22, 0.99, 1.23, 0.98, 1.19, + 0.97, 1.21, 1.30, 1.00, 1.37, 0.94, 1.07, 1.14, 0.98, 0.96, 0.86, 0.91, 0.83, 0.88, 0.82, 0.89, 1.11, 0.94, 1.07, 0.73, + 0.76, 0.79, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.46, 1.65, 1.60, 1.82, 1.80, 1.71, 1.93, 1.91, 1.99, 1.94, 1.98, 1.74, 1.76, 1.89, 1.89, 1.42, 1.34, 1.61, 1.11, 1.22, + 1.36, 1.50, 1.61, 1.81, 1.75, 1.15, 1.17, 1.41, 1.18, 1.19, 1.42, 1.44, 1.65, 1.83, 1.67, 1.94, 1.81, 1.88, 1.32, 1.58, + 1.45, 1.57, 1.74, 1.53, 1.70, 1.98, 1.69, 1.87, 1.77, 1.79, 1.92, 1.45, 1.27, 1.55, 0.97, 1.07, 1.11, 1.34, 1.37, 1.59, + 1.60, 1.35, 1.07, 1.18, 0.86, 0.93, 0.87, 0.96, 0.90, 0.93, 0.99, 1.03, 0.95, 1.15, 0.90, 0.99, 1.27, 0.98, 0.90, 0.92, + 0.78, 0.83, 0.77, 0.84, 0.79, 0.82, 0.86, 0.73, 0.71, 0.73, 0.72, 0.70, 0.73, 0.72, 0.76, 0.81, 0.76, 0.76, 0.82, 0.77, + 0.89, 0.85, 0.82, 0.75, 0.80, 0.80, 0.94, 0.88, 0.94, 0.87, 0.95, 0.96, 0.88, 0.72, 0.74, 0.76, 0.83, 0.78, 0.84, 0.79, + 0.87, 0.91, 0.83, 0.89, 0.98, 0.92, 1.23, 1.34, 1.05, 1.16, 0.99, 0.96, 1.46, 1.57, 1.54, 1.33, 1.05, 1.26, 1.08, 1.37, + 1.10, 0.98, 1.03, 0.92, 1.14, 0.86, 0.95, 0.97, 0.90, 0.89, 0.79, 0.84, 0.77, 0.82, 0.76, 0.82, 0.97, 0.89, 0.98, 0.71, + 0.72, 0.74, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.34, 1.51, 1.53, 1.67, 1.72, 1.71, 1.80, 1.77, 1.91, 1.86, 1.98, 1.86, 1.82, 1.95, 1.89, 1.24, 1.10, 1.41, 0.95, 0.99, + 1.09, 1.25, 1.37, 1.63, 1.55, 0.96, 0.98, 1.16, 1.05, 1.00, 1.27, 1.23, 1.50, 1.69, 1.46, 1.86, 1.72, 1.87, 1.24, 1.49, + 1.44, 1.69, 1.80, 1.59, 1.69, 1.97, 1.82, 1.94, 1.91, 1.92, 1.99, 1.63, 1.50, 1.74, 1.16, 1.33, 1.38, 1.58, 1.60, 1.77, + 1.80, 1.48, 1.21, 1.37, 0.90, 0.97, 0.93, 1.05, 0.97, 1.04, 1.21, 0.99, 0.95, 1.14, 0.92, 1.02, 1.34, 0.94, 0.86, 0.90, + 0.74, 0.79, 0.75, 0.81, 0.79, 0.84, 0.86, 0.71, 0.71, 0.73, 0.76, 0.73, 0.77, 0.74, 0.80, 0.85, 0.78, 0.81, 0.89, 0.84, + 0.97, 0.92, 0.88, 0.79, 0.85, 0.86, 0.98, 0.92, 1.00, 0.93, 1.06, 1.12, 0.95, 0.74, 0.74, 0.78, 0.79, 0.76, 0.82, 0.79, + 0.87, 0.93, 0.85, 0.85, 0.94, 0.90, 1.09, 1.27, 0.99, 1.17, 1.05, 0.96, 1.46, 1.71, 1.62, 1.48, 1.20, 1.34, 1.28, 1.57, + 1.35, 0.90, 0.94, 0.85, 0.98, 0.81, 0.89, 0.89, 0.83, 0.82, 0.75, 0.78, 0.73, 0.77, 0.72, 0.76, 0.89, 0.83, 0.91, 0.71, + 0.70, 0.72, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 }, + { 1.26, 1.39, 1.48, 1.51, 1.64, 1.71, 1.60, 1.58, 1.77, 1.74, 1.91, 1.94, 1.87, 1.97, 1.85, 1.10, 0.97, 1.22, 0.88, 0.92, + 0.95, 1.01, 1.11, 1.39, 1.32, 0.88, 0.90, 0.97, 0.96, 0.93, 1.05, 0.99, 1.27, 1.47, 1.20, 1.70, 1.54, 1.76, 1.08, 1.31, + 1.33, 1.70, 1.76, 1.55, 1.57, 1.88, 1.85, 1.91, 1.97, 1.99, 1.99, 1.70, 1.65, 1.85, 1.41, 1.54, 1.61, 1.76, 1.80, 1.91, + 1.93, 1.52, 1.26, 1.48, 0.92, 0.99, 0.97, 1.18, 1.09, 1.28, 1.39, 0.94, 0.93, 1.05, 0.92, 1.01, 1.31, 0.88, 0.81, 0.86, + 0.72, 0.75, 0.74, 0.79, 0.79, 0.86, 0.85, 0.71, 0.73, 0.75, 0.82, 0.77, 0.83, 0.78, 0.85, 0.88, 0.81, 0.88, 0.97, 0.90, + 1.18, 1.00, 0.93, 0.86, 0.92, 0.94, 1.14, 0.99, 1.24, 1.03, 1.33, 1.39, 1.11, 0.79, 0.77, 0.84, 0.79, 0.77, 0.84, 0.83, + 0.90, 0.98, 0.91, 0.85, 0.92, 0.91, 1.02, 1.26, 1.00, 1.23, 1.19, 0.99, 1.50, 1.84, 1.71, 1.64, 1.38, 1.46, 1.51, 1.76, + 1.59, 0.84, 0.88, 0.80, 0.94, 0.79, 0.86, 0.82, 0.77, 0.76, 0.74, 0.74, 0.71, 0.73, 0.70, 0.72, 0.82, 0.77, 0.85, 0.74, + 0.70, 0.73, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, + 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00, 1.00 } + diff --git a/src/constants/warpsin.h b/src/constants/warpsin.h new file mode 100644 index 0000000..01f478b --- /dev/null +++ b/src/constants/warpsin.h @@ -0,0 +1,72 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * Precalculated sinus warps + * + * ======================================================================= + */ + +#if 0 +// In case you wonder how these were generated/what they mean: +for (int i = 0; i < 256; i++) { + r_turbsin[i] = 8.0 * sin( (float)i / TURBSCALE ); // TURBSCALE = 256.0 / (2.0 * M_PI) +} +// so r_turbsin[i] is 8 * sin( i/TURBSCALE ); +// however, the values are multiplied with 0.5 in RI_Init() +// so what's actually used is 4 * sin(i/TURBSCALE) +// in R_EmitWaterPolys() something like +s = os + r_turbsin [ (int) ( ( ot * 0.125 + r_newrefdef.time ) * TURBSCALE ) & 255 ]; +// is used; which should (except for rounding errors from lookup table) be equivalent to +s = os + 4.0*sin( ot*0.125 + r_newrefdef.time ); +#endif // 0 + +0, 0.19633, 0.392541, 0.588517, 0.784137, 0.979285, 1.17384, 1.3677, +1.56072, 1.75281, 1.94384, 2.1337, 2.32228, 2.50945, 2.69512, 2.87916, +3.06147, 3.24193, 3.42044, 3.59689, 3.77117, 3.94319, 4.11282, 4.27998, +4.44456, 4.60647, 4.76559, 4.92185, 5.07515, 5.22538, 5.37247, 5.51632, +5.65685, 5.79398, 5.92761, 6.05767, 6.18408, 6.30677, 6.42566, 6.54068, +6.65176, 6.75883, 6.86183, 6.9607, 7.05537, 7.14579, 7.23191, 7.31368, +7.39104, 7.46394, 7.53235, 7.59623, 7.65552, 7.71021, 7.76025, 7.80562, +7.84628, 7.88222, 7.91341, 7.93984, 7.96148, 7.97832, 7.99036, 7.99759, +8, 7.99759, 7.99036, 7.97832, 7.96148, 7.93984, 7.91341, 7.88222, +7.84628, 7.80562, 7.76025, 7.71021, 7.65552, 7.59623, 7.53235, 7.46394, +7.39104, 7.31368, 7.23191, 7.14579, 7.05537, 6.9607, 6.86183, 6.75883, +6.65176, 6.54068, 6.42566, 6.30677, 6.18408, 6.05767, 5.92761, 5.79398, +5.65685, 5.51632, 5.37247, 5.22538, 5.07515, 4.92185, 4.76559, 4.60647, +4.44456, 4.27998, 4.11282, 3.94319, 3.77117, 3.59689, 3.42044, 3.24193, +3.06147, 2.87916, 2.69512, 2.50945, 2.32228, 2.1337, 1.94384, 1.75281, +1.56072, 1.3677, 1.17384, 0.979285, 0.784137, 0.588517, 0.392541, 0.19633, +9.79717e-16, -0.19633, -0.392541, -0.588517, -0.784137, -0.979285, -1.17384, -1.3677, +-1.56072, -1.75281, -1.94384, -2.1337, -2.32228, -2.50945, -2.69512, -2.87916, +-3.06147, -3.24193, -3.42044, -3.59689, -3.77117, -3.94319, -4.11282, -4.27998, +-4.44456, -4.60647, -4.76559, -4.92185, -5.07515, -5.22538, -5.37247, -5.51632, +-5.65685, -5.79398, -5.92761, -6.05767, -6.18408, -6.30677, -6.42566, -6.54068, +-6.65176, -6.75883, -6.86183, -6.9607, -7.05537, -7.14579, -7.23191, -7.31368, +-7.39104, -7.46394, -7.53235, -7.59623, -7.65552, -7.71021, -7.76025, -7.80562, +-7.84628, -7.88222, -7.91341, -7.93984, -7.96148, -7.97832, -7.99036, -7.99759, +-8, -7.99759, -7.99036, -7.97832, -7.96148, -7.93984, -7.91341, -7.88222, +-7.84628, -7.80562, -7.76025, -7.71021, -7.65552, -7.59623, -7.53235, -7.46394, +-7.39104, -7.31368, -7.23191, -7.14579, -7.05537, -6.9607, -6.86183, -6.75883, +-6.65176, -6.54068, -6.42566, -6.30677, -6.18408, -6.05767, -5.92761, -5.79398, +-5.65685, -5.51632, -5.37247, -5.22538, -5.07515, -4.92185, -4.76559, -4.60647, +-4.44456, -4.27998, -4.11282, -3.94319, -3.77117, -3.59689, -3.42044, -3.24193, +-3.06147, -2.87916, -2.69512, -2.50945, -2.32228, -2.1337, -1.94384, -1.75281, +-1.56072, -1.3677, -1.17384, -0.979285, -0.784137, -0.588517, -0.392541, -0.19633, diff --git a/src/files/pcx.c b/src/files/pcx.c new file mode 100644 index 0000000..fb9bcbf --- /dev/null +++ b/src/files/pcx.c @@ -0,0 +1,317 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * The PCX file format + * + * ======================================================================= + */ + +#include "../common/header/ref_shared.h" + +// Fix Jennell Jaquays' name in the Quitscreen +// this is 98x11 pixels, each value an index +// into the standard baseq2/pak0/pics/quit.pcx colormap +static unsigned char quitscreenfix[] = { + 191,191,191,47,28,39,4,4,39,1,47,28,47,28,29,1, + 28,28,47,31,31,1,29,31,1,28,47,47,47,47,29,28, + 47,31,30,28,40,40,4,28,28,40,39,40,29,102,102,245, + 28,39,4,4,39,103,40,40,1,1,102,94,47,47,1,94, + 94,94,94,47,102,245,103,103,103,47,1,102,1,102,29,29, + 29,29,47,28,245,31,31,31,47,1,28,1,28,47,1,102, 102,102, + 191,191,142,47,4,8,8,8,8,4,47,28,1,28,29,28, + 29,29,31,1,47,245,47,47,28,28,31,47,28,1,31,1, + 1,245,47,39,8,8,8,40,39,8,8,8,39,1,1,47, + 4,8,8,8,8,4,47,29,28,31,28,28,29,28,28,28, + 29,28,31,28,47,29,1,28,31,47,1,28,1,1,29,29, + 29,47,28,1,28,28,245,28,28,28,28,47,29,28,47,102,102,103, + 191,191,142,31,29,36,8,8,36,31,40,39,40,4,1,1, + 39,40,39,40,40,31,28,40,40,4,39,40,28,47,31,40, + 39,40,4,1,36,8,8,4,47,36,8,8,39,1,1,1, + 29,36,8,8,36,4,4,39,40,4,47,1,47,40,40,39, + 39,40,28,40,40,47,45,39,40,28,4,39,40,4,39,1, + 28,4,40,28,28,4,39,28,47,40,40,39,40,39,28,28,1,103, + 1,142,29,142,28,39,8,8,36,36,8,8,8,8,36,1, + 8,8,8,8,8,36,39,8,8,8,8,8,36,40,36,8, + 8,8,8,36,40,8,8,40,1,4,8,8,40,1,1,31, + 28,39,8,8,36,8,8,8,8,8,36,31,36,8,8,8, + 8,8,36,8,8,4,40,8,8,36,8,8,8,8,8,36, + 40,8,8,40,39,8,8,40,36,8,8,8,8,8,39,29,28,29, + 103,191,142,47,28,40,8,8,40,8,8,33,33,8,8,36, + 8,8,36,36,8,8,36,8,8,36,36,8,8,36,8,8, + 33,33,8,8,36,8,8,4,47,40,8,8,39,47,28,245, + 28,40,8,8,40,40,36,36,33,8,8,36,8,8,36,36, + 8,8,36,8,8,40,40,8,8,40,4,36,36,33,8,8, + 36,8,8,39,39,8,8,36,8,8,33,36,36,39,28,1,47,28, + 103,246,1,47,1,39,8,8,40,8,8,8,8,8,8,36, + 8,8,4,40,8,8,36,8,8,40,4,8,8,36,8,8, + 8,8,8,8,36,8,8,40,29,39,8,8,39,1,1,47, + 1,39,8,8,40,36,8,8,8,8,8,36,8,8,4,40, + 8,8,36,8,8,40,39,8,8,40,36,8,8,8,8,8, + 36,8,8,39,40,8,8,40,36,8,8,8,8,36,28,1,1,29, + 103,47,40,40,4,36,8,8,36,8,8,33,36,36,36,4, + 8,8,39,4,8,8,36,8,8,4,40,8,8,36,8,8, + 33,36,36,36,36,8,8,40,31,40,8,8,40,47,40,40, + 4,36,8,8,36,8,8,33,33,8,8,36,8,8,36,36, + 8,8,36,8,8,36,36,8,8,36,8,8,33,33,8,8, + 36,8,8,36,36,8,8,4,39,36,36,33,8,8,4,40,4,31, + 191,40,8,8,8,8,8,36,29,36,8,8,8,8,8,40, + 8,8,40,4,8,8,36,8,8,40,39,8,8,39,36,8, + 8,8,8,8,39,8,8,39,45,4,8,8,40,40,8,8, + 8,8,8,36,29,36,8,8,8,8,8,40,36,8,8,8, + 8,8,40,36,8,8,8,8,8,40,36,8,8,8,8,8, + 40,36,8,8,8,8,8,36,8,8,8,8,8,36,4,8,8,4, + 47,45,40,39,40,39,39,245,246,1,40,40,40,39,4,47, + 40,4,28,29,39,40,30,39,39,1,28,40,4,28,1,40, + 40,40,39,4,29,40,39,1,1,1,4,4,47,45,40,39, + 40,39,39,245,246,29,39,40,40,40,4,47,28,39,39,36, + 8,8,4,1,39,40,4,40,40,1,29,4,39,4,40,39, + 1,39,36,36,33,8,8,4,39,4,39,4,40,47,36,8,8,40, + 1,28,47,28,28,29,1,28,47,28,31,28,28,27,47,28, + 45,246,30,28,245,29,47,47,29,30,28,47,27,1,246,47, + 47,47,1,28,47,28,47,1,47,47,1,29,29,47,47,28, + 28,29,1,47,1,47,47,28,31,47,47,31,47,47,47,4, + 8,8,39,245,1,47,28,245,28,47,31,28,47,28,28,28, + 40,8,8,8,8,8,36,47,28,1,246,47,1,40,8,8,36,1, + 47,1,102,1,102,102,47,94,94,102,47,47,102,102,102,102, + 94,1,94,47,102,1,102,47,30,30,102,27,47,102,94,1, + 102,47,1,94,102,103,1,102,103,103,47,47,47,29,1,29, + 28,28,29,28,1,47,28,31,29,1,47,29,28,1,1,47, + 4,39,1,47,47,1,28,28,28,47,1,28,45,28,47,47, + 1,40,4,4,40,4,29,28,31,45,47,28,47,47,4,40,28,28 +}; + +static void +fixQuitScreen(byte* px) +{ + // overwrite 11 lines, 98 pixels each, from quitscreenfix[] + // starting at line 140, column 188 + // quitscreen is 320x240 px + int r, qsIdx = 0; + + px += 140*320; // go to line 140 + px += 188; // to colum 188 + for(r=0; r<11; ++r) + { + memcpy(px, quitscreenfix+qsIdx, 98); + qsIdx += 98; + px += 320; + } +} + +void +LoadPCX(char *origname, byte **pic, byte **palette, int *width, int *height) +{ + byte *raw; + pcx_t *pcx; + int x, y; + int len, full_size; + int pcx_width, pcx_height; + qboolean image_issues = false; + int dataByte, runLength; + byte *out, *pix; + char filename[256]; + + Q_strlcpy(filename, origname, sizeof(filename)); + + /* Add the extension */ + if (strcmp(COM_FileExtension(filename), "pcx")) + { + Q_strlcat(filename, ".pcx", sizeof(filename)); + } + + *pic = NULL; + + if (palette) + { + *palette = NULL; + } + + /* load the file */ + len = ri.FS_LoadFile(filename, (void **)&raw); + + if (!raw || len < sizeof(pcx_t)) + { + R_Printf(PRINT_DEVELOPER, "Bad pcx file %s\n", filename); + return; + } + + /* parse the PCX file */ + pcx = (pcx_t *)raw; + + pcx->xmin = LittleShort(pcx->xmin); + pcx->ymin = LittleShort(pcx->ymin); + pcx->xmax = LittleShort(pcx->xmax); + pcx->ymax = LittleShort(pcx->ymax); + pcx->hres = LittleShort(pcx->hres); + pcx->vres = LittleShort(pcx->vres); + pcx->bytes_per_line = LittleShort(pcx->bytes_per_line); + pcx->palette_type = LittleShort(pcx->palette_type); + + raw = &pcx->data; + + pcx_width = pcx->xmax - pcx->xmin; + pcx_height = pcx->ymax - pcx->ymin; + + if ((pcx->manufacturer != 0x0a) || (pcx->version != 5) || + (pcx->encoding != 1) || (pcx->bits_per_pixel != 8) || + (pcx_width >= 4096) || (pcx_height >= 4096)) + { + R_Printf(PRINT_ALL, "Bad pcx file %s\n", filename); + ri.FS_FreeFile(pcx); + return; + } + + full_size = (pcx_height + 1) * (pcx_width + 1); + out = malloc(full_size); + if (!out) + { + R_Printf(PRINT_ALL, "Can't allocate\n"); + ri.FS_FreeFile(pcx); + return; + } + + *pic = out; + + pix = out; + + if (palette) + { + *palette = malloc(768); + if (!(*palette)) + { + R_Printf(PRINT_ALL, "Can't allocate\n"); + free(out); + ri.FS_FreeFile(pcx); + return; + } + if (len > 768) + { + memcpy(*palette, (byte *)pcx + len - 768, 768); + } + else + { + image_issues = true; + } + } + + if (width) + { + *width = pcx_width + 1; + } + + if (height) + { + *height = pcx_height + 1; + } + + for (y = 0; y <= pcx_height; y++, pix += pcx_width + 1) + { + for (x = 0; x <= pcx_width; ) + { + if (raw - (byte *)pcx > len) + { + // no place for read + image_issues = true; + x = pcx_width; + break; + } + dataByte = *raw++; + + if ((dataByte & 0xC0) == 0xC0) + { + runLength = dataByte & 0x3F; + if (raw - (byte *)pcx > len) + { + // no place for read + image_issues = true; + x = pcx_width; + break; + } + dataByte = *raw++; + } + else + { + runLength = 1; + } + + while (runLength-- > 0) + { + if ((*pic + full_size) <= (pix + x)) + { + // no place for write + image_issues = true; + x += runLength; + runLength = 0; + } + else + { + pix[x++] = dataByte; + } + } + } + } + + if (raw - (byte *)pcx > len) + { + R_Printf(PRINT_DEVELOPER, "PCX file %s was malformed", filename); + free(*pic); + *pic = NULL; + } + else if(pcx_width == 319 && pcx_height == 239 + && Q_strcasecmp(origname, "pics/quit.pcx") == 0 + && Com_BlockChecksum(pcx, len) == 3329419434u) + { + // it's the quit screen, and the baseq2 one (identified by checksum) + // so fix it + fixQuitScreen(*pic); + } + + if (image_issues) + { + R_Printf(PRINT_ALL, "PCX file %s has possible size issues.\n", filename); + } + + ri.FS_FreeFile(pcx); +} + +void +GetPCXInfo(char *filename, int *width, int *height) +{ + pcx_t *pcx; + byte *raw; + + ri.FS_LoadFile(filename, (void **)&raw); + + if (!raw) + { + return; + } + + pcx = (pcx_t *)raw; + + *width = pcx->xmax + 1; + *height = pcx->ymax + 1; + + ri.FS_FreeFile(raw); + + return; +} + diff --git a/src/files/pvs.c b/src/files/pvs.c new file mode 100644 index 0000000..5318500 --- /dev/null +++ b/src/files/pvs.c @@ -0,0 +1,90 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * The PVS Decompress + * + * ======================================================================= + */ + +#include "../common/header/ref_shared.h" + + +/* +=================== +Mod_DecompressVis +=================== +*/ +byte * +Mod_DecompressVis(byte *in, int row) +{ + YQ2_ALIGNAS_TYPE(int) static byte decompressed[MAX_MAP_LEAFS / 8]; + int c; + byte *out; + + out = decompressed; + + if (!in) + { + /* no vis info, so make all visible */ + while (row) + { + *out++ = 0xff; + row--; + } + + return decompressed; + } + + do + { + if (*in) + { + *out++ = *in++; + continue; + } + + c = in[1]; + in += 2; + + while (c) + { + *out++ = 0; + c--; + } + } + while (out - decompressed < row); + + return decompressed; +} + +float +Mod_RadiusFromBounds(const vec3_t mins, const vec3_t maxs) +{ + int i; + vec3_t corner; + + for (i = 0; i < 3; i++) + { + corner[i] = fabs(mins[i]) > fabs(maxs[i]) ? fabs(mins[i]) : fabs(maxs[i]); + } + + return VectorLength(corner); +} diff --git a/src/files/stb.c b/src/files/stb.c new file mode 100644 index 0000000..b96a84c --- /dev/null +++ b/src/files/stb.c @@ -0,0 +1,425 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * Copyright (C) 2015 Daniel Gibson + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * File formats supported by stb_image, for now only tga, png, jpg + * See also https://github.com/nothings/stb + * + * ======================================================================= + */ + +#include + +#include "../common/header/ref_shared.h" + +// don't need HDR stuff +#define STBI_NO_LINEAR +#define STBI_NO_HDR +// make sure STB_image uses standard malloc(), as we'll use standard free() to deallocate +#define STBI_MALLOC(sz) malloc(sz) +#define STBI_REALLOC(p,sz) realloc(p,sz) +#define STBI_FREE(p) free(p) +// include implementation part of stb_image into this file +#define STB_IMAGE_IMPLEMENTATION +#include "stb_image.h" + +// include resize implementation +#define STB_IMAGE_RESIZE_IMPLEMENTATION +#include "stb_image_resize.h" + +/* + * origname: the filename to be opened, might be without extension + * type: extension of the type we wanna open ("jpg", "png" or "tga") + * pic: pointer RGBA pixel data will be assigned to + */ +qboolean +LoadSTB(const char *origname, const char* type, byte **pic, int *width, int *height) +{ + char filename[256]; + + Q_strlcpy(filename, origname, sizeof(filename)); + + /* Add the extension */ + if (strcmp(COM_FileExtension(filename), type) != 0) + { + Q_strlcat(filename, ".", sizeof(filename)); + Q_strlcat(filename, type, sizeof(filename)); + } + + *pic = NULL; + + byte* rawdata = NULL; + int rawsize = ri.FS_LoadFile(filename, (void **)&rawdata); + if (rawdata == NULL) + { + return false; + } + + int w, h, bytesPerPixel; + byte* data = NULL; + data = stbi_load_from_memory(rawdata, rawsize, &w, &h, &bytesPerPixel, STBI_rgb_alpha); + if (data == NULL) + { + R_Printf(PRINT_ALL, "%s couldn't load data from %s: %s!\n", __func__, filename, stbi_failure_reason()); + ri.FS_FreeFile(rawdata); + return false; + } + + ri.FS_FreeFile(rawdata); + + R_Printf(PRINT_DEVELOPER, "%s() loaded: %s\n", __func__, filename); + + *pic = data; + *width = w; + *height = h; + return true; +} + +qboolean +ResizeSTB(byte *input_pixels, int input_width, int input_height, + byte *output_pixels, int output_width, int output_height) +{ + if (stbir_resize_uint8(input_pixels, input_width, input_height, 0, + output_pixels, output_width, output_height, 0, 4)) + return true; + return false; +} + +// We have 16 color palette, 256 / 16 should be enough +#define COLOR_DISTANCE 16 + +void +SmoothColorImage(unsigned *dst, size_t size, size_t rstep) +{ + unsigned *full_size; + unsigned last_color; + unsigned *last_diff; + + // maximum step for apply + if (rstep < 2) + { + return; + } + + // step one pixel back as with check one pixel more + full_size = dst + size - rstep - 1; + last_diff = dst; + last_color = *dst; + + // skip current point + dst ++; + + while (dst < full_size) + { + if (last_color != *dst) + { + int step = dst - last_diff; + if (step > 1) + { + int a_beg, b_beg, c_beg, d_beg; + int a_end, b_end, c_end, d_end; + int a_step, b_step, c_step, d_step; + int k; + + // minimize effect size to rstep + if (step > rstep) + { + // change place for start effect + last_diff += (step - rstep); + step = rstep; + } + + // compare next pixels + for(k = 1; k <= step; k ++) + { + if (dst[k] != dst[0]) + { + break; + } + } + + // step back as pixel different after previous step + k --; + + // mirror steps + if (k < step) + { + // R_Printf(PRINT_ALL, "%s %d -> %d\n", __func__, k, step); + // change place for start effect + last_diff += (step - k); + step = k; + } + + // update step to correct value + step += k; + dst += k; + + // get colors + a_beg = (last_color >> 0 ) & 0xff; + b_beg = (last_color >> 8 ) & 0xff; + c_beg = (last_color >> 16) & 0xff; + d_beg = (last_color >> 24) & 0xff; + + a_end = (*dst >> 0 ) & 0xff; + b_end = (*dst >> 8 ) & 0xff; + c_end = (*dst >> 16) & 0xff; + d_end = (*dst >> 24) & 0xff; + + a_step = a_end - a_beg; + b_step = b_end - b_beg; + c_step = c_end - c_beg; + d_step = d_end - d_beg; + + if ((abs(a_step) <= COLOR_DISTANCE) && + (abs(b_step) <= COLOR_DISTANCE) && + (abs(c_step) <= COLOR_DISTANCE) && + (abs(d_step) <= COLOR_DISTANCE) && + step > 0) + { + // generate color change steps + a_step = (a_step << 16) / step; + b_step = (b_step << 16) / step; + c_step = (c_step << 16) / step; + d_step = (d_step << 16) / step; + + // apply color changes + for (k=0; k < step; k++) + { + *last_diff = (((a_beg + ((a_step * k) >> 16)) << 0) & 0x000000ff) | + (((b_beg + ((b_step * k) >> 16)) << 8) & 0x0000ff00) | + (((c_beg + ((c_step * k) >> 16)) << 16) & 0x00ff0000) | + (((d_beg + ((d_step * k) >> 16)) << 24) & 0xff000000); + last_diff++; + } + } + } + last_color = *dst; + last_diff = dst; + } + dst ++; + } +} + +/* https://en.wikipedia.org/wiki/Pixel-art_scaling_algorithms */ + +void +scale2x(byte *src, byte *dst, int width, int height) +{ + /* + EPX/Scale2×/AdvMAME2× + + x A x + C P B -> 1 2 + x D x 3 4 + + 1=P; 2=P; 3=P; 4=P; + IF C==A AND C!=D AND A!=B => 1=A + IF A==B AND A!=C AND B!=D => 2=B + IF D==C AND D!=B AND C!=A => 3=C + IF B==D AND B!=A AND D!=C => 4=D + */ + { + byte *in_buff = src; + byte *out_buff = dst; + byte *out_buff_full = dst + ((width * height) << 2); + while (out_buff < out_buff_full) + { + int x; + for (x = 0; x < width; x ++) + { + // copy one source byte to two destinatuion bytes + *out_buff = *in_buff; + out_buff ++; + *out_buff = *in_buff; + out_buff ++; + + // next source pixel + in_buff ++; + } + // copy last line one more time + memcpy(out_buff, out_buff - (width << 1), width << 1); + out_buff += width << 1; + } + } + + { + int y, h, w; + h = height - 1; + w = width - 1; + for (y = 0; y < height; y ++) + { + int x; + for (x = 0; x < width; x ++) + { + byte a, b, c, d, p; + + p = src[(width * (y )) + (x )]; + a = (y > 0) ? src[(width * (y - 1)) + (x )] : p; + b = (x < w) ? src[(width * (y )) + (x + 1)] : p; + c = (x > 0) ? src[(width * (y )) + (x - 1)] : p; + d = (y < h) ? src[(width * (y + 1)) + (x )] : p; + + if ((c == a) && (c != d) && (a != b)) + { + dst[(2 * width * ((y * 2) )) + ((x * 2) )] = a; + } + + if ((a == b) && (a != c) && (b != d)) + { + dst[(2 * width * ((y * 2) )) + ((x * 2) + 1)] = b; + } + + if ((d == c) && (d != b) && (c != a)) + { + dst[(2 * width * ((y * 2) + 1)) + ((x * 2) )] = c; + } + + if ((b == d) && (b != a) && (d != c)) + { + dst[(2 * width * ((y * 2) + 1)) + ((x * 2) + 1)] = d; + } + } + } + } +} + +void +scale3x(byte *src, byte *dst, int width, int height) +{ + /* + Scale3×/AdvMAME3× and ScaleFX + + A B C 1 2 3 + D E F -> 4 5 6 + G H I 7 8 9 + + + 1=E; 2=E; 3=E; 4=E; 5=E; 6=E; 7=E; 8=E; 9=E; + IF D==B AND D!=H AND B!=F => 1=D + IF (D==B AND D!=H AND B!=F AND E!=C) OR (B==F AND B!=D AND F!=H AND E!=A) => 2=B + IF B==F AND B!=D AND F!=H => 3=F + IF (H==D AND H!=F AND D!=B AND E!=A) OR (D==B AND D!=H AND B!=F AND E!=G) => 4=D + 5=E + IF (B==F AND B!=D AND F!=H AND E!=I) OR (F==H AND F!=B AND H!=D AND E!=C) => 6=F + IF H==D AND H!=F AND D!=B => 7=D + IF (F==H AND F!=B AND H!=D AND E!=G) OR (H==D AND H!=F AND D!=B AND E!=I) => 8=H + IF F==H AND F!=B AND H!=D => 9=F + */ + { + byte *in_buff = src; + byte *out_buff = dst; + byte *out_buff_full = dst + ((width * height) * 9); + while (out_buff < out_buff_full) + { + int x; + for (x = 0; x < width; x ++) + { + // copy one source byte to two destinatuion bytes + *out_buff = *in_buff; + out_buff ++; + *out_buff = *in_buff; + out_buff ++; + *out_buff = *in_buff; + out_buff ++; + + // next source pixel + in_buff ++; + } + // copy last line one more time + memcpy(out_buff, out_buff - (width * 3), width * 3); + out_buff += width * 3; + // copy last line one more time + memcpy(out_buff, out_buff - (width * 3), width * 3); + out_buff += width * 3; + } + } + + { + int y, z, w; + z = height - 1; + w = width - 1; + for (y = 0; y < height; y ++) + { + int x; + for (x = 0; x < width; x ++) + { + byte a, b, c, d, e, f, g, h, i; + + e = src[(width * y) + x]; + + a = ((y > 0) && (x > 0)) ? src[(width * (y - 1)) + (x - 1)] : e; + b = ((y > 0) && (x )) ? src[(width * (y - 1)) + (x )] : e; + c = ((y > 0) && (x < w)) ? src[(width * (y - 1)) + (x + 1)] : e; + + d = ( (x > 0)) ? src[(width * (y )) + (x - 1)] : e; + f = ( (x < w)) ? src[(width * (y )) + (x + 1)] : e; + + g = ((y < z) && (x > 0)) ? src[(width * (y + 1)) + (x - 1)] : e; + h = ((y < z) && (x )) ? src[(width * (y + 1)) + (x )] : e; + i = ((y < z) && (x < w)) ? src[(width * (y + 1)) + (x + 1)] : e; + + if ((d == b) && (d != h) && (b != f)) + { + dst[(3 * width * ((y * 3) )) + ((x * 3) )] = d; + } + + if (((d == b) && (d != h) && (b != f) && (e != c)) || + ((b == f) && (b != d) && (f != h) && (e != a))) + { + dst[(3 * width * ((y * 3) )) + ((x * 3) + 1)] = b; + } + + if ((b == f) && (b != d) && (f != h)) + { + dst[(3 * width * ((y * 3) )) + ((x * 3) + 2)] = f; + } + + if (((h == d) && (h != f) && (d != b) && (e != a)) || + ((d == b) && (d != h) && (b != f) && (e != g))) + { + dst[(3 * width * ((y * 3) + 1)) + ((x * 3) )] = d; + } + + if (((b == f) && (b != d) && (f != h) && (e != i)) || + ((f == h) && (f != b) && (h != d) && (e != c))) + { + dst[(3 * width * ((y * 3) + 1)) + ((x * 3) + 2)] = f; + } + + if ((h == d) && (h != f) && (d != b)) + { + dst[(3 * width * ((y * 3) + 2)) + ((x * 3) )] = d; + } + + if (((f == h) && (f != b) && (h != d) && (e != g)) || + ((h == d) && (h != f) && (d != b) && (e != i))) + { + dst[(3 * width * ((y * 3) + 2)) + ((x * 3) + 1)] = h; + } + + if ((f == h) && (f != b) && (h != d)) + { + dst[(3 * width * ((y * 3) + 2)) + ((x * 3) + 2)] = f; + } + } + } + } +} diff --git a/src/files/stb_image.h b/src/files/stb_image.h new file mode 100644 index 0000000..196dfd5 --- /dev/null +++ b/src/files/stb_image.h @@ -0,0 +1,7559 @@ +/* stb_image - v2.23 - public domain image loader - http://nothings.org/stb + no warranty implied; use at your own risk + + Do this: + #define STB_IMAGE_IMPLEMENTATION + before you include this file in *one* C or C++ file to create the implementation. + + // i.e. it should look like this: + #include ... + #include ... + #include ... + #define STB_IMAGE_IMPLEMENTATION + #include "stb_image.h" + + You can #define STBI_ASSERT(x) before the #include to avoid using assert.h. + And #define STBI_MALLOC, STBI_REALLOC, and STBI_FREE to avoid using malloc,realloc,free + + + QUICK NOTES: + Primarily of interest to game developers and other people who can + avoid problematic images and only need the trivial interface + + JPEG baseline & progressive (12 bpc/arithmetic not supported, same as stock IJG lib) + PNG 1/2/4/8/16-bit-per-channel + + TGA (not sure what subset, if a subset) + BMP non-1bpp, non-RLE + PSD (composited view only, no extra channels, 8/16 bit-per-channel) + + GIF (*comp always reports as 4-channel) + HDR (radiance rgbE format) + PIC (Softimage PIC) + PNM (PPM and PGM binary only) + + Animated GIF still needs a proper API, but here's one way to do it: + http://gist.github.com/urraka/685d9a6340b26b830d49 + + - decode from memory or through FILE (define STBI_NO_STDIO to remove code) + - decode from arbitrary I/O callbacks + - SIMD acceleration on x86/x64 (SSE2) and ARM (NEON) + + Full documentation under "DOCUMENTATION" below. + + +LICENSE + + See end of file for license information. + +RECENT REVISION HISTORY: + + 2.23 (2019-08-11) fix clang static analysis warning + 2.22 (2019-03-04) gif fixes, fix warnings + 2.21 (2019-02-25) fix typo in comment + 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs + 2.19 (2018-02-11) fix warning + 2.18 (2018-01-30) fix warnings + 2.17 (2018-01-29) bugfix, 1-bit BMP, 16-bitness query, fix warnings + 2.16 (2017-07-23) all functions have 16-bit variants; optimizations; bugfixes + 2.15 (2017-03-18) fix png-1,2,4; all Imagenet JPGs; no runtime SSE detection on GCC + 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs + 2.13 (2016-12-04) experimental 16-bit API, only for PNG so far; fixes + 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes + 2.11 (2016-04-02) 16-bit PNGS; enable SSE2 in non-gcc x64 + RGB-format JPEG; remove white matting in PSD; + allocate large structures on the stack; + correct channel count for PNG & BMP + 2.10 (2016-01-22) avoid warning introduced in 2.09 + 2.09 (2016-01-16) 16-bit TGA; comments in PNM files; STBI_REALLOC_SIZED + + See end of file for full revision history. + + + ============================ Contributors ========================= + + Image formats Extensions, features + Sean Barrett (jpeg, png, bmp) Jetro Lauha (stbi_info) + Nicolas Schulz (hdr, psd) Martin "SpartanJ" Golini (stbi_info) + Jonathan Dummer (tga) James "moose2000" Brown (iPhone PNG) + Jean-Marc Lienher (gif) Ben "Disch" Wenger (io callbacks) + Tom Seddon (pic) Omar Cornut (1/2/4-bit PNG) + Thatcher Ulrich (psd) Nicolas Guillemot (vertical flip) + Ken Miller (pgm, ppm) Richard Mitton (16-bit PSD) + github:urraka (animated gif) Junggon Kim (PNM comments) + Christopher Forseth (animated gif) Daniel Gibson (16-bit TGA) + socks-the-fox (16-bit PNG) + Jeremy Sawicki (handle all ImageNet JPGs) + Optimizations & bugfixes Mikhail Morozov (1-bit BMP) + Fabian "ryg" Giesen Anael Seghezzi (is-16-bit query) + Arseny Kapoulkine + John-Mark Allen + Carmelo J Fdez-Aguera + + Bug & warning fixes + Marc LeBlanc David Woo Guillaume George Martins Mozeiko + Christpher Lloyd Jerry Jansson Joseph Thomson Phil Jordan + Dave Moore Roy Eltham Hayaki Saito Nathan Reed + Won Chun Luke Graham Johan Duparc Nick Verigakis + the Horde3D community Thomas Ruf Ronny Chevalier github:rlyeh + Janez Zemva John Bartholomew Michal Cichon github:romigrou + Jonathan Blow Ken Hamada Tero Hanninen github:svdijk + Laurent Gomila Cort Stratton Sergio Gonzalez github:snagar + Aruelien Pocheville Thibault Reuille Cass Everitt github:Zelex + Ryamond Barbiero Paul Du Bois Engin Manap github:grim210 + Aldo Culquicondor Philipp Wiesemann Dale Weiler github:sammyhw + Oriol Ferrer Mesia Josh Tobin Matthew Gregan github:phprus + Julian Raschke Gregory Mullen Baldur Karlsson github:poppolopoppo + Christian Floisand Kevin Schmidt JR Smith github:darealshinji + Blazej Dariusz Roszkowski github:Michaelangel007 +*/ + +#ifndef STBI_INCLUDE_STB_IMAGE_H +#define STBI_INCLUDE_STB_IMAGE_H + +// DOCUMENTATION +// +// Limitations: +// - no 12-bit-per-channel JPEG +// - no JPEGs with arithmetic coding +// - GIF always returns *comp=4 +// +// Basic usage (see HDR discussion below for HDR usage): +// int x,y,n; +// unsigned char *data = stbi_load(filename, &x, &y, &n, 0); +// // ... process data if not NULL ... +// // ... x = width, y = height, n = # 8-bit components per pixel ... +// // ... replace '0' with '1'..'4' to force that many components per pixel +// // ... but 'n' will always be the number that it would have been if you said 0 +// stbi_image_free(data) +// +// Standard parameters: +// int *x -- outputs image width in pixels +// int *y -- outputs image height in pixels +// int *channels_in_file -- outputs # of image components in image file +// int desired_channels -- if non-zero, # of image components requested in result +// +// The return value from an image loader is an 'unsigned char *' which points +// to the pixel data, or NULL on an allocation failure or if the image is +// corrupt or invalid. The pixel data consists of *y scanlines of *x pixels, +// with each pixel consisting of N interleaved 8-bit components; the first +// pixel pointed to is top-left-most in the image. There is no padding between +// image scanlines or between pixels, regardless of format. The number of +// components N is 'desired_channels' if desired_channels is non-zero, or +// *channels_in_file otherwise. If desired_channels is non-zero, +// *channels_in_file has the number of components that _would_ have been +// output otherwise. E.g. if you set desired_channels to 4, you will always +// get RGBA output, but you can check *channels_in_file to see if it's trivially +// opaque because e.g. there were only 3 channels in the source image. +// +// An output image with N components has the following components interleaved +// in this order in each pixel: +// +// N=#comp components +// 1 grey +// 2 grey, alpha +// 3 red, green, blue +// 4 red, green, blue, alpha +// +// If image loading fails for any reason, the return value will be NULL, +// and *x, *y, *channels_in_file will be unchanged. The function +// stbi_failure_reason() can be queried for an extremely brief, end-user +// unfriendly explanation of why the load failed. Define STBI_NO_FAILURE_STRINGS +// to avoid compiling these strings at all, and STBI_FAILURE_USERMSG to get slightly +// more user-friendly ones. +// +// Paletted PNG, BMP, GIF, and PIC images are automatically depalettized. +// +// =========================================================================== +// +// UNICODE: +// +// If compiling for Windows and you wish to use Unicode filenames, compile +// with +// #define STBI_WINDOWS_UTF8 +// and pass utf8-encoded filenames. Call stbi_convert_wchar_to_utf8 to convert +// Windows wchar_t filenames to utf8. +// +// =========================================================================== +// +// Philosophy +// +// stb libraries are designed with the following priorities: +// +// 1. easy to use +// 2. easy to maintain +// 3. good performance +// +// Sometimes I let "good performance" creep up in priority over "easy to maintain", +// and for best performance I may provide less-easy-to-use APIs that give higher +// performance, in addition to the easy-to-use ones. Nevertheless, it's important +// to keep in mind that from the standpoint of you, a client of this library, +// all you care about is #1 and #3, and stb libraries DO NOT emphasize #3 above all. +// +// Some secondary priorities arise directly from the first two, some of which +// provide more explicit reasons why performance can't be emphasized. +// +// - Portable ("ease of use") +// - Small source code footprint ("easy to maintain") +// - No dependencies ("ease of use") +// +// =========================================================================== +// +// I/O callbacks +// +// I/O callbacks allow you to read from arbitrary sources, like packaged +// files or some other source. Data read from callbacks are processed +// through a small internal buffer (currently 128 bytes) to try to reduce +// overhead. +// +// The three functions you must define are "read" (reads some bytes of data), +// "skip" (skips some bytes of data), "eof" (reports if the stream is at the end). +// +// =========================================================================== +// +// SIMD support +// +// The JPEG decoder will try to automatically use SIMD kernels on x86 when +// supported by the compiler. For ARM Neon support, you must explicitly +// request it. +// +// (The old do-it-yourself SIMD API is no longer supported in the current +// code.) +// +// On x86, SSE2 will automatically be used when available based on a run-time +// test; if not, the generic C versions are used as a fall-back. On ARM targets, +// the typical path is to have separate builds for NEON and non-NEON devices +// (at least this is true for iOS and Android). Therefore, the NEON support is +// toggled by a build flag: define STBI_NEON to get NEON loops. +// +// If for some reason you do not want to use any of SIMD code, or if +// you have issues compiling it, you can disable it entirely by +// defining STBI_NO_SIMD. +// +// =========================================================================== +// +// HDR image support (disable by defining STBI_NO_HDR) +// +// stb_image supports loading HDR images in general, and currently the Radiance +// .HDR file format specifically. You can still load any file through the existing +// interface; if you attempt to load an HDR file, it will be automatically remapped +// to LDR, assuming gamma 2.2 and an arbitrary scale factor defaulting to 1; +// both of these constants can be reconfigured through this interface: +// +// stbi_hdr_to_ldr_gamma(2.2f); +// stbi_hdr_to_ldr_scale(1.0f); +// +// (note, do not use _inverse_ constants; stbi_image will invert them +// appropriately). +// +// Additionally, there is a new, parallel interface for loading files as +// (linear) floats to preserve the full dynamic range: +// +// float *data = stbi_loadf(filename, &x, &y, &n, 0); +// +// If you load LDR images through this interface, those images will +// be promoted to floating point values, run through the inverse of +// constants corresponding to the above: +// +// stbi_ldr_to_hdr_scale(1.0f); +// stbi_ldr_to_hdr_gamma(2.2f); +// +// Finally, given a filename (or an open file or memory block--see header +// file for details) containing image data, you can query for the "most +// appropriate" interface to use (that is, whether the image is HDR or +// not), using: +// +// stbi_is_hdr(char *filename); +// +// =========================================================================== +// +// iPhone PNG support: +// +// By default we convert iphone-formatted PNGs back to RGB, even though +// they are internally encoded differently. You can disable this conversion +// by calling stbi_convert_iphone_png_to_rgb(0), in which case +// you will always just get the native iphone "format" through (which +// is BGR stored in RGB). +// +// Call stbi_set_unpremultiply_on_load(1) as well to force a divide per +// pixel to remove any premultiplied alpha *only* if the image file explicitly +// says there's premultiplied data (currently only happens in iPhone images, +// and only if iPhone convert-to-rgb processing is on). +// +// =========================================================================== +// +// ADDITIONAL CONFIGURATION +// +// - You can suppress implementation of any of the decoders to reduce +// your code footprint by #defining one or more of the following +// symbols before creating the implementation. +// +// STBI_NO_JPEG +// STBI_NO_PNG +// STBI_NO_BMP +// STBI_NO_PSD +// STBI_NO_TGA +// STBI_NO_GIF +// STBI_NO_HDR +// STBI_NO_PIC +// STBI_NO_PNM (.ppm and .pgm) +// +// - You can request *only* certain decoders and suppress all other ones +// (this will be more forward-compatible, as addition of new decoders +// doesn't require you to disable them explicitly): +// +// STBI_ONLY_JPEG +// STBI_ONLY_PNG +// STBI_ONLY_BMP +// STBI_ONLY_PSD +// STBI_ONLY_TGA +// STBI_ONLY_GIF +// STBI_ONLY_HDR +// STBI_ONLY_PIC +// STBI_ONLY_PNM (.ppm and .pgm) +// +// - If you use STBI_NO_PNG (or _ONLY_ without PNG), and you still +// want the zlib decoder to be available, #define STBI_SUPPORT_ZLIB +// + + +#ifndef STBI_NO_STDIO +#include +#endif // STBI_NO_STDIO + +#define STBI_VERSION 1 + +enum +{ + STBI_default = 0, // only used for desired_channels + + STBI_grey = 1, + STBI_grey_alpha = 2, + STBI_rgb = 3, + STBI_rgb_alpha = 4 +}; + +#include +typedef unsigned char stbi_uc; +typedef unsigned short stbi_us; + +#ifdef __cplusplus +extern "C" { +#endif + +#ifndef STBIDEF +#ifdef STB_IMAGE_STATIC +#define STBIDEF static +#else +#define STBIDEF extern +#endif +#endif + +////////////////////////////////////////////////////////////////////////////// +// +// PRIMARY API - works on images of any type +// + +// +// load image by filename, open file, or memory buffer +// + +typedef struct +{ + int (*read) (void *user,char *data,int size); // fill 'data' with 'size' bytes. return number of bytes actually read + void (*skip) (void *user,int n); // skip the next 'n' bytes, or 'unget' the last -n bytes if negative + int (*eof) (void *user); // returns nonzero if we are at end of file/data +} stbi_io_callbacks; + +//////////////////////////////////// +// +// 8-bits-per-channel interface +// + +STBIDEF stbi_uc *stbi_load_from_memory (stbi_uc const *buffer, int len , int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk , void *user, int *x, int *y, int *channels_in_file, int desired_channels); + +#ifndef STBI_NO_STDIO +STBIDEF stbi_uc *stbi_load (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_uc *stbi_load_from_file (FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); +// for stbi_load_from_file, file pointer is left pointing immediately after image +#endif + +#ifndef STBI_NO_GIF +STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, int **delays, int *x, int *y, int *z, int *comp, int req_comp); +#endif + +#ifdef STBI_WINDOWS_UTF8 +STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input); +#endif + +//////////////////////////////////// +// +// 16-bits-per-channel interface +// + +STBIDEF stbi_us *stbi_load_16_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels); + +#ifndef STBI_NO_STDIO +STBIDEF stbi_us *stbi_load_16 (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_us *stbi_load_from_file_16(FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); +#endif + +//////////////////////////////////// +// +// float-per-channel interface +// +#ifndef STBI_NO_LINEAR + STBIDEF float *stbi_loadf_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels); + STBIDEF float *stbi_loadf_from_callbacks (stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels); + + #ifndef STBI_NO_STDIO + STBIDEF float *stbi_loadf (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); + STBIDEF float *stbi_loadf_from_file (FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); + #endif +#endif + +#ifndef STBI_NO_HDR + STBIDEF void stbi_hdr_to_ldr_gamma(float gamma); + STBIDEF void stbi_hdr_to_ldr_scale(float scale); +#endif // STBI_NO_HDR + +#ifndef STBI_NO_LINEAR + STBIDEF void stbi_ldr_to_hdr_gamma(float gamma); + STBIDEF void stbi_ldr_to_hdr_scale(float scale); +#endif // STBI_NO_LINEAR + +// stbi_is_hdr is always defined, but always returns false if STBI_NO_HDR +STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user); +STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len); +#ifndef STBI_NO_STDIO +STBIDEF int stbi_is_hdr (char const *filename); +STBIDEF int stbi_is_hdr_from_file(FILE *f); +#endif // STBI_NO_STDIO + + +// get a VERY brief reason for failure +// NOT THREADSAFE +STBIDEF const char *stbi_failure_reason (void); + +// free the loaded image -- this is just free() +STBIDEF void stbi_image_free (void *retval_from_stbi_load); + +// get image dimensions & components without fully decoding +STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp); +STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp); +STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len); +STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *clbk, void *user); + +#ifndef STBI_NO_STDIO +STBIDEF int stbi_info (char const *filename, int *x, int *y, int *comp); +STBIDEF int stbi_info_from_file (FILE *f, int *x, int *y, int *comp); +STBIDEF int stbi_is_16_bit (char const *filename); +STBIDEF int stbi_is_16_bit_from_file(FILE *f); +#endif + + + +// for image formats that explicitly notate that they have premultiplied alpha, +// we just return the colors as stored in the file. set this flag to force +// unpremultiplication. results are undefined if the unpremultiply overflow. +STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply); + +// indicate whether we should process iphone images back to canonical format, +// or just pass them through "as-is" +STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert); + +// flip the image vertically, so the first pixel in the output array is the bottom left +STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip); + +// ZLIB client - used by PNG, available for other purposes + +STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen); +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header); +STBIDEF char *stbi_zlib_decode_malloc(const char *buffer, int len, int *outlen); +STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); + +STBIDEF char *stbi_zlib_decode_noheader_malloc(const char *buffer, int len, int *outlen); +STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); + + +#ifdef __cplusplus +} +#endif + +// +// +//// end header file ///////////////////////////////////////////////////// +#endif // STBI_INCLUDE_STB_IMAGE_H + +#ifdef STB_IMAGE_IMPLEMENTATION + +#if defined(STBI_ONLY_JPEG) || defined(STBI_ONLY_PNG) || defined(STBI_ONLY_BMP) \ + || defined(STBI_ONLY_TGA) || defined(STBI_ONLY_GIF) || defined(STBI_ONLY_PSD) \ + || defined(STBI_ONLY_HDR) || defined(STBI_ONLY_PIC) || defined(STBI_ONLY_PNM) \ + || defined(STBI_ONLY_ZLIB) + #ifndef STBI_ONLY_JPEG + #define STBI_NO_JPEG + #endif + #ifndef STBI_ONLY_PNG + #define STBI_NO_PNG + #endif + #ifndef STBI_ONLY_BMP + #define STBI_NO_BMP + #endif + #ifndef STBI_ONLY_PSD + #define STBI_NO_PSD + #endif + #ifndef STBI_ONLY_TGA + #define STBI_NO_TGA + #endif + #ifndef STBI_ONLY_GIF + #define STBI_NO_GIF + #endif + #ifndef STBI_ONLY_HDR + #define STBI_NO_HDR + #endif + #ifndef STBI_ONLY_PIC + #define STBI_NO_PIC + #endif + #ifndef STBI_ONLY_PNM + #define STBI_NO_PNM + #endif +#endif + +#if defined(STBI_NO_PNG) && !defined(STBI_SUPPORT_ZLIB) && !defined(STBI_NO_ZLIB) +#define STBI_NO_ZLIB +#endif + + +#include +#include // ptrdiff_t on osx +#include +#include +#include + +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +#include // ldexp, pow +#endif + +#ifndef STBI_NO_STDIO +#include +#endif + +#ifndef STBI_ASSERT +#include +#define STBI_ASSERT(x) assert(x) +#endif + +#ifdef __cplusplus +#define STBI_EXTERN extern "C" +#else +#define STBI_EXTERN extern +#endif + + +#ifndef _MSC_VER + #ifdef __cplusplus + #define stbi_inline inline + #else + #define stbi_inline + #endif +#else + #define stbi_inline __forceinline +#endif + + +#ifdef _MSC_VER +typedef unsigned short stbi__uint16; +typedef signed short stbi__int16; +typedef unsigned int stbi__uint32; +typedef signed int stbi__int32; +#else +#include +typedef uint16_t stbi__uint16; +typedef int16_t stbi__int16; +typedef uint32_t stbi__uint32; +typedef int32_t stbi__int32; +#endif + +// should produce compiler error if size is wrong +typedef unsigned char validate_uint32[sizeof(stbi__uint32)==4 ? 1 : -1]; + +#ifdef _MSC_VER +#define STBI_NOTUSED(v) (void)(v) +#else +#define STBI_NOTUSED(v) (void)sizeof(v) +#endif + +#ifdef _MSC_VER +#define STBI_HAS_LROTL +#endif + +#ifdef STBI_HAS_LROTL + #define stbi_lrot(x,y) _lrotl(x,y) +#else + #define stbi_lrot(x,y) (((x) << (y)) | ((x) >> (32 - (y)))) +#endif + +#if defined(STBI_MALLOC) && defined(STBI_FREE) && (defined(STBI_REALLOC) || defined(STBI_REALLOC_SIZED)) +// ok +#elif !defined(STBI_MALLOC) && !defined(STBI_FREE) && !defined(STBI_REALLOC) && !defined(STBI_REALLOC_SIZED) +// ok +#else +#error "Must define all or none of STBI_MALLOC, STBI_FREE, and STBI_REALLOC (or STBI_REALLOC_SIZED)." +#endif + +#ifndef STBI_MALLOC +#define STBI_MALLOC(sz) malloc(sz) +#define STBI_REALLOC(p,newsz) realloc(p,newsz) +#define STBI_FREE(p) free(p) +#endif + +#ifndef STBI_REALLOC_SIZED +#define STBI_REALLOC_SIZED(p,oldsz,newsz) STBI_REALLOC(p,newsz) +#endif + +// x86/x64 detection +#if defined(__x86_64__) || defined(_M_X64) +#define STBI__X64_TARGET +#elif defined(__i386) || defined(_M_IX86) +#define STBI__X86_TARGET +#endif + +#if defined(__GNUC__) && defined(STBI__X86_TARGET) && !defined(__SSE2__) && !defined(STBI_NO_SIMD) +// gcc doesn't support sse2 intrinsics unless you compile with -msse2, +// which in turn means it gets to use SSE2 everywhere. This is unfortunate, +// but previous attempts to provide the SSE2 functions with runtime +// detection caused numerous issues. The way architecture extensions are +// exposed in GCC/Clang is, sadly, not really suited for one-file libs. +// New behavior: if compiled with -msse2, we use SSE2 without any +// detection; if not, we don't use it at all. +#define STBI_NO_SIMD +#endif + +#if defined(__MINGW32__) && defined(STBI__X86_TARGET) && !defined(STBI_MINGW_ENABLE_SSE2) && !defined(STBI_NO_SIMD) +// Note that __MINGW32__ doesn't actually mean 32-bit, so we have to avoid STBI__X64_TARGET +// +// 32-bit MinGW wants ESP to be 16-byte aligned, but this is not in the +// Windows ABI and VC++ as well as Windows DLLs don't maintain that invariant. +// As a result, enabling SSE2 on 32-bit MinGW is dangerous when not +// simultaneously enabling "-mstackrealign". +// +// See https://github.com/nothings/stb/issues/81 for more information. +// +// So default to no SSE2 on 32-bit MinGW. If you've read this far and added +// -mstackrealign to your build settings, feel free to #define STBI_MINGW_ENABLE_SSE2. +#define STBI_NO_SIMD +#endif + +#if !defined(STBI_NO_SIMD) && (defined(STBI__X86_TARGET) || defined(STBI__X64_TARGET)) +#define STBI_SSE2 +#include + +#ifdef _MSC_VER + +#if _MSC_VER >= 1400 // not VC6 +#include // __cpuid +static int stbi__cpuid3(void) +{ + int info[4]; + __cpuid(info,1); + return info[3]; +} +#else +static int stbi__cpuid3(void) +{ + int res; + __asm { + mov eax,1 + cpuid + mov res,edx + } + return res; +} +#endif + +#define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name + +#if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) +static int stbi__sse2_available(void) +{ + int info3 = stbi__cpuid3(); + return ((info3 >> 26) & 1) != 0; +} +#endif + +#else // assume GCC-style if not VC++ +#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) + +#if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) +static int stbi__sse2_available(void) +{ + // If we're even attempting to compile this on GCC/Clang, that means + // -msse2 is on, which means the compiler is allowed to use SSE2 + // instructions at will, and so are we. + return 1; +} +#endif + +#endif +#endif + +// ARM NEON +#if defined(STBI_NO_SIMD) && defined(STBI_NEON) +#undef STBI_NEON +#endif + +#ifdef STBI_NEON +#include +// assume GCC or Clang on ARM targets +#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) +#endif + +#ifndef STBI_SIMD_ALIGN +#define STBI_SIMD_ALIGN(type, name) type name +#endif + +/////////////////////////////////////////////// +// +// stbi__context struct and start_xxx functions + +// stbi__context structure is our basic context used by all images, so it +// contains all the IO context, plus some basic image information +typedef struct +{ + stbi__uint32 img_x, img_y; + int img_n, img_out_n; + + stbi_io_callbacks io; + void *io_user_data; + + int read_from_callbacks; + int buflen; + stbi_uc buffer_start[128]; + + stbi_uc *img_buffer, *img_buffer_end; + stbi_uc *img_buffer_original, *img_buffer_original_end; +} stbi__context; + + +static void stbi__refill_buffer(stbi__context *s); + +// initialize a memory-decode context +static void stbi__start_mem(stbi__context *s, stbi_uc const *buffer, int len) +{ + s->io.read = NULL; + s->read_from_callbacks = 0; + s->img_buffer = s->img_buffer_original = (stbi_uc *) buffer; + s->img_buffer_end = s->img_buffer_original_end = (stbi_uc *) buffer+len; +} + +// initialize a callback-based context +static void stbi__start_callbacks(stbi__context *s, stbi_io_callbacks *c, void *user) +{ + s->io = *c; + s->io_user_data = user; + s->buflen = sizeof(s->buffer_start); + s->read_from_callbacks = 1; + s->img_buffer_original = s->buffer_start; + stbi__refill_buffer(s); + s->img_buffer_original_end = s->img_buffer_end; +} + +#ifndef STBI_NO_STDIO + +static int stbi__stdio_read(void *user, char *data, int size) +{ + return (int) fread(data,1,size,(FILE*) user); +} + +static void stbi__stdio_skip(void *user, int n) +{ + fseek((FILE*) user, n, SEEK_CUR); +} + +static int stbi__stdio_eof(void *user) +{ + return feof((FILE*) user); +} + +static stbi_io_callbacks stbi__stdio_callbacks = +{ + stbi__stdio_read, + stbi__stdio_skip, + stbi__stdio_eof, +}; + +static void stbi__start_file(stbi__context *s, FILE *f) +{ + stbi__start_callbacks(s, &stbi__stdio_callbacks, (void *) f); +} + +//static void stop_file(stbi__context *s) { } + +#endif // !STBI_NO_STDIO + +static void stbi__rewind(stbi__context *s) +{ + // conceptually rewind SHOULD rewind to the beginning of the stream, + // but we just rewind to the beginning of the initial buffer, because + // we only use it after doing 'test', which only ever looks at at most 92 bytes + s->img_buffer = s->img_buffer_original; + s->img_buffer_end = s->img_buffer_original_end; +} + +enum +{ + STBI_ORDER_RGB, + STBI_ORDER_BGR +}; + +typedef struct +{ + int bits_per_channel; + int num_channels; + int channel_order; +} stbi__result_info; + +#ifndef STBI_NO_JPEG +static int stbi__jpeg_test(stbi__context *s); +static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_PNG +static int stbi__png_test(stbi__context *s); +static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__png_is16(stbi__context *s); +#endif + +#ifndef STBI_NO_BMP +static int stbi__bmp_test(stbi__context *s); +static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_TGA +static int stbi__tga_test(stbi__context *s); +static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context *s); +static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc); +static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__psd_is16(stbi__context *s); +#endif + +#ifndef STBI_NO_HDR +static int stbi__hdr_test(stbi__context *s); +static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_PIC +static int stbi__pic_test(stbi__context *s); +static void *stbi__pic_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_GIF +static int stbi__gif_test(stbi__context *s); +static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, int *z, int *comp, int req_comp); +static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_PNM +static int stbi__pnm_test(stbi__context *s); +static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +// this is not threadsafe +static const char *stbi__g_failure_reason; + +STBIDEF const char *stbi_failure_reason(void) +{ + return stbi__g_failure_reason; +} + +static int stbi__err(const char *str) +{ + stbi__g_failure_reason = str; + return 0; +} + +static void *stbi__malloc(size_t size) +{ + return STBI_MALLOC(size); +} + +// stb_image uses ints pervasively, including for offset calculations. +// therefore the largest decoded image size we can support with the +// current code, even on 64-bit targets, is INT_MAX. this is not a +// significant limitation for the intended use case. +// +// we do, however, need to make sure our size calculations don't +// overflow. hence a few helper functions for size calculations that +// multiply integers together, making sure that they're non-negative +// and no overflow occurs. + +// return 1 if the sum is valid, 0 on overflow. +// negative terms are considered invalid. +static int stbi__addsizes_valid(int a, int b) +{ + if (b < 0) return 0; + // now 0 <= b <= INT_MAX, hence also + // 0 <= INT_MAX - b <= INTMAX. + // And "a + b <= INT_MAX" (which might overflow) is the + // same as a <= INT_MAX - b (no overflow) + return a <= INT_MAX - b; +} + +// returns 1 if the product is valid, 0 on overflow. +// negative factors are considered invalid. +static int stbi__mul2sizes_valid(int a, int b) +{ + if (a < 0 || b < 0) return 0; + if (b == 0) return 1; // mul-by-0 is always safe + // portable way to check for no overflows in a*b + return a <= INT_MAX/b; +} + +// returns 1 if "a*b + add" has no negative terms/factors and doesn't overflow +static int stbi__mad2sizes_valid(int a, int b, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__addsizes_valid(a*b, add); +} + +// returns 1 if "a*b*c + add" has no negative terms/factors and doesn't overflow +static int stbi__mad3sizes_valid(int a, int b, int c, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a*b, c) && + stbi__addsizes_valid(a*b*c, add); +} + +// returns 1 if "a*b*c*d + add" has no negative terms/factors and doesn't overflow +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +static int stbi__mad4sizes_valid(int a, int b, int c, int d, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a*b, c) && + stbi__mul2sizes_valid(a*b*c, d) && stbi__addsizes_valid(a*b*c*d, add); +} +#endif + +// mallocs with size overflow checking +static void *stbi__malloc_mad2(int a, int b, int add) +{ + if (!stbi__mad2sizes_valid(a, b, add)) return NULL; + return stbi__malloc(a*b + add); +} + +static void *stbi__malloc_mad3(int a, int b, int c, int add) +{ + if (!stbi__mad3sizes_valid(a, b, c, add)) return NULL; + return stbi__malloc(a*b*c + add); +} + +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +static void *stbi__malloc_mad4(int a, int b, int c, int d, int add) +{ + if (!stbi__mad4sizes_valid(a, b, c, d, add)) return NULL; + return stbi__malloc(a*b*c*d + add); +} +#endif + +// stbi__err - error +// stbi__errpf - error returning pointer to float +// stbi__errpuc - error returning pointer to unsigned char + +#ifdef STBI_NO_FAILURE_STRINGS + #define stbi__err(x,y) 0 +#elif defined(STBI_FAILURE_USERMSG) + #define stbi__err(x,y) stbi__err(y) +#else + #define stbi__err(x,y) stbi__err(x) +#endif + +#define stbi__errpf(x,y) ((float *)(size_t) (stbi__err(x,y)?NULL:NULL)) +#define stbi__errpuc(x,y) ((unsigned char *)(size_t) (stbi__err(x,y)?NULL:NULL)) + +STBIDEF void stbi_image_free(void *retval_from_stbi_load) +{ + STBI_FREE(retval_from_stbi_load); +} + +#ifndef STBI_NO_LINEAR +static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp); +#endif + +#ifndef STBI_NO_HDR +static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp); +#endif + +static int stbi__vertically_flip_on_load = 0; + +STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip) +{ + stbi__vertically_flip_on_load = flag_true_if_should_flip; +} + +static void *stbi__load_main(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc) +{ + memset(ri, 0, sizeof(*ri)); // make sure it's initialized if we add new fields + ri->bits_per_channel = 8; // default is 8 so most paths don't have to be changed + ri->channel_order = STBI_ORDER_RGB; // all current input & output are this, but this is here so we can add BGR order + ri->num_channels = 0; + + #ifndef STBI_NO_JPEG + if (stbi__jpeg_test(s)) return stbi__jpeg_load(s,x,y,comp,req_comp, ri); + #endif + #ifndef STBI_NO_PNG + if (stbi__png_test(s)) return stbi__png_load(s,x,y,comp,req_comp, ri); + #endif + #ifndef STBI_NO_BMP + if (stbi__bmp_test(s)) return stbi__bmp_load(s,x,y,comp,req_comp, ri); + #endif + #ifndef STBI_NO_GIF + if (stbi__gif_test(s)) return stbi__gif_load(s,x,y,comp,req_comp, ri); + #endif + #ifndef STBI_NO_PSD + if (stbi__psd_test(s)) return stbi__psd_load(s,x,y,comp,req_comp, ri, bpc); + #endif + #ifndef STBI_NO_PIC + if (stbi__pic_test(s)) return stbi__pic_load(s,x,y,comp,req_comp, ri); + #endif + #ifndef STBI_NO_PNM + if (stbi__pnm_test(s)) return stbi__pnm_load(s,x,y,comp,req_comp, ri); + #endif + + #ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + float *hdr = stbi__hdr_load(s, x,y,comp,req_comp, ri); + return stbi__hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp); + } + #endif + + #ifndef STBI_NO_TGA + // test tga last because it's a crappy test! + if (stbi__tga_test(s)) + return stbi__tga_load(s,x,y,comp,req_comp, ri); + #endif + + return stbi__errpuc("unknown image type", "Image not of any known type, or corrupt"); +} + +static stbi_uc *stbi__convert_16_to_8(stbi__uint16 *orig, int w, int h, int channels) +{ + int i; + int img_len = w * h * channels; + stbi_uc *reduced; + + reduced = (stbi_uc *) stbi__malloc(img_len); + if (reduced == NULL) return stbi__errpuc("outofmem", "Out of memory"); + + for (i = 0; i < img_len; ++i) + reduced[i] = (stbi_uc)((orig[i] >> 8) & 0xFF); // top half of each byte is sufficient approx of 16->8 bit scaling + + STBI_FREE(orig); + return reduced; +} + +static stbi__uint16 *stbi__convert_8_to_16(stbi_uc *orig, int w, int h, int channels) +{ + int i; + int img_len = w * h * channels; + stbi__uint16 *enlarged; + + enlarged = (stbi__uint16 *) stbi__malloc(img_len*2); + if (enlarged == NULL) return (stbi__uint16 *) stbi__errpuc("outofmem", "Out of memory"); + + for (i = 0; i < img_len; ++i) + enlarged[i] = (stbi__uint16)((orig[i] << 8) + orig[i]); // replicate to high and low byte, maps 0->0, 255->0xffff + + STBI_FREE(orig); + return enlarged; +} + +static void stbi__vertical_flip(void *image, int w, int h, int bytes_per_pixel) +{ + int row; + size_t bytes_per_row = (size_t)w * bytes_per_pixel; + stbi_uc temp[2048]; + stbi_uc *bytes = (stbi_uc *)image; + + for (row = 0; row < (h>>1); row++) { + stbi_uc *row0 = bytes + row*bytes_per_row; + stbi_uc *row1 = bytes + (h - row - 1)*bytes_per_row; + // swap row0 with row1 + size_t bytes_left = bytes_per_row; + while (bytes_left) { + size_t bytes_copy = (bytes_left < sizeof(temp)) ? bytes_left : sizeof(temp); + memcpy(temp, row0, bytes_copy); + memcpy(row0, row1, bytes_copy); + memcpy(row1, temp, bytes_copy); + row0 += bytes_copy; + row1 += bytes_copy; + bytes_left -= bytes_copy; + } + } +} + +#ifndef STBI_NO_GIF +static void stbi__vertical_flip_slices(void *image, int w, int h, int z, int bytes_per_pixel) +{ + int slice; + int slice_size = w * h * bytes_per_pixel; + + stbi_uc *bytes = (stbi_uc *)image; + for (slice = 0; slice < z; ++slice) { + stbi__vertical_flip(bytes, w, h, bytes_per_pixel); + bytes += slice_size; + } +} +#endif + +static unsigned char *stbi__load_and_postprocess_8bit(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + stbi__result_info ri; + void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 8); + + if (result == NULL) + return NULL; + + if (ri.bits_per_channel != 8) { + STBI_ASSERT(ri.bits_per_channel == 16); + result = stbi__convert_16_to_8((stbi__uint16 *) result, *x, *y, req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 8; + } + + // @TODO: move stbi__convert_format to here + + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi_uc)); + } + + return (unsigned char *) result; +} + +static stbi__uint16 *stbi__load_and_postprocess_16bit(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + stbi__result_info ri; + void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 16); + + if (result == NULL) + return NULL; + + if (ri.bits_per_channel != 16) { + STBI_ASSERT(ri.bits_per_channel == 8); + result = stbi__convert_8_to_16((stbi_uc *) result, *x, *y, req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 16; + } + + // @TODO: move stbi__convert_format16 to here + // @TODO: special case RGB-to-Y (and RGBA-to-YA) for 8-bit-to-16-bit case to keep more precision + + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi__uint16)); + } + + return (stbi__uint16 *) result; +} + +#if !defined(STBI_NO_HDR) && !defined(STBI_NO_LINEAR) +static void stbi__float_postprocess(float *result, int *x, int *y, int *comp, int req_comp) +{ + if (stbi__vertically_flip_on_load && result != NULL) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(float)); + } +} +#endif + +#ifndef STBI_NO_STDIO + +#if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8) +STBI_EXTERN __declspec(dllimport) int __stdcall MultiByteToWideChar(unsigned int cp, unsigned long flags, const char *str, int cbmb, wchar_t *widestr, int cchwide); +STBI_EXTERN __declspec(dllimport) int __stdcall WideCharToMultiByte(unsigned int cp, unsigned long flags, const wchar_t *widestr, int cchwide, char *str, int cbmb, const char *defchar, int *used_default); +#endif + +#if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8) +STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input) +{ + return WideCharToMultiByte(65001 /* UTF8 */, 0, input, -1, buffer, (int) bufferlen, NULL, NULL); +} +#endif + +static FILE *stbi__fopen(char const *filename, char const *mode) +{ + FILE *f; +#if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8) + wchar_t wMode[64]; + wchar_t wFilename[1024]; + if (0 == MultiByteToWideChar(65001 /* UTF8 */, 0, filename, -1, wFilename, sizeof(wFilename))) + return 0; + + if (0 == MultiByteToWideChar(65001 /* UTF8 */, 0, mode, -1, wMode, sizeof(wMode))) + return 0; + +#if _MSC_VER >= 1400 + if (0 != _wfopen_s(&f, wFilename, wMode)) + f = 0; +#else + f = _wfopen(wFilename, wMode); +#endif + +#elif defined(_MSC_VER) && _MSC_VER >= 1400 + if (0 != fopen_s(&f, filename, mode)) + f=0; +#else + f = fopen(filename, mode); +#endif + return f; +} + + +STBIDEF stbi_uc *stbi_load(char const *filename, int *x, int *y, int *comp, int req_comp) +{ + FILE *f = stbi__fopen(filename, "rb"); + unsigned char *result; + if (!f) return stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file(f,x,y,comp,req_comp); + fclose(f); + return result; +} + +STBIDEF stbi_uc *stbi_load_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) +{ + unsigned char *result; + stbi__context s; + stbi__start_file(&s,f); + result = stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); + if (result) { + // need to 'unget' all the characters in the IO buffer + fseek(f, - (int) (s.img_buffer_end - s.img_buffer), SEEK_CUR); + } + return result; +} + +STBIDEF stbi__uint16 *stbi_load_from_file_16(FILE *f, int *x, int *y, int *comp, int req_comp) +{ + stbi__uint16 *result; + stbi__context s; + stbi__start_file(&s,f); + result = stbi__load_and_postprocess_16bit(&s,x,y,comp,req_comp); + if (result) { + // need to 'unget' all the characters in the IO buffer + fseek(f, - (int) (s.img_buffer_end - s.img_buffer), SEEK_CUR); + } + return result; +} + +STBIDEF stbi_us *stbi_load_16(char const *filename, int *x, int *y, int *comp, int req_comp) +{ + FILE *f = stbi__fopen(filename, "rb"); + stbi__uint16 *result; + if (!f) return (stbi_us *) stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file_16(f,x,y,comp,req_comp); + fclose(f); + return result; +} + + +#endif //!STBI_NO_STDIO + +STBIDEF stbi_us *stbi_load_16_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__load_and_postprocess_16bit(&s,x,y,channels_in_file,desired_channels); +} + +STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__load_and_postprocess_16bit(&s,x,y,channels_in_file,desired_channels); +} + +STBIDEF stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); +} + +STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); +} + +#ifndef STBI_NO_GIF +STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, int **delays, int *x, int *y, int *z, int *comp, int req_comp) +{ + unsigned char *result; + stbi__context s; + stbi__start_mem(&s,buffer,len); + + result = (unsigned char*) stbi__load_gif_main(&s, delays, x, y, z, comp, req_comp); + if (stbi__vertically_flip_on_load) { + stbi__vertical_flip_slices( result, *x, *y, *z, *comp ); + } + + return result; +} +#endif + +#ifndef STBI_NO_LINEAR +static float *stbi__loadf_main(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + unsigned char *data; + #ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + stbi__result_info ri; + float *hdr_data = stbi__hdr_load(s,x,y,comp,req_comp, &ri); + if (hdr_data) + stbi__float_postprocess(hdr_data,x,y,comp,req_comp); + return hdr_data; + } + #endif + data = stbi__load_and_postprocess_8bit(s, x, y, comp, req_comp); + if (data) + return stbi__ldr_to_hdr(data, *x, *y, req_comp ? req_comp : *comp); + return stbi__errpf("unknown image type", "Image not of any known type, or corrupt"); +} + +STBIDEF float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__loadf_main(&s,x,y,comp,req_comp); +} + +STBIDEF float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi__loadf_main(&s,x,y,comp,req_comp); +} + +#ifndef STBI_NO_STDIO +STBIDEF float *stbi_loadf(char const *filename, int *x, int *y, int *comp, int req_comp) +{ + float *result; + FILE *f = stbi__fopen(filename, "rb"); + if (!f) return stbi__errpf("can't fopen", "Unable to open file"); + result = stbi_loadf_from_file(f,x,y,comp,req_comp); + fclose(f); + return result; +} + +STBIDEF float *stbi_loadf_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_file(&s,f); + return stbi__loadf_main(&s,x,y,comp,req_comp); +} +#endif // !STBI_NO_STDIO + +#endif // !STBI_NO_LINEAR + +// these is-hdr-or-not is defined independent of whether STBI_NO_LINEAR is +// defined, for API simplicity; if STBI_NO_LINEAR is defined, it always +// reports false! + +STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len) +{ + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__hdr_test(&s); + #else + STBI_NOTUSED(buffer); + STBI_NOTUSED(len); + return 0; + #endif +} + +#ifndef STBI_NO_STDIO +STBIDEF int stbi_is_hdr (char const *filename) +{ + FILE *f = stbi__fopen(filename, "rb"); + int result=0; + if (f) { + result = stbi_is_hdr_from_file(f); + fclose(f); + } + return result; +} + +STBIDEF int stbi_is_hdr_from_file(FILE *f) +{ + #ifndef STBI_NO_HDR + long pos = ftell(f); + int res; + stbi__context s; + stbi__start_file(&s,f); + res = stbi__hdr_test(&s); + fseek(f, pos, SEEK_SET); + return res; + #else + STBI_NOTUSED(f); + return 0; + #endif +} +#endif // !STBI_NO_STDIO + +STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user) +{ + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi__hdr_test(&s); + #else + STBI_NOTUSED(clbk); + STBI_NOTUSED(user); + return 0; + #endif +} + +#ifndef STBI_NO_LINEAR +static float stbi__l2h_gamma=2.2f, stbi__l2h_scale=1.0f; + +STBIDEF void stbi_ldr_to_hdr_gamma(float gamma) { stbi__l2h_gamma = gamma; } +STBIDEF void stbi_ldr_to_hdr_scale(float scale) { stbi__l2h_scale = scale; } +#endif + +static float stbi__h2l_gamma_i=1.0f/2.2f, stbi__h2l_scale_i=1.0f; + +STBIDEF void stbi_hdr_to_ldr_gamma(float gamma) { stbi__h2l_gamma_i = 1/gamma; } +STBIDEF void stbi_hdr_to_ldr_scale(float scale) { stbi__h2l_scale_i = 1/scale; } + + +////////////////////////////////////////////////////////////////////////////// +// +// Common code used by all image loaders +// + +enum +{ + STBI__SCAN_load=0, + STBI__SCAN_type, + STBI__SCAN_header +}; + +static void stbi__refill_buffer(stbi__context *s) +{ + int n = (s->io.read)(s->io_user_data,(char*)s->buffer_start,s->buflen); + if (n == 0) { + // at end of file, treat same as if from memory, but need to handle case + // where s->img_buffer isn't pointing to safe memory, e.g. 0-byte file + s->read_from_callbacks = 0; + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start+1; + *s->img_buffer = 0; + } else { + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start + n; + } +} + +stbi_inline static stbi_uc stbi__get8(stbi__context *s) +{ + if (s->img_buffer < s->img_buffer_end) + return *s->img_buffer++; + if (s->read_from_callbacks) { + stbi__refill_buffer(s); + return *s->img_buffer++; + } + return 0; +} + +stbi_inline static int stbi__at_eof(stbi__context *s) +{ + if (s->io.read) { + if (!(s->io.eof)(s->io_user_data)) return 0; + // if feof() is true, check if buffer = end + // special case: we've only got the special 0 character at the end + if (s->read_from_callbacks == 0) return 1; + } + + return s->img_buffer >= s->img_buffer_end; +} + +static void stbi__skip(stbi__context *s, int n) +{ + if (n < 0) { + s->img_buffer = s->img_buffer_end; + return; + } + if (s->io.read) { + int blen = (int) (s->img_buffer_end - s->img_buffer); + if (blen < n) { + s->img_buffer = s->img_buffer_end; + (s->io.skip)(s->io_user_data, n - blen); + return; + } + } + s->img_buffer += n; +} + +static int stbi__getn(stbi__context *s, stbi_uc *buffer, int n) +{ + if (s->io.read) { + int blen = (int) (s->img_buffer_end - s->img_buffer); + if (blen < n) { + int res, count; + + memcpy(buffer, s->img_buffer, blen); + + count = (s->io.read)(s->io_user_data, (char*) buffer + blen, n - blen); + res = (count == (n-blen)); + s->img_buffer = s->img_buffer_end; + return res; + } + } + + if (s->img_buffer+n <= s->img_buffer_end) { + memcpy(buffer, s->img_buffer, n); + s->img_buffer += n; + return 1; + } else + return 0; +} + +static int stbi__get16be(stbi__context *s) +{ + int z = stbi__get8(s); + return (z << 8) + stbi__get8(s); +} + +static stbi__uint32 stbi__get32be(stbi__context *s) +{ + stbi__uint32 z = stbi__get16be(s); + return (z << 16) + stbi__get16be(s); +} + +#if defined(STBI_NO_BMP) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) +// nothing +#else +static int stbi__get16le(stbi__context *s) +{ + int z = stbi__get8(s); + return z + (stbi__get8(s) << 8); +} +#endif + +#ifndef STBI_NO_BMP +static stbi__uint32 stbi__get32le(stbi__context *s) +{ + stbi__uint32 z = stbi__get16le(s); + return z + (stbi__get16le(s) << 16); +} +#endif + +#define STBI__BYTECAST(x) ((stbi_uc) ((x) & 255)) // truncate int to byte without warnings + + +////////////////////////////////////////////////////////////////////////////// +// +// generic converter from built-in img_n to req_comp +// individual types do this automatically as much as possible (e.g. jpeg +// does all cases internally since it needs to colorspace convert anyway, +// and it never has alpha, so very few cases ). png can automatically +// interleave an alpha=255 channel, but falls back to this for other cases +// +// assume data buffer is malloced, so malloc a new one and free that one +// only failure mode is malloc failing + +static stbi_uc stbi__compute_y(int r, int g, int b) +{ + return (stbi_uc) (((r*77) + (g*150) + (29*b)) >> 8); +} + +static unsigned char *stbi__convert_format(unsigned char *data, int img_n, int req_comp, unsigned int x, unsigned int y) +{ + int i,j; + unsigned char *good; + + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + + good = (unsigned char *) stbi__malloc_mad3(req_comp, x, y, 0); + if (good == NULL) { + STBI_FREE(data); + return stbi__errpuc("outofmem", "Out of memory"); + } + + for (j=0; j < (int) y; ++j) { + unsigned char *src = data + j * x * img_n ; + unsigned char *dest = good + j * x * req_comp; + + #define STBI__COMBO(a,b) ((a)*8+(b)) + #define STBI__CASE(a,b) case STBI__COMBO(a,b): for(i=x-1; i >= 0; --i, src += a, dest += b) + // convert source image with img_n components to one with req_comp components; + // avoid switch per pixel, so use switch per scanline and massive macros + switch (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1,2) { dest[0]=src[0]; dest[1]=255; } break; + STBI__CASE(1,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; + STBI__CASE(1,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=255; } break; + STBI__CASE(2,1) { dest[0]=src[0]; } break; + STBI__CASE(2,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; + STBI__CASE(2,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=src[1]; } break; + STBI__CASE(3,4) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2];dest[3]=255; } break; + STBI__CASE(3,1) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); } break; + STBI__CASE(3,2) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); dest[1] = 255; } break; + STBI__CASE(4,1) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); } break; + STBI__CASE(4,2) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); dest[1] = src[3]; } break; + STBI__CASE(4,3) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2]; } break; + default: STBI_ASSERT(0); + } + #undef STBI__CASE + } + + STBI_FREE(data); + return good; +} + +static stbi__uint16 stbi__compute_y_16(int r, int g, int b) +{ + return (stbi__uint16) (((r*77) + (g*150) + (29*b)) >> 8); +} + +static stbi__uint16 *stbi__convert_format16(stbi__uint16 *data, int img_n, int req_comp, unsigned int x, unsigned int y) +{ + int i,j; + stbi__uint16 *good; + + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + + good = (stbi__uint16 *) stbi__malloc(req_comp * x * y * 2); + if (good == NULL) { + STBI_FREE(data); + return (stbi__uint16 *) stbi__errpuc("outofmem", "Out of memory"); + } + + for (j=0; j < (int) y; ++j) { + stbi__uint16 *src = data + j * x * img_n ; + stbi__uint16 *dest = good + j * x * req_comp; + + #define STBI__COMBO(a,b) ((a)*8+(b)) + #define STBI__CASE(a,b) case STBI__COMBO(a,b): for(i=x-1; i >= 0; --i, src += a, dest += b) + // convert source image with img_n components to one with req_comp components; + // avoid switch per pixel, so use switch per scanline and massive macros + switch (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1,2) { dest[0]=src[0]; dest[1]=0xffff; } break; + STBI__CASE(1,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; + STBI__CASE(1,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=0xffff; } break; + STBI__CASE(2,1) { dest[0]=src[0]; } break; + STBI__CASE(2,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; + STBI__CASE(2,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=src[1]; } break; + STBI__CASE(3,4) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2];dest[3]=0xffff; } break; + STBI__CASE(3,1) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); } break; + STBI__CASE(3,2) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); dest[1] = 0xffff; } break; + STBI__CASE(4,1) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); } break; + STBI__CASE(4,2) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); dest[1] = src[3]; } break; + STBI__CASE(4,3) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2]; } break; + default: STBI_ASSERT(0); + } + #undef STBI__CASE + } + + STBI_FREE(data); + return good; +} + +#ifndef STBI_NO_LINEAR +static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp) +{ + int i,k,n; + float *output; + if (!data) return NULL; + output = (float *) stbi__malloc_mad4(x, y, comp, sizeof(float), 0); + if (output == NULL) { STBI_FREE(data); return stbi__errpf("outofmem", "Out of memory"); } + // compute number of non-alpha components + if (comp & 1) n = comp; else n = comp-1; + for (i=0; i < x*y; ++i) { + for (k=0; k < n; ++k) { + output[i*comp + k] = (float) (pow(data[i*comp+k]/255.0f, stbi__l2h_gamma) * stbi__l2h_scale); + } + } + if (n < comp) { + for (i=0; i < x*y; ++i) { + output[i*comp + n] = data[i*comp + n]/255.0f; + } + } + STBI_FREE(data); + return output; +} +#endif + +#ifndef STBI_NO_HDR +#define stbi__float2int(x) ((int) (x)) +static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp) +{ + int i,k,n; + stbi_uc *output; + if (!data) return NULL; + output = (stbi_uc *) stbi__malloc_mad3(x, y, comp, 0); + if (output == NULL) { STBI_FREE(data); return stbi__errpuc("outofmem", "Out of memory"); } + // compute number of non-alpha components + if (comp & 1) n = comp; else n = comp-1; + for (i=0; i < x*y; ++i) { + for (k=0; k < n; ++k) { + float z = (float) pow(data[i*comp+k]*stbi__h2l_scale_i, stbi__h2l_gamma_i) * 255 + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i*comp + k] = (stbi_uc) stbi__float2int(z); + } + if (k < comp) { + float z = data[i*comp+k] * 255 + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i*comp + k] = (stbi_uc) stbi__float2int(z); + } + } + STBI_FREE(data); + return output; +} +#endif + +////////////////////////////////////////////////////////////////////////////// +// +// "baseline" JPEG/JFIF decoder +// +// simple implementation +// - doesn't support delayed output of y-dimension +// - simple interface (only one output format: 8-bit interleaved RGB) +// - doesn't try to recover corrupt jpegs +// - doesn't allow partial loading, loading multiple at once +// - still fast on x86 (copying globals into locals doesn't help x86) +// - allocates lots of intermediate memory (full size of all components) +// - non-interleaved case requires this anyway +// - allows good upsampling (see next) +// high-quality +// - upsampled channels are bilinearly interpolated, even across blocks +// - quality integer IDCT derived from IJG's 'slow' +// performance +// - fast huffman; reasonable integer IDCT +// - some SIMD kernels for common paths on targets with SSE2/NEON +// - uses a lot of intermediate memory, could cache poorly + +#ifndef STBI_NO_JPEG + +// huffman decoding acceleration +#define FAST_BITS 9 // larger handles more cases; smaller stomps less cache + +typedef struct +{ + stbi_uc fast[1 << FAST_BITS]; + // weirdly, repacking this into AoS is a 10% speed loss, instead of a win + stbi__uint16 code[256]; + stbi_uc values[256]; + stbi_uc size[257]; + unsigned int maxcode[18]; + int delta[17]; // old 'firstsymbol' - old 'firstcode' +} stbi__huffman; + +typedef struct +{ + stbi__context *s; + stbi__huffman huff_dc[4]; + stbi__huffman huff_ac[4]; + stbi__uint16 dequant[4][64]; + stbi__int16 fast_ac[4][1 << FAST_BITS]; + +// sizes for components, interleaved MCUs + int img_h_max, img_v_max; + int img_mcu_x, img_mcu_y; + int img_mcu_w, img_mcu_h; + +// definition of jpeg image component + struct + { + int id; + int h,v; + int tq; + int hd,ha; + int dc_pred; + + int x,y,w2,h2; + stbi_uc *data; + void *raw_data, *raw_coeff; + stbi_uc *linebuf; + short *coeff; // progressive only + int coeff_w, coeff_h; // number of 8x8 coefficient blocks + } img_comp[4]; + + stbi__uint32 code_buffer; // jpeg entropy-coded buffer + int code_bits; // number of valid bits + unsigned char marker; // marker seen while filling entropy buffer + int nomore; // flag if we saw a marker so must stop + + int progressive; + int spec_start; + int spec_end; + int succ_high; + int succ_low; + int eob_run; + int jfif; + int app14_color_transform; // Adobe APP14 tag + int rgb; + + int scan_n, order[4]; + int restart_interval, todo; + +// kernels + void (*idct_block_kernel)(stbi_uc *out, int out_stride, short data[64]); + void (*YCbCr_to_RGB_kernel)(stbi_uc *out, const stbi_uc *y, const stbi_uc *pcb, const stbi_uc *pcr, int count, int step); + stbi_uc *(*resample_row_hv_2_kernel)(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs); +} stbi__jpeg; + +static int stbi__build_huffman(stbi__huffman *h, int *count) +{ + int i,j,k=0; + unsigned int code; + // build size list for each symbol (from JPEG spec) + for (i=0; i < 16; ++i) + for (j=0; j < count[i]; ++j) + h->size[k++] = (stbi_uc) (i+1); + h->size[k] = 0; + + // compute actual symbols (from jpeg spec) + code = 0; + k = 0; + for(j=1; j <= 16; ++j) { + // compute delta to add to code to compute symbol id + h->delta[j] = k - code; + if (h->size[k] == j) { + while (h->size[k] == j) + h->code[k++] = (stbi__uint16) (code++); + if (code-1 >= (1u << j)) return stbi__err("bad code lengths","Corrupt JPEG"); + } + // compute largest code + 1 for this size, preshifted as needed later + h->maxcode[j] = code << (16-j); + code <<= 1; + } + h->maxcode[j] = 0xffffffff; + + // build non-spec acceleration table; 255 is flag for not-accelerated + memset(h->fast, 255, 1 << FAST_BITS); + for (i=0; i < k; ++i) { + int s = h->size[i]; + if (s <= FAST_BITS) { + int c = h->code[i] << (FAST_BITS-s); + int m = 1 << (FAST_BITS-s); + for (j=0; j < m; ++j) { + h->fast[c+j] = (stbi_uc) i; + } + } + } + return 1; +} + +// build a table that decodes both magnitude and value of small ACs in +// one go. +static void stbi__build_fast_ac(stbi__int16 *fast_ac, stbi__huffman *h) +{ + int i; + for (i=0; i < (1 << FAST_BITS); ++i) { + stbi_uc fast = h->fast[i]; + fast_ac[i] = 0; + if (fast < 255) { + int rs = h->values[fast]; + int run = (rs >> 4) & 15; + int magbits = rs & 15; + int len = h->size[fast]; + + if (magbits && len + magbits <= FAST_BITS) { + // magnitude code followed by receive_extend code + int k = ((i << len) & ((1 << FAST_BITS) - 1)) >> (FAST_BITS - magbits); + int m = 1 << (magbits - 1); + if (k < m) k += (~0U << magbits) + 1; + // if the result is small enough, we can fit it in fast_ac table + if (k >= -128 && k <= 127) + fast_ac[i] = (stbi__int16) ((k * 256) + (run * 16) + (len + magbits)); + } + } + } +} + +static void stbi__grow_buffer_unsafe(stbi__jpeg *j) +{ + do { + unsigned int b = j->nomore ? 0 : stbi__get8(j->s); + if (b == 0xff) { + int c = stbi__get8(j->s); + while (c == 0xff) c = stbi__get8(j->s); // consume fill bytes + if (c != 0) { + j->marker = (unsigned char) c; + j->nomore = 1; + return; + } + } + j->code_buffer |= b << (24 - j->code_bits); + j->code_bits += 8; + } while (j->code_bits <= 24); +} + +// (1 << n) - 1 +static const stbi__uint32 stbi__bmask[17]={0,1,3,7,15,31,63,127,255,511,1023,2047,4095,8191,16383,32767,65535}; + +// decode a jpeg huffman value from the bitstream +stbi_inline static int stbi__jpeg_huff_decode(stbi__jpeg *j, stbi__huffman *h) +{ + unsigned int temp; + int c,k; + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + + // look at the top FAST_BITS and determine what symbol ID it is, + // if the code is <= FAST_BITS + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); + k = h->fast[c]; + if (k < 255) { + int s = h->size[k]; + if (s > j->code_bits) + return -1; + j->code_buffer <<= s; + j->code_bits -= s; + return h->values[k]; + } + + // naive test is to shift the code_buffer down so k bits are + // valid, then test against maxcode. To speed this up, we've + // preshifted maxcode left so that it has (16-k) 0s at the + // end; in other words, regardless of the number of bits, it + // wants to be compared against something shifted to have 16; + // that way we don't need to shift inside the loop. + temp = j->code_buffer >> 16; + for (k=FAST_BITS+1 ; ; ++k) + if (temp < h->maxcode[k]) + break; + if (k == 17) { + // error! code not found + j->code_bits -= 16; + return -1; + } + + if (k > j->code_bits) + return -1; + + // convert the huffman code to the symbol id + c = ((j->code_buffer >> (32 - k)) & stbi__bmask[k]) + h->delta[k]; + STBI_ASSERT((((j->code_buffer) >> (32 - h->size[c])) & stbi__bmask[h->size[c]]) == h->code[c]); + + // convert the id to a symbol + j->code_bits -= k; + j->code_buffer <<= k; + return h->values[c]; +} + +// bias[n] = (-1<code_bits < n) stbi__grow_buffer_unsafe(j); + + sgn = (stbi__int32)j->code_buffer >> 31; // sign bit is always in MSB + k = stbi_lrot(j->code_buffer, n); + STBI_ASSERT(n >= 0 && n < (int) (sizeof(stbi__bmask)/sizeof(*stbi__bmask))); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k + (stbi__jbias[n] & ~sgn); +} + +// get some unsigned bits +stbi_inline static int stbi__jpeg_get_bits(stbi__jpeg *j, int n) +{ + unsigned int k; + if (j->code_bits < n) stbi__grow_buffer_unsafe(j); + k = stbi_lrot(j->code_buffer, n); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k; +} + +stbi_inline static int stbi__jpeg_get_bit(stbi__jpeg *j) +{ + unsigned int k; + if (j->code_bits < 1) stbi__grow_buffer_unsafe(j); + k = j->code_buffer; + j->code_buffer <<= 1; + --j->code_bits; + return k & 0x80000000; +} + +// given a value that's at position X in the zigzag stream, +// where does it appear in the 8x8 matrix coded as row-major? +static const stbi_uc stbi__jpeg_dezigzag[64+15] = +{ + 0, 1, 8, 16, 9, 2, 3, 10, + 17, 24, 32, 25, 18, 11, 4, 5, + 12, 19, 26, 33, 40, 48, 41, 34, + 27, 20, 13, 6, 7, 14, 21, 28, + 35, 42, 49, 56, 57, 50, 43, 36, + 29, 22, 15, 23, 30, 37, 44, 51, + 58, 59, 52, 45, 38, 31, 39, 46, + 53, 60, 61, 54, 47, 55, 62, 63, + // let corrupt input sample past end + 63, 63, 63, 63, 63, 63, 63, 63, + 63, 63, 63, 63, 63, 63, 63 +}; + +// decode one 64-entry block-- +static int stbi__jpeg_decode_block(stbi__jpeg *j, short data[64], stbi__huffman *hdc, stbi__huffman *hac, stbi__int16 *fac, int b, stbi__uint16 *dequant) +{ + int diff,dc,k; + int t; + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + t = stbi__jpeg_huff_decode(j, hdc); + if (t < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + + // 0 all the ac values now so we can do it 32-bits at a time + memset(data,0,64*sizeof(data[0])); + + diff = t ? stbi__extend_receive(j, t) : 0; + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + data[0] = (short) (dc * dequant[0]); + + // decode AC components, see JPEG spec + k = 1; + do { + unsigned int zig; + int c,r,s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + j->code_buffer <<= s; + j->code_bits -= s; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) ((r >> 8) * dequant[zig]); + } else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (rs != 0xf0) break; // end block + k += 16; + } else { + k += r; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) (stbi__extend_receive(j,s) * dequant[zig]); + } + } + } while (k < 64); + return 1; +} + +static int stbi__jpeg_decode_block_prog_dc(stbi__jpeg *j, short data[64], stbi__huffman *hdc, int b) +{ + int diff,dc; + int t; + if (j->spec_end != 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + + if (j->succ_high == 0) { + // first scan for DC coefficient, must be first + memset(data,0,64*sizeof(data[0])); // 0 all the ac values now + t = stbi__jpeg_huff_decode(j, hdc); + diff = t ? stbi__extend_receive(j, t) : 0; + + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + data[0] = (short) (dc << j->succ_low); + } else { + // refinement scan for DC coefficient + if (stbi__jpeg_get_bit(j)) + data[0] += (short) (1 << j->succ_low); + } + return 1; +} + +// @OPTIMIZE: store non-zigzagged during the decode passes, +// and only de-zigzag when dequantizing +static int stbi__jpeg_decode_block_prog_ac(stbi__jpeg *j, short data[64], stbi__huffman *hac, stbi__int16 *fac) +{ + int k; + if (j->spec_start == 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + + if (j->succ_high == 0) { + int shift = j->succ_low; + + if (j->eob_run) { + --j->eob_run; + return 1; + } + + k = j->spec_start; + do { + unsigned int zig; + int c,r,s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + j->code_buffer <<= s; + j->code_bits -= s; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) ((r >> 8) << shift); + } else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r); + if (r) + j->eob_run += stbi__jpeg_get_bits(j, r); + --j->eob_run; + break; + } + k += 16; + } else { + k += r; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) (stbi__extend_receive(j,s) << shift); + } + } + } while (k <= j->spec_end); + } else { + // refinement scan for these AC coefficients + + short bit = (short) (1 << j->succ_low); + + if (j->eob_run) { + --j->eob_run; + for (k = j->spec_start; k <= j->spec_end; ++k) { + short *p = &data[stbi__jpeg_dezigzag[k]]; + if (*p != 0) + if (stbi__jpeg_get_bit(j)) + if ((*p & bit)==0) { + if (*p > 0) + *p += bit; + else + *p -= bit; + } + } + } else { + k = j->spec_start; + do { + int r,s; + int rs = stbi__jpeg_huff_decode(j, hac); // @OPTIMIZE see if we can use the fast path here, advance-by-r is so slow, eh + if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r) - 1; + if (r) + j->eob_run += stbi__jpeg_get_bits(j, r); + r = 64; // force end of block + } else { + // r=15 s=0 should write 16 0s, so we just do + // a run of 15 0s and then write s (which is 0), + // so we don't have to do anything special here + } + } else { + if (s != 1) return stbi__err("bad huffman code", "Corrupt JPEG"); + // sign bit + if (stbi__jpeg_get_bit(j)) + s = bit; + else + s = -bit; + } + + // advance by r + while (k <= j->spec_end) { + short *p = &data[stbi__jpeg_dezigzag[k++]]; + if (*p != 0) { + if (stbi__jpeg_get_bit(j)) + if ((*p & bit)==0) { + if (*p > 0) + *p += bit; + else + *p -= bit; + } + } else { + if (r == 0) { + *p = (short) s; + break; + } + --r; + } + } + } while (k <= j->spec_end); + } + } + return 1; +} + +// take a -128..127 value and stbi__clamp it and convert to 0..255 +stbi_inline static stbi_uc stbi__clamp(int x) +{ + // trick to use a single test to catch both cases + if ((unsigned int) x > 255) { + if (x < 0) return 0; + if (x > 255) return 255; + } + return (stbi_uc) x; +} + +#define stbi__f2f(x) ((int) (((x) * 4096 + 0.5))) +#define stbi__fsh(x) ((x) * 4096) + +// derived from jidctint -- DCT_ISLOW +#define STBI__IDCT_1D(s0,s1,s2,s3,s4,s5,s6,s7) \ + int t0,t1,t2,t3,p1,p2,p3,p4,p5,x0,x1,x2,x3; \ + p2 = s2; \ + p3 = s6; \ + p1 = (p2+p3) * stbi__f2f(0.5411961f); \ + t2 = p1 + p3*stbi__f2f(-1.847759065f); \ + t3 = p1 + p2*stbi__f2f( 0.765366865f); \ + p2 = s0; \ + p3 = s4; \ + t0 = stbi__fsh(p2+p3); \ + t1 = stbi__fsh(p2-p3); \ + x0 = t0+t3; \ + x3 = t0-t3; \ + x1 = t1+t2; \ + x2 = t1-t2; \ + t0 = s7; \ + t1 = s5; \ + t2 = s3; \ + t3 = s1; \ + p3 = t0+t2; \ + p4 = t1+t3; \ + p1 = t0+t3; \ + p2 = t1+t2; \ + p5 = (p3+p4)*stbi__f2f( 1.175875602f); \ + t0 = t0*stbi__f2f( 0.298631336f); \ + t1 = t1*stbi__f2f( 2.053119869f); \ + t2 = t2*stbi__f2f( 3.072711026f); \ + t3 = t3*stbi__f2f( 1.501321110f); \ + p1 = p5 + p1*stbi__f2f(-0.899976223f); \ + p2 = p5 + p2*stbi__f2f(-2.562915447f); \ + p3 = p3*stbi__f2f(-1.961570560f); \ + p4 = p4*stbi__f2f(-0.390180644f); \ + t3 += p1+p4; \ + t2 += p2+p3; \ + t1 += p2+p4; \ + t0 += p1+p3; + +static void stbi__idct_block(stbi_uc *out, int out_stride, short data[64]) +{ + int i,val[64],*v=val; + stbi_uc *o; + short *d = data; + + // columns + for (i=0; i < 8; ++i,++d, ++v) { + // if all zeroes, shortcut -- this avoids dequantizing 0s and IDCTing + if (d[ 8]==0 && d[16]==0 && d[24]==0 && d[32]==0 + && d[40]==0 && d[48]==0 && d[56]==0) { + // no shortcut 0 seconds + // (1|2|3|4|5|6|7)==0 0 seconds + // all separate -0.047 seconds + // 1 && 2|3 && 4|5 && 6|7: -0.047 seconds + int dcterm = d[0]*4; + v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm; + } else { + STBI__IDCT_1D(d[ 0],d[ 8],d[16],d[24],d[32],d[40],d[48],d[56]) + // constants scaled things up by 1<<12; let's bring them back + // down, but keep 2 extra bits of precision + x0 += 512; x1 += 512; x2 += 512; x3 += 512; + v[ 0] = (x0+t3) >> 10; + v[56] = (x0-t3) >> 10; + v[ 8] = (x1+t2) >> 10; + v[48] = (x1-t2) >> 10; + v[16] = (x2+t1) >> 10; + v[40] = (x2-t1) >> 10; + v[24] = (x3+t0) >> 10; + v[32] = (x3-t0) >> 10; + } + } + + for (i=0, v=val, o=out; i < 8; ++i,v+=8,o+=out_stride) { + // no fast case since the first 1D IDCT spread components out + STBI__IDCT_1D(v[0],v[1],v[2],v[3],v[4],v[5],v[6],v[7]) + // constants scaled things up by 1<<12, plus we had 1<<2 from first + // loop, plus horizontal and vertical each scale by sqrt(8) so together + // we've got an extra 1<<3, so 1<<17 total we need to remove. + // so we want to round that, which means adding 0.5 * 1<<17, + // aka 65536. Also, we'll end up with -128 to 127 that we want + // to encode as 0..255 by adding 128, so we'll add that before the shift + x0 += 65536 + (128<<17); + x1 += 65536 + (128<<17); + x2 += 65536 + (128<<17); + x3 += 65536 + (128<<17); + // tried computing the shifts into temps, or'ing the temps to see + // if any were out of range, but that was slower + o[0] = stbi__clamp((x0+t3) >> 17); + o[7] = stbi__clamp((x0-t3) >> 17); + o[1] = stbi__clamp((x1+t2) >> 17); + o[6] = stbi__clamp((x1-t2) >> 17); + o[2] = stbi__clamp((x2+t1) >> 17); + o[5] = stbi__clamp((x2-t1) >> 17); + o[3] = stbi__clamp((x3+t0) >> 17); + o[4] = stbi__clamp((x3-t0) >> 17); + } +} + +#ifdef STBI_SSE2 +// sse2 integer IDCT. not the fastest possible implementation but it +// produces bit-identical results to the generic C version so it's +// fully "transparent". +static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) +{ + // This is constructed to match our regular (generic) integer IDCT exactly. + __m128i row0, row1, row2, row3, row4, row5, row6, row7; + __m128i tmp; + + // dot product constant: even elems=x, odd elems=y + #define dct_const(x,y) _mm_setr_epi16((x),(y),(x),(y),(x),(y),(x),(y)) + + // out(0) = c0[even]*x + c0[odd]*y (c0, x, y 16-bit, out 32-bit) + // out(1) = c1[even]*x + c1[odd]*y + #define dct_rot(out0,out1, x,y,c0,c1) \ + __m128i c0##lo = _mm_unpacklo_epi16((x),(y)); \ + __m128i c0##hi = _mm_unpackhi_epi16((x),(y)); \ + __m128i out0##_l = _mm_madd_epi16(c0##lo, c0); \ + __m128i out0##_h = _mm_madd_epi16(c0##hi, c0); \ + __m128i out1##_l = _mm_madd_epi16(c0##lo, c1); \ + __m128i out1##_h = _mm_madd_epi16(c0##hi, c1) + + // out = in << 12 (in 16-bit, out 32-bit) + #define dct_widen(out, in) \ + __m128i out##_l = _mm_srai_epi32(_mm_unpacklo_epi16(_mm_setzero_si128(), (in)), 4); \ + __m128i out##_h = _mm_srai_epi32(_mm_unpackhi_epi16(_mm_setzero_si128(), (in)), 4) + + // wide add + #define dct_wadd(out, a, b) \ + __m128i out##_l = _mm_add_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_add_epi32(a##_h, b##_h) + + // wide sub + #define dct_wsub(out, a, b) \ + __m128i out##_l = _mm_sub_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_sub_epi32(a##_h, b##_h) + + // butterfly a/b, add bias, then shift by "s" and pack + #define dct_bfly32o(out0, out1, a,b,bias,s) \ + { \ + __m128i abiased_l = _mm_add_epi32(a##_l, bias); \ + __m128i abiased_h = _mm_add_epi32(a##_h, bias); \ + dct_wadd(sum, abiased, b); \ + dct_wsub(dif, abiased, b); \ + out0 = _mm_packs_epi32(_mm_srai_epi32(sum_l, s), _mm_srai_epi32(sum_h, s)); \ + out1 = _mm_packs_epi32(_mm_srai_epi32(dif_l, s), _mm_srai_epi32(dif_h, s)); \ + } + + // 8-bit interleave step (for transposes) + #define dct_interleave8(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi8(a, b); \ + b = _mm_unpackhi_epi8(tmp, b) + + // 16-bit interleave step (for transposes) + #define dct_interleave16(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi16(a, b); \ + b = _mm_unpackhi_epi16(tmp, b) + + #define dct_pass(bias,shift) \ + { \ + /* even part */ \ + dct_rot(t2e,t3e, row2,row6, rot0_0,rot0_1); \ + __m128i sum04 = _mm_add_epi16(row0, row4); \ + __m128i dif04 = _mm_sub_epi16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + dct_rot(y0o,y2o, row7,row3, rot2_0,rot2_1); \ + dct_rot(y1o,y3o, row5,row1, rot3_0,rot3_1); \ + __m128i sum17 = _mm_add_epi16(row1, row7); \ + __m128i sum35 = _mm_add_epi16(row3, row5); \ + dct_rot(y4o,y5o, sum17,sum35, rot1_0,rot1_1); \ + dct_wadd(x4, y0o, y4o); \ + dct_wadd(x5, y1o, y5o); \ + dct_wadd(x6, y2o, y5o); \ + dct_wadd(x7, y3o, y4o); \ + dct_bfly32o(row0,row7, x0,x7,bias,shift); \ + dct_bfly32o(row1,row6, x1,x6,bias,shift); \ + dct_bfly32o(row2,row5, x2,x5,bias,shift); \ + dct_bfly32o(row3,row4, x3,x4,bias,shift); \ + } + + __m128i rot0_0 = dct_const(stbi__f2f(0.5411961f), stbi__f2f(0.5411961f) + stbi__f2f(-1.847759065f)); + __m128i rot0_1 = dct_const(stbi__f2f(0.5411961f) + stbi__f2f( 0.765366865f), stbi__f2f(0.5411961f)); + __m128i rot1_0 = dct_const(stbi__f2f(1.175875602f) + stbi__f2f(-0.899976223f), stbi__f2f(1.175875602f)); + __m128i rot1_1 = dct_const(stbi__f2f(1.175875602f), stbi__f2f(1.175875602f) + stbi__f2f(-2.562915447f)); + __m128i rot2_0 = dct_const(stbi__f2f(-1.961570560f) + stbi__f2f( 0.298631336f), stbi__f2f(-1.961570560f)); + __m128i rot2_1 = dct_const(stbi__f2f(-1.961570560f), stbi__f2f(-1.961570560f) + stbi__f2f( 3.072711026f)); + __m128i rot3_0 = dct_const(stbi__f2f(-0.390180644f) + stbi__f2f( 2.053119869f), stbi__f2f(-0.390180644f)); + __m128i rot3_1 = dct_const(stbi__f2f(-0.390180644f), stbi__f2f(-0.390180644f) + stbi__f2f( 1.501321110f)); + + // rounding biases in column/row passes, see stbi__idct_block for explanation. + __m128i bias_0 = _mm_set1_epi32(512); + __m128i bias_1 = _mm_set1_epi32(65536 + (128<<17)); + + // load + row0 = _mm_load_si128((const __m128i *) (data + 0*8)); + row1 = _mm_load_si128((const __m128i *) (data + 1*8)); + row2 = _mm_load_si128((const __m128i *) (data + 2*8)); + row3 = _mm_load_si128((const __m128i *) (data + 3*8)); + row4 = _mm_load_si128((const __m128i *) (data + 4*8)); + row5 = _mm_load_si128((const __m128i *) (data + 5*8)); + row6 = _mm_load_si128((const __m128i *) (data + 6*8)); + row7 = _mm_load_si128((const __m128i *) (data + 7*8)); + + // column pass + dct_pass(bias_0, 10); + + { + // 16bit 8x8 transpose pass 1 + dct_interleave16(row0, row4); + dct_interleave16(row1, row5); + dct_interleave16(row2, row6); + dct_interleave16(row3, row7); + + // transpose pass 2 + dct_interleave16(row0, row2); + dct_interleave16(row1, row3); + dct_interleave16(row4, row6); + dct_interleave16(row5, row7); + + // transpose pass 3 + dct_interleave16(row0, row1); + dct_interleave16(row2, row3); + dct_interleave16(row4, row5); + dct_interleave16(row6, row7); + } + + // row pass + dct_pass(bias_1, 17); + + { + // pack + __m128i p0 = _mm_packus_epi16(row0, row1); // a0a1a2a3...a7b0b1b2b3...b7 + __m128i p1 = _mm_packus_epi16(row2, row3); + __m128i p2 = _mm_packus_epi16(row4, row5); + __m128i p3 = _mm_packus_epi16(row6, row7); + + // 8bit 8x8 transpose pass 1 + dct_interleave8(p0, p2); // a0e0a1e1... + dct_interleave8(p1, p3); // c0g0c1g1... + + // transpose pass 2 + dct_interleave8(p0, p1); // a0c0e0g0... + dct_interleave8(p2, p3); // b0d0f0h0... + + // transpose pass 3 + dct_interleave8(p0, p2); // a0b0c0d0... + dct_interleave8(p1, p3); // a4b4c4d4... + + // store + _mm_storel_epi64((__m128i *) out, p0); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p0, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i *) out, p2); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p2, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i *) out, p1); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p1, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i *) out, p3); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p3, 0x4e)); + } + +#undef dct_const +#undef dct_rot +#undef dct_widen +#undef dct_wadd +#undef dct_wsub +#undef dct_bfly32o +#undef dct_interleave8 +#undef dct_interleave16 +#undef dct_pass +} + +#endif // STBI_SSE2 + +#ifdef STBI_NEON + +// NEON integer IDCT. should produce bit-identical +// results to the generic C version. +static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) +{ + int16x8_t row0, row1, row2, row3, row4, row5, row6, row7; + + int16x4_t rot0_0 = vdup_n_s16(stbi__f2f(0.5411961f)); + int16x4_t rot0_1 = vdup_n_s16(stbi__f2f(-1.847759065f)); + int16x4_t rot0_2 = vdup_n_s16(stbi__f2f( 0.765366865f)); + int16x4_t rot1_0 = vdup_n_s16(stbi__f2f( 1.175875602f)); + int16x4_t rot1_1 = vdup_n_s16(stbi__f2f(-0.899976223f)); + int16x4_t rot1_2 = vdup_n_s16(stbi__f2f(-2.562915447f)); + int16x4_t rot2_0 = vdup_n_s16(stbi__f2f(-1.961570560f)); + int16x4_t rot2_1 = vdup_n_s16(stbi__f2f(-0.390180644f)); + int16x4_t rot3_0 = vdup_n_s16(stbi__f2f( 0.298631336f)); + int16x4_t rot3_1 = vdup_n_s16(stbi__f2f( 2.053119869f)); + int16x4_t rot3_2 = vdup_n_s16(stbi__f2f( 3.072711026f)); + int16x4_t rot3_3 = vdup_n_s16(stbi__f2f( 1.501321110f)); + +#define dct_long_mul(out, inq, coeff) \ + int32x4_t out##_l = vmull_s16(vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmull_s16(vget_high_s16(inq), coeff) + +#define dct_long_mac(out, acc, inq, coeff) \ + int32x4_t out##_l = vmlal_s16(acc##_l, vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmlal_s16(acc##_h, vget_high_s16(inq), coeff) + +#define dct_widen(out, inq) \ + int32x4_t out##_l = vshll_n_s16(vget_low_s16(inq), 12); \ + int32x4_t out##_h = vshll_n_s16(vget_high_s16(inq), 12) + +// wide add +#define dct_wadd(out, a, b) \ + int32x4_t out##_l = vaddq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vaddq_s32(a##_h, b##_h) + +// wide sub +#define dct_wsub(out, a, b) \ + int32x4_t out##_l = vsubq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vsubq_s32(a##_h, b##_h) + +// butterfly a/b, then shift using "shiftop" by "s" and pack +#define dct_bfly32o(out0,out1, a,b,shiftop,s) \ + { \ + dct_wadd(sum, a, b); \ + dct_wsub(dif, a, b); \ + out0 = vcombine_s16(shiftop(sum_l, s), shiftop(sum_h, s)); \ + out1 = vcombine_s16(shiftop(dif_l, s), shiftop(dif_h, s)); \ + } + +#define dct_pass(shiftop, shift) \ + { \ + /* even part */ \ + int16x8_t sum26 = vaddq_s16(row2, row6); \ + dct_long_mul(p1e, sum26, rot0_0); \ + dct_long_mac(t2e, p1e, row6, rot0_1); \ + dct_long_mac(t3e, p1e, row2, rot0_2); \ + int16x8_t sum04 = vaddq_s16(row0, row4); \ + int16x8_t dif04 = vsubq_s16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + int16x8_t sum15 = vaddq_s16(row1, row5); \ + int16x8_t sum17 = vaddq_s16(row1, row7); \ + int16x8_t sum35 = vaddq_s16(row3, row5); \ + int16x8_t sum37 = vaddq_s16(row3, row7); \ + int16x8_t sumodd = vaddq_s16(sum17, sum35); \ + dct_long_mul(p5o, sumodd, rot1_0); \ + dct_long_mac(p1o, p5o, sum17, rot1_1); \ + dct_long_mac(p2o, p5o, sum35, rot1_2); \ + dct_long_mul(p3o, sum37, rot2_0); \ + dct_long_mul(p4o, sum15, rot2_1); \ + dct_wadd(sump13o, p1o, p3o); \ + dct_wadd(sump24o, p2o, p4o); \ + dct_wadd(sump23o, p2o, p3o); \ + dct_wadd(sump14o, p1o, p4o); \ + dct_long_mac(x4, sump13o, row7, rot3_0); \ + dct_long_mac(x5, sump24o, row5, rot3_1); \ + dct_long_mac(x6, sump23o, row3, rot3_2); \ + dct_long_mac(x7, sump14o, row1, rot3_3); \ + dct_bfly32o(row0,row7, x0,x7,shiftop,shift); \ + dct_bfly32o(row1,row6, x1,x6,shiftop,shift); \ + dct_bfly32o(row2,row5, x2,x5,shiftop,shift); \ + dct_bfly32o(row3,row4, x3,x4,shiftop,shift); \ + } + + // load + row0 = vld1q_s16(data + 0*8); + row1 = vld1q_s16(data + 1*8); + row2 = vld1q_s16(data + 2*8); + row3 = vld1q_s16(data + 3*8); + row4 = vld1q_s16(data + 4*8); + row5 = vld1q_s16(data + 5*8); + row6 = vld1q_s16(data + 6*8); + row7 = vld1q_s16(data + 7*8); + + // add DC bias + row0 = vaddq_s16(row0, vsetq_lane_s16(1024, vdupq_n_s16(0), 0)); + + // column pass + dct_pass(vrshrn_n_s32, 10); + + // 16bit 8x8 transpose + { +// these three map to a single VTRN.16, VTRN.32, and VSWP, respectively. +// whether compilers actually get this is another story, sadly. +#define dct_trn16(x, y) { int16x8x2_t t = vtrnq_s16(x, y); x = t.val[0]; y = t.val[1]; } +#define dct_trn32(x, y) { int32x4x2_t t = vtrnq_s32(vreinterpretq_s32_s16(x), vreinterpretq_s32_s16(y)); x = vreinterpretq_s16_s32(t.val[0]); y = vreinterpretq_s16_s32(t.val[1]); } +#define dct_trn64(x, y) { int16x8_t x0 = x; int16x8_t y0 = y; x = vcombine_s16(vget_low_s16(x0), vget_low_s16(y0)); y = vcombine_s16(vget_high_s16(x0), vget_high_s16(y0)); } + + // pass 1 + dct_trn16(row0, row1); // a0b0a2b2a4b4a6b6 + dct_trn16(row2, row3); + dct_trn16(row4, row5); + dct_trn16(row6, row7); + + // pass 2 + dct_trn32(row0, row2); // a0b0c0d0a4b4c4d4 + dct_trn32(row1, row3); + dct_trn32(row4, row6); + dct_trn32(row5, row7); + + // pass 3 + dct_trn64(row0, row4); // a0b0c0d0e0f0g0h0 + dct_trn64(row1, row5); + dct_trn64(row2, row6); + dct_trn64(row3, row7); + +#undef dct_trn16 +#undef dct_trn32 +#undef dct_trn64 + } + + // row pass + // vrshrn_n_s32 only supports shifts up to 16, we need + // 17. so do a non-rounding shift of 16 first then follow + // up with a rounding shift by 1. + dct_pass(vshrn_n_s32, 16); + + { + // pack and round + uint8x8_t p0 = vqrshrun_n_s16(row0, 1); + uint8x8_t p1 = vqrshrun_n_s16(row1, 1); + uint8x8_t p2 = vqrshrun_n_s16(row2, 1); + uint8x8_t p3 = vqrshrun_n_s16(row3, 1); + uint8x8_t p4 = vqrshrun_n_s16(row4, 1); + uint8x8_t p5 = vqrshrun_n_s16(row5, 1); + uint8x8_t p6 = vqrshrun_n_s16(row6, 1); + uint8x8_t p7 = vqrshrun_n_s16(row7, 1); + + // again, these can translate into one instruction, but often don't. +#define dct_trn8_8(x, y) { uint8x8x2_t t = vtrn_u8(x, y); x = t.val[0]; y = t.val[1]; } +#define dct_trn8_16(x, y) { uint16x4x2_t t = vtrn_u16(vreinterpret_u16_u8(x), vreinterpret_u16_u8(y)); x = vreinterpret_u8_u16(t.val[0]); y = vreinterpret_u8_u16(t.val[1]); } +#define dct_trn8_32(x, y) { uint32x2x2_t t = vtrn_u32(vreinterpret_u32_u8(x), vreinterpret_u32_u8(y)); x = vreinterpret_u8_u32(t.val[0]); y = vreinterpret_u8_u32(t.val[1]); } + + // sadly can't use interleaved stores here since we only write + // 8 bytes to each scan line! + + // 8x8 8-bit transpose pass 1 + dct_trn8_8(p0, p1); + dct_trn8_8(p2, p3); + dct_trn8_8(p4, p5); + dct_trn8_8(p6, p7); + + // pass 2 + dct_trn8_16(p0, p2); + dct_trn8_16(p1, p3); + dct_trn8_16(p4, p6); + dct_trn8_16(p5, p7); + + // pass 3 + dct_trn8_32(p0, p4); + dct_trn8_32(p1, p5); + dct_trn8_32(p2, p6); + dct_trn8_32(p3, p7); + + // store + vst1_u8(out, p0); out += out_stride; + vst1_u8(out, p1); out += out_stride; + vst1_u8(out, p2); out += out_stride; + vst1_u8(out, p3); out += out_stride; + vst1_u8(out, p4); out += out_stride; + vst1_u8(out, p5); out += out_stride; + vst1_u8(out, p6); out += out_stride; + vst1_u8(out, p7); + +#undef dct_trn8_8 +#undef dct_trn8_16 +#undef dct_trn8_32 + } + +#undef dct_long_mul +#undef dct_long_mac +#undef dct_widen +#undef dct_wadd +#undef dct_wsub +#undef dct_bfly32o +#undef dct_pass +} + +#endif // STBI_NEON + +#define STBI__MARKER_none 0xff +// if there's a pending marker from the entropy stream, return that +// otherwise, fetch from the stream and get a marker. if there's no +// marker, return 0xff, which is never a valid marker value +static stbi_uc stbi__get_marker(stbi__jpeg *j) +{ + stbi_uc x; + if (j->marker != STBI__MARKER_none) { x = j->marker; j->marker = STBI__MARKER_none; return x; } + x = stbi__get8(j->s); + if (x != 0xff) return STBI__MARKER_none; + while (x == 0xff) + x = stbi__get8(j->s); // consume repeated 0xff fill bytes + return x; +} + +// in each scan, we'll have scan_n components, and the order +// of the components is specified by order[] +#define STBI__RESTART(x) ((x) >= 0xd0 && (x) <= 0xd7) + +// after a restart interval, stbi__jpeg_reset the entropy decoder and +// the dc prediction +static void stbi__jpeg_reset(stbi__jpeg *j) +{ + j->code_bits = 0; + j->code_buffer = 0; + j->nomore = 0; + j->img_comp[0].dc_pred = j->img_comp[1].dc_pred = j->img_comp[2].dc_pred = j->img_comp[3].dc_pred = 0; + j->marker = STBI__MARKER_none; + j->todo = j->restart_interval ? j->restart_interval : 0x7fffffff; + j->eob_run = 0; + // no more than 1<<31 MCUs if no restart_interal? that's plenty safe, + // since we don't even allow 1<<30 pixels +} + +static int stbi__parse_entropy_coded_data(stbi__jpeg *z) +{ + stbi__jpeg_reset(z); + if (!z->progressive) { + if (z->scan_n == 1) { + int i,j; + STBI_SIMD_ALIGN(short, data[64]); + int n = z->order[0]; + // non-interleaved data, we just need to process one block at a time, + // in trivial scanline order + // number of blocks to do just depends on how many actual "pixels" this + // component has, independent of interleaved MCU blocking and such + int w = (z->img_comp[n].x+7) >> 3; + int h = (z->img_comp[n].y+7) >> 3; + for (j=0; j < h; ++j) { + for (i=0; i < w; ++i) { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; + z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data); + // every data block is an MCU, so countdown the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + // if it's NOT a restart, then just bail, so we get corrupt data + // rather than no data + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } else { // interleaved + int i,j,k,x,y; + STBI_SIMD_ALIGN(short, data[64]); + for (j=0; j < z->img_mcu_y; ++j) { + for (i=0; i < z->img_mcu_x; ++i) { + // scan an interleaved mcu... process scan_n components in order + for (k=0; k < z->scan_n; ++k) { + int n = z->order[k]; + // scan out an mcu's worth of this component; that's just determined + // by the basic H and V specified for the component + for (y=0; y < z->img_comp[n].v; ++y) { + for (x=0; x < z->img_comp[n].h; ++x) { + int x2 = (i*z->img_comp[n].h + x)*8; + int y2 = (j*z->img_comp[n].v + y)*8; + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; + z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*y2+x2, z->img_comp[n].w2, data); + } + } + } + // after all interleaved components, that's an interleaved MCU, + // so now count down the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } else { + if (z->scan_n == 1) { + int i,j; + int n = z->order[0]; + // non-interleaved data, we just need to process one block at a time, + // in trivial scanline order + // number of blocks to do just depends on how many actual "pixels" this + // component has, independent of interleaved MCU blocking and such + int w = (z->img_comp[n].x+7) >> 3; + int h = (z->img_comp[n].y+7) >> 3; + for (j=0; j < h; ++j) { + for (i=0; i < w; ++i) { + short *data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + if (z->spec_start == 0) { + if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } else { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block_prog_ac(z, data, &z->huff_ac[ha], z->fast_ac[ha])) + return 0; + } + // every data block is an MCU, so countdown the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } else { // interleaved + int i,j,k,x,y; + for (j=0; j < z->img_mcu_y; ++j) { + for (i=0; i < z->img_mcu_x; ++i) { + // scan an interleaved mcu... process scan_n components in order + for (k=0; k < z->scan_n; ++k) { + int n = z->order[k]; + // scan out an mcu's worth of this component; that's just determined + // by the basic H and V specified for the component + for (y=0; y < z->img_comp[n].v; ++y) { + for (x=0; x < z->img_comp[n].h; ++x) { + int x2 = (i*z->img_comp[n].h + x); + int y2 = (j*z->img_comp[n].v + y); + short *data = z->img_comp[n].coeff + 64 * (x2 + y2 * z->img_comp[n].coeff_w); + if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } + } + } + // after all interleaved components, that's an interleaved MCU, + // so now count down the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } +} + +static void stbi__jpeg_dequantize(short *data, stbi__uint16 *dequant) +{ + int i; + for (i=0; i < 64; ++i) + data[i] *= dequant[i]; +} + +static void stbi__jpeg_finish(stbi__jpeg *z) +{ + if (z->progressive) { + // dequantize and idct the data + int i,j,n; + for (n=0; n < z->s->img_n; ++n) { + int w = (z->img_comp[n].x+7) >> 3; + int h = (z->img_comp[n].y+7) >> 3; + for (j=0; j < h; ++j) { + for (i=0; i < w; ++i) { + short *data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + stbi__jpeg_dequantize(data, z->dequant[z->img_comp[n].tq]); + z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data); + } + } + } + } +} + +static int stbi__process_marker(stbi__jpeg *z, int m) +{ + int L; + switch (m) { + case STBI__MARKER_none: // no marker found + return stbi__err("expected marker","Corrupt JPEG"); + + case 0xDD: // DRI - specify restart interval + if (stbi__get16be(z->s) != 4) return stbi__err("bad DRI len","Corrupt JPEG"); + z->restart_interval = stbi__get16be(z->s); + return 1; + + case 0xDB: // DQT - define quantization table + L = stbi__get16be(z->s)-2; + while (L > 0) { + int q = stbi__get8(z->s); + int p = q >> 4, sixteen = (p != 0); + int t = q & 15,i; + if (p != 0 && p != 1) return stbi__err("bad DQT type","Corrupt JPEG"); + if (t > 3) return stbi__err("bad DQT table","Corrupt JPEG"); + + for (i=0; i < 64; ++i) + z->dequant[t][stbi__jpeg_dezigzag[i]] = (stbi__uint16)(sixteen ? stbi__get16be(z->s) : stbi__get8(z->s)); + L -= (sixteen ? 129 : 65); + } + return L==0; + + case 0xC4: // DHT - define huffman table + L = stbi__get16be(z->s)-2; + while (L > 0) { + stbi_uc *v; + int sizes[16],i,n=0; + int q = stbi__get8(z->s); + int tc = q >> 4; + int th = q & 15; + if (tc > 1 || th > 3) return stbi__err("bad DHT header","Corrupt JPEG"); + for (i=0; i < 16; ++i) { + sizes[i] = stbi__get8(z->s); + n += sizes[i]; + } + L -= 17; + if (tc == 0) { + if (!stbi__build_huffman(z->huff_dc+th, sizes)) return 0; + v = z->huff_dc[th].values; + } else { + if (!stbi__build_huffman(z->huff_ac+th, sizes)) return 0; + v = z->huff_ac[th].values; + } + for (i=0; i < n; ++i) + v[i] = stbi__get8(z->s); + if (tc != 0) + stbi__build_fast_ac(z->fast_ac[th], z->huff_ac + th); + L -= n; + } + return L==0; + } + + // check for comment block or APP blocks + if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) { + L = stbi__get16be(z->s); + if (L < 2) { + if (m == 0xFE) + return stbi__err("bad COM len","Corrupt JPEG"); + else + return stbi__err("bad APP len","Corrupt JPEG"); + } + L -= 2; + + if (m == 0xE0 && L >= 5) { // JFIF APP0 segment + static const unsigned char tag[5] = {'J','F','I','F','\0'}; + int ok = 1; + int i; + for (i=0; i < 5; ++i) + if (stbi__get8(z->s) != tag[i]) + ok = 0; + L -= 5; + if (ok) + z->jfif = 1; + } else if (m == 0xEE && L >= 12) { // Adobe APP14 segment + static const unsigned char tag[6] = {'A','d','o','b','e','\0'}; + int ok = 1; + int i; + for (i=0; i < 6; ++i) + if (stbi__get8(z->s) != tag[i]) + ok = 0; + L -= 6; + if (ok) { + stbi__get8(z->s); // version + stbi__get16be(z->s); // flags0 + stbi__get16be(z->s); // flags1 + z->app14_color_transform = stbi__get8(z->s); // color transform + L -= 6; + } + } + + stbi__skip(z->s, L); + return 1; + } + + return stbi__err("unknown marker","Corrupt JPEG"); +} + +// after we see SOS +static int stbi__process_scan_header(stbi__jpeg *z) +{ + int i; + int Ls = stbi__get16be(z->s); + z->scan_n = stbi__get8(z->s); + if (z->scan_n < 1 || z->scan_n > 4 || z->scan_n > (int) z->s->img_n) return stbi__err("bad SOS component count","Corrupt JPEG"); + if (Ls != 6+2*z->scan_n) return stbi__err("bad SOS len","Corrupt JPEG"); + for (i=0; i < z->scan_n; ++i) { + int id = stbi__get8(z->s), which; + int q = stbi__get8(z->s); + for (which = 0; which < z->s->img_n; ++which) + if (z->img_comp[which].id == id) + break; + if (which == z->s->img_n) return 0; // no match + z->img_comp[which].hd = q >> 4; if (z->img_comp[which].hd > 3) return stbi__err("bad DC huff","Corrupt JPEG"); + z->img_comp[which].ha = q & 15; if (z->img_comp[which].ha > 3) return stbi__err("bad AC huff","Corrupt JPEG"); + z->order[i] = which; + } + + { + int aa; + z->spec_start = stbi__get8(z->s); + z->spec_end = stbi__get8(z->s); // should be 63, but might be 0 + aa = stbi__get8(z->s); + z->succ_high = (aa >> 4); + z->succ_low = (aa & 15); + if (z->progressive) { + if (z->spec_start > 63 || z->spec_end > 63 || z->spec_start > z->spec_end || z->succ_high > 13 || z->succ_low > 13) + return stbi__err("bad SOS", "Corrupt JPEG"); + } else { + if (z->spec_start != 0) return stbi__err("bad SOS","Corrupt JPEG"); + if (z->succ_high != 0 || z->succ_low != 0) return stbi__err("bad SOS","Corrupt JPEG"); + z->spec_end = 63; + } + } + + return 1; +} + +static int stbi__free_jpeg_components(stbi__jpeg *z, int ncomp, int why) +{ + int i; + for (i=0; i < ncomp; ++i) { + if (z->img_comp[i].raw_data) { + STBI_FREE(z->img_comp[i].raw_data); + z->img_comp[i].raw_data = NULL; + z->img_comp[i].data = NULL; + } + if (z->img_comp[i].raw_coeff) { + STBI_FREE(z->img_comp[i].raw_coeff); + z->img_comp[i].raw_coeff = 0; + z->img_comp[i].coeff = 0; + } + if (z->img_comp[i].linebuf) { + STBI_FREE(z->img_comp[i].linebuf); + z->img_comp[i].linebuf = NULL; + } + } + return why; +} + +static int stbi__process_frame_header(stbi__jpeg *z, int scan) +{ + stbi__context *s = z->s; + int Lf,p,i,q, h_max=1,v_max=1,c; + Lf = stbi__get16be(s); if (Lf < 11) return stbi__err("bad SOF len","Corrupt JPEG"); // JPEG + p = stbi__get8(s); if (p != 8) return stbi__err("only 8-bit","JPEG format not supported: 8-bit only"); // JPEG baseline + s->img_y = stbi__get16be(s); if (s->img_y == 0) return stbi__err("no header height", "JPEG format not supported: delayed height"); // Legal, but we don't handle it--but neither does IJG + s->img_x = stbi__get16be(s); if (s->img_x == 0) return stbi__err("0 width","Corrupt JPEG"); // JPEG requires + c = stbi__get8(s); + if (c != 3 && c != 1 && c != 4) return stbi__err("bad component count","Corrupt JPEG"); + s->img_n = c; + for (i=0; i < c; ++i) { + z->img_comp[i].data = NULL; + z->img_comp[i].linebuf = NULL; + } + + if (Lf != 8+3*s->img_n) return stbi__err("bad SOF len","Corrupt JPEG"); + + z->rgb = 0; + for (i=0; i < s->img_n; ++i) { + static const unsigned char rgb[3] = { 'R', 'G', 'B' }; + z->img_comp[i].id = stbi__get8(s); + if (s->img_n == 3 && z->img_comp[i].id == rgb[i]) + ++z->rgb; + q = stbi__get8(s); + z->img_comp[i].h = (q >> 4); if (!z->img_comp[i].h || z->img_comp[i].h > 4) return stbi__err("bad H","Corrupt JPEG"); + z->img_comp[i].v = q & 15; if (!z->img_comp[i].v || z->img_comp[i].v > 4) return stbi__err("bad V","Corrupt JPEG"); + z->img_comp[i].tq = stbi__get8(s); if (z->img_comp[i].tq > 3) return stbi__err("bad TQ","Corrupt JPEG"); + } + + if (scan != STBI__SCAN_load) return 1; + + if (!stbi__mad3sizes_valid(s->img_x, s->img_y, s->img_n, 0)) return stbi__err("too large", "Image too large to decode"); + + for (i=0; i < s->img_n; ++i) { + if (z->img_comp[i].h > h_max) h_max = z->img_comp[i].h; + if (z->img_comp[i].v > v_max) v_max = z->img_comp[i].v; + } + + // compute interleaved mcu info + z->img_h_max = h_max; + z->img_v_max = v_max; + z->img_mcu_w = h_max * 8; + z->img_mcu_h = v_max * 8; + // these sizes can't be more than 17 bits + z->img_mcu_x = (s->img_x + z->img_mcu_w-1) / z->img_mcu_w; + z->img_mcu_y = (s->img_y + z->img_mcu_h-1) / z->img_mcu_h; + + for (i=0; i < s->img_n; ++i) { + // number of effective pixels (e.g. for non-interleaved MCU) + z->img_comp[i].x = (s->img_x * z->img_comp[i].h + h_max-1) / h_max; + z->img_comp[i].y = (s->img_y * z->img_comp[i].v + v_max-1) / v_max; + // to simplify generation, we'll allocate enough memory to decode + // the bogus oversized data from using interleaved MCUs and their + // big blocks (e.g. a 16x16 iMCU on an image of width 33); we won't + // discard the extra data until colorspace conversion + // + // img_mcu_x, img_mcu_y: <=17 bits; comp[i].h and .v are <=4 (checked earlier) + // so these muls can't overflow with 32-bit ints (which we require) + z->img_comp[i].w2 = z->img_mcu_x * z->img_comp[i].h * 8; + z->img_comp[i].h2 = z->img_mcu_y * z->img_comp[i].v * 8; + z->img_comp[i].coeff = 0; + z->img_comp[i].raw_coeff = 0; + z->img_comp[i].linebuf = NULL; + z->img_comp[i].raw_data = stbi__malloc_mad2(z->img_comp[i].w2, z->img_comp[i].h2, 15); + if (z->img_comp[i].raw_data == NULL) + return stbi__free_jpeg_components(z, i+1, stbi__err("outofmem", "Out of memory")); + // align blocks for idct using mmx/sse + z->img_comp[i].data = (stbi_uc*) (((size_t) z->img_comp[i].raw_data + 15) & ~15); + if (z->progressive) { + // w2, h2 are multiples of 8 (see above) + z->img_comp[i].coeff_w = z->img_comp[i].w2 / 8; + z->img_comp[i].coeff_h = z->img_comp[i].h2 / 8; + z->img_comp[i].raw_coeff = stbi__malloc_mad3(z->img_comp[i].w2, z->img_comp[i].h2, sizeof(short), 15); + if (z->img_comp[i].raw_coeff == NULL) + return stbi__free_jpeg_components(z, i+1, stbi__err("outofmem", "Out of memory")); + z->img_comp[i].coeff = (short*) (((size_t) z->img_comp[i].raw_coeff + 15) & ~15); + } + } + + return 1; +} + +// use comparisons since in some cases we handle more than one case (e.g. SOF) +#define stbi__DNL(x) ((x) == 0xdc) +#define stbi__SOI(x) ((x) == 0xd8) +#define stbi__EOI(x) ((x) == 0xd9) +#define stbi__SOF(x) ((x) == 0xc0 || (x) == 0xc1 || (x) == 0xc2) +#define stbi__SOS(x) ((x) == 0xda) + +#define stbi__SOF_progressive(x) ((x) == 0xc2) + +static int stbi__decode_jpeg_header(stbi__jpeg *z, int scan) +{ + int m; + z->jfif = 0; + z->app14_color_transform = -1; // valid values are 0,1,2 + z->marker = STBI__MARKER_none; // initialize cached marker to empty + m = stbi__get_marker(z); + if (!stbi__SOI(m)) return stbi__err("no SOI","Corrupt JPEG"); + if (scan == STBI__SCAN_type) return 1; + m = stbi__get_marker(z); + while (!stbi__SOF(m)) { + if (!stbi__process_marker(z,m)) return 0; + m = stbi__get_marker(z); + while (m == STBI__MARKER_none) { + // some files have extra padding after their blocks, so ok, we'll scan + if (stbi__at_eof(z->s)) return stbi__err("no SOF", "Corrupt JPEG"); + m = stbi__get_marker(z); + } + } + z->progressive = stbi__SOF_progressive(m); + if (!stbi__process_frame_header(z, scan)) return 0; + return 1; +} + +// decode image to YCbCr format +static int stbi__decode_jpeg_image(stbi__jpeg *j) +{ + int m; + for (m = 0; m < 4; m++) { + j->img_comp[m].raw_data = NULL; + j->img_comp[m].raw_coeff = NULL; + } + j->restart_interval = 0; + if (!stbi__decode_jpeg_header(j, STBI__SCAN_load)) return 0; + m = stbi__get_marker(j); + while (!stbi__EOI(m)) { + if (stbi__SOS(m)) { + if (!stbi__process_scan_header(j)) return 0; + if (!stbi__parse_entropy_coded_data(j)) return 0; + if (j->marker == STBI__MARKER_none ) { + // handle 0s at the end of image data from IP Kamera 9060 + while (!stbi__at_eof(j->s)) { + int x = stbi__get8(j->s); + if (x == 255) { + j->marker = stbi__get8(j->s); + break; + } + } + // if we reach eof without hitting a marker, stbi__get_marker() below will fail and we'll eventually return 0 + } + } else if (stbi__DNL(m)) { + int Ld = stbi__get16be(j->s); + stbi__uint32 NL = stbi__get16be(j->s); + if (Ld != 4) return stbi__err("bad DNL len", "Corrupt JPEG"); + if (NL != j->s->img_y) return stbi__err("bad DNL height", "Corrupt JPEG"); + } else { + if (!stbi__process_marker(j, m)) return 0; + } + m = stbi__get_marker(j); + } + if (j->progressive) + stbi__jpeg_finish(j); + return 1; +} + +// static jfif-centered resampling (across block boundaries) + +typedef stbi_uc *(*resample_row_func)(stbi_uc *out, stbi_uc *in0, stbi_uc *in1, + int w, int hs); + +#define stbi__div4(x) ((stbi_uc) ((x) >> 2)) + +static stbi_uc *resample_row_1(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + STBI_NOTUSED(out); + STBI_NOTUSED(in_far); + STBI_NOTUSED(w); + STBI_NOTUSED(hs); + return in_near; +} + +static stbi_uc* stbi__resample_row_v_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate two samples vertically for every one in input + int i; + STBI_NOTUSED(hs); + for (i=0; i < w; ++i) + out[i] = stbi__div4(3*in_near[i] + in_far[i] + 2); + return out; +} + +static stbi_uc* stbi__resample_row_h_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate two samples horizontally for every one in input + int i; + stbi_uc *input = in_near; + + if (w == 1) { + // if only one sample, can't do any interpolation + out[0] = out[1] = input[0]; + return out; + } + + out[0] = input[0]; + out[1] = stbi__div4(input[0]*3 + input[1] + 2); + for (i=1; i < w-1; ++i) { + int n = 3*input[i]+2; + out[i*2+0] = stbi__div4(n+input[i-1]); + out[i*2+1] = stbi__div4(n+input[i+1]); + } + out[i*2+0] = stbi__div4(input[w-2]*3 + input[w-1] + 2); + out[i*2+1] = input[w-1]; + + STBI_NOTUSED(in_far); + STBI_NOTUSED(hs); + + return out; +} + +#define stbi__div16(x) ((stbi_uc) ((x) >> 4)) + +static stbi_uc *stbi__resample_row_hv_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate 2x2 samples for every one in input + int i,t0,t1; + if (w == 1) { + out[0] = out[1] = stbi__div4(3*in_near[0] + in_far[0] + 2); + return out; + } + + t1 = 3*in_near[0] + in_far[0]; + out[0] = stbi__div4(t1+2); + for (i=1; i < w; ++i) { + t0 = t1; + t1 = 3*in_near[i]+in_far[i]; + out[i*2-1] = stbi__div16(3*t0 + t1 + 8); + out[i*2 ] = stbi__div16(3*t1 + t0 + 8); + } + out[w*2-1] = stbi__div4(t1+2); + + STBI_NOTUSED(hs); + + return out; +} + +#if defined(STBI_SSE2) || defined(STBI_NEON) +static stbi_uc *stbi__resample_row_hv_2_simd(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate 2x2 samples for every one in input + int i=0,t0,t1; + + if (w == 1) { + out[0] = out[1] = stbi__div4(3*in_near[0] + in_far[0] + 2); + return out; + } + + t1 = 3*in_near[0] + in_far[0]; + // process groups of 8 pixels for as long as we can. + // note we can't handle the last pixel in a row in this loop + // because we need to handle the filter boundary conditions. + for (; i < ((w-1) & ~7); i += 8) { +#if defined(STBI_SSE2) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + __m128i zero = _mm_setzero_si128(); + __m128i farb = _mm_loadl_epi64((__m128i *) (in_far + i)); + __m128i nearb = _mm_loadl_epi64((__m128i *) (in_near + i)); + __m128i farw = _mm_unpacklo_epi8(farb, zero); + __m128i nearw = _mm_unpacklo_epi8(nearb, zero); + __m128i diff = _mm_sub_epi16(farw, nearw); + __m128i nears = _mm_slli_epi16(nearw, 2); + __m128i curr = _mm_add_epi16(nears, diff); // current row + + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + __m128i prv0 = _mm_slli_si128(curr, 2); + __m128i nxt0 = _mm_srli_si128(curr, 2); + __m128i prev = _mm_insert_epi16(prv0, t1, 0); + __m128i next = _mm_insert_epi16(nxt0, 3*in_near[i+8] + in_far[i+8], 7); + + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + __m128i bias = _mm_set1_epi16(8); + __m128i curs = _mm_slli_epi16(curr, 2); + __m128i prvd = _mm_sub_epi16(prev, curr); + __m128i nxtd = _mm_sub_epi16(next, curr); + __m128i curb = _mm_add_epi16(curs, bias); + __m128i even = _mm_add_epi16(prvd, curb); + __m128i odd = _mm_add_epi16(nxtd, curb); + + // interleave even and odd pixels, then undo scaling. + __m128i int0 = _mm_unpacklo_epi16(even, odd); + __m128i int1 = _mm_unpackhi_epi16(even, odd); + __m128i de0 = _mm_srli_epi16(int0, 4); + __m128i de1 = _mm_srli_epi16(int1, 4); + + // pack and write output + __m128i outv = _mm_packus_epi16(de0, de1); + _mm_storeu_si128((__m128i *) (out + i*2), outv); +#elif defined(STBI_NEON) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + uint8x8_t farb = vld1_u8(in_far + i); + uint8x8_t nearb = vld1_u8(in_near + i); + int16x8_t diff = vreinterpretq_s16_u16(vsubl_u8(farb, nearb)); + int16x8_t nears = vreinterpretq_s16_u16(vshll_n_u8(nearb, 2)); + int16x8_t curr = vaddq_s16(nears, diff); // current row + + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + int16x8_t prv0 = vextq_s16(curr, curr, 7); + int16x8_t nxt0 = vextq_s16(curr, curr, 1); + int16x8_t prev = vsetq_lane_s16(t1, prv0, 0); + int16x8_t next = vsetq_lane_s16(3*in_near[i+8] + in_far[i+8], nxt0, 7); + + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + int16x8_t curs = vshlq_n_s16(curr, 2); + int16x8_t prvd = vsubq_s16(prev, curr); + int16x8_t nxtd = vsubq_s16(next, curr); + int16x8_t even = vaddq_s16(curs, prvd); + int16x8_t odd = vaddq_s16(curs, nxtd); + + // undo scaling and round, then store with even/odd phases interleaved + uint8x8x2_t o; + o.val[0] = vqrshrun_n_s16(even, 4); + o.val[1] = vqrshrun_n_s16(odd, 4); + vst2_u8(out + i*2, o); +#endif + + // "previous" value for next iter + t1 = 3*in_near[i+7] + in_far[i+7]; + } + + t0 = t1; + t1 = 3*in_near[i] + in_far[i]; + out[i*2] = stbi__div16(3*t1 + t0 + 8); + + for (++i; i < w; ++i) { + t0 = t1; + t1 = 3*in_near[i]+in_far[i]; + out[i*2-1] = stbi__div16(3*t0 + t1 + 8); + out[i*2 ] = stbi__div16(3*t1 + t0 + 8); + } + out[w*2-1] = stbi__div4(t1+2); + + STBI_NOTUSED(hs); + + return out; +} +#endif + +static stbi_uc *stbi__resample_row_generic(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // resample with nearest-neighbor + int i,j; + STBI_NOTUSED(in_far); + for (i=0; i < w; ++i) + for (j=0; j < hs; ++j) + out[i*hs+j] = in_near[i]; + return out; +} + +// this is a reduced-precision calculation of YCbCr-to-RGB introduced +// to make sure the code produces the same results in both SIMD and scalar +#define stbi__float2fixed(x) (((int) ((x) * 4096.0f + 0.5f)) << 8) +static void stbi__YCbCr_to_RGB_row(stbi_uc *out, const stbi_uc *y, const stbi_uc *pcb, const stbi_uc *pcr, int count, int step) +{ + int i; + for (i=0; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1<<19); // rounding + int r,g,b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr* stbi__float2fixed(1.40200f); + g = y_fixed + (cr*-stbi__float2fixed(0.71414f)) + ((cb*-stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb* stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; } + if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; } + if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } +} + +#if defined(STBI_SSE2) || defined(STBI_NEON) +static void stbi__YCbCr_to_RGB_simd(stbi_uc *out, stbi_uc const *y, stbi_uc const *pcb, stbi_uc const *pcr, int count, int step) +{ + int i = 0; + +#ifdef STBI_SSE2 + // step == 3 is pretty ugly on the final interleave, and i'm not convinced + // it's useful in practice (you wouldn't use it for textures, for example). + // so just accelerate step == 4 case. + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + __m128i signflip = _mm_set1_epi8(-0x80); + __m128i cr_const0 = _mm_set1_epi16( (short) ( 1.40200f*4096.0f+0.5f)); + __m128i cr_const1 = _mm_set1_epi16( - (short) ( 0.71414f*4096.0f+0.5f)); + __m128i cb_const0 = _mm_set1_epi16( - (short) ( 0.34414f*4096.0f+0.5f)); + __m128i cb_const1 = _mm_set1_epi16( (short) ( 1.77200f*4096.0f+0.5f)); + __m128i y_bias = _mm_set1_epi8((char) (unsigned char) 128); + __m128i xw = _mm_set1_epi16(255); // alpha channel + + for (; i+7 < count; i += 8) { + // load + __m128i y_bytes = _mm_loadl_epi64((__m128i *) (y+i)); + __m128i cr_bytes = _mm_loadl_epi64((__m128i *) (pcr+i)); + __m128i cb_bytes = _mm_loadl_epi64((__m128i *) (pcb+i)); + __m128i cr_biased = _mm_xor_si128(cr_bytes, signflip); // -128 + __m128i cb_biased = _mm_xor_si128(cb_bytes, signflip); // -128 + + // unpack to short (and left-shift cr, cb by 8) + __m128i yw = _mm_unpacklo_epi8(y_bias, y_bytes); + __m128i crw = _mm_unpacklo_epi8(_mm_setzero_si128(), cr_biased); + __m128i cbw = _mm_unpacklo_epi8(_mm_setzero_si128(), cb_biased); + + // color transform + __m128i yws = _mm_srli_epi16(yw, 4); + __m128i cr0 = _mm_mulhi_epi16(cr_const0, crw); + __m128i cb0 = _mm_mulhi_epi16(cb_const0, cbw); + __m128i cb1 = _mm_mulhi_epi16(cbw, cb_const1); + __m128i cr1 = _mm_mulhi_epi16(crw, cr_const1); + __m128i rws = _mm_add_epi16(cr0, yws); + __m128i gwt = _mm_add_epi16(cb0, yws); + __m128i bws = _mm_add_epi16(yws, cb1); + __m128i gws = _mm_add_epi16(gwt, cr1); + + // descale + __m128i rw = _mm_srai_epi16(rws, 4); + __m128i bw = _mm_srai_epi16(bws, 4); + __m128i gw = _mm_srai_epi16(gws, 4); + + // back to byte, set up for transpose + __m128i brb = _mm_packus_epi16(rw, bw); + __m128i gxb = _mm_packus_epi16(gw, xw); + + // transpose to interleave channels + __m128i t0 = _mm_unpacklo_epi8(brb, gxb); + __m128i t1 = _mm_unpackhi_epi8(brb, gxb); + __m128i o0 = _mm_unpacklo_epi16(t0, t1); + __m128i o1 = _mm_unpackhi_epi16(t0, t1); + + // store + _mm_storeu_si128((__m128i *) (out + 0), o0); + _mm_storeu_si128((__m128i *) (out + 16), o1); + out += 32; + } + } +#endif + +#ifdef STBI_NEON + // in this version, step=3 support would be easy to add. but is there demand? + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + uint8x8_t signflip = vdup_n_u8(0x80); + int16x8_t cr_const0 = vdupq_n_s16( (short) ( 1.40200f*4096.0f+0.5f)); + int16x8_t cr_const1 = vdupq_n_s16( - (short) ( 0.71414f*4096.0f+0.5f)); + int16x8_t cb_const0 = vdupq_n_s16( - (short) ( 0.34414f*4096.0f+0.5f)); + int16x8_t cb_const1 = vdupq_n_s16( (short) ( 1.77200f*4096.0f+0.5f)); + + for (; i+7 < count; i += 8) { + // load + uint8x8_t y_bytes = vld1_u8(y + i); + uint8x8_t cr_bytes = vld1_u8(pcr + i); + uint8x8_t cb_bytes = vld1_u8(pcb + i); + int8x8_t cr_biased = vreinterpret_s8_u8(vsub_u8(cr_bytes, signflip)); + int8x8_t cb_biased = vreinterpret_s8_u8(vsub_u8(cb_bytes, signflip)); + + // expand to s16 + int16x8_t yws = vreinterpretq_s16_u16(vshll_n_u8(y_bytes, 4)); + int16x8_t crw = vshll_n_s8(cr_biased, 7); + int16x8_t cbw = vshll_n_s8(cb_biased, 7); + + // color transform + int16x8_t cr0 = vqdmulhq_s16(crw, cr_const0); + int16x8_t cb0 = vqdmulhq_s16(cbw, cb_const0); + int16x8_t cr1 = vqdmulhq_s16(crw, cr_const1); + int16x8_t cb1 = vqdmulhq_s16(cbw, cb_const1); + int16x8_t rws = vaddq_s16(yws, cr0); + int16x8_t gws = vaddq_s16(vaddq_s16(yws, cb0), cr1); + int16x8_t bws = vaddq_s16(yws, cb1); + + // undo scaling, round, convert to byte + uint8x8x4_t o; + o.val[0] = vqrshrun_n_s16(rws, 4); + o.val[1] = vqrshrun_n_s16(gws, 4); + o.val[2] = vqrshrun_n_s16(bws, 4); + o.val[3] = vdup_n_u8(255); + + // store, interleaving r/g/b/a + vst4_u8(out, o); + out += 8*4; + } + } +#endif + + for (; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1<<19); // rounding + int r,g,b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr* stbi__float2fixed(1.40200f); + g = y_fixed + cr*-stbi__float2fixed(0.71414f) + ((cb*-stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb* stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; } + if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; } + if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } +} +#endif + +// set up the kernels +static void stbi__setup_jpeg(stbi__jpeg *j) +{ + j->idct_block_kernel = stbi__idct_block; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_row; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2; + +#ifdef STBI_SSE2 + if (stbi__sse2_available()) { + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; + } +#endif + +#ifdef STBI_NEON + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; +#endif +} + +// clean up the temporary component buffers +static void stbi__cleanup_jpeg(stbi__jpeg *j) +{ + stbi__free_jpeg_components(j, j->s->img_n, 0); +} + +typedef struct +{ + resample_row_func resample; + stbi_uc *line0,*line1; + int hs,vs; // expansion factor in each axis + int w_lores; // horizontal pixels pre-expansion + int ystep; // how far through vertical expansion we are + int ypos; // which pre-expansion row we're on +} stbi__resample; + +// fast 0..255 * 0..255 => 0..255 rounded multiplication +static stbi_uc stbi__blinn_8x8(stbi_uc x, stbi_uc y) +{ + unsigned int t = x*y + 128; + return (stbi_uc) ((t + (t >>8)) >> 8); +} + +static stbi_uc *load_jpeg_image(stbi__jpeg *z, int *out_x, int *out_y, int *comp, int req_comp) +{ + int n, decode_n, is_rgb; + z->s->img_n = 0; // make stbi__cleanup_jpeg safe + + // validate req_comp + if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); + + // load a jpeg image from whichever source, but leave in YCbCr format + if (!stbi__decode_jpeg_image(z)) { stbi__cleanup_jpeg(z); return NULL; } + + // determine actual number of components to generate + n = req_comp ? req_comp : z->s->img_n >= 3 ? 3 : 1; + + is_rgb = z->s->img_n == 3 && (z->rgb == 3 || (z->app14_color_transform == 0 && !z->jfif)); + + if (z->s->img_n == 3 && n < 3 && !is_rgb) + decode_n = 1; + else + decode_n = z->s->img_n; + + // resample and color-convert + { + int k; + unsigned int i,j; + stbi_uc *output; + stbi_uc *coutput[4] = { NULL, NULL, NULL, NULL }; + + stbi__resample res_comp[4]; + + for (k=0; k < decode_n; ++k) { + stbi__resample *r = &res_comp[k]; + + // allocate line buffer big enough for upsampling off the edges + // with upsample factor of 4 + z->img_comp[k].linebuf = (stbi_uc *) stbi__malloc(z->s->img_x + 3); + if (!z->img_comp[k].linebuf) { stbi__cleanup_jpeg(z); return stbi__errpuc("outofmem", "Out of memory"); } + + r->hs = z->img_h_max / z->img_comp[k].h; + r->vs = z->img_v_max / z->img_comp[k].v; + r->ystep = r->vs >> 1; + r->w_lores = (z->s->img_x + r->hs-1) / r->hs; + r->ypos = 0; + r->line0 = r->line1 = z->img_comp[k].data; + + if (r->hs == 1 && r->vs == 1) r->resample = resample_row_1; + else if (r->hs == 1 && r->vs == 2) r->resample = stbi__resample_row_v_2; + else if (r->hs == 2 && r->vs == 1) r->resample = stbi__resample_row_h_2; + else if (r->hs == 2 && r->vs == 2) r->resample = z->resample_row_hv_2_kernel; + else r->resample = stbi__resample_row_generic; + } + + // can't error after this so, this is safe + output = (stbi_uc *) stbi__malloc_mad3(n, z->s->img_x, z->s->img_y, 1); + if (!output) { stbi__cleanup_jpeg(z); return stbi__errpuc("outofmem", "Out of memory"); } + + // now go ahead and resample + for (j=0; j < z->s->img_y; ++j) { + stbi_uc *out = output + n * z->s->img_x * j; + for (k=0; k < decode_n; ++k) { + stbi__resample *r = &res_comp[k]; + int y_bot = r->ystep >= (r->vs >> 1); + coutput[k] = r->resample(z->img_comp[k].linebuf, + y_bot ? r->line1 : r->line0, + y_bot ? r->line0 : r->line1, + r->w_lores, r->hs); + if (++r->ystep >= r->vs) { + r->ystep = 0; + r->line0 = r->line1; + if (++r->ypos < z->img_comp[k].y) + r->line1 += z->img_comp[k].w2; + } + } + if (n >= 3) { + stbi_uc *y = coutput[0]; + if (z->s->img_n == 3) { + if (is_rgb) { + for (i=0; i < z->s->img_x; ++i) { + out[0] = y[i]; + out[1] = coutput[1][i]; + out[2] = coutput[2][i]; + out[3] = 255; + out += n; + } + } else { + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + } + } else if (z->s->img_n == 4) { + if (z->app14_color_transform == 0) { // CMYK + for (i=0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(coutput[0][i], m); + out[1] = stbi__blinn_8x8(coutput[1][i], m); + out[2] = stbi__blinn_8x8(coutput[2][i], m); + out[3] = 255; + out += n; + } + } else if (z->app14_color_transform == 2) { // YCCK + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + for (i=0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(255 - out[0], m); + out[1] = stbi__blinn_8x8(255 - out[1], m); + out[2] = stbi__blinn_8x8(255 - out[2], m); + out += n; + } + } else { // YCbCr + alpha? Ignore the fourth channel for now + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + } + } else + for (i=0; i < z->s->img_x; ++i) { + out[0] = out[1] = out[2] = y[i]; + out[3] = 255; // not used if n==3 + out += n; + } + } else { + if (is_rgb) { + if (n == 1) + for (i=0; i < z->s->img_x; ++i) + *out++ = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + else { + for (i=0; i < z->s->img_x; ++i, out += 2) { + out[0] = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + out[1] = 255; + } + } + } else if (z->s->img_n == 4 && z->app14_color_transform == 0) { + for (i=0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + stbi_uc r = stbi__blinn_8x8(coutput[0][i], m); + stbi_uc g = stbi__blinn_8x8(coutput[1][i], m); + stbi_uc b = stbi__blinn_8x8(coutput[2][i], m); + out[0] = stbi__compute_y(r, g, b); + out[1] = 255; + out += n; + } + } else if (z->s->img_n == 4 && z->app14_color_transform == 2) { + for (i=0; i < z->s->img_x; ++i) { + out[0] = stbi__blinn_8x8(255 - coutput[0][i], coutput[3][i]); + out[1] = 255; + out += n; + } + } else { + stbi_uc *y = coutput[0]; + if (n == 1) + for (i=0; i < z->s->img_x; ++i) out[i] = y[i]; + else + for (i=0; i < z->s->img_x; ++i) { *out++ = y[i]; *out++ = 255; } + } + } + } + stbi__cleanup_jpeg(z); + *out_x = z->s->img_x; + *out_y = z->s->img_y; + if (comp) *comp = z->s->img_n >= 3 ? 3 : 1; // report original components, not output + return output; + } +} + +static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + unsigned char* result; + stbi__jpeg* j = (stbi__jpeg*) stbi__malloc(sizeof(stbi__jpeg)); + STBI_NOTUSED(ri); + j->s = s; + stbi__setup_jpeg(j); + result = load_jpeg_image(j, x,y,comp,req_comp); + STBI_FREE(j); + return result; +} + +static int stbi__jpeg_test(stbi__context *s) +{ + int r; + stbi__jpeg* j = (stbi__jpeg*)stbi__malloc(sizeof(stbi__jpeg)); + j->s = s; + stbi__setup_jpeg(j); + r = stbi__decode_jpeg_header(j, STBI__SCAN_type); + stbi__rewind(s); + STBI_FREE(j); + return r; +} + +static int stbi__jpeg_info_raw(stbi__jpeg *j, int *x, int *y, int *comp) +{ + if (!stbi__decode_jpeg_header(j, STBI__SCAN_header)) { + stbi__rewind( j->s ); + return 0; + } + if (x) *x = j->s->img_x; + if (y) *y = j->s->img_y; + if (comp) *comp = j->s->img_n >= 3 ? 3 : 1; + return 1; +} + +static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp) +{ + int result; + stbi__jpeg* j = (stbi__jpeg*) (stbi__malloc(sizeof(stbi__jpeg))); + j->s = s; + result = stbi__jpeg_info_raw(j, x, y, comp); + STBI_FREE(j); + return result; +} +#endif + +// public domain zlib decode v0.2 Sean Barrett 2006-11-18 +// simple implementation +// - all input must be provided in an upfront buffer +// - all output is written to a single output buffer (can malloc/realloc) +// performance +// - fast huffman + +#ifndef STBI_NO_ZLIB + +// fast-way is faster to check than jpeg huffman, but slow way is slower +#define STBI__ZFAST_BITS 9 // accelerate all cases in default tables +#define STBI__ZFAST_MASK ((1 << STBI__ZFAST_BITS) - 1) + +// zlib-style huffman encoding +// (jpegs packs from left, zlib from right, so can't share code) +typedef struct +{ + stbi__uint16 fast[1 << STBI__ZFAST_BITS]; + stbi__uint16 firstcode[16]; + int maxcode[17]; + stbi__uint16 firstsymbol[16]; + stbi_uc size[288]; + stbi__uint16 value[288]; +} stbi__zhuffman; + +stbi_inline static int stbi__bitreverse16(int n) +{ + n = ((n & 0xAAAA) >> 1) | ((n & 0x5555) << 1); + n = ((n & 0xCCCC) >> 2) | ((n & 0x3333) << 2); + n = ((n & 0xF0F0) >> 4) | ((n & 0x0F0F) << 4); + n = ((n & 0xFF00) >> 8) | ((n & 0x00FF) << 8); + return n; +} + +stbi_inline static int stbi__bit_reverse(int v, int bits) +{ + STBI_ASSERT(bits <= 16); + // to bit reverse n bits, reverse 16 and shift + // e.g. 11 bits, bit reverse and shift away 5 + return stbi__bitreverse16(v) >> (16-bits); +} + +static int stbi__zbuild_huffman(stbi__zhuffman *z, const stbi_uc *sizelist, int num) +{ + int i,k=0; + int code, next_code[16], sizes[17]; + + // DEFLATE spec for generating codes + memset(sizes, 0, sizeof(sizes)); + memset(z->fast, 0, sizeof(z->fast)); + for (i=0; i < num; ++i) + ++sizes[sizelist[i]]; + sizes[0] = 0; + for (i=1; i < 16; ++i) + if (sizes[i] > (1 << i)) + return stbi__err("bad sizes", "Corrupt PNG"); + code = 0; + for (i=1; i < 16; ++i) { + next_code[i] = code; + z->firstcode[i] = (stbi__uint16) code; + z->firstsymbol[i] = (stbi__uint16) k; + code = (code + sizes[i]); + if (sizes[i]) + if (code-1 >= (1 << i)) return stbi__err("bad codelengths","Corrupt PNG"); + z->maxcode[i] = code << (16-i); // preshift for inner loop + code <<= 1; + k += sizes[i]; + } + z->maxcode[16] = 0x10000; // sentinel + for (i=0; i < num; ++i) { + int s = sizelist[i]; + if (s) { + int c = next_code[s] - z->firstcode[s] + z->firstsymbol[s]; + stbi__uint16 fastv = (stbi__uint16) ((s << 9) | i); + z->size [c] = (stbi_uc ) s; + z->value[c] = (stbi__uint16) i; + if (s <= STBI__ZFAST_BITS) { + int j = stbi__bit_reverse(next_code[s],s); + while (j < (1 << STBI__ZFAST_BITS)) { + z->fast[j] = fastv; + j += (1 << s); + } + } + ++next_code[s]; + } + } + return 1; +} + +// zlib-from-memory implementation for PNG reading +// because PNG allows splitting the zlib stream arbitrarily, +// and it's annoying structurally to have PNG call ZLIB call PNG, +// we require PNG read all the IDATs and combine them into a single +// memory buffer + +typedef struct +{ + stbi_uc *zbuffer, *zbuffer_end; + int num_bits; + stbi__uint32 code_buffer; + + char *zout; + char *zout_start; + char *zout_end; + int z_expandable; + + stbi__zhuffman z_length, z_distance; +} stbi__zbuf; + +stbi_inline static stbi_uc stbi__zget8(stbi__zbuf *z) +{ + if (z->zbuffer >= z->zbuffer_end) return 0; + return *z->zbuffer++; +} + +static void stbi__fill_bits(stbi__zbuf *z) +{ + do { + STBI_ASSERT(z->code_buffer < (1U << z->num_bits)); + z->code_buffer |= (unsigned int) stbi__zget8(z) << z->num_bits; + z->num_bits += 8; + } while (z->num_bits <= 24); +} + +stbi_inline static unsigned int stbi__zreceive(stbi__zbuf *z, int n) +{ + unsigned int k; + if (z->num_bits < n) stbi__fill_bits(z); + k = z->code_buffer & ((1 << n) - 1); + z->code_buffer >>= n; + z->num_bits -= n; + return k; +} + +static int stbi__zhuffman_decode_slowpath(stbi__zbuf *a, stbi__zhuffman *z) +{ + int b,s,k; + // not resolved by fast table, so compute it the slow way + // use jpeg approach, which requires MSbits at top + k = stbi__bit_reverse(a->code_buffer, 16); + for (s=STBI__ZFAST_BITS+1; ; ++s) + if (k < z->maxcode[s]) + break; + if (s == 16) return -1; // invalid code! + // code size is s, so: + b = (k >> (16-s)) - z->firstcode[s] + z->firstsymbol[s]; + STBI_ASSERT(z->size[b] == s); + a->code_buffer >>= s; + a->num_bits -= s; + return z->value[b]; +} + +stbi_inline static int stbi__zhuffman_decode(stbi__zbuf *a, stbi__zhuffman *z) +{ + int b,s; + if (a->num_bits < 16) stbi__fill_bits(a); + b = z->fast[a->code_buffer & STBI__ZFAST_MASK]; + if (b) { + s = b >> 9; + a->code_buffer >>= s; + a->num_bits -= s; + return b & 511; + } + return stbi__zhuffman_decode_slowpath(a, z); +} + +static int stbi__zexpand(stbi__zbuf *z, char *zout, int n) // need to make room for n bytes +{ + char *q; + int cur, limit, old_limit; + z->zout = zout; + if (!z->z_expandable) return stbi__err("output buffer limit","Corrupt PNG"); + cur = (int) (z->zout - z->zout_start); + limit = old_limit = (int) (z->zout_end - z->zout_start); + while (cur + n > limit) + limit *= 2; + q = (char *) STBI_REALLOC_SIZED(z->zout_start, old_limit, limit); + STBI_NOTUSED(old_limit); + if (q == NULL) return stbi__err("outofmem", "Out of memory"); + z->zout_start = q; + z->zout = q + cur; + z->zout_end = q + limit; + return 1; +} + +static const int stbi__zlength_base[31] = { + 3,4,5,6,7,8,9,10,11,13, + 15,17,19,23,27,31,35,43,51,59, + 67,83,99,115,131,163,195,227,258,0,0 }; + +static const int stbi__zlength_extra[31]= +{ 0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0,0,0 }; + +static const int stbi__zdist_base[32] = { 1,2,3,4,5,7,9,13,17,25,33,49,65,97,129,193, +257,385,513,769,1025,1537,2049,3073,4097,6145,8193,12289,16385,24577,0,0}; + +static const int stbi__zdist_extra[32] = +{ 0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13}; + +static int stbi__parse_huffman_block(stbi__zbuf *a) +{ + char *zout = a->zout; + for(;;) { + int z = stbi__zhuffman_decode(a, &a->z_length); + if (z < 256) { + if (z < 0) return stbi__err("bad huffman code","Corrupt PNG"); // error in huffman codes + if (zout >= a->zout_end) { + if (!stbi__zexpand(a, zout, 1)) return 0; + zout = a->zout; + } + *zout++ = (char) z; + } else { + stbi_uc *p; + int len,dist; + if (z == 256) { + a->zout = zout; + return 1; + } + z -= 257; + len = stbi__zlength_base[z]; + if (stbi__zlength_extra[z]) len += stbi__zreceive(a, stbi__zlength_extra[z]); + z = stbi__zhuffman_decode(a, &a->z_distance); + if (z < 0) return stbi__err("bad huffman code","Corrupt PNG"); + dist = stbi__zdist_base[z]; + if (stbi__zdist_extra[z]) dist += stbi__zreceive(a, stbi__zdist_extra[z]); + if (zout - a->zout_start < dist) return stbi__err("bad dist","Corrupt PNG"); + if (zout + len > a->zout_end) { + if (!stbi__zexpand(a, zout, len)) return 0; + zout = a->zout; + } + p = (stbi_uc *) (zout - dist); + if (dist == 1) { // run of one byte; common in images. + stbi_uc v = *p; + if (len) { do *zout++ = v; while (--len); } + } else { + if (len) { do *zout++ = *p++; while (--len); } + } + } + } +} + +static int stbi__compute_huffman_codes(stbi__zbuf *a) +{ + static const stbi_uc length_dezigzag[19] = { 16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15 }; + stbi__zhuffman z_codelength; + stbi_uc lencodes[286+32+137];//padding for maximum single op + stbi_uc codelength_sizes[19]; + int i,n; + + int hlit = stbi__zreceive(a,5) + 257; + int hdist = stbi__zreceive(a,5) + 1; + int hclen = stbi__zreceive(a,4) + 4; + int ntot = hlit + hdist; + + memset(codelength_sizes, 0, sizeof(codelength_sizes)); + for (i=0; i < hclen; ++i) { + int s = stbi__zreceive(a,3); + codelength_sizes[length_dezigzag[i]] = (stbi_uc) s; + } + if (!stbi__zbuild_huffman(&z_codelength, codelength_sizes, 19)) return 0; + + n = 0; + while (n < ntot) { + int c = stbi__zhuffman_decode(a, &z_codelength); + if (c < 0 || c >= 19) return stbi__err("bad codelengths", "Corrupt PNG"); + if (c < 16) + lencodes[n++] = (stbi_uc) c; + else { + stbi_uc fill = 0; + if (c == 16) { + c = stbi__zreceive(a,2)+3; + if (n == 0) return stbi__err("bad codelengths", "Corrupt PNG"); + fill = lencodes[n-1]; + } else if (c == 17) + c = stbi__zreceive(a,3)+3; + else { + STBI_ASSERT(c == 18); + c = stbi__zreceive(a,7)+11; + } + if (ntot - n < c) return stbi__err("bad codelengths", "Corrupt PNG"); + memset(lencodes+n, fill, c); + n += c; + } + } + if (n != ntot) return stbi__err("bad codelengths","Corrupt PNG"); + if (!stbi__zbuild_huffman(&a->z_length, lencodes, hlit)) return 0; + if (!stbi__zbuild_huffman(&a->z_distance, lencodes+hlit, hdist)) return 0; + return 1; +} + +static int stbi__parse_uncompressed_block(stbi__zbuf *a) +{ + stbi_uc header[4]; + int len,nlen,k; + if (a->num_bits & 7) + stbi__zreceive(a, a->num_bits & 7); // discard + // drain the bit-packed data into header + k = 0; + while (a->num_bits > 0) { + header[k++] = (stbi_uc) (a->code_buffer & 255); // suppress MSVC run-time check + a->code_buffer >>= 8; + a->num_bits -= 8; + } + STBI_ASSERT(a->num_bits == 0); + // now fill header the normal way + while (k < 4) + header[k++] = stbi__zget8(a); + len = header[1] * 256 + header[0]; + nlen = header[3] * 256 + header[2]; + if (nlen != (len ^ 0xffff)) return stbi__err("zlib corrupt","Corrupt PNG"); + if (a->zbuffer + len > a->zbuffer_end) return stbi__err("read past buffer","Corrupt PNG"); + if (a->zout + len > a->zout_end) + if (!stbi__zexpand(a, a->zout, len)) return 0; + memcpy(a->zout, a->zbuffer, len); + a->zbuffer += len; + a->zout += len; + return 1; +} + +static int stbi__parse_zlib_header(stbi__zbuf *a) +{ + int cmf = stbi__zget8(a); + int cm = cmf & 15; + /* int cinfo = cmf >> 4; */ + int flg = stbi__zget8(a); + if ((cmf*256+flg) % 31 != 0) return stbi__err("bad zlib header","Corrupt PNG"); // zlib spec + if (flg & 32) return stbi__err("no preset dict","Corrupt PNG"); // preset dictionary not allowed in png + if (cm != 8) return stbi__err("bad compression","Corrupt PNG"); // DEFLATE required for png + // window = 1 << (8 + cinfo)... but who cares, we fully buffer output + return 1; +} + +static const stbi_uc stbi__zdefault_length[288] = +{ + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, 7,7,7,7,7,7,7,7,8,8,8,8,8,8,8,8 +}; +static const stbi_uc stbi__zdefault_distance[32] = +{ + 5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5 +}; +/* +Init algorithm: +{ + int i; // use <= to match clearly with spec + for (i=0; i <= 143; ++i) stbi__zdefault_length[i] = 8; + for ( ; i <= 255; ++i) stbi__zdefault_length[i] = 9; + for ( ; i <= 279; ++i) stbi__zdefault_length[i] = 7; + for ( ; i <= 287; ++i) stbi__zdefault_length[i] = 8; + + for (i=0; i <= 31; ++i) stbi__zdefault_distance[i] = 5; +} +*/ + +static int stbi__parse_zlib(stbi__zbuf *a, int parse_header) +{ + int final, type; + if (parse_header) + if (!stbi__parse_zlib_header(a)) return 0; + a->num_bits = 0; + a->code_buffer = 0; + do { + final = stbi__zreceive(a,1); + type = stbi__zreceive(a,2); + if (type == 0) { + if (!stbi__parse_uncompressed_block(a)) return 0; + } else if (type == 3) { + return 0; + } else { + if (type == 1) { + // use fixed code lengths + if (!stbi__zbuild_huffman(&a->z_length , stbi__zdefault_length , 288)) return 0; + if (!stbi__zbuild_huffman(&a->z_distance, stbi__zdefault_distance, 32)) return 0; + } else { + if (!stbi__compute_huffman_codes(a)) return 0; + } + if (!stbi__parse_huffman_block(a)) return 0; + } + } while (!final); + return 1; +} + +static int stbi__do_zlib(stbi__zbuf *a, char *obuf, int olen, int exp, int parse_header) +{ + a->zout_start = obuf; + a->zout = obuf; + a->zout_end = obuf + olen; + a->z_expandable = exp; + + return stbi__parse_zlib(a, parse_header); +} + +STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen) +{ + stbi__zbuf a; + char *p = (char *) stbi__malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *) buffer; + a.zbuffer_end = (stbi_uc *) buffer + len; + if (stbi__do_zlib(&a, p, initial_size, 1, 1)) { + if (outlen) *outlen = (int) (a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } +} + +STBIDEF char *stbi_zlib_decode_malloc(char const *buffer, int len, int *outlen) +{ + return stbi_zlib_decode_malloc_guesssize(buffer, len, 16384, outlen); +} + +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header) +{ + stbi__zbuf a; + char *p = (char *) stbi__malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *) buffer; + a.zbuffer_end = (stbi_uc *) buffer + len; + if (stbi__do_zlib(&a, p, initial_size, 1, parse_header)) { + if (outlen) *outlen = (int) (a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } +} + +STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, char const *ibuffer, int ilen) +{ + stbi__zbuf a; + a.zbuffer = (stbi_uc *) ibuffer; + a.zbuffer_end = (stbi_uc *) ibuffer + ilen; + if (stbi__do_zlib(&a, obuffer, olen, 0, 1)) + return (int) (a.zout - a.zout_start); + else + return -1; +} + +STBIDEF char *stbi_zlib_decode_noheader_malloc(char const *buffer, int len, int *outlen) +{ + stbi__zbuf a; + char *p = (char *) stbi__malloc(16384); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *) buffer; + a.zbuffer_end = (stbi_uc *) buffer+len; + if (stbi__do_zlib(&a, p, 16384, 1, 0)) { + if (outlen) *outlen = (int) (a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } +} + +STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen) +{ + stbi__zbuf a; + a.zbuffer = (stbi_uc *) ibuffer; + a.zbuffer_end = (stbi_uc *) ibuffer + ilen; + if (stbi__do_zlib(&a, obuffer, olen, 0, 0)) + return (int) (a.zout - a.zout_start); + else + return -1; +} +#endif + +// public domain "baseline" PNG decoder v0.10 Sean Barrett 2006-11-18 +// simple implementation +// - only 8-bit samples +// - no CRC checking +// - allocates lots of intermediate memory +// - avoids problem of streaming data between subsystems +// - avoids explicit window management +// performance +// - uses stb_zlib, a PD zlib implementation with fast huffman decoding + +#ifndef STBI_NO_PNG +typedef struct +{ + stbi__uint32 length; + stbi__uint32 type; +} stbi__pngchunk; + +static stbi__pngchunk stbi__get_chunk_header(stbi__context *s) +{ + stbi__pngchunk c; + c.length = stbi__get32be(s); + c.type = stbi__get32be(s); + return c; +} + +static int stbi__check_png_header(stbi__context *s) +{ + static const stbi_uc png_sig[8] = { 137,80,78,71,13,10,26,10 }; + int i; + for (i=0; i < 8; ++i) + if (stbi__get8(s) != png_sig[i]) return stbi__err("bad png sig","Not a PNG"); + return 1; +} + +typedef struct +{ + stbi__context *s; + stbi_uc *idata, *expanded, *out; + int depth; +} stbi__png; + + +enum { + STBI__F_none=0, + STBI__F_sub=1, + STBI__F_up=2, + STBI__F_avg=3, + STBI__F_paeth=4, + // synthetic filters used for first scanline to avoid needing a dummy row of 0s + STBI__F_avg_first, + STBI__F_paeth_first +}; + +static stbi_uc first_row_filter[5] = +{ + STBI__F_none, + STBI__F_sub, + STBI__F_none, + STBI__F_avg_first, + STBI__F_paeth_first +}; + +static int stbi__paeth(int a, int b, int c) +{ + int p = a + b - c; + int pa = abs(p-a); + int pb = abs(p-b); + int pc = abs(p-c); + if (pa <= pb && pa <= pc) return a; + if (pb <= pc) return b; + return c; +} + +static const stbi_uc stbi__depth_scale_table[9] = { 0, 0xff, 0x55, 0, 0x11, 0,0,0, 0x01 }; + +// create the png data from post-deflated data +static int stbi__create_png_image_raw(stbi__png *a, stbi_uc *raw, stbi__uint32 raw_len, int out_n, stbi__uint32 x, stbi__uint32 y, int depth, int color) +{ + int bytes = (depth == 16? 2 : 1); + stbi__context *s = a->s; + stbi__uint32 i,j,stride = x*out_n*bytes; + stbi__uint32 img_len, img_width_bytes; + int k; + int img_n = s->img_n; // copy it into a local for later + + int output_bytes = out_n*bytes; + int filter_bytes = img_n*bytes; + int width = x; + + STBI_ASSERT(out_n == s->img_n || out_n == s->img_n+1); + a->out = (stbi_uc *) stbi__malloc_mad3(x, y, output_bytes, 0); // extra bytes to write off the end into + if (!a->out) return stbi__err("outofmem", "Out of memory"); + + if (!stbi__mad3sizes_valid(img_n, x, depth, 7)) return stbi__err("too large", "Corrupt PNG"); + img_width_bytes = (((img_n * x * depth) + 7) >> 3); + img_len = (img_width_bytes + 1) * y; + + // we used to check for exact match between raw_len and img_len on non-interlaced PNGs, + // but issue #276 reported a PNG in the wild that had extra data at the end (all zeros), + // so just check for raw_len < img_len always. + if (raw_len < img_len) return stbi__err("not enough pixels","Corrupt PNG"); + + for (j=0; j < y; ++j) { + stbi_uc *cur = a->out + stride*j; + stbi_uc *prior; + int filter = *raw++; + + if (filter > 4) + return stbi__err("invalid filter","Corrupt PNG"); + + if (depth < 8) { + STBI_ASSERT(img_width_bytes <= x); + cur += x*out_n - img_width_bytes; // store output to the rightmost img_len bytes, so we can decode in place + filter_bytes = 1; + width = img_width_bytes; + } + prior = cur - stride; // bugfix: need to compute this after 'cur +=' computation above + + // if first row, use special filter that doesn't sample previous row + if (j == 0) filter = first_row_filter[filter]; + + // handle first byte explicitly + for (k=0; k < filter_bytes; ++k) { + switch (filter) { + case STBI__F_none : cur[k] = raw[k]; break; + case STBI__F_sub : cur[k] = raw[k]; break; + case STBI__F_up : cur[k] = STBI__BYTECAST(raw[k] + prior[k]); break; + case STBI__F_avg : cur[k] = STBI__BYTECAST(raw[k] + (prior[k]>>1)); break; + case STBI__F_paeth : cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(0,prior[k],0)); break; + case STBI__F_avg_first : cur[k] = raw[k]; break; + case STBI__F_paeth_first: cur[k] = raw[k]; break; + } + } + + if (depth == 8) { + if (img_n != out_n) + cur[img_n] = 255; // first pixel + raw += img_n; + cur += out_n; + prior += out_n; + } else if (depth == 16) { + if (img_n != out_n) { + cur[filter_bytes] = 255; // first pixel top byte + cur[filter_bytes+1] = 255; // first pixel bottom byte + } + raw += filter_bytes; + cur += output_bytes; + prior += output_bytes; + } else { + raw += 1; + cur += 1; + prior += 1; + } + + // this is a little gross, so that we don't switch per-pixel or per-component + if (depth < 8 || img_n == out_n) { + int nk = (width - 1)*filter_bytes; + #define STBI__CASE(f) \ + case f: \ + for (k=0; k < nk; ++k) + switch (filter) { + // "none" filter turns into a memcpy here; make that explicit. + case STBI__F_none: memcpy(cur, raw, nk); break; + STBI__CASE(STBI__F_sub) { cur[k] = STBI__BYTECAST(raw[k] + cur[k-filter_bytes]); } break; + STBI__CASE(STBI__F_up) { cur[k] = STBI__BYTECAST(raw[k] + prior[k]); } break; + STBI__CASE(STBI__F_avg) { cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k-filter_bytes])>>1)); } break; + STBI__CASE(STBI__F_paeth) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes],prior[k],prior[k-filter_bytes])); } break; + STBI__CASE(STBI__F_avg_first) { cur[k] = STBI__BYTECAST(raw[k] + (cur[k-filter_bytes] >> 1)); } break; + STBI__CASE(STBI__F_paeth_first) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes],0,0)); } break; + } + #undef STBI__CASE + raw += nk; + } else { + STBI_ASSERT(img_n+1 == out_n); + #define STBI__CASE(f) \ + case f: \ + for (i=x-1; i >= 1; --i, cur[filter_bytes]=255,raw+=filter_bytes,cur+=output_bytes,prior+=output_bytes) \ + for (k=0; k < filter_bytes; ++k) + switch (filter) { + STBI__CASE(STBI__F_none) { cur[k] = raw[k]; } break; + STBI__CASE(STBI__F_sub) { cur[k] = STBI__BYTECAST(raw[k] + cur[k- output_bytes]); } break; + STBI__CASE(STBI__F_up) { cur[k] = STBI__BYTECAST(raw[k] + prior[k]); } break; + STBI__CASE(STBI__F_avg) { cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k- output_bytes])>>1)); } break; + STBI__CASE(STBI__F_paeth) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k- output_bytes],prior[k],prior[k- output_bytes])); } break; + STBI__CASE(STBI__F_avg_first) { cur[k] = STBI__BYTECAST(raw[k] + (cur[k- output_bytes] >> 1)); } break; + STBI__CASE(STBI__F_paeth_first) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k- output_bytes],0,0)); } break; + } + #undef STBI__CASE + + // the loop above sets the high byte of the pixels' alpha, but for + // 16 bit png files we also need the low byte set. we'll do that here. + if (depth == 16) { + cur = a->out + stride*j; // start at the beginning of the row again + for (i=0; i < x; ++i,cur+=output_bytes) { + cur[filter_bytes+1] = 255; + } + } + } + } + + // we make a separate pass to expand bits to pixels; for performance, + // this could run two scanlines behind the above code, so it won't + // intefere with filtering but will still be in the cache. + if (depth < 8) { + for (j=0; j < y; ++j) { + stbi_uc *cur = a->out + stride*j; + stbi_uc *in = a->out + stride*j + x*out_n - img_width_bytes; + // unpack 1/2/4-bit into a 8-bit buffer. allows us to keep the common 8-bit path optimal at minimal cost for 1/2/4-bit + // png guarante byte alignment, if width is not multiple of 8/4/2 we'll decode dummy trailing data that will be skipped in the later loop + stbi_uc scale = (color == 0) ? stbi__depth_scale_table[depth] : 1; // scale grayscale values to 0..255 range + + // note that the final byte might overshoot and write more data than desired. + // we can allocate enough data that this never writes out of memory, but it + // could also overwrite the next scanline. can it overwrite non-empty data + // on the next scanline? yes, consider 1-pixel-wide scanlines with 1-bit-per-pixel. + // so we need to explicitly clamp the final ones + + if (depth == 4) { + for (k=x*img_n; k >= 2; k-=2, ++in) { + *cur++ = scale * ((*in >> 4) ); + *cur++ = scale * ((*in ) & 0x0f); + } + if (k > 0) *cur++ = scale * ((*in >> 4) ); + } else if (depth == 2) { + for (k=x*img_n; k >= 4; k-=4, ++in) { + *cur++ = scale * ((*in >> 6) ); + *cur++ = scale * ((*in >> 4) & 0x03); + *cur++ = scale * ((*in >> 2) & 0x03); + *cur++ = scale * ((*in ) & 0x03); + } + if (k > 0) *cur++ = scale * ((*in >> 6) ); + if (k > 1) *cur++ = scale * ((*in >> 4) & 0x03); + if (k > 2) *cur++ = scale * ((*in >> 2) & 0x03); + } else if (depth == 1) { + for (k=x*img_n; k >= 8; k-=8, ++in) { + *cur++ = scale * ((*in >> 7) ); + *cur++ = scale * ((*in >> 6) & 0x01); + *cur++ = scale * ((*in >> 5) & 0x01); + *cur++ = scale * ((*in >> 4) & 0x01); + *cur++ = scale * ((*in >> 3) & 0x01); + *cur++ = scale * ((*in >> 2) & 0x01); + *cur++ = scale * ((*in >> 1) & 0x01); + *cur++ = scale * ((*in ) & 0x01); + } + if (k > 0) *cur++ = scale * ((*in >> 7) ); + if (k > 1) *cur++ = scale * ((*in >> 6) & 0x01); + if (k > 2) *cur++ = scale * ((*in >> 5) & 0x01); + if (k > 3) *cur++ = scale * ((*in >> 4) & 0x01); + if (k > 4) *cur++ = scale * ((*in >> 3) & 0x01); + if (k > 5) *cur++ = scale * ((*in >> 2) & 0x01); + if (k > 6) *cur++ = scale * ((*in >> 1) & 0x01); + } + if (img_n != out_n) { + int q; + // insert alpha = 255 + cur = a->out + stride*j; + if (img_n == 1) { + for (q=x-1; q >= 0; --q) { + cur[q*2+1] = 255; + cur[q*2+0] = cur[q]; + } + } else { + STBI_ASSERT(img_n == 3); + for (q=x-1; q >= 0; --q) { + cur[q*4+3] = 255; + cur[q*4+2] = cur[q*3+2]; + cur[q*4+1] = cur[q*3+1]; + cur[q*4+0] = cur[q*3+0]; + } + } + } + } + } else if (depth == 16) { + // force the image data from big-endian to platform-native. + // this is done in a separate pass due to the decoding relying + // on the data being untouched, but could probably be done + // per-line during decode if care is taken. + stbi_uc *cur = a->out; + stbi__uint16 *cur16 = (stbi__uint16*)cur; + + for(i=0; i < x*y*out_n; ++i,cur16++,cur+=2) { + *cur16 = (cur[0] << 8) | cur[1]; + } + } + + return 1; +} + +static int stbi__create_png_image(stbi__png *a, stbi_uc *image_data, stbi__uint32 image_data_len, int out_n, int depth, int color, int interlaced) +{ + int bytes = (depth == 16 ? 2 : 1); + int out_bytes = out_n * bytes; + stbi_uc *final; + int p; + if (!interlaced) + return stbi__create_png_image_raw(a, image_data, image_data_len, out_n, a->s->img_x, a->s->img_y, depth, color); + + // de-interlacing + final = (stbi_uc *) stbi__malloc_mad3(a->s->img_x, a->s->img_y, out_bytes, 0); + for (p=0; p < 7; ++p) { + int xorig[] = { 0,4,0,2,0,1,0 }; + int yorig[] = { 0,0,4,0,2,0,1 }; + int xspc[] = { 8,8,4,4,2,2,1 }; + int yspc[] = { 8,8,8,4,4,2,2 }; + int i,j,x,y; + // pass1_x[4] = 0, pass1_x[5] = 1, pass1_x[12] = 1 + x = (a->s->img_x - xorig[p] + xspc[p]-1) / xspc[p]; + y = (a->s->img_y - yorig[p] + yspc[p]-1) / yspc[p]; + if (x && y) { + stbi__uint32 img_len = ((((a->s->img_n * x * depth) + 7) >> 3) + 1) * y; + if (!stbi__create_png_image_raw(a, image_data, image_data_len, out_n, x, y, depth, color)) { + STBI_FREE(final); + return 0; + } + for (j=0; j < y; ++j) { + for (i=0; i < x; ++i) { + int out_y = j*yspc[p]+yorig[p]; + int out_x = i*xspc[p]+xorig[p]; + memcpy(final + out_y*a->s->img_x*out_bytes + out_x*out_bytes, + a->out + (j*x+i)*out_bytes, out_bytes); + } + } + STBI_FREE(a->out); + image_data += img_len; + image_data_len -= img_len; + } + } + a->out = final; + + return 1; +} + +static int stbi__compute_transparency(stbi__png *z, stbi_uc tc[3], int out_n) +{ + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi_uc *p = z->out; + + // compute color-based transparency, assuming we've + // already got 255 as the alpha value in the output + STBI_ASSERT(out_n == 2 || out_n == 4); + + if (out_n == 2) { + for (i=0; i < pixel_count; ++i) { + p[1] = (p[0] == tc[0] ? 0 : 255); + p += 2; + } + } else { + for (i=0; i < pixel_count; ++i) { + if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) + p[3] = 0; + p += 4; + } + } + return 1; +} + +static int stbi__compute_transparency16(stbi__png *z, stbi__uint16 tc[3], int out_n) +{ + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi__uint16 *p = (stbi__uint16*) z->out; + + // compute color-based transparency, assuming we've + // already got 65535 as the alpha value in the output + STBI_ASSERT(out_n == 2 || out_n == 4); + + if (out_n == 2) { + for (i = 0; i < pixel_count; ++i) { + p[1] = (p[0] == tc[0] ? 0 : 65535); + p += 2; + } + } else { + for (i = 0; i < pixel_count; ++i) { + if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) + p[3] = 0; + p += 4; + } + } + return 1; +} + +static int stbi__expand_png_palette(stbi__png *a, stbi_uc *palette, int len, int pal_img_n) +{ + stbi__uint32 i, pixel_count = a->s->img_x * a->s->img_y; + stbi_uc *p, *temp_out, *orig = a->out; + + p = (stbi_uc *) stbi__malloc_mad2(pixel_count, pal_img_n, 0); + if (p == NULL) return stbi__err("outofmem", "Out of memory"); + + // between here and free(out) below, exitting would leak + temp_out = p; + + if (pal_img_n == 3) { + for (i=0; i < pixel_count; ++i) { + int n = orig[i]*4; + p[0] = palette[n ]; + p[1] = palette[n+1]; + p[2] = palette[n+2]; + p += 3; + } + } else { + for (i=0; i < pixel_count; ++i) { + int n = orig[i]*4; + p[0] = palette[n ]; + p[1] = palette[n+1]; + p[2] = palette[n+2]; + p[3] = palette[n+3]; + p += 4; + } + } + STBI_FREE(a->out); + a->out = temp_out; + + STBI_NOTUSED(len); + + return 1; +} + +static int stbi__unpremultiply_on_load = 0; +static int stbi__de_iphone_flag = 0; + +STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply) +{ + stbi__unpremultiply_on_load = flag_true_if_should_unpremultiply; +} + +STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert) +{ + stbi__de_iphone_flag = flag_true_if_should_convert; +} + +static void stbi__de_iphone(stbi__png *z) +{ + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi_uc *p = z->out; + + if (s->img_out_n == 3) { // convert bgr to rgb + for (i=0; i < pixel_count; ++i) { + stbi_uc t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 3; + } + } else { + STBI_ASSERT(s->img_out_n == 4); + if (stbi__unpremultiply_on_load) { + // convert bgr to rgb and unpremultiply + for (i=0; i < pixel_count; ++i) { + stbi_uc a = p[3]; + stbi_uc t = p[0]; + if (a) { + stbi_uc half = a / 2; + p[0] = (p[2] * 255 + half) / a; + p[1] = (p[1] * 255 + half) / a; + p[2] = ( t * 255 + half) / a; + } else { + p[0] = p[2]; + p[2] = t; + } + p += 4; + } + } else { + // convert bgr to rgb + for (i=0; i < pixel_count; ++i) { + stbi_uc t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 4; + } + } + } +} + +#define STBI__PNG_TYPE(a,b,c,d) (((unsigned) (a) << 24) + ((unsigned) (b) << 16) + ((unsigned) (c) << 8) + (unsigned) (d)) + +static int stbi__parse_png_file(stbi__png *z, int scan, int req_comp) +{ + stbi_uc palette[1024], pal_img_n=0; + stbi_uc has_trans=0, tc[3]={0}; + stbi__uint16 tc16[3]; + stbi__uint32 ioff=0, idata_limit=0, i, pal_len=0; + int first=1,k,interlace=0, color=0, is_iphone=0; + stbi__context *s = z->s; + + z->expanded = NULL; + z->idata = NULL; + z->out = NULL; + + if (!stbi__check_png_header(s)) return 0; + + if (scan == STBI__SCAN_type) return 1; + + for (;;) { + stbi__pngchunk c = stbi__get_chunk_header(s); + switch (c.type) { + case STBI__PNG_TYPE('C','g','B','I'): + is_iphone = 1; + stbi__skip(s, c.length); + break; + case STBI__PNG_TYPE('I','H','D','R'): { + int comp,filter; + if (!first) return stbi__err("multiple IHDR","Corrupt PNG"); + first = 0; + if (c.length != 13) return stbi__err("bad IHDR len","Corrupt PNG"); + s->img_x = stbi__get32be(s); if (s->img_x > (1 << 24)) return stbi__err("too large","Very large image (corrupt?)"); + s->img_y = stbi__get32be(s); if (s->img_y > (1 << 24)) return stbi__err("too large","Very large image (corrupt?)"); + z->depth = stbi__get8(s); if (z->depth != 1 && z->depth != 2 && z->depth != 4 && z->depth != 8 && z->depth != 16) return stbi__err("1/2/4/8/16-bit only","PNG not supported: 1/2/4/8/16-bit only"); + color = stbi__get8(s); if (color > 6) return stbi__err("bad ctype","Corrupt PNG"); + if (color == 3 && z->depth == 16) return stbi__err("bad ctype","Corrupt PNG"); + if (color == 3) pal_img_n = 3; else if (color & 1) return stbi__err("bad ctype","Corrupt PNG"); + comp = stbi__get8(s); if (comp) return stbi__err("bad comp method","Corrupt PNG"); + filter= stbi__get8(s); if (filter) return stbi__err("bad filter method","Corrupt PNG"); + interlace = stbi__get8(s); if (interlace>1) return stbi__err("bad interlace method","Corrupt PNG"); + if (!s->img_x || !s->img_y) return stbi__err("0-pixel image","Corrupt PNG"); + if (!pal_img_n) { + s->img_n = (color & 2 ? 3 : 1) + (color & 4 ? 1 : 0); + if ((1 << 30) / s->img_x / s->img_n < s->img_y) return stbi__err("too large", "Image too large to decode"); + if (scan == STBI__SCAN_header) return 1; + } else { + // if paletted, then pal_n is our final components, and + // img_n is # components to decompress/filter. + s->img_n = 1; + if ((1 << 30) / s->img_x / 4 < s->img_y) return stbi__err("too large","Corrupt PNG"); + // if SCAN_header, have to scan to see if we have a tRNS + } + break; + } + + case STBI__PNG_TYPE('P','L','T','E'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (c.length > 256*3) return stbi__err("invalid PLTE","Corrupt PNG"); + pal_len = c.length / 3; + if (pal_len * 3 != c.length) return stbi__err("invalid PLTE","Corrupt PNG"); + for (i=0; i < pal_len; ++i) { + palette[i*4+0] = stbi__get8(s); + palette[i*4+1] = stbi__get8(s); + palette[i*4+2] = stbi__get8(s); + palette[i*4+3] = 255; + } + break; + } + + case STBI__PNG_TYPE('t','R','N','S'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (z->idata) return stbi__err("tRNS after IDAT","Corrupt PNG"); + if (pal_img_n) { + if (scan == STBI__SCAN_header) { s->img_n = 4; return 1; } + if (pal_len == 0) return stbi__err("tRNS before PLTE","Corrupt PNG"); + if (c.length > pal_len) return stbi__err("bad tRNS len","Corrupt PNG"); + pal_img_n = 4; + for (i=0; i < c.length; ++i) + palette[i*4+3] = stbi__get8(s); + } else { + if (!(s->img_n & 1)) return stbi__err("tRNS with alpha","Corrupt PNG"); + if (c.length != (stbi__uint32) s->img_n*2) return stbi__err("bad tRNS len","Corrupt PNG"); + has_trans = 1; + if (z->depth == 16) { + for (k = 0; k < s->img_n; ++k) tc16[k] = (stbi__uint16)stbi__get16be(s); // copy the values as-is + } else { + for (k = 0; k < s->img_n; ++k) tc[k] = (stbi_uc)(stbi__get16be(s) & 255) * stbi__depth_scale_table[z->depth]; // non 8-bit images will be larger + } + } + break; + } + + case STBI__PNG_TYPE('I','D','A','T'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (pal_img_n && !pal_len) return stbi__err("no PLTE","Corrupt PNG"); + if (scan == STBI__SCAN_header) { s->img_n = pal_img_n; return 1; } + if ((int)(ioff + c.length) < (int)ioff) return 0; + if (ioff + c.length > idata_limit) { + stbi__uint32 idata_limit_old = idata_limit; + stbi_uc *p; + if (idata_limit == 0) idata_limit = c.length > 4096 ? c.length : 4096; + while (ioff + c.length > idata_limit) + idata_limit *= 2; + STBI_NOTUSED(idata_limit_old); + p = (stbi_uc *) STBI_REALLOC_SIZED(z->idata, idata_limit_old, idata_limit); if (p == NULL) return stbi__err("outofmem", "Out of memory"); + z->idata = p; + } + if (!stbi__getn(s, z->idata+ioff,c.length)) return stbi__err("outofdata","Corrupt PNG"); + ioff += c.length; + break; + } + + case STBI__PNG_TYPE('I','E','N','D'): { + stbi__uint32 raw_len, bpl; + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (scan != STBI__SCAN_load) return 1; + if (z->idata == NULL) return stbi__err("no IDAT","Corrupt PNG"); + // initial guess for decoded data size to avoid unnecessary reallocs + bpl = (s->img_x * z->depth + 7) / 8; // bytes per line, per component + raw_len = bpl * s->img_y * s->img_n /* pixels */ + s->img_y /* filter mode per row */; + z->expanded = (stbi_uc *) stbi_zlib_decode_malloc_guesssize_headerflag((char *) z->idata, ioff, raw_len, (int *) &raw_len, !is_iphone); + if (z->expanded == NULL) return 0; // zlib should set error + STBI_FREE(z->idata); z->idata = NULL; + if ((req_comp == s->img_n+1 && req_comp != 3 && !pal_img_n) || has_trans) + s->img_out_n = s->img_n+1; + else + s->img_out_n = s->img_n; + if (!stbi__create_png_image(z, z->expanded, raw_len, s->img_out_n, z->depth, color, interlace)) return 0; + if (has_trans) { + if (z->depth == 16) { + if (!stbi__compute_transparency16(z, tc16, s->img_out_n)) return 0; + } else { + if (!stbi__compute_transparency(z, tc, s->img_out_n)) return 0; + } + } + if (is_iphone && stbi__de_iphone_flag && s->img_out_n > 2) + stbi__de_iphone(z); + if (pal_img_n) { + // pal_img_n == 3 or 4 + s->img_n = pal_img_n; // record the actual colors we had + s->img_out_n = pal_img_n; + if (req_comp >= 3) s->img_out_n = req_comp; + if (!stbi__expand_png_palette(z, palette, pal_len, s->img_out_n)) + return 0; + } else if (has_trans) { + // non-paletted image with tRNS -> source image has (constant) alpha + ++s->img_n; + } + STBI_FREE(z->expanded); z->expanded = NULL; + return 1; + } + + default: + // if critical, fail + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if ((c.type & (1 << 29)) == 0) { + #ifndef STBI_NO_FAILURE_STRINGS + // not threadsafe + static char invalid_chunk[] = "XXXX PNG chunk not known"; + invalid_chunk[0] = STBI__BYTECAST(c.type >> 24); + invalid_chunk[1] = STBI__BYTECAST(c.type >> 16); + invalid_chunk[2] = STBI__BYTECAST(c.type >> 8); + invalid_chunk[3] = STBI__BYTECAST(c.type >> 0); + #endif + return stbi__err(invalid_chunk, "PNG not supported: unknown PNG chunk type"); + } + stbi__skip(s, c.length); + break; + } + // end of PNG chunk, read and skip CRC + stbi__get32be(s); + } +} + +static void *stbi__do_png(stbi__png *p, int *x, int *y, int *n, int req_comp, stbi__result_info *ri) +{ + void *result=NULL; + if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); + if (stbi__parse_png_file(p, STBI__SCAN_load, req_comp)) { + if (p->depth < 8) + ri->bits_per_channel = 8; + else + ri->bits_per_channel = p->depth; + result = p->out; + p->out = NULL; + if (req_comp && req_comp != p->s->img_out_n) { + if (ri->bits_per_channel == 8) + result = stbi__convert_format((unsigned char *) result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); + else + result = stbi__convert_format16((stbi__uint16 *) result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); + p->s->img_out_n = req_comp; + if (result == NULL) return result; + } + *x = p->s->img_x; + *y = p->s->img_y; + if (n) *n = p->s->img_n; + } + STBI_FREE(p->out); p->out = NULL; + STBI_FREE(p->expanded); p->expanded = NULL; + STBI_FREE(p->idata); p->idata = NULL; + + return result; +} + +static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + stbi__png p; + p.s = s; + return stbi__do_png(&p, x,y,comp,req_comp, ri); +} + +static int stbi__png_test(stbi__context *s) +{ + int r; + r = stbi__check_png_header(s); + stbi__rewind(s); + return r; +} + +static int stbi__png_info_raw(stbi__png *p, int *x, int *y, int *comp) +{ + if (!stbi__parse_png_file(p, STBI__SCAN_header, 0)) { + stbi__rewind( p->s ); + return 0; + } + if (x) *x = p->s->img_x; + if (y) *y = p->s->img_y; + if (comp) *comp = p->s->img_n; + return 1; +} + +static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp) +{ + stbi__png p; + p.s = s; + return stbi__png_info_raw(&p, x, y, comp); +} + +static int stbi__png_is16(stbi__context *s) +{ + stbi__png p; + p.s = s; + if (!stbi__png_info_raw(&p, NULL, NULL, NULL)) + return 0; + if (p.depth != 16) { + stbi__rewind(p.s); + return 0; + } + return 1; +} +#endif + +// Microsoft/Windows BMP image + +#ifndef STBI_NO_BMP +static int stbi__bmp_test_raw(stbi__context *s) +{ + int r; + int sz; + if (stbi__get8(s) != 'B') return 0; + if (stbi__get8(s) != 'M') return 0; + stbi__get32le(s); // discard filesize + stbi__get16le(s); // discard reserved + stbi__get16le(s); // discard reserved + stbi__get32le(s); // discard data offset + sz = stbi__get32le(s); + r = (sz == 12 || sz == 40 || sz == 56 || sz == 108 || sz == 124); + return r; +} + +static int stbi__bmp_test(stbi__context *s) +{ + int r = stbi__bmp_test_raw(s); + stbi__rewind(s); + return r; +} + + +// returns 0..31 for the highest set bit +static int stbi__high_bit(unsigned int z) +{ + int n=0; + if (z == 0) return -1; + if (z >= 0x10000) { n += 16; z >>= 16; } + if (z >= 0x00100) { n += 8; z >>= 8; } + if (z >= 0x00010) { n += 4; z >>= 4; } + if (z >= 0x00004) { n += 2; z >>= 2; } + if (z >= 0x00002) { n += 1;/* >>= 1;*/ } + return n; +} + +static int stbi__bitcount(unsigned int a) +{ + a = (a & 0x55555555) + ((a >> 1) & 0x55555555); // max 2 + a = (a & 0x33333333) + ((a >> 2) & 0x33333333); // max 4 + a = (a + (a >> 4)) & 0x0f0f0f0f; // max 8 per 4, now 8 bits + a = (a + (a >> 8)); // max 16 per 8 bits + a = (a + (a >> 16)); // max 32 per 8 bits + return a & 0xff; +} + +// extract an arbitrarily-aligned N-bit value (N=bits) +// from v, and then make it 8-bits long and fractionally +// extend it to full full range. +static int stbi__shiftsigned(unsigned int v, int shift, int bits) +{ + static unsigned int mul_table[9] = { + 0, + 0xff/*0b11111111*/, 0x55/*0b01010101*/, 0x49/*0b01001001*/, 0x11/*0b00010001*/, + 0x21/*0b00100001*/, 0x41/*0b01000001*/, 0x81/*0b10000001*/, 0x01/*0b00000001*/, + }; + static unsigned int shift_table[9] = { + 0, 0,0,1,0,2,4,6,0, + }; + if (shift < 0) + v <<= -shift; + else + v >>= shift; + STBI_ASSERT(v >= 0 && v < 256); + v >>= (8-bits); + STBI_ASSERT(bits >= 0 && bits <= 8); + return (int) ((unsigned) v * mul_table[bits]) >> shift_table[bits]; +} + +typedef struct +{ + int bpp, offset, hsz; + unsigned int mr,mg,mb,ma, all_a; +} stbi__bmp_data; + +static void *stbi__bmp_parse_header(stbi__context *s, stbi__bmp_data *info) +{ + int hsz; + if (stbi__get8(s) != 'B' || stbi__get8(s) != 'M') return stbi__errpuc("not BMP", "Corrupt BMP"); + stbi__get32le(s); // discard filesize + stbi__get16le(s); // discard reserved + stbi__get16le(s); // discard reserved + info->offset = stbi__get32le(s); + info->hsz = hsz = stbi__get32le(s); + info->mr = info->mg = info->mb = info->ma = 0; + + if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108 && hsz != 124) return stbi__errpuc("unknown BMP", "BMP type not supported: unknown"); + if (hsz == 12) { + s->img_x = stbi__get16le(s); + s->img_y = stbi__get16le(s); + } else { + s->img_x = stbi__get32le(s); + s->img_y = stbi__get32le(s); + } + if (stbi__get16le(s) != 1) return stbi__errpuc("bad BMP", "bad BMP"); + info->bpp = stbi__get16le(s); + if (hsz != 12) { + int compress = stbi__get32le(s); + if (compress == 1 || compress == 2) return stbi__errpuc("BMP RLE", "BMP type not supported: RLE"); + stbi__get32le(s); // discard sizeof + stbi__get32le(s); // discard hres + stbi__get32le(s); // discard vres + stbi__get32le(s); // discard colorsused + stbi__get32le(s); // discard max important + if (hsz == 40 || hsz == 56) { + if (hsz == 56) { + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + } + if (info->bpp == 16 || info->bpp == 32) { + if (compress == 0) { + if (info->bpp == 32) { + info->mr = 0xffu << 16; + info->mg = 0xffu << 8; + info->mb = 0xffu << 0; + info->ma = 0xffu << 24; + info->all_a = 0; // if all_a is 0 at end, then we loaded alpha channel but it was all 0 + } else { + info->mr = 31u << 10; + info->mg = 31u << 5; + info->mb = 31u << 0; + } + } else if (compress == 3) { + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + // not documented, but generated by photoshop and handled by mspaint + if (info->mr == info->mg && info->mg == info->mb) { + // ?!?!? + return stbi__errpuc("bad BMP", "bad BMP"); + } + } else + return stbi__errpuc("bad BMP", "bad BMP"); + } + } else { + int i; + if (hsz != 108 && hsz != 124) + return stbi__errpuc("bad BMP", "bad BMP"); + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + info->ma = stbi__get32le(s); + stbi__get32le(s); // discard color space + for (i=0; i < 12; ++i) + stbi__get32le(s); // discard color space parameters + if (hsz == 124) { + stbi__get32le(s); // discard rendering intent + stbi__get32le(s); // discard offset of profile data + stbi__get32le(s); // discard size of profile data + stbi__get32le(s); // discard reserved + } + } + } + return (void *) 1; +} + + +static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + stbi_uc *out; + unsigned int mr=0,mg=0,mb=0,ma=0, all_a; + stbi_uc pal[256][4]; + int psize=0,i,j,width; + int flip_vertically, pad, target; + stbi__bmp_data info; + STBI_NOTUSED(ri); + + info.all_a = 255; + if (stbi__bmp_parse_header(s, &info) == NULL) + return NULL; // error code already set + + flip_vertically = ((int) s->img_y) > 0; + s->img_y = abs((int) s->img_y); + + mr = info.mr; + mg = info.mg; + mb = info.mb; + ma = info.ma; + all_a = info.all_a; + + if (info.hsz == 12) { + if (info.bpp < 24) + psize = (info.offset - 14 - 24) / 3; + } else { + if (info.bpp < 16) + psize = (info.offset - 14 - info.hsz) >> 2; + } + + if (info.bpp == 24 && ma == 0xff000000) + s->img_n = 3; + else + s->img_n = ma ? 4 : 3; + if (req_comp && req_comp >= 3) // we can directly decode 3 or 4 + target = req_comp; + else + target = s->img_n; // if they want monochrome, we'll post-convert + + // sanity-check size + if (!stbi__mad3sizes_valid(target, s->img_x, s->img_y, 0)) + return stbi__errpuc("too large", "Corrupt BMP"); + + out = (stbi_uc *) stbi__malloc_mad3(target, s->img_x, s->img_y, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + if (info.bpp < 16) { + int z=0; + if (psize == 0 || psize > 256) { STBI_FREE(out); return stbi__errpuc("invalid", "Corrupt BMP"); } + for (i=0; i < psize; ++i) { + pal[i][2] = stbi__get8(s); + pal[i][1] = stbi__get8(s); + pal[i][0] = stbi__get8(s); + if (info.hsz != 12) stbi__get8(s); + pal[i][3] = 255; + } + stbi__skip(s, info.offset - 14 - info.hsz - psize * (info.hsz == 12 ? 3 : 4)); + if (info.bpp == 1) width = (s->img_x + 7) >> 3; + else if (info.bpp == 4) width = (s->img_x + 1) >> 1; + else if (info.bpp == 8) width = s->img_x; + else { STBI_FREE(out); return stbi__errpuc("bad bpp", "Corrupt BMP"); } + pad = (-width)&3; + if (info.bpp == 1) { + for (j=0; j < (int) s->img_y; ++j) { + int bit_offset = 7, v = stbi__get8(s); + for (i=0; i < (int) s->img_x; ++i) { + int color = (v>>bit_offset)&0x1; + out[z++] = pal[color][0]; + out[z++] = pal[color][1]; + out[z++] = pal[color][2]; + if (target == 4) out[z++] = 255; + if (i+1 == (int) s->img_x) break; + if((--bit_offset) < 0) { + bit_offset = 7; + v = stbi__get8(s); + } + } + stbi__skip(s, pad); + } + } else { + for (j=0; j < (int) s->img_y; ++j) { + for (i=0; i < (int) s->img_x; i += 2) { + int v=stbi__get8(s),v2=0; + if (info.bpp == 4) { + v2 = v & 15; + v >>= 4; + } + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + if (i+1 == (int) s->img_x) break; + v = (info.bpp == 8) ? stbi__get8(s) : v2; + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + } + stbi__skip(s, pad); + } + } + } else { + int rshift=0,gshift=0,bshift=0,ashift=0,rcount=0,gcount=0,bcount=0,acount=0; + int z = 0; + int easy=0; + stbi__skip(s, info.offset - 14 - info.hsz); + if (info.bpp == 24) width = 3 * s->img_x; + else if (info.bpp == 16) width = 2*s->img_x; + else /* bpp = 32 and pad = 0 */ width=0; + pad = (-width) & 3; + if (info.bpp == 24) { + easy = 1; + } else if (info.bpp == 32) { + if (mb == 0xff && mg == 0xff00 && mr == 0x00ff0000 && ma == 0xff000000) + easy = 2; + } + if (!easy) { + if (!mr || !mg || !mb) { STBI_FREE(out); return stbi__errpuc("bad masks", "Corrupt BMP"); } + // right shift amt to put high bit in position #7 + rshift = stbi__high_bit(mr)-7; rcount = stbi__bitcount(mr); + gshift = stbi__high_bit(mg)-7; gcount = stbi__bitcount(mg); + bshift = stbi__high_bit(mb)-7; bcount = stbi__bitcount(mb); + ashift = stbi__high_bit(ma)-7; acount = stbi__bitcount(ma); + } + for (j=0; j < (int) s->img_y; ++j) { + if (easy) { + for (i=0; i < (int) s->img_x; ++i) { + unsigned char a; + out[z+2] = stbi__get8(s); + out[z+1] = stbi__get8(s); + out[z+0] = stbi__get8(s); + z += 3; + a = (easy == 2 ? stbi__get8(s) : 255); + all_a |= a; + if (target == 4) out[z++] = a; + } + } else { + int bpp = info.bpp; + for (i=0; i < (int) s->img_x; ++i) { + stbi__uint32 v = (bpp == 16 ? (stbi__uint32) stbi__get16le(s) : stbi__get32le(s)); + unsigned int a; + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mr, rshift, rcount)); + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mg, gshift, gcount)); + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mb, bshift, bcount)); + a = (ma ? stbi__shiftsigned(v & ma, ashift, acount) : 255); + all_a |= a; + if (target == 4) out[z++] = STBI__BYTECAST(a); + } + } + stbi__skip(s, pad); + } + } + + // if alpha channel is all 0s, replace with all 255s + if (target == 4 && all_a == 0) + for (i=4*s->img_x*s->img_y-1; i >= 0; i -= 4) + out[i] = 255; + + if (flip_vertically) { + stbi_uc t; + for (j=0; j < (int) s->img_y>>1; ++j) { + stbi_uc *p1 = out + j *s->img_x*target; + stbi_uc *p2 = out + (s->img_y-1-j)*s->img_x*target; + for (i=0; i < (int) s->img_x*target; ++i) { + t = p1[i]; p1[i] = p2[i]; p2[i] = t; + } + } + } + + if (req_comp && req_comp != target) { + out = stbi__convert_format(out, target, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; + return out; +} +#endif + +// Targa Truevision - TGA +// by Jonathan Dummer +#ifndef STBI_NO_TGA +// returns STBI_rgb or whatever, 0 on error +static int stbi__tga_get_comp(int bits_per_pixel, int is_grey, int* is_rgb16) +{ + // only RGB or RGBA (incl. 16bit) or grey allowed + if (is_rgb16) *is_rgb16 = 0; + switch(bits_per_pixel) { + case 8: return STBI_grey; + case 16: if(is_grey) return STBI_grey_alpha; + // fallthrough + case 15: if(is_rgb16) *is_rgb16 = 1; + return STBI_rgb; + case 24: // fallthrough + case 32: return bits_per_pixel/8; + default: return 0; + } +} + +static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp) +{ + int tga_w, tga_h, tga_comp, tga_image_type, tga_bits_per_pixel, tga_colormap_bpp; + int sz, tga_colormap_type; + stbi__get8(s); // discard Offset + tga_colormap_type = stbi__get8(s); // colormap type + if( tga_colormap_type > 1 ) { + stbi__rewind(s); + return 0; // only RGB or indexed allowed + } + tga_image_type = stbi__get8(s); // image type + if ( tga_colormap_type == 1 ) { // colormapped (paletted) image + if (tga_image_type != 1 && tga_image_type != 9) { + stbi__rewind(s); + return 0; + } + stbi__skip(s,4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) { + stbi__rewind(s); + return 0; + } + stbi__skip(s,4); // skip image x and y origin + tga_colormap_bpp = sz; + } else { // "normal" image w/o colormap - only RGB or grey allowed, +/- RLE + if ( (tga_image_type != 2) && (tga_image_type != 3) && (tga_image_type != 10) && (tga_image_type != 11) ) { + stbi__rewind(s); + return 0; // only RGB or grey allowed, +/- RLE + } + stbi__skip(s,9); // skip colormap specification and image x/y origin + tga_colormap_bpp = 0; + } + tga_w = stbi__get16le(s); + if( tga_w < 1 ) { + stbi__rewind(s); + return 0; // test width + } + tga_h = stbi__get16le(s); + if( tga_h < 1 ) { + stbi__rewind(s); + return 0; // test height + } + tga_bits_per_pixel = stbi__get8(s); // bits per pixel + stbi__get8(s); // ignore alpha bits + if (tga_colormap_bpp != 0) { + if((tga_bits_per_pixel != 8) && (tga_bits_per_pixel != 16)) { + // when using a colormap, tga_bits_per_pixel is the size of the indexes + // I don't think anything but 8 or 16bit indexes makes sense + stbi__rewind(s); + return 0; + } + tga_comp = stbi__tga_get_comp(tga_colormap_bpp, 0, NULL); + } else { + tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3) || (tga_image_type == 11), NULL); + } + if(!tga_comp) { + stbi__rewind(s); + return 0; + } + if (x) *x = tga_w; + if (y) *y = tga_h; + if (comp) *comp = tga_comp; + return 1; // seems to have passed everything +} + +static int stbi__tga_test(stbi__context *s) +{ + int res = 0; + int sz, tga_color_type; + stbi__get8(s); // discard Offset + tga_color_type = stbi__get8(s); // color type + if ( tga_color_type > 1 ) goto errorEnd; // only RGB or indexed allowed + sz = stbi__get8(s); // image type + if ( tga_color_type == 1 ) { // colormapped (paletted) image + if (sz != 1 && sz != 9) goto errorEnd; // colortype 1 demands image type 1 or 9 + stbi__skip(s,4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) goto errorEnd; + stbi__skip(s,4); // skip image x and y origin + } else { // "normal" image w/o colormap + if ( (sz != 2) && (sz != 3) && (sz != 10) && (sz != 11) ) goto errorEnd; // only RGB or grey allowed, +/- RLE + stbi__skip(s,9); // skip colormap specification and image x/y origin + } + if ( stbi__get16le(s) < 1 ) goto errorEnd; // test width + if ( stbi__get16le(s) < 1 ) goto errorEnd; // test height + sz = stbi__get8(s); // bits per pixel + if ( (tga_color_type == 1) && (sz != 8) && (sz != 16) ) goto errorEnd; // for colormapped images, bpp is size of an index + if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) goto errorEnd; + + res = 1; // if we got this far, everything's good and we can return 1 instead of 0 + +errorEnd: + stbi__rewind(s); + return res; +} + +// read 16bit value and convert to 24bit RGB +static void stbi__tga_read_rgb16(stbi__context *s, stbi_uc* out) +{ + stbi__uint16 px = (stbi__uint16)stbi__get16le(s); + stbi__uint16 fiveBitMask = 31; + // we have 3 channels with 5bits each + int r = (px >> 10) & fiveBitMask; + int g = (px >> 5) & fiveBitMask; + int b = px & fiveBitMask; + // Note that this saves the data in RGB(A) order, so it doesn't need to be swapped later + out[0] = (stbi_uc)((r * 255)/31); + out[1] = (stbi_uc)((g * 255)/31); + out[2] = (stbi_uc)((b * 255)/31); + + // some people claim that the most significant bit might be used for alpha + // (possibly if an alpha-bit is set in the "image descriptor byte") + // but that only made 16bit test images completely translucent.. + // so let's treat all 15 and 16bit TGAs as RGB with no alpha. +} + +static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + // read in the TGA header stuff + int tga_offset = stbi__get8(s); + int tga_indexed = stbi__get8(s); + int tga_image_type = stbi__get8(s); + int tga_is_RLE = 0; + int tga_palette_start = stbi__get16le(s); + int tga_palette_len = stbi__get16le(s); + int tga_palette_bits = stbi__get8(s); + int tga_x_origin = stbi__get16le(s); + int tga_y_origin = stbi__get16le(s); + int tga_width = stbi__get16le(s); + int tga_height = stbi__get16le(s); + int tga_bits_per_pixel = stbi__get8(s); + int tga_comp, tga_rgb16=0; + int tga_inverted = stbi__get8(s); + // int tga_alpha_bits = tga_inverted & 15; // the 4 lowest bits - unused (useless?) + // image data + unsigned char *tga_data; + unsigned char *tga_palette = NULL; + int i, j; + unsigned char raw_data[4] = {0}; + int RLE_count = 0; + int RLE_repeating = 0; + int read_next_pixel = 1; + STBI_NOTUSED(ri); + STBI_NOTUSED(tga_x_origin); // @TODO + STBI_NOTUSED(tga_y_origin); // @TODO + + // do a tiny bit of precessing + if ( tga_image_type >= 8 ) + { + tga_image_type -= 8; + tga_is_RLE = 1; + } + tga_inverted = 1 - ((tga_inverted >> 5) & 1); + + // If I'm paletted, then I'll use the number of bits from the palette + if ( tga_indexed ) tga_comp = stbi__tga_get_comp(tga_palette_bits, 0, &tga_rgb16); + else tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3), &tga_rgb16); + + if(!tga_comp) // shouldn't really happen, stbi__tga_test() should have ensured basic consistency + return stbi__errpuc("bad format", "Can't find out TGA pixelformat"); + + // tga info + *x = tga_width; + *y = tga_height; + if (comp) *comp = tga_comp; + + if (!stbi__mad3sizes_valid(tga_width, tga_height, tga_comp, 0)) + return stbi__errpuc("too large", "Corrupt TGA"); + + tga_data = (unsigned char*)stbi__malloc_mad3(tga_width, tga_height, tga_comp, 0); + if (!tga_data) return stbi__errpuc("outofmem", "Out of memory"); + + // skip to the data's starting position (offset usually = 0) + stbi__skip(s, tga_offset ); + + if ( !tga_indexed && !tga_is_RLE && !tga_rgb16 ) { + for (i=0; i < tga_height; ++i) { + int row = tga_inverted ? tga_height -i - 1 : i; + stbi_uc *tga_row = tga_data + row*tga_width*tga_comp; + stbi__getn(s, tga_row, tga_width * tga_comp); + } + } else { + // do I need to load a palette? + if ( tga_indexed) + { + // any data to skip? (offset usually = 0) + stbi__skip(s, tga_palette_start ); + // load the palette + tga_palette = (unsigned char*)stbi__malloc_mad2(tga_palette_len, tga_comp, 0); + if (!tga_palette) { + STBI_FREE(tga_data); + return stbi__errpuc("outofmem", "Out of memory"); + } + if (tga_rgb16) { + stbi_uc *pal_entry = tga_palette; + STBI_ASSERT(tga_comp == STBI_rgb); + for (i=0; i < tga_palette_len; ++i) { + stbi__tga_read_rgb16(s, pal_entry); + pal_entry += tga_comp; + } + } else if (!stbi__getn(s, tga_palette, tga_palette_len * tga_comp)) { + STBI_FREE(tga_data); + STBI_FREE(tga_palette); + return stbi__errpuc("bad palette", "Corrupt TGA"); + } + } + // load the data + for (i=0; i < tga_width * tga_height; ++i) + { + // if I'm in RLE mode, do I need to get a RLE stbi__pngchunk? + if ( tga_is_RLE ) + { + if ( RLE_count == 0 ) + { + // yep, get the next byte as a RLE command + int RLE_cmd = stbi__get8(s); + RLE_count = 1 + (RLE_cmd & 127); + RLE_repeating = RLE_cmd >> 7; + read_next_pixel = 1; + } else if ( !RLE_repeating ) + { + read_next_pixel = 1; + } + } else + { + read_next_pixel = 1; + } + // OK, if I need to read a pixel, do it now + if ( read_next_pixel ) + { + // load however much data we did have + if ( tga_indexed ) + { + // read in index, then perform the lookup + int pal_idx = (tga_bits_per_pixel == 8) ? stbi__get8(s) : stbi__get16le(s); + if ( pal_idx >= tga_palette_len ) { + // invalid index + pal_idx = 0; + } + pal_idx *= tga_comp; + for (j = 0; j < tga_comp; ++j) { + raw_data[j] = tga_palette[pal_idx+j]; + } + } else if(tga_rgb16) { + STBI_ASSERT(tga_comp == STBI_rgb); + stbi__tga_read_rgb16(s, raw_data); + } else { + // read in the data raw + for (j = 0; j < tga_comp; ++j) { + raw_data[j] = stbi__get8(s); + } + } + // clear the reading flag for the next pixel + read_next_pixel = 0; + } // end of reading a pixel + + // copy data + for (j = 0; j < tga_comp; ++j) + tga_data[i*tga_comp+j] = raw_data[j]; + + // in case we're in RLE mode, keep counting down + --RLE_count; + } + // do I need to invert the image? + if ( tga_inverted ) + { + for (j = 0; j*2 < tga_height; ++j) + { + int index1 = j * tga_width * tga_comp; + int index2 = (tga_height - 1 - j) * tga_width * tga_comp; + for (i = tga_width * tga_comp; i > 0; --i) + { + unsigned char temp = tga_data[index1]; + tga_data[index1] = tga_data[index2]; + tga_data[index2] = temp; + ++index1; + ++index2; + } + } + } + // clear my palette, if I had one + if ( tga_palette != NULL ) + { + STBI_FREE( tga_palette ); + } + } + + // swap RGB - if the source data was RGB16, it already is in the right order + if (tga_comp >= 3 && !tga_rgb16) + { + unsigned char* tga_pixel = tga_data; + for (i=0; i < tga_width * tga_height; ++i) + { + unsigned char temp = tga_pixel[0]; + tga_pixel[0] = tga_pixel[2]; + tga_pixel[2] = temp; + tga_pixel += tga_comp; + } + } + + // convert to target component count + if (req_comp && req_comp != tga_comp) + tga_data = stbi__convert_format(tga_data, tga_comp, req_comp, tga_width, tga_height); + + // the things I do to get rid of an error message, and yet keep + // Microsoft's C compilers happy... [8^( + tga_palette_start = tga_palette_len = tga_palette_bits = + tga_x_origin = tga_y_origin = 0; + STBI_NOTUSED(tga_palette_start); + // OK, done + return tga_data; +} +#endif + +// ************************************************************************************************* +// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicolas Schulz, tweaked by STB + +#ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context *s) +{ + int r = (stbi__get32be(s) == 0x38425053); + stbi__rewind(s); + return r; +} + +static int stbi__psd_decode_rle(stbi__context *s, stbi_uc *p, int pixelCount) +{ + int count, nleft, len; + + count = 0; + while ((nleft = pixelCount - count) > 0) { + len = stbi__get8(s); + if (len == 128) { + // No-op. + } else if (len < 128) { + // Copy next len+1 bytes literally. + len++; + if (len > nleft) return 0; // corrupt data + count += len; + while (len) { + *p = stbi__get8(s); + p += 4; + len--; + } + } else if (len > 128) { + stbi_uc val; + // Next -len+1 bytes in the dest are replicated from next source byte. + // (Interpret len as a negative 8-bit int.) + len = 257 - len; + if (len > nleft) return 0; // corrupt data + val = stbi__get8(s); + count += len; + while (len) { + *p = val; + p += 4; + len--; + } + } + } + + return 1; +} + +static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc) +{ + int pixelCount; + int channelCount, compression; + int channel, i; + int bitdepth; + int w,h; + stbi_uc *out; + STBI_NOTUSED(ri); + + // Check identifier + if (stbi__get32be(s) != 0x38425053) // "8BPS" + return stbi__errpuc("not PSD", "Corrupt PSD image"); + + // Check file type version. + if (stbi__get16be(s) != 1) + return stbi__errpuc("wrong version", "Unsupported version of PSD image"); + + // Skip 6 reserved bytes. + stbi__skip(s, 6 ); + + // Read the number of channels (R, G, B, A, etc). + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) + return stbi__errpuc("wrong channel count", "Unsupported number of channels in PSD image"); + + // Read the rows and columns of the image. + h = stbi__get32be(s); + w = stbi__get32be(s); + + // Make sure the depth is 8 bits. + bitdepth = stbi__get16be(s); + if (bitdepth != 8 && bitdepth != 16) + return stbi__errpuc("unsupported bit depth", "PSD bit depth is not 8 or 16 bit"); + + // Make sure the color mode is RGB. + // Valid options are: + // 0: Bitmap + // 1: Grayscale + // 2: Indexed color + // 3: RGB color + // 4: CMYK color + // 7: Multichannel + // 8: Duotone + // 9: Lab color + if (stbi__get16be(s) != 3) + return stbi__errpuc("wrong color format", "PSD is not in RGB color format"); + + // Skip the Mode Data. (It's the palette for indexed color; other info for other modes.) + stbi__skip(s,stbi__get32be(s) ); + + // Skip the image resources. (resolution, pen tool paths, etc) + stbi__skip(s, stbi__get32be(s) ); + + // Skip the reserved data. + stbi__skip(s, stbi__get32be(s) ); + + // Find out if the data is compressed. + // Known values: + // 0: no compression + // 1: RLE compressed + compression = stbi__get16be(s); + if (compression > 1) + return stbi__errpuc("bad compression", "PSD has an unknown compression format"); + + // Check size + if (!stbi__mad3sizes_valid(4, w, h, 0)) + return stbi__errpuc("too large", "Corrupt PSD"); + + // Create the destination image. + + if (!compression && bitdepth == 16 && bpc == 16) { + out = (stbi_uc *) stbi__malloc_mad3(8, w, h, 0); + ri->bits_per_channel = 16; + } else + out = (stbi_uc *) stbi__malloc(4 * w*h); + + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + pixelCount = w*h; + + // Initialize the data to zero. + //memset( out, 0, pixelCount * 4 ); + + // Finally, the image data. + if (compression) { + // RLE as used by .PSD and .TIFF + // Loop until you get the number of unpacked bytes you are expecting: + // Read the next source byte into n. + // If n is between 0 and 127 inclusive, copy the next n+1 bytes literally. + // Else if n is between -127 and -1 inclusive, copy the next byte -n+1 times. + // Else if n is 128, noop. + // Endloop + + // The RLE-compressed data is preceded by a 2-byte data count for each row in the data, + // which we're going to just skip. + stbi__skip(s, h * channelCount * 2 ); + + // Read the RLE data by channel. + for (channel = 0; channel < 4; channel++) { + stbi_uc *p; + + p = out+channel; + if (channel >= channelCount) { + // Fill this channel with default data. + for (i = 0; i < pixelCount; i++, p += 4) + *p = (channel == 3 ? 255 : 0); + } else { + // Read the RLE data. + if (!stbi__psd_decode_rle(s, p, pixelCount)) { + STBI_FREE(out); + return stbi__errpuc("corrupt", "bad RLE data"); + } + } + } + + } else { + // We're at the raw image data. It's each channel in order (Red, Green, Blue, Alpha, ...) + // where each channel consists of an 8-bit (or 16-bit) value for each pixel in the image. + + // Read the data by channel. + for (channel = 0; channel < 4; channel++) { + if (channel >= channelCount) { + // Fill this channel with default data. + if (bitdepth == 16 && bpc == 16) { + stbi__uint16 *q = ((stbi__uint16 *) out) + channel; + stbi__uint16 val = channel == 3 ? 65535 : 0; + for (i = 0; i < pixelCount; i++, q += 4) + *q = val; + } else { + stbi_uc *p = out+channel; + stbi_uc val = channel == 3 ? 255 : 0; + for (i = 0; i < pixelCount; i++, p += 4) + *p = val; + } + } else { + if (ri->bits_per_channel == 16) { // output bpc + stbi__uint16 *q = ((stbi__uint16 *) out) + channel; + for (i = 0; i < pixelCount; i++, q += 4) + *q = (stbi__uint16) stbi__get16be(s); + } else { + stbi_uc *p = out+channel; + if (bitdepth == 16) { // input bpc + for (i = 0; i < pixelCount; i++, p += 4) + *p = (stbi_uc) (stbi__get16be(s) >> 8); + } else { + for (i = 0; i < pixelCount; i++, p += 4) + *p = stbi__get8(s); + } + } + } + } + } + + // remove weird white matte from PSD + if (channelCount >= 4) { + if (ri->bits_per_channel == 16) { + for (i=0; i < w*h; ++i) { + stbi__uint16 *pixel = (stbi__uint16 *) out + 4*i; + if (pixel[3] != 0 && pixel[3] != 65535) { + float a = pixel[3] / 65535.0f; + float ra = 1.0f / a; + float inv_a = 65535.0f * (1 - ra); + pixel[0] = (stbi__uint16) (pixel[0]*ra + inv_a); + pixel[1] = (stbi__uint16) (pixel[1]*ra + inv_a); + pixel[2] = (stbi__uint16) (pixel[2]*ra + inv_a); + } + } + } else { + for (i=0; i < w*h; ++i) { + unsigned char *pixel = out + 4*i; + if (pixel[3] != 0 && pixel[3] != 255) { + float a = pixel[3] / 255.0f; + float ra = 1.0f / a; + float inv_a = 255.0f * (1 - ra); + pixel[0] = (unsigned char) (pixel[0]*ra + inv_a); + pixel[1] = (unsigned char) (pixel[1]*ra + inv_a); + pixel[2] = (unsigned char) (pixel[2]*ra + inv_a); + } + } + } + } + + // convert to desired output format + if (req_comp && req_comp != 4) { + if (ri->bits_per_channel == 16) + out = (stbi_uc *) stbi__convert_format16((stbi__uint16 *) out, 4, req_comp, w, h); + else + out = stbi__convert_format(out, 4, req_comp, w, h); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + + if (comp) *comp = 4; + *y = h; + *x = w; + + return out; +} +#endif + +// ************************************************************************************************* +// Softimage PIC loader +// by Tom Seddon +// +// See http://softimage.wiki.softimage.com/index.php/INFO:_PIC_file_format +// See http://ozviz.wasp.uwa.edu.au/~pbourke/dataformats/softimagepic/ + +#ifndef STBI_NO_PIC +static int stbi__pic_is4(stbi__context *s,const char *str) +{ + int i; + for (i=0; i<4; ++i) + if (stbi__get8(s) != (stbi_uc)str[i]) + return 0; + + return 1; +} + +static int stbi__pic_test_core(stbi__context *s) +{ + int i; + + if (!stbi__pic_is4(s,"\x53\x80\xF6\x34")) + return 0; + + for(i=0;i<84;++i) + stbi__get8(s); + + if (!stbi__pic_is4(s,"PICT")) + return 0; + + return 1; +} + +typedef struct +{ + stbi_uc size,type,channel; +} stbi__pic_packet; + +static stbi_uc *stbi__readval(stbi__context *s, int channel, stbi_uc *dest) +{ + int mask=0x80, i; + + for (i=0; i<4; ++i, mask>>=1) { + if (channel & mask) { + if (stbi__at_eof(s)) return stbi__errpuc("bad file","PIC file too short"); + dest[i]=stbi__get8(s); + } + } + + return dest; +} + +static void stbi__copyval(int channel,stbi_uc *dest,const stbi_uc *src) +{ + int mask=0x80,i; + + for (i=0;i<4; ++i, mask>>=1) + if (channel&mask) + dest[i]=src[i]; +} + +static stbi_uc *stbi__pic_load_core(stbi__context *s,int width,int height,int *comp, stbi_uc *result) +{ + int act_comp=0,num_packets=0,y,chained; + stbi__pic_packet packets[10]; + + // this will (should...) cater for even some bizarre stuff like having data + // for the same channel in multiple packets. + do { + stbi__pic_packet *packet; + + if (num_packets==sizeof(packets)/sizeof(packets[0])) + return stbi__errpuc("bad format","too many packets"); + + packet = &packets[num_packets++]; + + chained = stbi__get8(s); + packet->size = stbi__get8(s); + packet->type = stbi__get8(s); + packet->channel = stbi__get8(s); + + act_comp |= packet->channel; + + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (reading packets)"); + if (packet->size != 8) return stbi__errpuc("bad format","packet isn't 8bpp"); + } while (chained); + + *comp = (act_comp & 0x10 ? 4 : 3); // has alpha channel? + + for(y=0; ytype) { + default: + return stbi__errpuc("bad format","packet has bad compression type"); + + case 0: {//uncompressed + int x; + + for(x=0;xchannel,dest)) + return 0; + break; + } + + case 1://Pure RLE + { + int left=width, i; + + while (left>0) { + stbi_uc count,value[4]; + + count=stbi__get8(s); + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (pure read count)"); + + if (count > left) + count = (stbi_uc) left; + + if (!stbi__readval(s,packet->channel,value)) return 0; + + for(i=0; ichannel,dest,value); + left -= count; + } + } + break; + + case 2: {//Mixed RLE + int left=width; + while (left>0) { + int count = stbi__get8(s), i; + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (mixed read count)"); + + if (count >= 128) { // Repeated + stbi_uc value[4]; + + if (count==128) + count = stbi__get16be(s); + else + count -= 127; + if (count > left) + return stbi__errpuc("bad file","scanline overrun"); + + if (!stbi__readval(s,packet->channel,value)) + return 0; + + for(i=0;ichannel,dest,value); + } else { // Raw + ++count; + if (count>left) return stbi__errpuc("bad file","scanline overrun"); + + for(i=0;ichannel,dest)) + return 0; + } + left-=count; + } + break; + } + } + } + } + + return result; +} + +static void *stbi__pic_load(stbi__context *s,int *px,int *py,int *comp,int req_comp, stbi__result_info *ri) +{ + stbi_uc *result; + int i, x,y, internal_comp; + STBI_NOTUSED(ri); + + if (!comp) comp = &internal_comp; + + for (i=0; i<92; ++i) + stbi__get8(s); + + x = stbi__get16be(s); + y = stbi__get16be(s); + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (pic header)"); + if (!stbi__mad3sizes_valid(x, y, 4, 0)) return stbi__errpuc("too large", "PIC image too large to decode"); + + stbi__get32be(s); //skip `ratio' + stbi__get16be(s); //skip `fields' + stbi__get16be(s); //skip `pad' + + // intermediate buffer is RGBA + result = (stbi_uc *) stbi__malloc_mad3(x, y, 4, 0); + memset(result, 0xff, x*y*4); + + if (!stbi__pic_load_core(s,x,y,comp, result)) { + STBI_FREE(result); + result=0; + } + *px = x; + *py = y; + if (req_comp == 0) req_comp = *comp; + result=stbi__convert_format(result,4,req_comp,x,y); + + return result; +} + +static int stbi__pic_test(stbi__context *s) +{ + int r = stbi__pic_test_core(s); + stbi__rewind(s); + return r; +} +#endif + +// ************************************************************************************************* +// GIF loader -- public domain by Jean-Marc Lienher -- simplified/shrunk by stb + +#ifndef STBI_NO_GIF +typedef struct +{ + stbi__int16 prefix; + stbi_uc first; + stbi_uc suffix; +} stbi__gif_lzw; + +typedef struct +{ + int w,h; + stbi_uc *out; // output buffer (always 4 components) + stbi_uc *background; // The current "background" as far as a gif is concerned + stbi_uc *history; + int flags, bgindex, ratio, transparent, eflags; + stbi_uc pal[256][4]; + stbi_uc lpal[256][4]; + stbi__gif_lzw codes[8192]; + stbi_uc *color_table; + int parse, step; + int lflags; + int start_x, start_y; + int max_x, max_y; + int cur_x, cur_y; + int line_size; + int delay; +} stbi__gif; + +static int stbi__gif_test_raw(stbi__context *s) +{ + int sz; + if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || stbi__get8(s) != '8') return 0; + sz = stbi__get8(s); + if (sz != '9' && sz != '7') return 0; + if (stbi__get8(s) != 'a') return 0; + return 1; +} + +static int stbi__gif_test(stbi__context *s) +{ + int r = stbi__gif_test_raw(s); + stbi__rewind(s); + return r; +} + +static void stbi__gif_parse_colortable(stbi__context *s, stbi_uc pal[256][4], int num_entries, int transp) +{ + int i; + for (i=0; i < num_entries; ++i) { + pal[i][2] = stbi__get8(s); + pal[i][1] = stbi__get8(s); + pal[i][0] = stbi__get8(s); + pal[i][3] = transp == i ? 0 : 255; + } +} + +static int stbi__gif_header(stbi__context *s, stbi__gif *g, int *comp, int is_info) +{ + stbi_uc version; + if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || stbi__get8(s) != '8') + return stbi__err("not GIF", "Corrupt GIF"); + + version = stbi__get8(s); + if (version != '7' && version != '9') return stbi__err("not GIF", "Corrupt GIF"); + if (stbi__get8(s) != 'a') return stbi__err("not GIF", "Corrupt GIF"); + + stbi__g_failure_reason = ""; + g->w = stbi__get16le(s); + g->h = stbi__get16le(s); + g->flags = stbi__get8(s); + g->bgindex = stbi__get8(s); + g->ratio = stbi__get8(s); + g->transparent = -1; + + if (comp != 0) *comp = 4; // can't actually tell whether it's 3 or 4 until we parse the comments + + if (is_info) return 1; + + if (g->flags & 0x80) + stbi__gif_parse_colortable(s,g->pal, 2 << (g->flags & 7), -1); + + return 1; +} + +static int stbi__gif_info_raw(stbi__context *s, int *x, int *y, int *comp) +{ + stbi__gif* g = (stbi__gif*) stbi__malloc(sizeof(stbi__gif)); + if (!stbi__gif_header(s, g, comp, 1)) { + STBI_FREE(g); + stbi__rewind( s ); + return 0; + } + if (x) *x = g->w; + if (y) *y = g->h; + STBI_FREE(g); + return 1; +} + +static void stbi__out_gif_code(stbi__gif *g, stbi__uint16 code) +{ + stbi_uc *p, *c; + int idx; + + // recurse to decode the prefixes, since the linked-list is backwards, + // and working backwards through an interleaved image would be nasty + if (g->codes[code].prefix >= 0) + stbi__out_gif_code(g, g->codes[code].prefix); + + if (g->cur_y >= g->max_y) return; + + idx = g->cur_x + g->cur_y; + p = &g->out[idx]; + g->history[idx / 4] = 1; + + c = &g->color_table[g->codes[code].suffix * 4]; + if (c[3] > 128) { // don't render transparent pixels; + p[0] = c[2]; + p[1] = c[1]; + p[2] = c[0]; + p[3] = c[3]; + } + g->cur_x += 4; + + if (g->cur_x >= g->max_x) { + g->cur_x = g->start_x; + g->cur_y += g->step; + + while (g->cur_y >= g->max_y && g->parse > 0) { + g->step = (1 << g->parse) * g->line_size; + g->cur_y = g->start_y + (g->step >> 1); + --g->parse; + } + } +} + +static stbi_uc *stbi__process_gif_raster(stbi__context *s, stbi__gif *g) +{ + stbi_uc lzw_cs; + stbi__int32 len, init_code; + stbi__uint32 first; + stbi__int32 codesize, codemask, avail, oldcode, bits, valid_bits, clear; + stbi__gif_lzw *p; + + lzw_cs = stbi__get8(s); + if (lzw_cs > 12) return NULL; + clear = 1 << lzw_cs; + first = 1; + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + bits = 0; + valid_bits = 0; + for (init_code = 0; init_code < clear; init_code++) { + g->codes[init_code].prefix = -1; + g->codes[init_code].first = (stbi_uc) init_code; + g->codes[init_code].suffix = (stbi_uc) init_code; + } + + // support no starting clear code + avail = clear+2; + oldcode = -1; + + len = 0; + for(;;) { + if (valid_bits < codesize) { + if (len == 0) { + len = stbi__get8(s); // start new block + if (len == 0) + return g->out; + } + --len; + bits |= (stbi__int32) stbi__get8(s) << valid_bits; + valid_bits += 8; + } else { + stbi__int32 code = bits & codemask; + bits >>= codesize; + valid_bits -= codesize; + // @OPTIMIZE: is there some way we can accelerate the non-clear path? + if (code == clear) { // clear code + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + avail = clear + 2; + oldcode = -1; + first = 0; + } else if (code == clear + 1) { // end of stream code + stbi__skip(s, len); + while ((len = stbi__get8(s)) > 0) + stbi__skip(s,len); + return g->out; + } else if (code <= avail) { + if (first) { + return stbi__errpuc("no clear code", "Corrupt GIF"); + } + + if (oldcode >= 0) { + p = &g->codes[avail++]; + if (avail > 8192) { + return stbi__errpuc("too many codes", "Corrupt GIF"); + } + + p->prefix = (stbi__int16) oldcode; + p->first = g->codes[oldcode].first; + p->suffix = (code == avail) ? p->first : g->codes[code].first; + } else if (code == avail) + return stbi__errpuc("illegal code in raster", "Corrupt GIF"); + + stbi__out_gif_code(g, (stbi__uint16) code); + + if ((avail & codemask) == 0 && avail <= 0x0FFF) { + codesize++; + codemask = (1 << codesize) - 1; + } + + oldcode = code; + } else { + return stbi__errpuc("illegal code in raster", "Corrupt GIF"); + } + } + } +} + +// this function is designed to support animated gifs, although stb_image doesn't support it +// two back is the image from two frames ago, used for a very specific disposal format +static stbi_uc *stbi__gif_load_next(stbi__context *s, stbi__gif *g, int *comp, int req_comp, stbi_uc *two_back) +{ + int dispose; + int first_frame; + int pi; + int pcount; + STBI_NOTUSED(req_comp); + + // on first frame, any non-written pixels get the background colour (non-transparent) + first_frame = 0; + if (g->out == 0) { + if (!stbi__gif_header(s, g, comp,0)) return 0; // stbi__g_failure_reason set by stbi__gif_header + if (!stbi__mad3sizes_valid(4, g->w, g->h, 0)) + return stbi__errpuc("too large", "GIF image is too large"); + pcount = g->w * g->h; + g->out = (stbi_uc *) stbi__malloc(4 * pcount); + g->background = (stbi_uc *) stbi__malloc(4 * pcount); + g->history = (stbi_uc *) stbi__malloc(pcount); + if (!g->out || !g->background || !g->history) + return stbi__errpuc("outofmem", "Out of memory"); + + // image is treated as "transparent" at the start - ie, nothing overwrites the current background; + // background colour is only used for pixels that are not rendered first frame, after that "background" + // color refers to the color that was there the previous frame. + memset(g->out, 0x00, 4 * pcount); + memset(g->background, 0x00, 4 * pcount); // state of the background (starts transparent) + memset(g->history, 0x00, pcount); // pixels that were affected previous frame + first_frame = 1; + } else { + // second frame - how do we dispoase of the previous one? + dispose = (g->eflags & 0x1C) >> 2; + pcount = g->w * g->h; + + if ((dispose == 3) && (two_back == 0)) { + dispose = 2; // if I don't have an image to revert back to, default to the old background + } + + if (dispose == 3) { // use previous graphic + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { + memcpy( &g->out[pi * 4], &two_back[pi * 4], 4 ); + } + } + } else if (dispose == 2) { + // restore what was changed last frame to background before that frame; + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { + memcpy( &g->out[pi * 4], &g->background[pi * 4], 4 ); + } + } + } else { + // This is a non-disposal case eithe way, so just + // leave the pixels as is, and they will become the new background + // 1: do not dispose + // 0: not specified. + } + + // background is what out is after the undoing of the previou frame; + memcpy( g->background, g->out, 4 * g->w * g->h ); + } + + // clear my history; + memset( g->history, 0x00, g->w * g->h ); // pixels that were affected previous frame + + for (;;) { + int tag = stbi__get8(s); + switch (tag) { + case 0x2C: /* Image Descriptor */ + { + stbi__int32 x, y, w, h; + stbi_uc *o; + + x = stbi__get16le(s); + y = stbi__get16le(s); + w = stbi__get16le(s); + h = stbi__get16le(s); + if (((x + w) > (g->w)) || ((y + h) > (g->h))) + return stbi__errpuc("bad Image Descriptor", "Corrupt GIF"); + + g->line_size = g->w * 4; + g->start_x = x * 4; + g->start_y = y * g->line_size; + g->max_x = g->start_x + w * 4; + g->max_y = g->start_y + h * g->line_size; + g->cur_x = g->start_x; + g->cur_y = g->start_y; + + // if the width of the specified rectangle is 0, that means + // we may not see *any* pixels or the image is malformed; + // to make sure this is caught, move the current y down to + // max_y (which is what out_gif_code checks). + if (w == 0) + g->cur_y = g->max_y; + + g->lflags = stbi__get8(s); + + if (g->lflags & 0x40) { + g->step = 8 * g->line_size; // first interlaced spacing + g->parse = 3; + } else { + g->step = g->line_size; + g->parse = 0; + } + + if (g->lflags & 0x80) { + stbi__gif_parse_colortable(s,g->lpal, 2 << (g->lflags & 7), g->eflags & 0x01 ? g->transparent : -1); + g->color_table = (stbi_uc *) g->lpal; + } else if (g->flags & 0x80) { + g->color_table = (stbi_uc *) g->pal; + } else + return stbi__errpuc("missing color table", "Corrupt GIF"); + + o = stbi__process_gif_raster(s, g); + if (!o) return NULL; + + // if this was the first frame, + pcount = g->w * g->h; + if (first_frame && (g->bgindex > 0)) { + // if first frame, any pixel not drawn to gets the background color + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi] == 0) { + g->pal[g->bgindex][3] = 255; // just in case it was made transparent, undo that; It will be reset next frame if need be; + memcpy( &g->out[pi * 4], &g->pal[g->bgindex], 4 ); + } + } + } + + return o; + } + + case 0x21: // Comment Extension. + { + int len; + int ext = stbi__get8(s); + if (ext == 0xF9) { // Graphic Control Extension. + len = stbi__get8(s); + if (len == 4) { + g->eflags = stbi__get8(s); + g->delay = 10 * stbi__get16le(s); // delay - 1/100th of a second, saving as 1/1000ths. + + // unset old transparent + if (g->transparent >= 0) { + g->pal[g->transparent][3] = 255; + } + if (g->eflags & 0x01) { + g->transparent = stbi__get8(s); + if (g->transparent >= 0) { + g->pal[g->transparent][3] = 0; + } + } else { + // don't need transparent + stbi__skip(s, 1); + g->transparent = -1; + } + } else { + stbi__skip(s, len); + break; + } + } + while ((len = stbi__get8(s)) != 0) { + stbi__skip(s, len); + } + break; + } + + case 0x3B: // gif stream termination code + return (stbi_uc *) s; // using '1' causes warning on some compilers + + default: + return stbi__errpuc("unknown code", "Corrupt GIF"); + } + } +} + +static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, int *z, int *comp, int req_comp) +{ + if (stbi__gif_test(s)) { + int layers = 0; + stbi_uc *u = 0; + stbi_uc *out = 0; + stbi_uc *two_back = 0; + stbi__gif g; + int stride; + memset(&g, 0, sizeof(g)); + if (delays) { + *delays = 0; + } + + do { + u = stbi__gif_load_next(s, &g, comp, req_comp, two_back); + if (u == (stbi_uc *) s) u = 0; // end of animated gif marker + + if (u) { + *x = g.w; + *y = g.h; + ++layers; + stride = g.w * g.h * 4; + + if (out) { + out = (stbi_uc*) STBI_REALLOC( out, layers * stride ); + if (delays) { + *delays = (int*) STBI_REALLOC( *delays, sizeof(int) * layers ); + } + } else { + out = (stbi_uc*)stbi__malloc( layers * stride ); + if (delays) { + *delays = (int*) stbi__malloc( layers * sizeof(int) ); + } + } + memcpy( out + ((layers - 1) * stride), u, stride ); + if (layers >= 2) { + two_back = out - 2 * stride; + } + + if (delays) { + (*delays)[layers - 1U] = g.delay; + } + } + } while (u != 0); + + // free temp buffer; + STBI_FREE(g.out); + STBI_FREE(g.history); + STBI_FREE(g.background); + + // do the final conversion after loading everything; + if (req_comp && req_comp != 4) + out = stbi__convert_format(out, 4, req_comp, layers * g.w, g.h); + + *z = layers; + return out; + } else { + return stbi__errpuc("not GIF", "Image was not as a gif type."); + } +} + +static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + stbi_uc *u = 0; + stbi__gif g; + memset(&g, 0, sizeof(g)); + STBI_NOTUSED(ri); + + u = stbi__gif_load_next(s, &g, comp, req_comp, 0); + if (u == (stbi_uc *) s) u = 0; // end of animated gif marker + if (u) { + *x = g.w; + *y = g.h; + + // moved conversion to after successful load so that the same + // can be done for multiple frames. + if (req_comp && req_comp != 4) + u = stbi__convert_format(u, 4, req_comp, g.w, g.h); + } else if (g.out) { + // if there was an error and we allocated an image buffer, free it! + STBI_FREE(g.out); + } + + // free buffers needed for multiple frame loading; + STBI_FREE(g.history); + STBI_FREE(g.background); + + return u; +} + +static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp) +{ + return stbi__gif_info_raw(s,x,y,comp); +} +#endif + +// ************************************************************************************************* +// Radiance RGBE HDR loader +// originally by Nicolas Schulz +#ifndef STBI_NO_HDR +static int stbi__hdr_test_core(stbi__context *s, const char *signature) +{ + int i; + for (i=0; signature[i]; ++i) + if (stbi__get8(s) != signature[i]) + return 0; + stbi__rewind(s); + return 1; +} + +static int stbi__hdr_test(stbi__context* s) +{ + int r = stbi__hdr_test_core(s, "#?RADIANCE\n"); + stbi__rewind(s); + if(!r) { + r = stbi__hdr_test_core(s, "#?RGBE\n"); + stbi__rewind(s); + } + return r; +} + +#define STBI__HDR_BUFLEN 1024 +static char *stbi__hdr_gettoken(stbi__context *z, char *buffer) +{ + int len=0; + char c = '\0'; + + c = (char) stbi__get8(z); + + while (!stbi__at_eof(z) && c != '\n') { + buffer[len++] = c; + if (len == STBI__HDR_BUFLEN-1) { + // flush to end of line + while (!stbi__at_eof(z) && stbi__get8(z) != '\n') + ; + break; + } + c = (char) stbi__get8(z); + } + + buffer[len] = 0; + return buffer; +} + +static void stbi__hdr_convert(float *output, stbi_uc *input, int req_comp) +{ + if ( input[3] != 0 ) { + float f1; + // Exponent + f1 = (float) ldexp(1.0f, input[3] - (int)(128 + 8)); + if (req_comp <= 2) + output[0] = (input[0] + input[1] + input[2]) * f1 / 3; + else { + output[0] = input[0] * f1; + output[1] = input[1] * f1; + output[2] = input[2] * f1; + } + if (req_comp == 2) output[1] = 1; + if (req_comp == 4) output[3] = 1; + } else { + switch (req_comp) { + case 4: output[3] = 1; /* fallthrough */ + case 3: output[0] = output[1] = output[2] = 0; + break; + case 2: output[1] = 1; /* fallthrough */ + case 1: output[0] = 0; + break; + } + } +} + +static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + char buffer[STBI__HDR_BUFLEN]; + char *token; + int valid = 0; + int width, height; + stbi_uc *scanline; + float *hdr_data; + int len; + unsigned char count, value; + int i, j, k, c1,c2, z; + const char *headerToken; + STBI_NOTUSED(ri); + + // Check identifier + headerToken = stbi__hdr_gettoken(s,buffer); + if (strcmp(headerToken, "#?RADIANCE") != 0 && strcmp(headerToken, "#?RGBE") != 0) + return stbi__errpf("not HDR", "Corrupt HDR image"); + + // Parse header + for(;;) { + token = stbi__hdr_gettoken(s,buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) return stbi__errpf("unsupported format", "Unsupported HDR format"); + + // Parse width and height + // can't use sscanf() if we're not using stdio! + token = stbi__hdr_gettoken(s,buffer); + if (strncmp(token, "-Y ", 3)) return stbi__errpf("unsupported data layout", "Unsupported HDR format"); + token += 3; + height = (int) strtol(token, &token, 10); + while (*token == ' ') ++token; + if (strncmp(token, "+X ", 3)) return stbi__errpf("unsupported data layout", "Unsupported HDR format"); + token += 3; + width = (int) strtol(token, NULL, 10); + + *x = width; + *y = height; + + if (comp) *comp = 3; + if (req_comp == 0) req_comp = 3; + + if (!stbi__mad4sizes_valid(width, height, req_comp, sizeof(float), 0)) + return stbi__errpf("too large", "HDR image is too large"); + + // Read data + hdr_data = (float *) stbi__malloc_mad4(width, height, req_comp, sizeof(float), 0); + if (!hdr_data) + return stbi__errpf("outofmem", "Out of memory"); + + // Load image data + // image data is stored as some number of sca + if ( width < 8 || width >= 32768) { + // Read flat data + for (j=0; j < height; ++j) { + for (i=0; i < width; ++i) { + stbi_uc rgbe[4]; + main_decode_loop: + stbi__getn(s, rgbe, 4); + stbi__hdr_convert(hdr_data + j * width * req_comp + i * req_comp, rgbe, req_comp); + } + } + } else { + // Read RLE-encoded data + scanline = NULL; + + for (j = 0; j < height; ++j) { + c1 = stbi__get8(s); + c2 = stbi__get8(s); + len = stbi__get8(s); + if (c1 != 2 || c2 != 2 || (len & 0x80)) { + // not run-length encoded, so we have to actually use THIS data as a decoded + // pixel (note this can't be a valid pixel--one of RGB must be >= 128) + stbi_uc rgbe[4]; + rgbe[0] = (stbi_uc) c1; + rgbe[1] = (stbi_uc) c2; + rgbe[2] = (stbi_uc) len; + rgbe[3] = (stbi_uc) stbi__get8(s); + stbi__hdr_convert(hdr_data, rgbe, req_comp); + i = 1; + j = 0; + STBI_FREE(scanline); + goto main_decode_loop; // yes, this makes no sense + } + len <<= 8; + len |= stbi__get8(s); + if (len != width) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("invalid decoded scanline length", "corrupt HDR"); } + if (scanline == NULL) { + scanline = (stbi_uc *) stbi__malloc_mad2(width, 4, 0); + if (!scanline) { + STBI_FREE(hdr_data); + return stbi__errpf("outofmem", "Out of memory"); + } + } + + for (k = 0; k < 4; ++k) { + int nleft; + i = 0; + while ((nleft = width - i) > 0) { + count = stbi__get8(s); + if (count > 128) { + // Run + value = stbi__get8(s); + count -= 128; + if (count > nleft) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = value; + } else { + // Dump + if (count > nleft) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = stbi__get8(s); + } + } + } + for (i=0; i < width; ++i) + stbi__hdr_convert(hdr_data+(j*width + i)*req_comp, scanline + i*4, req_comp); + } + if (scanline) + STBI_FREE(scanline); + } + + return hdr_data; +} + +static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp) +{ + char buffer[STBI__HDR_BUFLEN]; + char *token; + int valid = 0; + int dummy; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + if (stbi__hdr_test(s) == 0) { + stbi__rewind( s ); + return 0; + } + + for(;;) { + token = stbi__hdr_gettoken(s,buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) { + stbi__rewind( s ); + return 0; + } + token = stbi__hdr_gettoken(s,buffer); + if (strncmp(token, "-Y ", 3)) { + stbi__rewind( s ); + return 0; + } + token += 3; + *y = (int) strtol(token, &token, 10); + while (*token == ' ') ++token; + if (strncmp(token, "+X ", 3)) { + stbi__rewind( s ); + return 0; + } + token += 3; + *x = (int) strtol(token, NULL, 10); + *comp = 3; + return 1; +} +#endif // STBI_NO_HDR + +#ifndef STBI_NO_BMP +static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp) +{ + void *p; + stbi__bmp_data info; + + info.all_a = 255; + p = stbi__bmp_parse_header(s, &info); + stbi__rewind( s ); + if (p == NULL) + return 0; + if (x) *x = s->img_x; + if (y) *y = s->img_y; + if (comp) { + if (info.bpp == 24 && info.ma == 0xff000000) + *comp = 3; + else + *comp = info.ma ? 4 : 3; + } + return 1; +} +#endif + +#ifndef STBI_NO_PSD +static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp) +{ + int channelCount, dummy, depth; + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + if (stbi__get32be(s) != 0x38425053) { + stbi__rewind( s ); + return 0; + } + if (stbi__get16be(s) != 1) { + stbi__rewind( s ); + return 0; + } + stbi__skip(s, 6); + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) { + stbi__rewind( s ); + return 0; + } + *y = stbi__get32be(s); + *x = stbi__get32be(s); + depth = stbi__get16be(s); + if (depth != 8 && depth != 16) { + stbi__rewind( s ); + return 0; + } + if (stbi__get16be(s) != 3) { + stbi__rewind( s ); + return 0; + } + *comp = 4; + return 1; +} + +static int stbi__psd_is16(stbi__context *s) +{ + int channelCount, depth; + if (stbi__get32be(s) != 0x38425053) { + stbi__rewind( s ); + return 0; + } + if (stbi__get16be(s) != 1) { + stbi__rewind( s ); + return 0; + } + stbi__skip(s, 6); + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) { + stbi__rewind( s ); + return 0; + } + (void) stbi__get32be(s); + (void) stbi__get32be(s); + depth = stbi__get16be(s); + if (depth != 16) { + stbi__rewind( s ); + return 0; + } + return 1; +} +#endif + +#ifndef STBI_NO_PIC +static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp) +{ + int act_comp=0,num_packets=0,chained,dummy; + stbi__pic_packet packets[10]; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + if (!stbi__pic_is4(s,"\x53\x80\xF6\x34")) { + stbi__rewind(s); + return 0; + } + + stbi__skip(s, 88); + + *x = stbi__get16be(s); + *y = stbi__get16be(s); + if (stbi__at_eof(s)) { + stbi__rewind( s); + return 0; + } + if ( (*x) != 0 && (1 << 28) / (*x) < (*y)) { + stbi__rewind( s ); + return 0; + } + + stbi__skip(s, 8); + + do { + stbi__pic_packet *packet; + + if (num_packets==sizeof(packets)/sizeof(packets[0])) + return 0; + + packet = &packets[num_packets++]; + chained = stbi__get8(s); + packet->size = stbi__get8(s); + packet->type = stbi__get8(s); + packet->channel = stbi__get8(s); + act_comp |= packet->channel; + + if (stbi__at_eof(s)) { + stbi__rewind( s ); + return 0; + } + if (packet->size != 8) { + stbi__rewind( s ); + return 0; + } + } while (chained); + + *comp = (act_comp & 0x10 ? 4 : 3); + + return 1; +} +#endif + +// ************************************************************************************************* +// Portable Gray Map and Portable Pixel Map loader +// by Ken Miller +// +// PGM: http://netpbm.sourceforge.net/doc/pgm.html +// PPM: http://netpbm.sourceforge.net/doc/ppm.html +// +// Known limitations: +// Does not support comments in the header section +// Does not support ASCII image data (formats P2 and P3) +// Does not support 16-bit-per-channel + +#ifndef STBI_NO_PNM + +static int stbi__pnm_test(stbi__context *s) +{ + char p, t; + p = (char) stbi__get8(s); + t = (char) stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind( s ); + return 0; + } + return 1; +} + +static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + stbi_uc *out; + STBI_NOTUSED(ri); + + if (!stbi__pnm_info(s, (int *)&s->img_x, (int *)&s->img_y, (int *)&s->img_n)) + return 0; + + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; + + if (!stbi__mad3sizes_valid(s->img_n, s->img_x, s->img_y, 0)) + return stbi__errpuc("too large", "PNM too large"); + + out = (stbi_uc *) stbi__malloc_mad3(s->img_n, s->img_x, s->img_y, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + stbi__getn(s, out, s->img_n * s->img_x * s->img_y); + + if (req_comp && req_comp != s->img_n) { + out = stbi__convert_format(out, s->img_n, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + return out; +} + +static int stbi__pnm_isspace(char c) +{ + return c == ' ' || c == '\t' || c == '\n' || c == '\v' || c == '\f' || c == '\r'; +} + +static void stbi__pnm_skip_whitespace(stbi__context *s, char *c) +{ + for (;;) { + while (!stbi__at_eof(s) && stbi__pnm_isspace(*c)) + *c = (char) stbi__get8(s); + + if (stbi__at_eof(s) || *c != '#') + break; + + while (!stbi__at_eof(s) && *c != '\n' && *c != '\r' ) + *c = (char) stbi__get8(s); + } +} + +static int stbi__pnm_isdigit(char c) +{ + return c >= '0' && c <= '9'; +} + +static int stbi__pnm_getinteger(stbi__context *s, char *c) +{ + int value = 0; + + while (!stbi__at_eof(s) && stbi__pnm_isdigit(*c)) { + value = value*10 + (*c - '0'); + *c = (char) stbi__get8(s); + } + + return value; +} + +static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp) +{ + int maxv, dummy; + char c, p, t; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + stbi__rewind(s); + + // Get identifier + p = (char) stbi__get8(s); + t = (char) stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind(s); + return 0; + } + + *comp = (t == '6') ? 3 : 1; // '5' is 1-component .pgm; '6' is 3-component .ppm + + c = (char) stbi__get8(s); + stbi__pnm_skip_whitespace(s, &c); + + *x = stbi__pnm_getinteger(s, &c); // read width + stbi__pnm_skip_whitespace(s, &c); + + *y = stbi__pnm_getinteger(s, &c); // read height + stbi__pnm_skip_whitespace(s, &c); + + maxv = stbi__pnm_getinteger(s, &c); // read max value + + if (maxv > 255) + return stbi__err("max value > 255", "PPM image not 8-bit"); + else + return 1; +} +#endif + +static int stbi__info_main(stbi__context *s, int *x, int *y, int *comp) +{ + #ifndef STBI_NO_JPEG + if (stbi__jpeg_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_PNG + if (stbi__png_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_GIF + if (stbi__gif_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_BMP + if (stbi__bmp_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_PSD + if (stbi__psd_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_PIC + if (stbi__pic_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_PNM + if (stbi__pnm_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_HDR + if (stbi__hdr_info(s, x, y, comp)) return 1; + #endif + + // test tga last because it's a crappy test! + #ifndef STBI_NO_TGA + if (stbi__tga_info(s, x, y, comp)) + return 1; + #endif + return stbi__err("unknown image type", "Image not of any known type, or corrupt"); +} + +static int stbi__is_16_main(stbi__context *s) +{ + #ifndef STBI_NO_PNG + if (stbi__png_is16(s)) return 1; + #endif + + #ifndef STBI_NO_PSD + if (stbi__psd_is16(s)) return 1; + #endif + + return 0; +} + +#ifndef STBI_NO_STDIO +STBIDEF int stbi_info(char const *filename, int *x, int *y, int *comp) +{ + FILE *f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_info_from_file(f, x, y, comp); + fclose(f); + return result; +} + +STBIDEF int stbi_info_from_file(FILE *f, int *x, int *y, int *comp) +{ + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__info_main(&s,x,y,comp); + fseek(f,pos,SEEK_SET); + return r; +} + +STBIDEF int stbi_is_16_bit(char const *filename) +{ + FILE *f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_is_16_bit_from_file(f); + fclose(f); + return result; +} + +STBIDEF int stbi_is_16_bit_from_file(FILE *f) +{ + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__is_16_main(&s); + fseek(f,pos,SEEK_SET); + return r; +} +#endif // !STBI_NO_STDIO + +STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__info_main(&s,x,y,comp); +} + +STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *c, void *user, int *x, int *y, int *comp) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) c, user); + return stbi__info_main(&s,x,y,comp); +} + +STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__is_16_main(&s); +} + +STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) c, user); + return stbi__is_16_main(&s); +} + +#endif // STB_IMAGE_IMPLEMENTATION + +/* + revision history: + 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs + 2.19 (2018-02-11) fix warning + 2.18 (2018-01-30) fix warnings + 2.17 (2018-01-29) change sbti__shiftsigned to avoid clang -O2 bug + 1-bit BMP + *_is_16_bit api + avoid warnings + 2.16 (2017-07-23) all functions have 16-bit variants; + STBI_NO_STDIO works again; + compilation fixes; + fix rounding in unpremultiply; + optimize vertical flip; + disable raw_len validation; + documentation fixes + 2.15 (2017-03-18) fix png-1,2,4 bug; now all Imagenet JPGs decode; + warning fixes; disable run-time SSE detection on gcc; + uniform handling of optional "return" values; + thread-safe initialization of zlib tables + 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs + 2.13 (2016-11-29) add 16-bit API, only supported for PNG right now + 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes + 2.11 (2016-04-02) allocate large structures on the stack + remove white matting for transparent PSD + fix reported channel count for PNG & BMP + re-enable SSE2 in non-gcc 64-bit + support RGB-formatted JPEG + read 16-bit PNGs (only as 8-bit) + 2.10 (2016-01-22) avoid warning introduced in 2.09 by STBI_REALLOC_SIZED + 2.09 (2016-01-16) allow comments in PNM files + 16-bit-per-pixel TGA (not bit-per-component) + info() for TGA could break due to .hdr handling + info() for BMP to shares code instead of sloppy parse + can use STBI_REALLOC_SIZED if allocator doesn't support realloc + code cleanup + 2.08 (2015-09-13) fix to 2.07 cleanup, reading RGB PSD as RGBA + 2.07 (2015-09-13) fix compiler warnings + partial animated GIF support + limited 16-bpc PSD support + #ifdef unused functions + bug with < 92 byte PIC,PNM,HDR,TGA + 2.06 (2015-04-19) fix bug where PSD returns wrong '*comp' value + 2.05 (2015-04-19) fix bug in progressive JPEG handling, fix warning + 2.04 (2015-04-15) try to re-enable SIMD on MinGW 64-bit + 2.03 (2015-04-12) extra corruption checking (mmozeiko) + stbi_set_flip_vertically_on_load (nguillemot) + fix NEON support; fix mingw support + 2.02 (2015-01-19) fix incorrect assert, fix warning + 2.01 (2015-01-17) fix various warnings; suppress SIMD on gcc 32-bit without -msse2 + 2.00b (2014-12-25) fix STBI_MALLOC in progressive JPEG + 2.00 (2014-12-25) optimize JPG, including x86 SSE2 & NEON SIMD (ryg) + progressive JPEG (stb) + PGM/PPM support (Ken Miller) + STBI_MALLOC,STBI_REALLOC,STBI_FREE + GIF bugfix -- seemingly never worked + STBI_NO_*, STBI_ONLY_* + 1.48 (2014-12-14) fix incorrectly-named assert() + 1.47 (2014-12-14) 1/2/4-bit PNG support, both direct and paletted (Omar Cornut & stb) + optimize PNG (ryg) + fix bug in interlaced PNG with user-specified channel count (stb) + 1.46 (2014-08-26) + fix broken tRNS chunk (colorkey-style transparency) in non-paletted PNG + 1.45 (2014-08-16) + fix MSVC-ARM internal compiler error by wrapping malloc + 1.44 (2014-08-07) + various warning fixes from Ronny Chevalier + 1.43 (2014-07-15) + fix MSVC-only compiler problem in code changed in 1.42 + 1.42 (2014-07-09) + don't define _CRT_SECURE_NO_WARNINGS (affects user code) + fixes to stbi__cleanup_jpeg path + added STBI_ASSERT to avoid requiring assert.h + 1.41 (2014-06-25) + fix search&replace from 1.36 that messed up comments/error messages + 1.40 (2014-06-22) + fix gcc struct-initialization warning + 1.39 (2014-06-15) + fix to TGA optimization when req_comp != number of components in TGA; + fix to GIF loading because BMP wasn't rewinding (whoops, no GIFs in my test suite) + add support for BMP version 5 (more ignored fields) + 1.38 (2014-06-06) + suppress MSVC warnings on integer casts truncating values + fix accidental rename of 'skip' field of I/O + 1.37 (2014-06-04) + remove duplicate typedef + 1.36 (2014-06-03) + convert to header file single-file library + if de-iphone isn't set, load iphone images color-swapped instead of returning NULL + 1.35 (2014-05-27) + various warnings + fix broken STBI_SIMD path + fix bug where stbi_load_from_file no longer left file pointer in correct place + fix broken non-easy path for 32-bit BMP (possibly never used) + TGA optimization by Arseny Kapoulkine + 1.34 (unknown) + use STBI_NOTUSED in stbi__resample_row_generic(), fix one more leak in tga failure case + 1.33 (2011-07-14) + make stbi_is_hdr work in STBI_NO_HDR (as specified), minor compiler-friendly improvements + 1.32 (2011-07-13) + support for "info" function for all supported filetypes (SpartanJ) + 1.31 (2011-06-20) + a few more leak fixes, bug in PNG handling (SpartanJ) + 1.30 (2011-06-11) + added ability to load files via callbacks to accomidate custom input streams (Ben Wenger) + removed deprecated format-specific test/load functions + removed support for installable file formats (stbi_loader) -- would have been broken for IO callbacks anyway + error cases in bmp and tga give messages and don't leak (Raymond Barbiero, grisha) + fix inefficiency in decoding 32-bit BMP (David Woo) + 1.29 (2010-08-16) + various warning fixes from Aurelien Pocheville + 1.28 (2010-08-01) + fix bug in GIF palette transparency (SpartanJ) + 1.27 (2010-08-01) + cast-to-stbi_uc to fix warnings + 1.26 (2010-07-24) + fix bug in file buffering for PNG reported by SpartanJ + 1.25 (2010-07-17) + refix trans_data warning (Won Chun) + 1.24 (2010-07-12) + perf improvements reading from files on platforms with lock-heavy fgetc() + minor perf improvements for jpeg + deprecated type-specific functions so we'll get feedback if they're needed + attempt to fix trans_data warning (Won Chun) + 1.23 fixed bug in iPhone support + 1.22 (2010-07-10) + removed image *writing* support + stbi_info support from Jetro Lauha + GIF support from Jean-Marc Lienher + iPhone PNG-extensions from James Brown + warning-fixes from Nicolas Schulz and Janez Zemva (i.stbi__err. Janez (U+017D)emva) + 1.21 fix use of 'stbi_uc' in header (reported by jon blow) + 1.20 added support for Softimage PIC, by Tom Seddon + 1.19 bug in interlaced PNG corruption check (found by ryg) + 1.18 (2008-08-02) + fix a threading bug (local mutable static) + 1.17 support interlaced PNG + 1.16 major bugfix - stbi__convert_format converted one too many pixels + 1.15 initialize some fields for thread safety + 1.14 fix threadsafe conversion bug + header-file-only version (#define STBI_HEADER_FILE_ONLY before including) + 1.13 threadsafe + 1.12 const qualifiers in the API + 1.11 Support installable IDCT, colorspace conversion routines + 1.10 Fixes for 64-bit (don't use "unsigned long") + optimized upsampling by Fabian "ryg" Giesen + 1.09 Fix format-conversion for PSD code (bad global variables!) + 1.08 Thatcher Ulrich's PSD code integrated by Nicolas Schulz + 1.07 attempt to fix C++ warning/errors again + 1.06 attempt to fix C++ warning/errors again + 1.05 fix TGA loading to return correct *comp and use good luminance calc + 1.04 default float alpha is 1, not 255; use 'void *' for stbi_image_free + 1.03 bugfixes to STBI_NO_STDIO, STBI_NO_HDR + 1.02 support for (subset of) HDR files, float interface for preferred access to them + 1.01 fix bug: possible bug in handling right-side up bmps... not sure + fix bug: the stbi__bmp_load() and stbi__tga_load() functions didn't work at all + 1.00 interface to zlib that skips zlib header + 0.99 correct handling of alpha in palette + 0.98 TGA loader by lonesock; dynamically add loaders (untested) + 0.97 jpeg errors on too large a file; also catch another malloc failure + 0.96 fix detection of invalid v value - particleman@mollyrocket forum + 0.95 during header scan, seek to markers in case of padding + 0.94 STBI_NO_STDIO to disable stdio usage; rename all #defines the same + 0.93 handle jpegtran output; verbose errors + 0.92 read 4,8,16,24,32-bit BMP files of several formats + 0.91 output 24-bit Windows 3.0 BMP files + 0.90 fix a few more warnings; bump version number to approach 1.0 + 0.61 bugfixes due to Marc LeBlanc, Christopher Lloyd + 0.60 fix compiling as c++ + 0.59 fix warnings: merge Dave Moore's -Wall fixes + 0.58 fix bug: zlib uncompressed mode len/nlen was wrong endian + 0.57 fix bug: jpg last huffman symbol before marker was >9 bits but less than 16 available + 0.56 fix bug: zlib uncompressed mode len vs. nlen + 0.55 fix bug: restart_interval not initialized to 0 + 0.54 allow NULL for 'int *comp' + 0.53 fix bug in png 3->4; speedup png decoding + 0.52 png handles req_comp=3,4 directly; minor cleanup; jpeg comments + 0.51 obey req_comp requests, 1-component jpegs return as 1-component, + on 'test' only check type, not whether we support this variant + 0.50 (2006-11-19) + first released version +*/ + + +/* +------------------------------------------------------------------------------ +This software is available under 2 licenses -- choose whichever you prefer. +------------------------------------------------------------------------------ +ALTERNATIVE A - MIT License +Copyright (c) 2017 Sean Barrett +Permission is hereby granted, free of charge, to any person obtaining a copy of +this software and associated documentation files (the "Software"), to deal in +the Software without restriction, including without limitation the rights to +use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies +of the Software, and to permit persons to whom the Software is furnished to do +so, subject to the following conditions: +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. +------------------------------------------------------------------------------ +ALTERNATIVE B - Public Domain (www.unlicense.org) +This is free and unencumbered software released into the public domain. +Anyone is free to copy, modify, publish, use, compile, sell, or distribute this +software, either in source code form or as a compiled binary, for any purpose, +commercial or non-commercial, and by any means. +In jurisdictions that recognize copyright laws, the author or authors of this +software dedicate any and all copyright interest in the software to the public +domain. We make this dedication for the benefit of the public at large and to +the detriment of our heirs and successors. We intend this dedication to be an +overt act of relinquishment in perpetuity of all present and future rights to +this software under copyright law. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN +ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION +WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. +------------------------------------------------------------------------------ +*/ diff --git a/src/files/stb_image_resize.h b/src/files/stb_image_resize.h new file mode 100644 index 0000000..4f6ad35 --- /dev/null +++ b/src/files/stb_image_resize.h @@ -0,0 +1,2630 @@ +/* stb_image_resize - v0.96 - public domain image resizing + by Jorge L Rodriguez (@VinoBS) - 2014 + http://github.com/nothings/stb + + Written with emphasis on usability, portability, and efficiency. (No + SIMD or threads, so it be easily outperformed by libs that use those.) + Only scaling and translation is supported, no rotations or shears. + Easy API downsamples w/Mitchell filter, upsamples w/cubic interpolation. + + COMPILING & LINKING + In one C/C++ file that #includes this file, do this: + #define STB_IMAGE_RESIZE_IMPLEMENTATION + before the #include. That will create the implementation in that file. + + QUICKSTART + stbir_resize_uint8( input_pixels , in_w , in_h , 0, + output_pixels, out_w, out_h, 0, num_channels) + stbir_resize_float(...) + stbir_resize_uint8_srgb( input_pixels , in_w , in_h , 0, + output_pixels, out_w, out_h, 0, + num_channels , alpha_chan , 0) + stbir_resize_uint8_srgb_edgemode( + input_pixels , in_w , in_h , 0, + output_pixels, out_w, out_h, 0, + num_channels , alpha_chan , 0, STBIR_EDGE_CLAMP) + // WRAP/REFLECT/ZERO + + FULL API + See the "header file" section of the source for API documentation. + + ADDITIONAL DOCUMENTATION + + SRGB & FLOATING POINT REPRESENTATION + The sRGB functions presume IEEE floating point. If you do not have + IEEE floating point, define STBIR_NON_IEEE_FLOAT. This will use + a slower implementation. + + MEMORY ALLOCATION + The resize functions here perform a single memory allocation using + malloc. To control the memory allocation, before the #include that + triggers the implementation, do: + + #define STBIR_MALLOC(size,context) ... + #define STBIR_FREE(ptr,context) ... + + Each resize function makes exactly one call to malloc/free, so to use + temp memory, store the temp memory in the context and return that. + + ASSERT + Define STBIR_ASSERT(boolval) to override assert() and not use assert.h + + OPTIMIZATION + Define STBIR_SATURATE_INT to compute clamp values in-range using + integer operations instead of float operations. This may be faster + on some platforms. + + DEFAULT FILTERS + For functions which don't provide explicit control over what filters + to use, you can change the compile-time defaults with + + #define STBIR_DEFAULT_FILTER_UPSAMPLE STBIR_FILTER_something + #define STBIR_DEFAULT_FILTER_DOWNSAMPLE STBIR_FILTER_something + + See stbir_filter in the header-file section for the list of filters. + + NEW FILTERS + A number of 1D filter kernels are used. For a list of + supported filters see the stbir_filter enum. To add a new filter, + write a filter function and add it to stbir__filter_info_table. + + PROGRESS + For interactive use with slow resize operations, you can install + a progress-report callback: + + #define STBIR_PROGRESS_REPORT(val) some_func(val) + + The parameter val is a float which goes from 0 to 1 as progress is made. + + For example: + + static void my_progress_report(float progress); + #define STBIR_PROGRESS_REPORT(val) my_progress_report(val) + + #define STB_IMAGE_RESIZE_IMPLEMENTATION + #include "stb_image_resize.h" + + static void my_progress_report(float progress) + { + printf("Progress: %f%%\n", progress*100); + } + + MAX CHANNELS + If your image has more than 64 channels, define STBIR_MAX_CHANNELS + to the max you'll have. + + ALPHA CHANNEL + Most of the resizing functions provide the ability to control how + the alpha channel of an image is processed. The important things + to know about this: + + 1. The best mathematically-behaved version of alpha to use is + called "premultiplied alpha", in which the other color channels + have had the alpha value multiplied in. If you use premultiplied + alpha, linear filtering (such as image resampling done by this + library, or performed in texture units on GPUs) does the "right + thing". While premultiplied alpha is standard in the movie CGI + industry, it is still uncommon in the videogame/real-time world. + + If you linearly filter non-premultiplied alpha, strange effects + occur. (For example, the 50/50 average of 99% transparent bright green + and 1% transparent black produces 50% transparent dark green when + non-premultiplied, whereas premultiplied it produces 50% + transparent near-black. The former introduces green energy + that doesn't exist in the source image.) + + 2. Artists should not edit premultiplied-alpha images; artists + want non-premultiplied alpha images. Thus, art tools generally output + non-premultiplied alpha images. + + 3. You will get best results in most cases by converting images + to premultiplied alpha before processing them mathematically. + + 4. If you pass the flag STBIR_FLAG_ALPHA_PREMULTIPLIED, the + resizer does not do anything special for the alpha channel; + it is resampled identically to other channels. This produces + the correct results for premultiplied-alpha images, but produces + less-than-ideal results for non-premultiplied-alpha images. + + 5. If you do not pass the flag STBIR_FLAG_ALPHA_PREMULTIPLIED, + then the resizer weights the contribution of input pixels + based on their alpha values, or, equivalently, it multiplies + the alpha value into the color channels, resamples, then divides + by the resultant alpha value. Input pixels which have alpha=0 do + not contribute at all to output pixels unless _all_ of the input + pixels affecting that output pixel have alpha=0, in which case + the result for that pixel is the same as it would be without + STBIR_FLAG_ALPHA_PREMULTIPLIED. However, this is only true for + input images in integer formats. For input images in float format, + input pixels with alpha=0 have no effect, and output pixels + which have alpha=0 will be 0 in all channels. (For float images, + you can manually achieve the same result by adding a tiny epsilon + value to the alpha channel of every image, and then subtracting + or clamping it at the end.) + + 6. You can suppress the behavior described in #5 and make + all-0-alpha pixels have 0 in all channels by #defining + STBIR_NO_ALPHA_EPSILON. + + 7. You can separately control whether the alpha channel is + interpreted as linear or affected by the colorspace. By default + it is linear; you almost never want to apply the colorspace. + (For example, graphics hardware does not apply sRGB conversion + to the alpha channel.) + + CONTRIBUTORS + Jorge L Rodriguez: Implementation + Sean Barrett: API design, optimizations + Aras Pranckevicius: bugfix + Nathan Reed: warning fixes + + REVISIONS + 0.96 (2019-03-04) fixed warnings + 0.95 (2017-07-23) fixed warnings + 0.94 (2017-03-18) fixed warnings + 0.93 (2017-03-03) fixed bug with certain combinations of heights + 0.92 (2017-01-02) fix integer overflow on large (>2GB) images + 0.91 (2016-04-02) fix warnings; fix handling of subpixel regions + 0.90 (2014-09-17) first released version + + LICENSE + See end of file for license information. + + TODO + Don't decode all of the image data when only processing a partial tile + Don't use full-width decode buffers when only processing a partial tile + When processing wide images, break processing into tiles so data fits in L1 cache + Installable filters? + Resize that respects alpha test coverage + (Reference code: FloatImage::alphaTestCoverage and FloatImage::scaleAlphaToCoverage: + https://code.google.com/p/nvidia-texture-tools/source/browse/trunk/src/nvimage/FloatImage.cpp ) +*/ + +#ifndef STBIR_INCLUDE_STB_IMAGE_RESIZE_H +#define STBIR_INCLUDE_STB_IMAGE_RESIZE_H + +#ifdef _MSC_VER +typedef unsigned char stbir_uint8; +typedef unsigned short stbir_uint16; +typedef unsigned int stbir_uint32; +#else +#include +typedef uint8_t stbir_uint8; +typedef uint16_t stbir_uint16; +typedef uint32_t stbir_uint32; +#endif + +#ifndef STBIRDEF +#ifdef STB_IMAGE_RESIZE_STATIC +#define STBIRDEF static +#else +#ifdef __cplusplus +#define STBIRDEF extern "C" +#else +#define STBIRDEF extern +#endif +#endif +#endif + +////////////////////////////////////////////////////////////////////////////// +// +// Easy-to-use API: +// +// * "input pixels" points to an array of image data with 'num_channels' channels (e.g. RGB=3, RGBA=4) +// * input_w is input image width (x-axis), input_h is input image height (y-axis) +// * stride is the offset between successive rows of image data in memory, in bytes. you can +// specify 0 to mean packed continuously in memory +// * alpha channel is treated identically to other channels. +// * colorspace is linear or sRGB as specified by function name +// * returned result is 1 for success or 0 in case of an error. +// #define STBIR_ASSERT() to trigger an assert on parameter validation errors. +// * Memory required grows approximately linearly with input and output size, but with +// discontinuities at input_w == output_w and input_h == output_h. +// * These functions use a "default" resampling filter defined at compile time. To change the filter, +// you can change the compile-time defaults by #defining STBIR_DEFAULT_FILTER_UPSAMPLE +// and STBIR_DEFAULT_FILTER_DOWNSAMPLE, or you can use the medium-complexity API. + +STBIRDEF int stbir_resize_uint8( const unsigned char *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + unsigned char *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + int num_channels); + +STBIRDEF int stbir_resize_float( const float *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + float *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + int num_channels); + + +// The following functions interpret image data as gamma-corrected sRGB. +// Specify STBIR_ALPHA_CHANNEL_NONE if you have no alpha channel, +// or otherwise provide the index of the alpha channel. Flags value +// of 0 will probably do the right thing if you're not sure what +// the flags mean. + +#define STBIR_ALPHA_CHANNEL_NONE -1 + +// Set this flag if your texture has premultiplied alpha. Otherwise, stbir will +// use alpha-weighted resampling (effectively premultiplying, resampling, +// then unpremultiplying). +#define STBIR_FLAG_ALPHA_PREMULTIPLIED (1 << 0) +// The specified alpha channel should be handled as gamma-corrected value even +// when doing sRGB operations. +#define STBIR_FLAG_ALPHA_USES_COLORSPACE (1 << 1) + +STBIRDEF int stbir_resize_uint8_srgb(const unsigned char *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + unsigned char *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + int num_channels, int alpha_channel, int flags); + + +typedef enum +{ + STBIR_EDGE_CLAMP = 1, + STBIR_EDGE_REFLECT = 2, + STBIR_EDGE_WRAP = 3, + STBIR_EDGE_ZERO = 4, +} stbir_edge; + +// This function adds the ability to specify how requests to sample off the edge of the image are handled. +STBIRDEF int stbir_resize_uint8_srgb_edgemode(const unsigned char *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + unsigned char *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + int num_channels, int alpha_channel, int flags, + stbir_edge edge_wrap_mode); + +////////////////////////////////////////////////////////////////////////////// +// +// Medium-complexity API +// +// This extends the easy-to-use API as follows: +// +// * Alpha-channel can be processed separately +// * If alpha_channel is not STBIR_ALPHA_CHANNEL_NONE +// * Alpha channel will not be gamma corrected (unless flags&STBIR_FLAG_GAMMA_CORRECT) +// * Filters will be weighted by alpha channel (unless flags&STBIR_FLAG_ALPHA_PREMULTIPLIED) +// * Filter can be selected explicitly +// * uint16 image type +// * sRGB colorspace available for all types +// * context parameter for passing to STBIR_MALLOC + +typedef enum +{ + STBIR_FILTER_DEFAULT = 0, // use same filter type that easy-to-use API chooses + STBIR_FILTER_BOX = 1, // A trapezoid w/1-pixel wide ramps, same result as box for integer scale ratios + STBIR_FILTER_TRIANGLE = 2, // On upsampling, produces same results as bilinear texture filtering + STBIR_FILTER_CUBICBSPLINE = 3, // The cubic b-spline (aka Mitchell-Netrevalli with B=1,C=0), gaussian-esque + STBIR_FILTER_CATMULLROM = 4, // An interpolating cubic spline + STBIR_FILTER_MITCHELL = 5, // Mitchell-Netrevalli filter with B=1/3, C=1/3 +} stbir_filter; + +typedef enum +{ + STBIR_COLORSPACE_LINEAR, + STBIR_COLORSPACE_SRGB, + + STBIR_MAX_COLORSPACES, +} stbir_colorspace; + +// The following functions are all identical except for the type of the image data + +STBIRDEF int stbir_resize_uint8_generic( const unsigned char *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + unsigned char *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + int num_channels, int alpha_channel, int flags, + stbir_edge edge_wrap_mode, stbir_filter filter, stbir_colorspace space, + void *alloc_context); + +STBIRDEF int stbir_resize_uint16_generic(const stbir_uint16 *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + stbir_uint16 *output_pixels , int output_w, int output_h, int output_stride_in_bytes, + int num_channels, int alpha_channel, int flags, + stbir_edge edge_wrap_mode, stbir_filter filter, stbir_colorspace space, + void *alloc_context); + +STBIRDEF int stbir_resize_float_generic( const float *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + float *output_pixels , int output_w, int output_h, int output_stride_in_bytes, + int num_channels, int alpha_channel, int flags, + stbir_edge edge_wrap_mode, stbir_filter filter, stbir_colorspace space, + void *alloc_context); + + + +////////////////////////////////////////////////////////////////////////////// +// +// Full-complexity API +// +// This extends the medium API as follows: +// +// * uint32 image type +// * not typesafe +// * separate filter types for each axis +// * separate edge modes for each axis +// * can specify scale explicitly for subpixel correctness +// * can specify image source tile using texture coordinates + +typedef enum +{ + STBIR_TYPE_UINT8 , + STBIR_TYPE_UINT16, + STBIR_TYPE_UINT32, + STBIR_TYPE_FLOAT , + + STBIR_MAX_TYPES +} stbir_datatype; + +STBIRDEF int stbir_resize( const void *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + void *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + stbir_datatype datatype, + int num_channels, int alpha_channel, int flags, + stbir_edge edge_mode_horizontal, stbir_edge edge_mode_vertical, + stbir_filter filter_horizontal, stbir_filter filter_vertical, + stbir_colorspace space, void *alloc_context); + +STBIRDEF int stbir_resize_subpixel(const void *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + void *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + stbir_datatype datatype, + int num_channels, int alpha_channel, int flags, + stbir_edge edge_mode_horizontal, stbir_edge edge_mode_vertical, + stbir_filter filter_horizontal, stbir_filter filter_vertical, + stbir_colorspace space, void *alloc_context, + float x_scale, float y_scale, + float x_offset, float y_offset); + +STBIRDEF int stbir_resize_region( const void *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + void *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + stbir_datatype datatype, + int num_channels, int alpha_channel, int flags, + stbir_edge edge_mode_horizontal, stbir_edge edge_mode_vertical, + stbir_filter filter_horizontal, stbir_filter filter_vertical, + stbir_colorspace space, void *alloc_context, + float s0, float t0, float s1, float t1); +// (s0, t0) & (s1, t1) are the top-left and bottom right corner (uv addressing style: [0, 1]x[0, 1]) of a region of the input image to use. + +// +// +//// end header file ///////////////////////////////////////////////////// +#endif // STBIR_INCLUDE_STB_IMAGE_RESIZE_H + + + + + +#ifdef STB_IMAGE_RESIZE_IMPLEMENTATION + +#ifndef STBIR_ASSERT +#include +#define STBIR_ASSERT(x) assert(x) +#endif + +// For memset +#include + +#include + +#ifndef STBIR_MALLOC +#include +// use comma operator to evaluate c, to avoid "unused parameter" warnings +#define STBIR_MALLOC(size,c) ((void)(c), malloc(size)) +#define STBIR_FREE(ptr,c) ((void)(c), free(ptr)) +#endif + +#ifndef _MSC_VER +#ifdef __cplusplus +#define stbir__inline inline +#else +#define stbir__inline +#endif +#else +#define stbir__inline __forceinline +#endif + + +// should produce compiler error if size is wrong +typedef unsigned char stbir__validate_uint32[sizeof(stbir_uint32) == 4 ? 1 : -1]; + +#ifdef _MSC_VER +#define STBIR__NOTUSED(v) (void)(v) +#else +#define STBIR__NOTUSED(v) (void)sizeof(v) +#endif + +#define STBIR__ARRAY_SIZE(a) (sizeof((a))/sizeof((a)[0])) + +#ifndef STBIR_DEFAULT_FILTER_UPSAMPLE +#define STBIR_DEFAULT_FILTER_UPSAMPLE STBIR_FILTER_CATMULLROM +#endif + +#ifndef STBIR_DEFAULT_FILTER_DOWNSAMPLE +#define STBIR_DEFAULT_FILTER_DOWNSAMPLE STBIR_FILTER_MITCHELL +#endif + +#ifndef STBIR_PROGRESS_REPORT +#define STBIR_PROGRESS_REPORT(float_0_to_1) +#endif + +#ifndef STBIR_MAX_CHANNELS +#define STBIR_MAX_CHANNELS 64 +#endif + +#if STBIR_MAX_CHANNELS > 65536 +#error "Too many channels; STBIR_MAX_CHANNELS must be no more than 65536." +// because we store the indices in 16-bit variables +#endif + +// This value is added to alpha just before premultiplication to avoid +// zeroing out color values. It is equivalent to 2^-80. If you don't want +// that behavior (it may interfere if you have floating point images with +// very small alpha values) then you can define STBIR_NO_ALPHA_EPSILON to +// disable it. +#ifndef STBIR_ALPHA_EPSILON +#define STBIR_ALPHA_EPSILON ((float)1 / (1 << 20) / (1 << 20) / (1 << 20) / (1 << 20)) +#endif + + + +#ifdef _MSC_VER +#define STBIR__UNUSED_PARAM(v) (void)(v) +#else +#define STBIR__UNUSED_PARAM(v) (void)sizeof(v) +#endif + +// must match stbir_datatype +static unsigned char stbir__type_size[] = { + 1, // STBIR_TYPE_UINT8 + 2, // STBIR_TYPE_UINT16 + 4, // STBIR_TYPE_UINT32 + 4, // STBIR_TYPE_FLOAT +}; + +// Kernel function centered at 0 +typedef float (stbir__kernel_fn)(float x, float scale); +typedef float (stbir__support_fn)(float scale); + +typedef struct +{ + stbir__kernel_fn* kernel; + stbir__support_fn* support; +} stbir__filter_info; + +// When upsampling, the contributors are which source pixels contribute. +// When downsampling, the contributors are which destination pixels are contributed to. +typedef struct +{ + int n0; // First contributing pixel + int n1; // Last contributing pixel +} stbir__contributors; + +typedef struct +{ + const void* input_data; + int input_w; + int input_h; + int input_stride_bytes; + + void* output_data; + int output_w; + int output_h; + int output_stride_bytes; + + float s0, t0, s1, t1; + + float horizontal_shift; // Units: output pixels + float vertical_shift; // Units: output pixels + float horizontal_scale; + float vertical_scale; + + int channels; + int alpha_channel; + stbir_uint32 flags; + stbir_datatype type; + stbir_filter horizontal_filter; + stbir_filter vertical_filter; + stbir_edge edge_horizontal; + stbir_edge edge_vertical; + stbir_colorspace colorspace; + + stbir__contributors* horizontal_contributors; + float* horizontal_coefficients; + + stbir__contributors* vertical_contributors; + float* vertical_coefficients; + + int decode_buffer_pixels; + float* decode_buffer; + + float* horizontal_buffer; + + // cache these because ceil/floor are inexplicably showing up in profile + int horizontal_coefficient_width; + int vertical_coefficient_width; + int horizontal_filter_pixel_width; + int vertical_filter_pixel_width; + int horizontal_filter_pixel_margin; + int vertical_filter_pixel_margin; + int horizontal_num_contributors; + int vertical_num_contributors; + + int ring_buffer_length_bytes; // The length of an individual entry in the ring buffer. The total number of ring buffers is stbir__get_filter_pixel_width(filter) + int ring_buffer_num_entries; // Total number of entries in the ring buffer. + int ring_buffer_first_scanline; + int ring_buffer_last_scanline; + int ring_buffer_begin_index; // first_scanline is at this index in the ring buffer + float* ring_buffer; + + float* encode_buffer; // A temporary buffer to store floats so we don't lose precision while we do multiply-adds. + + int horizontal_contributors_size; + int horizontal_coefficients_size; + int vertical_contributors_size; + int vertical_coefficients_size; + int decode_buffer_size; + int horizontal_buffer_size; + int ring_buffer_size; + int encode_buffer_size; +} stbir__info; + + +static const float stbir__max_uint8_as_float = 255.0f; +static const float stbir__max_uint16_as_float = 65535.0f; +static const double stbir__max_uint32_as_float = 4294967295.0; + + +static stbir__inline int stbir__min(int a, int b) +{ + return a < b ? a : b; +} + +static stbir__inline float stbir__saturate(float x) +{ + if (x < 0) + return 0; + + if (x > 1) + return 1; + + return x; +} + +#ifdef STBIR_SATURATE_INT +static stbir__inline stbir_uint8 stbir__saturate8(int x) +{ + if ((unsigned int) x <= 255) + return x; + + if (x < 0) + return 0; + + return 255; +} + +static stbir__inline stbir_uint16 stbir__saturate16(int x) +{ + if ((unsigned int) x <= 65535) + return x; + + if (x < 0) + return 0; + + return 65535; +} +#endif + +static float stbir__srgb_uchar_to_linear_float[256] = { + 0.000000f, 0.000304f, 0.000607f, 0.000911f, 0.001214f, 0.001518f, 0.001821f, 0.002125f, 0.002428f, 0.002732f, 0.003035f, + 0.003347f, 0.003677f, 0.004025f, 0.004391f, 0.004777f, 0.005182f, 0.005605f, 0.006049f, 0.006512f, 0.006995f, 0.007499f, + 0.008023f, 0.008568f, 0.009134f, 0.009721f, 0.010330f, 0.010960f, 0.011612f, 0.012286f, 0.012983f, 0.013702f, 0.014444f, + 0.015209f, 0.015996f, 0.016807f, 0.017642f, 0.018500f, 0.019382f, 0.020289f, 0.021219f, 0.022174f, 0.023153f, 0.024158f, + 0.025187f, 0.026241f, 0.027321f, 0.028426f, 0.029557f, 0.030713f, 0.031896f, 0.033105f, 0.034340f, 0.035601f, 0.036889f, + 0.038204f, 0.039546f, 0.040915f, 0.042311f, 0.043735f, 0.045186f, 0.046665f, 0.048172f, 0.049707f, 0.051269f, 0.052861f, + 0.054480f, 0.056128f, 0.057805f, 0.059511f, 0.061246f, 0.063010f, 0.064803f, 0.066626f, 0.068478f, 0.070360f, 0.072272f, + 0.074214f, 0.076185f, 0.078187f, 0.080220f, 0.082283f, 0.084376f, 0.086500f, 0.088656f, 0.090842f, 0.093059f, 0.095307f, + 0.097587f, 0.099899f, 0.102242f, 0.104616f, 0.107023f, 0.109462f, 0.111932f, 0.114435f, 0.116971f, 0.119538f, 0.122139f, + 0.124772f, 0.127438f, 0.130136f, 0.132868f, 0.135633f, 0.138432f, 0.141263f, 0.144128f, 0.147027f, 0.149960f, 0.152926f, + 0.155926f, 0.158961f, 0.162029f, 0.165132f, 0.168269f, 0.171441f, 0.174647f, 0.177888f, 0.181164f, 0.184475f, 0.187821f, + 0.191202f, 0.194618f, 0.198069f, 0.201556f, 0.205079f, 0.208637f, 0.212231f, 0.215861f, 0.219526f, 0.223228f, 0.226966f, + 0.230740f, 0.234551f, 0.238398f, 0.242281f, 0.246201f, 0.250158f, 0.254152f, 0.258183f, 0.262251f, 0.266356f, 0.270498f, + 0.274677f, 0.278894f, 0.283149f, 0.287441f, 0.291771f, 0.296138f, 0.300544f, 0.304987f, 0.309469f, 0.313989f, 0.318547f, + 0.323143f, 0.327778f, 0.332452f, 0.337164f, 0.341914f, 0.346704f, 0.351533f, 0.356400f, 0.361307f, 0.366253f, 0.371238f, + 0.376262f, 0.381326f, 0.386430f, 0.391573f, 0.396755f, 0.401978f, 0.407240f, 0.412543f, 0.417885f, 0.423268f, 0.428691f, + 0.434154f, 0.439657f, 0.445201f, 0.450786f, 0.456411f, 0.462077f, 0.467784f, 0.473532f, 0.479320f, 0.485150f, 0.491021f, + 0.496933f, 0.502887f, 0.508881f, 0.514918f, 0.520996f, 0.527115f, 0.533276f, 0.539480f, 0.545725f, 0.552011f, 0.558340f, + 0.564712f, 0.571125f, 0.577581f, 0.584078f, 0.590619f, 0.597202f, 0.603827f, 0.610496f, 0.617207f, 0.623960f, 0.630757f, + 0.637597f, 0.644480f, 0.651406f, 0.658375f, 0.665387f, 0.672443f, 0.679543f, 0.686685f, 0.693872f, 0.701102f, 0.708376f, + 0.715694f, 0.723055f, 0.730461f, 0.737911f, 0.745404f, 0.752942f, 0.760525f, 0.768151f, 0.775822f, 0.783538f, 0.791298f, + 0.799103f, 0.806952f, 0.814847f, 0.822786f, 0.830770f, 0.838799f, 0.846873f, 0.854993f, 0.863157f, 0.871367f, 0.879622f, + 0.887923f, 0.896269f, 0.904661f, 0.913099f, 0.921582f, 0.930111f, 0.938686f, 0.947307f, 0.955974f, 0.964686f, 0.973445f, + 0.982251f, 0.991102f, 1.0f +}; + +static float stbir__srgb_to_linear(float f) +{ + if (f <= 0.04045f) + return f / 12.92f; + else + return (float)pow((f + 0.055f) / 1.055f, 2.4f); +} + +static float stbir__linear_to_srgb(float f) +{ + if (f <= 0.0031308f) + return f * 12.92f; + else + return 1.055f * (float)pow(f, 1 / 2.4f) - 0.055f; +} + +#ifndef STBIR_NON_IEEE_FLOAT +// From https://gist.github.com/rygorous/2203834 + +typedef union +{ + stbir_uint32 u; + float f; +} stbir__FP32; + +static const stbir_uint32 fp32_to_srgb8_tab4[104] = { + 0x0073000d, 0x007a000d, 0x0080000d, 0x0087000d, 0x008d000d, 0x0094000d, 0x009a000d, 0x00a1000d, + 0x00a7001a, 0x00b4001a, 0x00c1001a, 0x00ce001a, 0x00da001a, 0x00e7001a, 0x00f4001a, 0x0101001a, + 0x010e0033, 0x01280033, 0x01410033, 0x015b0033, 0x01750033, 0x018f0033, 0x01a80033, 0x01c20033, + 0x01dc0067, 0x020f0067, 0x02430067, 0x02760067, 0x02aa0067, 0x02dd0067, 0x03110067, 0x03440067, + 0x037800ce, 0x03df00ce, 0x044600ce, 0x04ad00ce, 0x051400ce, 0x057b00c5, 0x05dd00bc, 0x063b00b5, + 0x06970158, 0x07420142, 0x07e30130, 0x087b0120, 0x090b0112, 0x09940106, 0x0a1700fc, 0x0a9500f2, + 0x0b0f01cb, 0x0bf401ae, 0x0ccb0195, 0x0d950180, 0x0e56016e, 0x0f0d015e, 0x0fbc0150, 0x10630143, + 0x11070264, 0x1238023e, 0x1357021d, 0x14660201, 0x156601e9, 0x165a01d3, 0x174401c0, 0x182401af, + 0x18fe0331, 0x1a9602fe, 0x1c1502d2, 0x1d7e02ad, 0x1ed4028d, 0x201a0270, 0x21520256, 0x227d0240, + 0x239f0443, 0x25c003fe, 0x27bf03c4, 0x29a10392, 0x2b6a0367, 0x2d1d0341, 0x2ebe031f, 0x304d0300, + 0x31d105b0, 0x34a80555, 0x37520507, 0x39d504c5, 0x3c37048b, 0x3e7c0458, 0x40a8042a, 0x42bd0401, + 0x44c20798, 0x488e071e, 0x4c1c06b6, 0x4f76065d, 0x52a50610, 0x55ac05cc, 0x5892058f, 0x5b590559, + 0x5e0c0a23, 0x631c0980, 0x67db08f6, 0x6c55087f, 0x70940818, 0x74a007bd, 0x787d076c, 0x7c330723, +}; + +static stbir_uint8 stbir__linear_to_srgb_uchar(float in) +{ + static const stbir__FP32 almostone = { 0x3f7fffff }; // 1-eps + static const stbir__FP32 minval = { (127-13) << 23 }; + stbir_uint32 tab,bias,scale,t; + stbir__FP32 f; + + // Clamp to [2^(-13), 1-eps]; these two values map to 0 and 1, respectively. + // The tests are carefully written so that NaNs map to 0, same as in the reference + // implementation. + if (!(in > minval.f)) // written this way to catch NaNs + in = minval.f; + if (in > almostone.f) + in = almostone.f; + + // Do the table lookup and unpack bias, scale + f.f = in; + tab = fp32_to_srgb8_tab4[(f.u - minval.u) >> 20]; + bias = (tab >> 16) << 9; + scale = tab & 0xffff; + + // Grab next-highest mantissa bits and perform linear interpolation + t = (f.u >> 12) & 0xff; + return (unsigned char) ((bias + scale*t) >> 16); +} + +#else +// sRGB transition values, scaled by 1<<28 +static int stbir__srgb_offset_to_linear_scaled[256] = +{ + 0, 40738, 122216, 203693, 285170, 366648, 448125, 529603, + 611080, 692557, 774035, 855852, 942009, 1033024, 1128971, 1229926, + 1335959, 1447142, 1563542, 1685229, 1812268, 1944725, 2082664, 2226148, + 2375238, 2529996, 2690481, 2856753, 3028870, 3206888, 3390865, 3580856, + 3776916, 3979100, 4187460, 4402049, 4622919, 4850123, 5083710, 5323731, + 5570236, 5823273, 6082892, 6349140, 6622065, 6901714, 7188133, 7481369, + 7781466, 8088471, 8402427, 8723380, 9051372, 9386448, 9728650, 10078021, + 10434603, 10798439, 11169569, 11548036, 11933879, 12327139, 12727857, 13136073, + 13551826, 13975156, 14406100, 14844697, 15290987, 15745007, 16206795, 16676389, + 17153826, 17639142, 18132374, 18633560, 19142734, 19659934, 20185196, 20718552, + 21260042, 21809696, 22367554, 22933648, 23508010, 24090680, 24681686, 25281066, + 25888850, 26505076, 27129772, 27762974, 28404716, 29055026, 29713942, 30381490, + 31057708, 31742624, 32436272, 33138682, 33849884, 34569912, 35298800, 36036568, + 36783260, 37538896, 38303512, 39077136, 39859796, 40651528, 41452360, 42262316, + 43081432, 43909732, 44747252, 45594016, 46450052, 47315392, 48190064, 49074096, + 49967516, 50870356, 51782636, 52704392, 53635648, 54576432, 55526772, 56486700, + 57456236, 58435408, 59424248, 60422780, 61431036, 62449032, 63476804, 64514376, + 65561776, 66619028, 67686160, 68763192, 69850160, 70947088, 72053992, 73170912, + 74297864, 75434880, 76581976, 77739184, 78906536, 80084040, 81271736, 82469648, + 83677792, 84896192, 86124888, 87363888, 88613232, 89872928, 91143016, 92423512, + 93714432, 95015816, 96327688, 97650056, 98982952, 100326408, 101680440, 103045072, + 104420320, 105806224, 107202800, 108610064, 110028048, 111456776, 112896264, 114346544, + 115807632, 117279552, 118762328, 120255976, 121760536, 123276016, 124802440, 126339832, + 127888216, 129447616, 131018048, 132599544, 134192112, 135795792, 137410592, 139036528, + 140673648, 142321952, 143981456, 145652208, 147334208, 149027488, 150732064, 152447968, + 154175200, 155913792, 157663776, 159425168, 161197984, 162982240, 164777968, 166585184, + 168403904, 170234160, 172075968, 173929344, 175794320, 177670896, 179559120, 181458992, + 183370528, 185293776, 187228736, 189175424, 191133888, 193104112, 195086128, 197079968, + 199085648, 201103184, 203132592, 205173888, 207227120, 209292272, 211369392, 213458480, + 215559568, 217672656, 219797792, 221934976, 224084240, 226245600, 228419056, 230604656, + 232802400, 235012320, 237234432, 239468736, 241715280, 243974080, 246245120, 248528464, + 250824112, 253132064, 255452368, 257785040, 260130080, 262487520, 264857376, 267239664, +}; + +static stbir_uint8 stbir__linear_to_srgb_uchar(float f) +{ + int x = (int) (f * (1 << 28)); // has headroom so you don't need to clamp + int v = 0; + int i; + + // Refine the guess with a short binary search. + i = v + 128; if (x >= stbir__srgb_offset_to_linear_scaled[i]) v = i; + i = v + 64; if (x >= stbir__srgb_offset_to_linear_scaled[i]) v = i; + i = v + 32; if (x >= stbir__srgb_offset_to_linear_scaled[i]) v = i; + i = v + 16; if (x >= stbir__srgb_offset_to_linear_scaled[i]) v = i; + i = v + 8; if (x >= stbir__srgb_offset_to_linear_scaled[i]) v = i; + i = v + 4; if (x >= stbir__srgb_offset_to_linear_scaled[i]) v = i; + i = v + 2; if (x >= stbir__srgb_offset_to_linear_scaled[i]) v = i; + i = v + 1; if (x >= stbir__srgb_offset_to_linear_scaled[i]) v = i; + + return (stbir_uint8) v; +} +#endif + +static float stbir__filter_trapezoid(float x, float scale) +{ + float halfscale = scale / 2; + float t = 0.5f + halfscale; + STBIR_ASSERT(scale <= 1); + + x = (float)fabs(x); + + if (x >= t) + return 0; + else + { + float r = 0.5f - halfscale; + if (x <= r) + return 1; + else + return (t - x) / scale; + } +} + +static float stbir__support_trapezoid(float scale) +{ + STBIR_ASSERT(scale <= 1); + return 0.5f + scale / 2; +} + +static float stbir__filter_triangle(float x, float s) +{ + STBIR__UNUSED_PARAM(s); + + x = (float)fabs(x); + + if (x <= 1.0f) + return 1 - x; + else + return 0; +} + +static float stbir__filter_cubic(float x, float s) +{ + STBIR__UNUSED_PARAM(s); + + x = (float)fabs(x); + + if (x < 1.0f) + return (4 + x*x*(3*x - 6))/6; + else if (x < 2.0f) + return (8 + x*(-12 + x*(6 - x)))/6; + + return (0.0f); +} + +static float stbir__filter_catmullrom(float x, float s) +{ + STBIR__UNUSED_PARAM(s); + + x = (float)fabs(x); + + if (x < 1.0f) + return 1 - x*x*(2.5f - 1.5f*x); + else if (x < 2.0f) + return 2 - x*(4 + x*(0.5f*x - 2.5f)); + + return (0.0f); +} + +static float stbir__filter_mitchell(float x, float s) +{ + STBIR__UNUSED_PARAM(s); + + x = (float)fabs(x); + + if (x < 1.0f) + return (16 + x*x*(21 * x - 36))/18; + else if (x < 2.0f) + return (32 + x*(-60 + x*(36 - 7*x)))/18; + + return (0.0f); +} + +static float stbir__support_zero(float s) +{ + STBIR__UNUSED_PARAM(s); + return 0; +} + +static float stbir__support_one(float s) +{ + STBIR__UNUSED_PARAM(s); + return 1; +} + +static float stbir__support_two(float s) +{ + STBIR__UNUSED_PARAM(s); + return 2; +} + +static stbir__filter_info stbir__filter_info_table[] = { + { NULL, stbir__support_zero }, + { stbir__filter_trapezoid, stbir__support_trapezoid }, + { stbir__filter_triangle, stbir__support_one }, + { stbir__filter_cubic, stbir__support_two }, + { stbir__filter_catmullrom, stbir__support_two }, + { stbir__filter_mitchell, stbir__support_two }, +}; + +stbir__inline static int stbir__use_upsampling(float ratio) +{ + return ratio > 1; +} + +stbir__inline static int stbir__use_width_upsampling(stbir__info* stbir_info) +{ + return stbir__use_upsampling(stbir_info->horizontal_scale); +} + +stbir__inline static int stbir__use_height_upsampling(stbir__info* stbir_info) +{ + return stbir__use_upsampling(stbir_info->vertical_scale); +} + +// This is the maximum number of input samples that can affect an output sample +// with the given filter +static int stbir__get_filter_pixel_width(stbir_filter filter, float scale) +{ + STBIR_ASSERT(filter != 0); + STBIR_ASSERT(filter < STBIR__ARRAY_SIZE(stbir__filter_info_table)); + + if (stbir__use_upsampling(scale)) + return (int)ceil(stbir__filter_info_table[filter].support(1/scale) * 2); + else + return (int)ceil(stbir__filter_info_table[filter].support(scale) * 2 / scale); +} + +// This is how much to expand buffers to account for filters seeking outside +// the image boundaries. +static int stbir__get_filter_pixel_margin(stbir_filter filter, float scale) +{ + return stbir__get_filter_pixel_width(filter, scale) / 2; +} + +static int stbir__get_coefficient_width(stbir_filter filter, float scale) +{ + if (stbir__use_upsampling(scale)) + return (int)ceil(stbir__filter_info_table[filter].support(1 / scale) * 2); + else + return (int)ceil(stbir__filter_info_table[filter].support(scale) * 2); +} + +static int stbir__get_contributors(float scale, stbir_filter filter, int input_size, int output_size) +{ + if (stbir__use_upsampling(scale)) + return output_size; + else + return (input_size + stbir__get_filter_pixel_margin(filter, scale) * 2); +} + +static int stbir__get_total_horizontal_coefficients(stbir__info* info) +{ + return info->horizontal_num_contributors + * stbir__get_coefficient_width (info->horizontal_filter, info->horizontal_scale); +} + +static int stbir__get_total_vertical_coefficients(stbir__info* info) +{ + return info->vertical_num_contributors + * stbir__get_coefficient_width (info->vertical_filter, info->vertical_scale); +} + +static stbir__contributors* stbir__get_contributor(stbir__contributors* contributors, int n) +{ + return &contributors[n]; +} + +// For perf reasons this code is duplicated in stbir__resample_horizontal_upsample/downsample, +// if you change it here change it there too. +static float* stbir__get_coefficient(float* coefficients, stbir_filter filter, float scale, int n, int c) +{ + int width = stbir__get_coefficient_width(filter, scale); + return &coefficients[width*n + c]; +} + +static int stbir__edge_wrap_slow(stbir_edge edge, int n, int max) +{ + switch (edge) + { + case STBIR_EDGE_ZERO: + return 0; // we'll decode the wrong pixel here, and then overwrite with 0s later + + case STBIR_EDGE_CLAMP: + if (n < 0) + return 0; + + if (n >= max) + return max - 1; + + return n; // NOTREACHED + + case STBIR_EDGE_REFLECT: + { + if (n < 0) + { + if (n < max) + return -n; + else + return max - 1; + } + + if (n >= max) + { + int max2 = max * 2; + if (n >= max2) + return 0; + else + return max2 - n - 1; + } + + return n; // NOTREACHED + } + + case STBIR_EDGE_WRAP: + if (n >= 0) + return (n % max); + else + { + int m = (-n) % max; + + if (m != 0) + m = max - m; + + return (m); + } + // NOTREACHED + + default: + STBIR_ASSERT(!"Unimplemented edge type"); + return 0; + } +} + +stbir__inline static int stbir__edge_wrap(stbir_edge edge, int n, int max) +{ + // avoid per-pixel switch + if (n >= 0 && n < max) + return n; + return stbir__edge_wrap_slow(edge, n, max); +} + +// What input pixels contribute to this output pixel? +static void stbir__calculate_sample_range_upsample(int n, float out_filter_radius, float scale_ratio, float out_shift, int* in_first_pixel, int* in_last_pixel, float* in_center_of_out) +{ + float out_pixel_center = (float)n + 0.5f; + float out_pixel_influence_lowerbound = out_pixel_center - out_filter_radius; + float out_pixel_influence_upperbound = out_pixel_center + out_filter_radius; + + float in_pixel_influence_lowerbound = (out_pixel_influence_lowerbound + out_shift) / scale_ratio; + float in_pixel_influence_upperbound = (out_pixel_influence_upperbound + out_shift) / scale_ratio; + + *in_center_of_out = (out_pixel_center + out_shift) / scale_ratio; + *in_first_pixel = (int)(floor(in_pixel_influence_lowerbound + 0.5)); + *in_last_pixel = (int)(floor(in_pixel_influence_upperbound - 0.5)); +} + +// What output pixels does this input pixel contribute to? +static void stbir__calculate_sample_range_downsample(int n, float in_pixels_radius, float scale_ratio, float out_shift, int* out_first_pixel, int* out_last_pixel, float* out_center_of_in) +{ + float in_pixel_center = (float)n + 0.5f; + float in_pixel_influence_lowerbound = in_pixel_center - in_pixels_radius; + float in_pixel_influence_upperbound = in_pixel_center + in_pixels_radius; + + float out_pixel_influence_lowerbound = in_pixel_influence_lowerbound * scale_ratio - out_shift; + float out_pixel_influence_upperbound = in_pixel_influence_upperbound * scale_ratio - out_shift; + + *out_center_of_in = in_pixel_center * scale_ratio - out_shift; + *out_first_pixel = (int)(floor(out_pixel_influence_lowerbound + 0.5)); + *out_last_pixel = (int)(floor(out_pixel_influence_upperbound - 0.5)); +} + +static void stbir__calculate_coefficients_upsample(stbir_filter filter, float scale, int in_first_pixel, int in_last_pixel, float in_center_of_out, stbir__contributors* contributor, float* coefficient_group) +{ + int i; + float total_filter = 0; + float filter_scale; + + STBIR_ASSERT(in_last_pixel - in_first_pixel <= (int)ceil(stbir__filter_info_table[filter].support(1/scale) * 2)); // Taken directly from stbir__get_coefficient_width() which we can't call because we don't know if we're horizontal or vertical. + + contributor->n0 = in_first_pixel; + contributor->n1 = in_last_pixel; + + STBIR_ASSERT(contributor->n1 >= contributor->n0); + + for (i = 0; i <= in_last_pixel - in_first_pixel; i++) + { + float in_pixel_center = (float)(i + in_first_pixel) + 0.5f; + coefficient_group[i] = stbir__filter_info_table[filter].kernel(in_center_of_out - in_pixel_center, 1 / scale); + + // If the coefficient is zero, skip it. (Don't do the <0 check here, we want the influence of those outside pixels.) + if (i == 0 && !coefficient_group[i]) + { + contributor->n0 = ++in_first_pixel; + i--; + continue; + } + + total_filter += coefficient_group[i]; + } + + STBIR_ASSERT(stbir__filter_info_table[filter].kernel((float)(in_last_pixel + 1) + 0.5f - in_center_of_out, 1/scale) == 0); + + STBIR_ASSERT(total_filter > 0.9); + STBIR_ASSERT(total_filter < 1.1f); // Make sure it's not way off. + + // Make sure the sum of all coefficients is 1. + filter_scale = 1 / total_filter; + + for (i = 0; i <= in_last_pixel - in_first_pixel; i++) + coefficient_group[i] *= filter_scale; + + for (i = in_last_pixel - in_first_pixel; i >= 0; i--) + { + if (coefficient_group[i]) + break; + + // This line has no weight. We can skip it. + contributor->n1 = contributor->n0 + i - 1; + } +} + +static void stbir__calculate_coefficients_downsample(stbir_filter filter, float scale_ratio, int out_first_pixel, int out_last_pixel, float out_center_of_in, stbir__contributors* contributor, float* coefficient_group) +{ + int i; + + STBIR_ASSERT(out_last_pixel - out_first_pixel <= (int)ceil(stbir__filter_info_table[filter].support(scale_ratio) * 2)); // Taken directly from stbir__get_coefficient_width() which we can't call because we don't know if we're horizontal or vertical. + + contributor->n0 = out_first_pixel; + contributor->n1 = out_last_pixel; + + STBIR_ASSERT(contributor->n1 >= contributor->n0); + + for (i = 0; i <= out_last_pixel - out_first_pixel; i++) + { + float out_pixel_center = (float)(i + out_first_pixel) + 0.5f; + float x = out_pixel_center - out_center_of_in; + coefficient_group[i] = stbir__filter_info_table[filter].kernel(x, scale_ratio) * scale_ratio; + } + + STBIR_ASSERT(stbir__filter_info_table[filter].kernel((float)(out_last_pixel + 1) + 0.5f - out_center_of_in, scale_ratio) == 0); + + for (i = out_last_pixel - out_first_pixel; i >= 0; i--) + { + if (coefficient_group[i]) + break; + + // This line has no weight. We can skip it. + contributor->n1 = contributor->n0 + i - 1; + } +} + +static void stbir__normalize_downsample_coefficients(stbir__contributors* contributors, float* coefficients, stbir_filter filter, float scale_ratio, int input_size, int output_size) +{ + int num_contributors = stbir__get_contributors(scale_ratio, filter, input_size, output_size); + int num_coefficients = stbir__get_coefficient_width(filter, scale_ratio); + int i, j; + int skip; + + for (i = 0; i < output_size; i++) + { + float scale; + float total = 0; + + for (j = 0; j < num_contributors; j++) + { + if (i >= contributors[j].n0 && i <= contributors[j].n1) + { + float coefficient = *stbir__get_coefficient(coefficients, filter, scale_ratio, j, i - contributors[j].n0); + total += coefficient; + } + else if (i < contributors[j].n0) + break; + } + + STBIR_ASSERT(total > 0.9f); + STBIR_ASSERT(total < 1.1f); + + scale = 1 / total; + + for (j = 0; j < num_contributors; j++) + { + if (i >= contributors[j].n0 && i <= contributors[j].n1) + *stbir__get_coefficient(coefficients, filter, scale_ratio, j, i - contributors[j].n0) *= scale; + else if (i < contributors[j].n0) + break; + } + } + + // Optimize: Skip zero coefficients and contributions outside of image bounds. + // Do this after normalizing because normalization depends on the n0/n1 values. + for (j = 0; j < num_contributors; j++) + { + int range, max, width; + + skip = 0; + while (*stbir__get_coefficient(coefficients, filter, scale_ratio, j, skip) == 0) + skip++; + + contributors[j].n0 += skip; + + while (contributors[j].n0 < 0) + { + contributors[j].n0++; + skip++; + } + + range = contributors[j].n1 - contributors[j].n0 + 1; + max = stbir__min(num_coefficients, range); + + width = stbir__get_coefficient_width(filter, scale_ratio); + for (i = 0; i < max; i++) + { + if (i + skip >= width) + break; + + *stbir__get_coefficient(coefficients, filter, scale_ratio, j, i) = *stbir__get_coefficient(coefficients, filter, scale_ratio, j, i + skip); + } + + continue; + } + + // Using min to avoid writing into invalid pixels. + for (i = 0; i < num_contributors; i++) + contributors[i].n1 = stbir__min(contributors[i].n1, output_size - 1); +} + +// Each scan line uses the same kernel values so we should calculate the kernel +// values once and then we can use them for every scan line. +static void stbir__calculate_filters(stbir__contributors* contributors, float* coefficients, stbir_filter filter, float scale_ratio, float shift, int input_size, int output_size) +{ + int n; + int total_contributors = stbir__get_contributors(scale_ratio, filter, input_size, output_size); + + if (stbir__use_upsampling(scale_ratio)) + { + float out_pixels_radius = stbir__filter_info_table[filter].support(1 / scale_ratio) * scale_ratio; + + // Looping through out pixels + for (n = 0; n < total_contributors; n++) + { + float in_center_of_out; // Center of the current out pixel in the in pixel space + int in_first_pixel, in_last_pixel; + + stbir__calculate_sample_range_upsample(n, out_pixels_radius, scale_ratio, shift, &in_first_pixel, &in_last_pixel, &in_center_of_out); + + stbir__calculate_coefficients_upsample(filter, scale_ratio, in_first_pixel, in_last_pixel, in_center_of_out, stbir__get_contributor(contributors, n), stbir__get_coefficient(coefficients, filter, scale_ratio, n, 0)); + } + } + else + { + float in_pixels_radius = stbir__filter_info_table[filter].support(scale_ratio) / scale_ratio; + + // Looping through in pixels + for (n = 0; n < total_contributors; n++) + { + float out_center_of_in; // Center of the current out pixel in the in pixel space + int out_first_pixel, out_last_pixel; + int n_adjusted = n - stbir__get_filter_pixel_margin(filter, scale_ratio); + + stbir__calculate_sample_range_downsample(n_adjusted, in_pixels_radius, scale_ratio, shift, &out_first_pixel, &out_last_pixel, &out_center_of_in); + + stbir__calculate_coefficients_downsample(filter, scale_ratio, out_first_pixel, out_last_pixel, out_center_of_in, stbir__get_contributor(contributors, n), stbir__get_coefficient(coefficients, filter, scale_ratio, n, 0)); + } + + stbir__normalize_downsample_coefficients(contributors, coefficients, filter, scale_ratio, input_size, output_size); + } +} + +static float* stbir__get_decode_buffer(stbir__info* stbir_info) +{ + // The 0 index of the decode buffer starts after the margin. This makes + // it okay to use negative indexes on the decode buffer. + return &stbir_info->decode_buffer[stbir_info->horizontal_filter_pixel_margin * stbir_info->channels]; +} + +#define STBIR__DECODE(type, colorspace) ((type) * (STBIR_MAX_COLORSPACES) + (colorspace)) + +static void stbir__decode_scanline(stbir__info* stbir_info, int n) +{ + int c; + int channels = stbir_info->channels; + int alpha_channel = stbir_info->alpha_channel; + int type = stbir_info->type; + int colorspace = stbir_info->colorspace; + int input_w = stbir_info->input_w; + size_t input_stride_bytes = stbir_info->input_stride_bytes; + float* decode_buffer = stbir__get_decode_buffer(stbir_info); + stbir_edge edge_horizontal = stbir_info->edge_horizontal; + stbir_edge edge_vertical = stbir_info->edge_vertical; + size_t in_buffer_row_offset = stbir__edge_wrap(edge_vertical, n, stbir_info->input_h) * input_stride_bytes; + const void* input_data = (char *) stbir_info->input_data + in_buffer_row_offset; + int max_x = input_w + stbir_info->horizontal_filter_pixel_margin; + int decode = STBIR__DECODE(type, colorspace); + + int x = -stbir_info->horizontal_filter_pixel_margin; + + // special handling for STBIR_EDGE_ZERO because it needs to return an item that doesn't appear in the input, + // and we want to avoid paying overhead on every pixel if not STBIR_EDGE_ZERO + if (edge_vertical == STBIR_EDGE_ZERO && (n < 0 || n >= stbir_info->input_h)) + { + for (; x < max_x; x++) + for (c = 0; c < channels; c++) + decode_buffer[x*channels + c] = 0; + return; + } + + switch (decode) + { + case STBIR__DECODE(STBIR_TYPE_UINT8, STBIR_COLORSPACE_LINEAR): + for (; x < max_x; x++) + { + int decode_pixel_index = x * channels; + int input_pixel_index = stbir__edge_wrap(edge_horizontal, x, input_w) * channels; + for (c = 0; c < channels; c++) + decode_buffer[decode_pixel_index + c] = ((float)((const unsigned char*)input_data)[input_pixel_index + c]) / stbir__max_uint8_as_float; + } + break; + + case STBIR__DECODE(STBIR_TYPE_UINT8, STBIR_COLORSPACE_SRGB): + for (; x < max_x; x++) + { + int decode_pixel_index = x * channels; + int input_pixel_index = stbir__edge_wrap(edge_horizontal, x, input_w) * channels; + for (c = 0; c < channels; c++) + decode_buffer[decode_pixel_index + c] = stbir__srgb_uchar_to_linear_float[((const unsigned char*)input_data)[input_pixel_index + c]]; + + if (!(stbir_info->flags&STBIR_FLAG_ALPHA_USES_COLORSPACE)) + decode_buffer[decode_pixel_index + alpha_channel] = ((float)((const unsigned char*)input_data)[input_pixel_index + alpha_channel]) / stbir__max_uint8_as_float; + } + break; + + case STBIR__DECODE(STBIR_TYPE_UINT16, STBIR_COLORSPACE_LINEAR): + for (; x < max_x; x++) + { + int decode_pixel_index = x * channels; + int input_pixel_index = stbir__edge_wrap(edge_horizontal, x, input_w) * channels; + for (c = 0; c < channels; c++) + decode_buffer[decode_pixel_index + c] = ((float)((const unsigned short*)input_data)[input_pixel_index + c]) / stbir__max_uint16_as_float; + } + break; + + case STBIR__DECODE(STBIR_TYPE_UINT16, STBIR_COLORSPACE_SRGB): + for (; x < max_x; x++) + { + int decode_pixel_index = x * channels; + int input_pixel_index = stbir__edge_wrap(edge_horizontal, x, input_w) * channels; + for (c = 0; c < channels; c++) + decode_buffer[decode_pixel_index + c] = stbir__srgb_to_linear(((float)((const unsigned short*)input_data)[input_pixel_index + c]) / stbir__max_uint16_as_float); + + if (!(stbir_info->flags&STBIR_FLAG_ALPHA_USES_COLORSPACE)) + decode_buffer[decode_pixel_index + alpha_channel] = ((float)((const unsigned short*)input_data)[input_pixel_index + alpha_channel]) / stbir__max_uint16_as_float; + } + break; + + case STBIR__DECODE(STBIR_TYPE_UINT32, STBIR_COLORSPACE_LINEAR): + for (; x < max_x; x++) + { + int decode_pixel_index = x * channels; + int input_pixel_index = stbir__edge_wrap(edge_horizontal, x, input_w) * channels; + for (c = 0; c < channels; c++) + decode_buffer[decode_pixel_index + c] = (float)(((double)((const unsigned int*)input_data)[input_pixel_index + c]) / stbir__max_uint32_as_float); + } + break; + + case STBIR__DECODE(STBIR_TYPE_UINT32, STBIR_COLORSPACE_SRGB): + for (; x < max_x; x++) + { + int decode_pixel_index = x * channels; + int input_pixel_index = stbir__edge_wrap(edge_horizontal, x, input_w) * channels; + for (c = 0; c < channels; c++) + decode_buffer[decode_pixel_index + c] = stbir__srgb_to_linear((float)(((double)((const unsigned int*)input_data)[input_pixel_index + c]) / stbir__max_uint32_as_float)); + + if (!(stbir_info->flags&STBIR_FLAG_ALPHA_USES_COLORSPACE)) + decode_buffer[decode_pixel_index + alpha_channel] = (float)(((double)((const unsigned int*)input_data)[input_pixel_index + alpha_channel]) / stbir__max_uint32_as_float); + } + break; + + case STBIR__DECODE(STBIR_TYPE_FLOAT, STBIR_COLORSPACE_LINEAR): + for (; x < max_x; x++) + { + int decode_pixel_index = x * channels; + int input_pixel_index = stbir__edge_wrap(edge_horizontal, x, input_w) * channels; + for (c = 0; c < channels; c++) + decode_buffer[decode_pixel_index + c] = ((const float*)input_data)[input_pixel_index + c]; + } + break; + + case STBIR__DECODE(STBIR_TYPE_FLOAT, STBIR_COLORSPACE_SRGB): + for (; x < max_x; x++) + { + int decode_pixel_index = x * channels; + int input_pixel_index = stbir__edge_wrap(edge_horizontal, x, input_w) * channels; + for (c = 0; c < channels; c++) + decode_buffer[decode_pixel_index + c] = stbir__srgb_to_linear(((const float*)input_data)[input_pixel_index + c]); + + if (!(stbir_info->flags&STBIR_FLAG_ALPHA_USES_COLORSPACE)) + decode_buffer[decode_pixel_index + alpha_channel] = ((const float*)input_data)[input_pixel_index + alpha_channel]; + } + + break; + + default: + STBIR_ASSERT(!"Unknown type/colorspace/channels combination."); + break; + } + + if (!(stbir_info->flags & STBIR_FLAG_ALPHA_PREMULTIPLIED)) + { + for (x = -stbir_info->horizontal_filter_pixel_margin; x < max_x; x++) + { + int decode_pixel_index = x * channels; + + // If the alpha value is 0 it will clobber the color values. Make sure it's not. + float alpha = decode_buffer[decode_pixel_index + alpha_channel]; +#ifndef STBIR_NO_ALPHA_EPSILON + if (stbir_info->type != STBIR_TYPE_FLOAT) { + alpha += STBIR_ALPHA_EPSILON; + decode_buffer[decode_pixel_index + alpha_channel] = alpha; + } +#endif + for (c = 0; c < channels; c++) + { + if (c == alpha_channel) + continue; + + decode_buffer[decode_pixel_index + c] *= alpha; + } + } + } + + if (edge_horizontal == STBIR_EDGE_ZERO) + { + for (x = -stbir_info->horizontal_filter_pixel_margin; x < 0; x++) + { + for (c = 0; c < channels; c++) + decode_buffer[x*channels + c] = 0; + } + for (x = input_w; x < max_x; x++) + { + for (c = 0; c < channels; c++) + decode_buffer[x*channels + c] = 0; + } + } +} + +static float* stbir__get_ring_buffer_entry(float* ring_buffer, int index, int ring_buffer_length) +{ + return &ring_buffer[index * ring_buffer_length]; +} + +static float* stbir__add_empty_ring_buffer_entry(stbir__info* stbir_info, int n) +{ + int ring_buffer_index; + float* ring_buffer; + + stbir_info->ring_buffer_last_scanline = n; + + if (stbir_info->ring_buffer_begin_index < 0) + { + ring_buffer_index = stbir_info->ring_buffer_begin_index = 0; + stbir_info->ring_buffer_first_scanline = n; + } + else + { + ring_buffer_index = (stbir_info->ring_buffer_begin_index + (stbir_info->ring_buffer_last_scanline - stbir_info->ring_buffer_first_scanline)) % stbir_info->ring_buffer_num_entries; + STBIR_ASSERT(ring_buffer_index != stbir_info->ring_buffer_begin_index); + } + + ring_buffer = stbir__get_ring_buffer_entry(stbir_info->ring_buffer, ring_buffer_index, stbir_info->ring_buffer_length_bytes / sizeof(float)); + memset(ring_buffer, 0, stbir_info->ring_buffer_length_bytes); + + return ring_buffer; +} + + +static void stbir__resample_horizontal_upsample(stbir__info* stbir_info, float* output_buffer) +{ + int x, k; + int output_w = stbir_info->output_w; + int channels = stbir_info->channels; + float* decode_buffer = stbir__get_decode_buffer(stbir_info); + stbir__contributors* horizontal_contributors = stbir_info->horizontal_contributors; + float* horizontal_coefficients = stbir_info->horizontal_coefficients; + int coefficient_width = stbir_info->horizontal_coefficient_width; + + for (x = 0; x < output_w; x++) + { + int n0 = horizontal_contributors[x].n0; + int n1 = horizontal_contributors[x].n1; + + int out_pixel_index = x * channels; + int coefficient_group = coefficient_width * x; + int coefficient_counter = 0; + + STBIR_ASSERT(n1 >= n0); + STBIR_ASSERT(n0 >= -stbir_info->horizontal_filter_pixel_margin); + STBIR_ASSERT(n1 >= -stbir_info->horizontal_filter_pixel_margin); + STBIR_ASSERT(n0 < stbir_info->input_w + stbir_info->horizontal_filter_pixel_margin); + STBIR_ASSERT(n1 < stbir_info->input_w + stbir_info->horizontal_filter_pixel_margin); + + switch (channels) { + case 1: + for (k = n0; k <= n1; k++) + { + int in_pixel_index = k * 1; + float coefficient = horizontal_coefficients[coefficient_group + coefficient_counter++]; + STBIR_ASSERT(coefficient != 0); + output_buffer[out_pixel_index + 0] += decode_buffer[in_pixel_index + 0] * coefficient; + } + break; + case 2: + for (k = n0; k <= n1; k++) + { + int in_pixel_index = k * 2; + float coefficient = horizontal_coefficients[coefficient_group + coefficient_counter++]; + STBIR_ASSERT(coefficient != 0); + output_buffer[out_pixel_index + 0] += decode_buffer[in_pixel_index + 0] * coefficient; + output_buffer[out_pixel_index + 1] += decode_buffer[in_pixel_index + 1] * coefficient; + } + break; + case 3: + for (k = n0; k <= n1; k++) + { + int in_pixel_index = k * 3; + float coefficient = horizontal_coefficients[coefficient_group + coefficient_counter++]; + STBIR_ASSERT(coefficient != 0); + output_buffer[out_pixel_index + 0] += decode_buffer[in_pixel_index + 0] * coefficient; + output_buffer[out_pixel_index + 1] += decode_buffer[in_pixel_index + 1] * coefficient; + output_buffer[out_pixel_index + 2] += decode_buffer[in_pixel_index + 2] * coefficient; + } + break; + case 4: + for (k = n0; k <= n1; k++) + { + int in_pixel_index = k * 4; + float coefficient = horizontal_coefficients[coefficient_group + coefficient_counter++]; + STBIR_ASSERT(coefficient != 0); + output_buffer[out_pixel_index + 0] += decode_buffer[in_pixel_index + 0] * coefficient; + output_buffer[out_pixel_index + 1] += decode_buffer[in_pixel_index + 1] * coefficient; + output_buffer[out_pixel_index + 2] += decode_buffer[in_pixel_index + 2] * coefficient; + output_buffer[out_pixel_index + 3] += decode_buffer[in_pixel_index + 3] * coefficient; + } + break; + default: + for (k = n0; k <= n1; k++) + { + int in_pixel_index = k * channels; + float coefficient = horizontal_coefficients[coefficient_group + coefficient_counter++]; + int c; + STBIR_ASSERT(coefficient != 0); + for (c = 0; c < channels; c++) + output_buffer[out_pixel_index + c] += decode_buffer[in_pixel_index + c] * coefficient; + } + break; + } + } +} + +static void stbir__resample_horizontal_downsample(stbir__info* stbir_info, float* output_buffer) +{ + int x, k; + int input_w = stbir_info->input_w; + int channels = stbir_info->channels; + float* decode_buffer = stbir__get_decode_buffer(stbir_info); + stbir__contributors* horizontal_contributors = stbir_info->horizontal_contributors; + float* horizontal_coefficients = stbir_info->horizontal_coefficients; + int coefficient_width = stbir_info->horizontal_coefficient_width; + int filter_pixel_margin = stbir_info->horizontal_filter_pixel_margin; + int max_x = input_w + filter_pixel_margin * 2; + + STBIR_ASSERT(!stbir__use_width_upsampling(stbir_info)); + + switch (channels) { + case 1: + for (x = 0; x < max_x; x++) + { + int n0 = horizontal_contributors[x].n0; + int n1 = horizontal_contributors[x].n1; + + int in_x = x - filter_pixel_margin; + int in_pixel_index = in_x * 1; + int max_n = n1; + int coefficient_group = coefficient_width * x; + + for (k = n0; k <= max_n; k++) + { + int out_pixel_index = k * 1; + float coefficient = horizontal_coefficients[coefficient_group + k - n0]; + STBIR_ASSERT(coefficient != 0); + output_buffer[out_pixel_index + 0] += decode_buffer[in_pixel_index + 0] * coefficient; + } + } + break; + + case 2: + for (x = 0; x < max_x; x++) + { + int n0 = horizontal_contributors[x].n0; + int n1 = horizontal_contributors[x].n1; + + int in_x = x - filter_pixel_margin; + int in_pixel_index = in_x * 2; + int max_n = n1; + int coefficient_group = coefficient_width * x; + + for (k = n0; k <= max_n; k++) + { + int out_pixel_index = k * 2; + float coefficient = horizontal_coefficients[coefficient_group + k - n0]; + STBIR_ASSERT(coefficient != 0); + output_buffer[out_pixel_index + 0] += decode_buffer[in_pixel_index + 0] * coefficient; + output_buffer[out_pixel_index + 1] += decode_buffer[in_pixel_index + 1] * coefficient; + } + } + break; + + case 3: + for (x = 0; x < max_x; x++) + { + int n0 = horizontal_contributors[x].n0; + int n1 = horizontal_contributors[x].n1; + + int in_x = x - filter_pixel_margin; + int in_pixel_index = in_x * 3; + int max_n = n1; + int coefficient_group = coefficient_width * x; + + for (k = n0; k <= max_n; k++) + { + int out_pixel_index = k * 3; + float coefficient = horizontal_coefficients[coefficient_group + k - n0]; + STBIR_ASSERT(coefficient != 0); + output_buffer[out_pixel_index + 0] += decode_buffer[in_pixel_index + 0] * coefficient; + output_buffer[out_pixel_index + 1] += decode_buffer[in_pixel_index + 1] * coefficient; + output_buffer[out_pixel_index + 2] += decode_buffer[in_pixel_index + 2] * coefficient; + } + } + break; + + case 4: + for (x = 0; x < max_x; x++) + { + int n0 = horizontal_contributors[x].n0; + int n1 = horizontal_contributors[x].n1; + + int in_x = x - filter_pixel_margin; + int in_pixel_index = in_x * 4; + int max_n = n1; + int coefficient_group = coefficient_width * x; + + for (k = n0; k <= max_n; k++) + { + int out_pixel_index = k * 4; + float coefficient = horizontal_coefficients[coefficient_group + k - n0]; + STBIR_ASSERT(coefficient != 0); + output_buffer[out_pixel_index + 0] += decode_buffer[in_pixel_index + 0] * coefficient; + output_buffer[out_pixel_index + 1] += decode_buffer[in_pixel_index + 1] * coefficient; + output_buffer[out_pixel_index + 2] += decode_buffer[in_pixel_index + 2] * coefficient; + output_buffer[out_pixel_index + 3] += decode_buffer[in_pixel_index + 3] * coefficient; + } + } + break; + + default: + for (x = 0; x < max_x; x++) + { + int n0 = horizontal_contributors[x].n0; + int n1 = horizontal_contributors[x].n1; + + int in_x = x - filter_pixel_margin; + int in_pixel_index = in_x * channels; + int max_n = n1; + int coefficient_group = coefficient_width * x; + + for (k = n0; k <= max_n; k++) + { + int c; + int out_pixel_index = k * channels; + float coefficient = horizontal_coefficients[coefficient_group + k - n0]; + STBIR_ASSERT(coefficient != 0); + for (c = 0; c < channels; c++) + output_buffer[out_pixel_index + c] += decode_buffer[in_pixel_index + c] * coefficient; + } + } + break; + } +} + +static void stbir__decode_and_resample_upsample(stbir__info* stbir_info, int n) +{ + // Decode the nth scanline from the source image into the decode buffer. + stbir__decode_scanline(stbir_info, n); + + // Now resample it into the ring buffer. + if (stbir__use_width_upsampling(stbir_info)) + stbir__resample_horizontal_upsample(stbir_info, stbir__add_empty_ring_buffer_entry(stbir_info, n)); + else + stbir__resample_horizontal_downsample(stbir_info, stbir__add_empty_ring_buffer_entry(stbir_info, n)); + + // Now it's sitting in the ring buffer ready to be used as source for the vertical sampling. +} + +static void stbir__decode_and_resample_downsample(stbir__info* stbir_info, int n) +{ + // Decode the nth scanline from the source image into the decode buffer. + stbir__decode_scanline(stbir_info, n); + + memset(stbir_info->horizontal_buffer, 0, stbir_info->output_w * stbir_info->channels * sizeof(float)); + + // Now resample it into the horizontal buffer. + if (stbir__use_width_upsampling(stbir_info)) + stbir__resample_horizontal_upsample(stbir_info, stbir_info->horizontal_buffer); + else + stbir__resample_horizontal_downsample(stbir_info, stbir_info->horizontal_buffer); + + // Now it's sitting in the horizontal buffer ready to be distributed into the ring buffers. +} + +// Get the specified scan line from the ring buffer. +static float* stbir__get_ring_buffer_scanline(int get_scanline, float* ring_buffer, int begin_index, int first_scanline, int ring_buffer_num_entries, int ring_buffer_length) +{ + int ring_buffer_index = (begin_index + (get_scanline - first_scanline)) % ring_buffer_num_entries; + return stbir__get_ring_buffer_entry(ring_buffer, ring_buffer_index, ring_buffer_length); +} + + +static void stbir__encode_scanline(stbir__info* stbir_info, int num_pixels, void *output_buffer, float *encode_buffer, int channels, int alpha_channel, int decode) +{ + int x; + int n; + int num_nonalpha; + stbir_uint16 nonalpha[STBIR_MAX_CHANNELS]; + + if (!(stbir_info->flags&STBIR_FLAG_ALPHA_PREMULTIPLIED)) + { + for (x=0; x < num_pixels; ++x) + { + int pixel_index = x*channels; + + float alpha = encode_buffer[pixel_index + alpha_channel]; + float reciprocal_alpha = alpha ? 1.0f / alpha : 0; + + // unrolling this produced a 1% slowdown upscaling a large RGBA linear-space image on my machine - stb + for (n = 0; n < channels; n++) + if (n != alpha_channel) + encode_buffer[pixel_index + n] *= reciprocal_alpha; + + // We added in a small epsilon to prevent the color channel from being deleted with zero alpha. + // Because we only add it for integer types, it will automatically be discarded on integer + // conversion, so we don't need to subtract it back out (which would be problematic for + // numeric precision reasons). + } + } + + // build a table of all channels that need colorspace correction, so + // we don't perform colorspace correction on channels that don't need it. + for (x = 0, num_nonalpha = 0; x < channels; ++x) + { + if (x != alpha_channel || (stbir_info->flags & STBIR_FLAG_ALPHA_USES_COLORSPACE)) + { + nonalpha[num_nonalpha++] = (stbir_uint16)x; + } + } + + #define STBIR__ROUND_INT(f) ((int) ((f)+0.5)) + #define STBIR__ROUND_UINT(f) ((stbir_uint32) ((f)+0.5)) + + #ifdef STBIR__SATURATE_INT + #define STBIR__ENCODE_LINEAR8(f) stbir__saturate8 (STBIR__ROUND_INT((f) * stbir__max_uint8_as_float )) + #define STBIR__ENCODE_LINEAR16(f) stbir__saturate16(STBIR__ROUND_INT((f) * stbir__max_uint16_as_float)) + #else + #define STBIR__ENCODE_LINEAR8(f) (unsigned char ) STBIR__ROUND_INT(stbir__saturate(f) * stbir__max_uint8_as_float ) + #define STBIR__ENCODE_LINEAR16(f) (unsigned short) STBIR__ROUND_INT(stbir__saturate(f) * stbir__max_uint16_as_float) + #endif + + switch (decode) + { + case STBIR__DECODE(STBIR_TYPE_UINT8, STBIR_COLORSPACE_LINEAR): + for (x=0; x < num_pixels; ++x) + { + int pixel_index = x*channels; + + for (n = 0; n < channels; n++) + { + int index = pixel_index + n; + ((unsigned char*)output_buffer)[index] = STBIR__ENCODE_LINEAR8(encode_buffer[index]); + } + } + break; + + case STBIR__DECODE(STBIR_TYPE_UINT8, STBIR_COLORSPACE_SRGB): + for (x=0; x < num_pixels; ++x) + { + int pixel_index = x*channels; + + for (n = 0; n < num_nonalpha; n++) + { + int index = pixel_index + nonalpha[n]; + ((unsigned char*)output_buffer)[index] = stbir__linear_to_srgb_uchar(encode_buffer[index]); + } + + if (!(stbir_info->flags & STBIR_FLAG_ALPHA_USES_COLORSPACE)) + ((unsigned char *)output_buffer)[pixel_index + alpha_channel] = STBIR__ENCODE_LINEAR8(encode_buffer[pixel_index+alpha_channel]); + } + break; + + case STBIR__DECODE(STBIR_TYPE_UINT16, STBIR_COLORSPACE_LINEAR): + for (x=0; x < num_pixels; ++x) + { + int pixel_index = x*channels; + + for (n = 0; n < channels; n++) + { + int index = pixel_index + n; + ((unsigned short*)output_buffer)[index] = STBIR__ENCODE_LINEAR16(encode_buffer[index]); + } + } + break; + + case STBIR__DECODE(STBIR_TYPE_UINT16, STBIR_COLORSPACE_SRGB): + for (x=0; x < num_pixels; ++x) + { + int pixel_index = x*channels; + + for (n = 0; n < num_nonalpha; n++) + { + int index = pixel_index + nonalpha[n]; + ((unsigned short*)output_buffer)[index] = (unsigned short)STBIR__ROUND_INT(stbir__linear_to_srgb(stbir__saturate(encode_buffer[index])) * stbir__max_uint16_as_float); + } + + if (!(stbir_info->flags&STBIR_FLAG_ALPHA_USES_COLORSPACE)) + ((unsigned short*)output_buffer)[pixel_index + alpha_channel] = STBIR__ENCODE_LINEAR16(encode_buffer[pixel_index + alpha_channel]); + } + + break; + + case STBIR__DECODE(STBIR_TYPE_UINT32, STBIR_COLORSPACE_LINEAR): + for (x=0; x < num_pixels; ++x) + { + int pixel_index = x*channels; + + for (n = 0; n < channels; n++) + { + int index = pixel_index + n; + ((unsigned int*)output_buffer)[index] = (unsigned int)STBIR__ROUND_UINT(((double)stbir__saturate(encode_buffer[index])) * stbir__max_uint32_as_float); + } + } + break; + + case STBIR__DECODE(STBIR_TYPE_UINT32, STBIR_COLORSPACE_SRGB): + for (x=0; x < num_pixels; ++x) + { + int pixel_index = x*channels; + + for (n = 0; n < num_nonalpha; n++) + { + int index = pixel_index + nonalpha[n]; + ((unsigned int*)output_buffer)[index] = (unsigned int)STBIR__ROUND_UINT(((double)stbir__linear_to_srgb(stbir__saturate(encode_buffer[index]))) * stbir__max_uint32_as_float); + } + + if (!(stbir_info->flags&STBIR_FLAG_ALPHA_USES_COLORSPACE)) + ((unsigned int*)output_buffer)[pixel_index + alpha_channel] = (unsigned int)STBIR__ROUND_INT(((double)stbir__saturate(encode_buffer[pixel_index + alpha_channel])) * stbir__max_uint32_as_float); + } + break; + + case STBIR__DECODE(STBIR_TYPE_FLOAT, STBIR_COLORSPACE_LINEAR): + for (x=0; x < num_pixels; ++x) + { + int pixel_index = x*channels; + + for (n = 0; n < channels; n++) + { + int index = pixel_index + n; + ((float*)output_buffer)[index] = encode_buffer[index]; + } + } + break; + + case STBIR__DECODE(STBIR_TYPE_FLOAT, STBIR_COLORSPACE_SRGB): + for (x=0; x < num_pixels; ++x) + { + int pixel_index = x*channels; + + for (n = 0; n < num_nonalpha; n++) + { + int index = pixel_index + nonalpha[n]; + ((float*)output_buffer)[index] = stbir__linear_to_srgb(encode_buffer[index]); + } + + if (!(stbir_info->flags&STBIR_FLAG_ALPHA_USES_COLORSPACE)) + ((float*)output_buffer)[pixel_index + alpha_channel] = encode_buffer[pixel_index + alpha_channel]; + } + break; + + default: + STBIR_ASSERT(!"Unknown type/colorspace/channels combination."); + break; + } +} + +static void stbir__resample_vertical_upsample(stbir__info* stbir_info, int n) +{ + int x, k; + int output_w = stbir_info->output_w; + stbir__contributors* vertical_contributors = stbir_info->vertical_contributors; + float* vertical_coefficients = stbir_info->vertical_coefficients; + int channels = stbir_info->channels; + int alpha_channel = stbir_info->alpha_channel; + int type = stbir_info->type; + int colorspace = stbir_info->colorspace; + int ring_buffer_entries = stbir_info->ring_buffer_num_entries; + void* output_data = stbir_info->output_data; + float* encode_buffer = stbir_info->encode_buffer; + int decode = STBIR__DECODE(type, colorspace); + int coefficient_width = stbir_info->vertical_coefficient_width; + int coefficient_counter; + int contributor = n; + + float* ring_buffer = stbir_info->ring_buffer; + int ring_buffer_begin_index = stbir_info->ring_buffer_begin_index; + int ring_buffer_first_scanline = stbir_info->ring_buffer_first_scanline; + int ring_buffer_length = stbir_info->ring_buffer_length_bytes/sizeof(float); + + int n0,n1, output_row_start; + int coefficient_group = coefficient_width * contributor; + + n0 = vertical_contributors[contributor].n0; + n1 = vertical_contributors[contributor].n1; + + output_row_start = n * stbir_info->output_stride_bytes; + + STBIR_ASSERT(stbir__use_height_upsampling(stbir_info)); + + memset(encode_buffer, 0, output_w * sizeof(float) * channels); + + // I tried reblocking this for better cache usage of encode_buffer + // (using x_outer, k, x_inner), but it lost speed. -- stb + + coefficient_counter = 0; + switch (channels) { + case 1: + for (k = n0; k <= n1; k++) + { + int coefficient_index = coefficient_counter++; + float* ring_buffer_entry = stbir__get_ring_buffer_scanline(k, ring_buffer, ring_buffer_begin_index, ring_buffer_first_scanline, ring_buffer_entries, ring_buffer_length); + float coefficient = vertical_coefficients[coefficient_group + coefficient_index]; + for (x = 0; x < output_w; ++x) + { + int in_pixel_index = x * 1; + encode_buffer[in_pixel_index + 0] += ring_buffer_entry[in_pixel_index + 0] * coefficient; + } + } + break; + case 2: + for (k = n0; k <= n1; k++) + { + int coefficient_index = coefficient_counter++; + float* ring_buffer_entry = stbir__get_ring_buffer_scanline(k, ring_buffer, ring_buffer_begin_index, ring_buffer_first_scanline, ring_buffer_entries, ring_buffer_length); + float coefficient = vertical_coefficients[coefficient_group + coefficient_index]; + for (x = 0; x < output_w; ++x) + { + int in_pixel_index = x * 2; + encode_buffer[in_pixel_index + 0] += ring_buffer_entry[in_pixel_index + 0] * coefficient; + encode_buffer[in_pixel_index + 1] += ring_buffer_entry[in_pixel_index + 1] * coefficient; + } + } + break; + case 3: + for (k = n0; k <= n1; k++) + { + int coefficient_index = coefficient_counter++; + float* ring_buffer_entry = stbir__get_ring_buffer_scanline(k, ring_buffer, ring_buffer_begin_index, ring_buffer_first_scanline, ring_buffer_entries, ring_buffer_length); + float coefficient = vertical_coefficients[coefficient_group + coefficient_index]; + for (x = 0; x < output_w; ++x) + { + int in_pixel_index = x * 3; + encode_buffer[in_pixel_index + 0] += ring_buffer_entry[in_pixel_index + 0] * coefficient; + encode_buffer[in_pixel_index + 1] += ring_buffer_entry[in_pixel_index + 1] * coefficient; + encode_buffer[in_pixel_index + 2] += ring_buffer_entry[in_pixel_index + 2] * coefficient; + } + } + break; + case 4: + for (k = n0; k <= n1; k++) + { + int coefficient_index = coefficient_counter++; + float* ring_buffer_entry = stbir__get_ring_buffer_scanline(k, ring_buffer, ring_buffer_begin_index, ring_buffer_first_scanline, ring_buffer_entries, ring_buffer_length); + float coefficient = vertical_coefficients[coefficient_group + coefficient_index]; + for (x = 0; x < output_w; ++x) + { + int in_pixel_index = x * 4; + encode_buffer[in_pixel_index + 0] += ring_buffer_entry[in_pixel_index + 0] * coefficient; + encode_buffer[in_pixel_index + 1] += ring_buffer_entry[in_pixel_index + 1] * coefficient; + encode_buffer[in_pixel_index + 2] += ring_buffer_entry[in_pixel_index + 2] * coefficient; + encode_buffer[in_pixel_index + 3] += ring_buffer_entry[in_pixel_index + 3] * coefficient; + } + } + break; + default: + for (k = n0; k <= n1; k++) + { + int coefficient_index = coefficient_counter++; + float* ring_buffer_entry = stbir__get_ring_buffer_scanline(k, ring_buffer, ring_buffer_begin_index, ring_buffer_first_scanline, ring_buffer_entries, ring_buffer_length); + float coefficient = vertical_coefficients[coefficient_group + coefficient_index]; + for (x = 0; x < output_w; ++x) + { + int in_pixel_index = x * channels; + int c; + for (c = 0; c < channels; c++) + encode_buffer[in_pixel_index + c] += ring_buffer_entry[in_pixel_index + c] * coefficient; + } + } + break; + } + stbir__encode_scanline(stbir_info, output_w, (char *) output_data + output_row_start, encode_buffer, channels, alpha_channel, decode); +} + +static void stbir__resample_vertical_downsample(stbir__info* stbir_info, int n) +{ + int x, k; + int output_w = stbir_info->output_w; + stbir__contributors* vertical_contributors = stbir_info->vertical_contributors; + float* vertical_coefficients = stbir_info->vertical_coefficients; + int channels = stbir_info->channels; + int ring_buffer_entries = stbir_info->ring_buffer_num_entries; + float* horizontal_buffer = stbir_info->horizontal_buffer; + int coefficient_width = stbir_info->vertical_coefficient_width; + int contributor = n + stbir_info->vertical_filter_pixel_margin; + + float* ring_buffer = stbir_info->ring_buffer; + int ring_buffer_begin_index = stbir_info->ring_buffer_begin_index; + int ring_buffer_first_scanline = stbir_info->ring_buffer_first_scanline; + int ring_buffer_length = stbir_info->ring_buffer_length_bytes/sizeof(float); + int n0,n1; + + n0 = vertical_contributors[contributor].n0; + n1 = vertical_contributors[contributor].n1; + + STBIR_ASSERT(!stbir__use_height_upsampling(stbir_info)); + + for (k = n0; k <= n1; k++) + { + int coefficient_index = k - n0; + int coefficient_group = coefficient_width * contributor; + float coefficient = vertical_coefficients[coefficient_group + coefficient_index]; + + float* ring_buffer_entry = stbir__get_ring_buffer_scanline(k, ring_buffer, ring_buffer_begin_index, ring_buffer_first_scanline, ring_buffer_entries, ring_buffer_length); + + switch (channels) { + case 1: + for (x = 0; x < output_w; x++) + { + int in_pixel_index = x * 1; + ring_buffer_entry[in_pixel_index + 0] += horizontal_buffer[in_pixel_index + 0] * coefficient; + } + break; + case 2: + for (x = 0; x < output_w; x++) + { + int in_pixel_index = x * 2; + ring_buffer_entry[in_pixel_index + 0] += horizontal_buffer[in_pixel_index + 0] * coefficient; + ring_buffer_entry[in_pixel_index + 1] += horizontal_buffer[in_pixel_index + 1] * coefficient; + } + break; + case 3: + for (x = 0; x < output_w; x++) + { + int in_pixel_index = x * 3; + ring_buffer_entry[in_pixel_index + 0] += horizontal_buffer[in_pixel_index + 0] * coefficient; + ring_buffer_entry[in_pixel_index + 1] += horizontal_buffer[in_pixel_index + 1] * coefficient; + ring_buffer_entry[in_pixel_index + 2] += horizontal_buffer[in_pixel_index + 2] * coefficient; + } + break; + case 4: + for (x = 0; x < output_w; x++) + { + int in_pixel_index = x * 4; + ring_buffer_entry[in_pixel_index + 0] += horizontal_buffer[in_pixel_index + 0] * coefficient; + ring_buffer_entry[in_pixel_index + 1] += horizontal_buffer[in_pixel_index + 1] * coefficient; + ring_buffer_entry[in_pixel_index + 2] += horizontal_buffer[in_pixel_index + 2] * coefficient; + ring_buffer_entry[in_pixel_index + 3] += horizontal_buffer[in_pixel_index + 3] * coefficient; + } + break; + default: + for (x = 0; x < output_w; x++) + { + int in_pixel_index = x * channels; + + int c; + for (c = 0; c < channels; c++) + ring_buffer_entry[in_pixel_index + c] += horizontal_buffer[in_pixel_index + c] * coefficient; + } + break; + } + } +} + +static void stbir__buffer_loop_upsample(stbir__info* stbir_info) +{ + int y; + float scale_ratio = stbir_info->vertical_scale; + float out_scanlines_radius = stbir__filter_info_table[stbir_info->vertical_filter].support(1/scale_ratio) * scale_ratio; + + STBIR_ASSERT(stbir__use_height_upsampling(stbir_info)); + + for (y = 0; y < stbir_info->output_h; y++) + { + float in_center_of_out = 0; // Center of the current out scanline in the in scanline space + int in_first_scanline = 0, in_last_scanline = 0; + + stbir__calculate_sample_range_upsample(y, out_scanlines_radius, scale_ratio, stbir_info->vertical_shift, &in_first_scanline, &in_last_scanline, &in_center_of_out); + + STBIR_ASSERT(in_last_scanline - in_first_scanline + 1 <= stbir_info->ring_buffer_num_entries); + + if (stbir_info->ring_buffer_begin_index >= 0) + { + // Get rid of whatever we don't need anymore. + while (in_first_scanline > stbir_info->ring_buffer_first_scanline) + { + if (stbir_info->ring_buffer_first_scanline == stbir_info->ring_buffer_last_scanline) + { + // We just popped the last scanline off the ring buffer. + // Reset it to the empty state. + stbir_info->ring_buffer_begin_index = -1; + stbir_info->ring_buffer_first_scanline = 0; + stbir_info->ring_buffer_last_scanline = 0; + break; + } + else + { + stbir_info->ring_buffer_first_scanline++; + stbir_info->ring_buffer_begin_index = (stbir_info->ring_buffer_begin_index + 1) % stbir_info->ring_buffer_num_entries; + } + } + } + + // Load in new ones. + if (stbir_info->ring_buffer_begin_index < 0) + stbir__decode_and_resample_upsample(stbir_info, in_first_scanline); + + while (in_last_scanline > stbir_info->ring_buffer_last_scanline) + stbir__decode_and_resample_upsample(stbir_info, stbir_info->ring_buffer_last_scanline + 1); + + // Now all buffers should be ready to write a row of vertical sampling. + stbir__resample_vertical_upsample(stbir_info, y); + + STBIR_PROGRESS_REPORT((float)y / stbir_info->output_h); + } +} + +static void stbir__empty_ring_buffer(stbir__info* stbir_info, int first_necessary_scanline) +{ + int output_stride_bytes = stbir_info->output_stride_bytes; + int channels = stbir_info->channels; + int alpha_channel = stbir_info->alpha_channel; + int type = stbir_info->type; + int colorspace = stbir_info->colorspace; + int output_w = stbir_info->output_w; + void* output_data = stbir_info->output_data; + int decode = STBIR__DECODE(type, colorspace); + + float* ring_buffer = stbir_info->ring_buffer; + int ring_buffer_length = stbir_info->ring_buffer_length_bytes/sizeof(float); + + if (stbir_info->ring_buffer_begin_index >= 0) + { + // Get rid of whatever we don't need anymore. + while (first_necessary_scanline > stbir_info->ring_buffer_first_scanline) + { + if (stbir_info->ring_buffer_first_scanline >= 0 && stbir_info->ring_buffer_first_scanline < stbir_info->output_h) + { + int output_row_start = stbir_info->ring_buffer_first_scanline * output_stride_bytes; + float* ring_buffer_entry = stbir__get_ring_buffer_entry(ring_buffer, stbir_info->ring_buffer_begin_index, ring_buffer_length); + stbir__encode_scanline(stbir_info, output_w, (char *) output_data + output_row_start, ring_buffer_entry, channels, alpha_channel, decode); + STBIR_PROGRESS_REPORT((float)stbir_info->ring_buffer_first_scanline / stbir_info->output_h); + } + + if (stbir_info->ring_buffer_first_scanline == stbir_info->ring_buffer_last_scanline) + { + // We just popped the last scanline off the ring buffer. + // Reset it to the empty state. + stbir_info->ring_buffer_begin_index = -1; + stbir_info->ring_buffer_first_scanline = 0; + stbir_info->ring_buffer_last_scanline = 0; + break; + } + else + { + stbir_info->ring_buffer_first_scanline++; + stbir_info->ring_buffer_begin_index = (stbir_info->ring_buffer_begin_index + 1) % stbir_info->ring_buffer_num_entries; + } + } + } +} + +static void stbir__buffer_loop_downsample(stbir__info* stbir_info) +{ + int y; + float scale_ratio = stbir_info->vertical_scale; + int output_h = stbir_info->output_h; + float in_pixels_radius = stbir__filter_info_table[stbir_info->vertical_filter].support(scale_ratio) / scale_ratio; + int pixel_margin = stbir_info->vertical_filter_pixel_margin; + int max_y = stbir_info->input_h + pixel_margin; + + STBIR_ASSERT(!stbir__use_height_upsampling(stbir_info)); + + for (y = -pixel_margin; y < max_y; y++) + { + float out_center_of_in; // Center of the current out scanline in the in scanline space + int out_first_scanline, out_last_scanline; + + stbir__calculate_sample_range_downsample(y, in_pixels_radius, scale_ratio, stbir_info->vertical_shift, &out_first_scanline, &out_last_scanline, &out_center_of_in); + + STBIR_ASSERT(out_last_scanline - out_first_scanline + 1 <= stbir_info->ring_buffer_num_entries); + + if (out_last_scanline < 0 || out_first_scanline >= output_h) + continue; + + stbir__empty_ring_buffer(stbir_info, out_first_scanline); + + stbir__decode_and_resample_downsample(stbir_info, y); + + // Load in new ones. + if (stbir_info->ring_buffer_begin_index < 0) + stbir__add_empty_ring_buffer_entry(stbir_info, out_first_scanline); + + while (out_last_scanline > stbir_info->ring_buffer_last_scanline) + stbir__add_empty_ring_buffer_entry(stbir_info, stbir_info->ring_buffer_last_scanline + 1); + + // Now the horizontal buffer is ready to write to all ring buffer rows. + stbir__resample_vertical_downsample(stbir_info, y); + } + + stbir__empty_ring_buffer(stbir_info, stbir_info->output_h); +} + +static void stbir__setup(stbir__info *info, int input_w, int input_h, int output_w, int output_h, int channels) +{ + info->input_w = input_w; + info->input_h = input_h; + info->output_w = output_w; + info->output_h = output_h; + info->channels = channels; +} + +static void stbir__calculate_transform(stbir__info *info, float s0, float t0, float s1, float t1, float *transform) +{ + info->s0 = s0; + info->t0 = t0; + info->s1 = s1; + info->t1 = t1; + + if (transform) + { + info->horizontal_scale = transform[0]; + info->vertical_scale = transform[1]; + info->horizontal_shift = transform[2]; + info->vertical_shift = transform[3]; + } + else + { + info->horizontal_scale = ((float)info->output_w / info->input_w) / (s1 - s0); + info->vertical_scale = ((float)info->output_h / info->input_h) / (t1 - t0); + + info->horizontal_shift = s0 * info->output_w / (s1 - s0); + info->vertical_shift = t0 * info->output_h / (t1 - t0); + } +} + +static void stbir__choose_filter(stbir__info *info, stbir_filter h_filter, stbir_filter v_filter) +{ + if (h_filter == 0) + h_filter = stbir__use_upsampling(info->horizontal_scale) ? STBIR_DEFAULT_FILTER_UPSAMPLE : STBIR_DEFAULT_FILTER_DOWNSAMPLE; + if (v_filter == 0) + v_filter = stbir__use_upsampling(info->vertical_scale) ? STBIR_DEFAULT_FILTER_UPSAMPLE : STBIR_DEFAULT_FILTER_DOWNSAMPLE; + info->horizontal_filter = h_filter; + info->vertical_filter = v_filter; +} + +static stbir_uint32 stbir__calculate_memory(stbir__info *info) +{ + int pixel_margin = stbir__get_filter_pixel_margin(info->horizontal_filter, info->horizontal_scale); + int filter_height = stbir__get_filter_pixel_width(info->vertical_filter, info->vertical_scale); + + info->horizontal_num_contributors = stbir__get_contributors(info->horizontal_scale, info->horizontal_filter, info->input_w, info->output_w); + info->vertical_num_contributors = stbir__get_contributors(info->vertical_scale , info->vertical_filter , info->input_h, info->output_h); + + // One extra entry because floating point precision problems sometimes cause an extra to be necessary. + info->ring_buffer_num_entries = filter_height + 1; + + info->horizontal_contributors_size = info->horizontal_num_contributors * sizeof(stbir__contributors); + info->horizontal_coefficients_size = stbir__get_total_horizontal_coefficients(info) * sizeof(float); + info->vertical_contributors_size = info->vertical_num_contributors * sizeof(stbir__contributors); + info->vertical_coefficients_size = stbir__get_total_vertical_coefficients(info) * sizeof(float); + info->decode_buffer_size = (info->input_w + pixel_margin * 2) * info->channels * sizeof(float); + info->horizontal_buffer_size = info->output_w * info->channels * sizeof(float); + info->ring_buffer_size = info->output_w * info->channels * info->ring_buffer_num_entries * sizeof(float); + info->encode_buffer_size = info->output_w * info->channels * sizeof(float); + + STBIR_ASSERT(info->horizontal_filter != 0); + STBIR_ASSERT(info->horizontal_filter < STBIR__ARRAY_SIZE(stbir__filter_info_table)); // this now happens too late + STBIR_ASSERT(info->vertical_filter != 0); + STBIR_ASSERT(info->vertical_filter < STBIR__ARRAY_SIZE(stbir__filter_info_table)); // this now happens too late + + if (stbir__use_height_upsampling(info)) + // The horizontal buffer is for when we're downsampling the height and we + // can't output the result of sampling the decode buffer directly into the + // ring buffers. + info->horizontal_buffer_size = 0; + else + // The encode buffer is to retain precision in the height upsampling method + // and isn't used when height downsampling. + info->encode_buffer_size = 0; + + return info->horizontal_contributors_size + info->horizontal_coefficients_size + + info->vertical_contributors_size + info->vertical_coefficients_size + + info->decode_buffer_size + info->horizontal_buffer_size + + info->ring_buffer_size + info->encode_buffer_size; +} + +static int stbir__resize_allocated(stbir__info *info, + const void* input_data, int input_stride_in_bytes, + void* output_data, int output_stride_in_bytes, + int alpha_channel, stbir_uint32 flags, stbir_datatype type, + stbir_edge edge_horizontal, stbir_edge edge_vertical, stbir_colorspace colorspace, + void* tempmem, size_t tempmem_size_in_bytes) +{ + size_t memory_required = stbir__calculate_memory(info); + + int width_stride_input = input_stride_in_bytes ? input_stride_in_bytes : info->channels * info->input_w * stbir__type_size[type]; + int width_stride_output = output_stride_in_bytes ? output_stride_in_bytes : info->channels * info->output_w * stbir__type_size[type]; + +#ifdef STBIR_DEBUG_OVERWRITE_TEST +#define OVERWRITE_ARRAY_SIZE 8 + unsigned char overwrite_output_before_pre[OVERWRITE_ARRAY_SIZE]; + unsigned char overwrite_tempmem_before_pre[OVERWRITE_ARRAY_SIZE]; + unsigned char overwrite_output_after_pre[OVERWRITE_ARRAY_SIZE]; + unsigned char overwrite_tempmem_after_pre[OVERWRITE_ARRAY_SIZE]; + + size_t begin_forbidden = width_stride_output * (info->output_h - 1) + info->output_w * info->channels * stbir__type_size[type]; + memcpy(overwrite_output_before_pre, &((unsigned char*)output_data)[-OVERWRITE_ARRAY_SIZE], OVERWRITE_ARRAY_SIZE); + memcpy(overwrite_output_after_pre, &((unsigned char*)output_data)[begin_forbidden], OVERWRITE_ARRAY_SIZE); + memcpy(overwrite_tempmem_before_pre, &((unsigned char*)tempmem)[-OVERWRITE_ARRAY_SIZE], OVERWRITE_ARRAY_SIZE); + memcpy(overwrite_tempmem_after_pre, &((unsigned char*)tempmem)[tempmem_size_in_bytes], OVERWRITE_ARRAY_SIZE); +#endif + + STBIR_ASSERT(info->channels >= 0); + STBIR_ASSERT(info->channels <= STBIR_MAX_CHANNELS); + + if (info->channels < 0 || info->channels > STBIR_MAX_CHANNELS) + return 0; + + STBIR_ASSERT(info->horizontal_filter < STBIR__ARRAY_SIZE(stbir__filter_info_table)); + STBIR_ASSERT(info->vertical_filter < STBIR__ARRAY_SIZE(stbir__filter_info_table)); + + if (info->horizontal_filter >= STBIR__ARRAY_SIZE(stbir__filter_info_table)) + return 0; + if (info->vertical_filter >= STBIR__ARRAY_SIZE(stbir__filter_info_table)) + return 0; + + if (alpha_channel < 0) + flags |= STBIR_FLAG_ALPHA_USES_COLORSPACE | STBIR_FLAG_ALPHA_PREMULTIPLIED; + + if (!(flags&STBIR_FLAG_ALPHA_USES_COLORSPACE) || !(flags&STBIR_FLAG_ALPHA_PREMULTIPLIED)) { + STBIR_ASSERT(alpha_channel >= 0 && alpha_channel < info->channels); + } + + if (alpha_channel >= info->channels) + return 0; + + STBIR_ASSERT(tempmem); + + if (!tempmem) + return 0; + + STBIR_ASSERT(tempmem_size_in_bytes >= memory_required); + + if (tempmem_size_in_bytes < memory_required) + return 0; + + memset(tempmem, 0, tempmem_size_in_bytes); + + info->input_data = input_data; + info->input_stride_bytes = width_stride_input; + + info->output_data = output_data; + info->output_stride_bytes = width_stride_output; + + info->alpha_channel = alpha_channel; + info->flags = flags; + info->type = type; + info->edge_horizontal = edge_horizontal; + info->edge_vertical = edge_vertical; + info->colorspace = colorspace; + + info->horizontal_coefficient_width = stbir__get_coefficient_width (info->horizontal_filter, info->horizontal_scale); + info->vertical_coefficient_width = stbir__get_coefficient_width (info->vertical_filter , info->vertical_scale ); + info->horizontal_filter_pixel_width = stbir__get_filter_pixel_width (info->horizontal_filter, info->horizontal_scale); + info->vertical_filter_pixel_width = stbir__get_filter_pixel_width (info->vertical_filter , info->vertical_scale ); + info->horizontal_filter_pixel_margin = stbir__get_filter_pixel_margin(info->horizontal_filter, info->horizontal_scale); + info->vertical_filter_pixel_margin = stbir__get_filter_pixel_margin(info->vertical_filter , info->vertical_scale ); + + info->ring_buffer_length_bytes = info->output_w * info->channels * sizeof(float); + info->decode_buffer_pixels = info->input_w + info->horizontal_filter_pixel_margin * 2; + +#define STBIR__NEXT_MEMPTR(current, newtype) (newtype*)(((unsigned char*)current) + current##_size) + + info->horizontal_contributors = (stbir__contributors *) tempmem; + info->horizontal_coefficients = STBIR__NEXT_MEMPTR(info->horizontal_contributors, float); + info->vertical_contributors = STBIR__NEXT_MEMPTR(info->horizontal_coefficients, stbir__contributors); + info->vertical_coefficients = STBIR__NEXT_MEMPTR(info->vertical_contributors, float); + info->decode_buffer = STBIR__NEXT_MEMPTR(info->vertical_coefficients, float); + + if (stbir__use_height_upsampling(info)) + { + info->horizontal_buffer = NULL; + info->ring_buffer = STBIR__NEXT_MEMPTR(info->decode_buffer, float); + info->encode_buffer = STBIR__NEXT_MEMPTR(info->ring_buffer, float); + + STBIR_ASSERT((size_t)STBIR__NEXT_MEMPTR(info->encode_buffer, unsigned char) == (size_t)tempmem + tempmem_size_in_bytes); + } + else + { + info->horizontal_buffer = STBIR__NEXT_MEMPTR(info->decode_buffer, float); + info->ring_buffer = STBIR__NEXT_MEMPTR(info->horizontal_buffer, float); + info->encode_buffer = NULL; + + STBIR_ASSERT((size_t)STBIR__NEXT_MEMPTR(info->ring_buffer, unsigned char) == (size_t)tempmem + tempmem_size_in_bytes); + } + +#undef STBIR__NEXT_MEMPTR + + // This signals that the ring buffer is empty + info->ring_buffer_begin_index = -1; + + stbir__calculate_filters(info->horizontal_contributors, info->horizontal_coefficients, info->horizontal_filter, info->horizontal_scale, info->horizontal_shift, info->input_w, info->output_w); + stbir__calculate_filters(info->vertical_contributors, info->vertical_coefficients, info->vertical_filter, info->vertical_scale, info->vertical_shift, info->input_h, info->output_h); + + STBIR_PROGRESS_REPORT(0); + + if (stbir__use_height_upsampling(info)) + stbir__buffer_loop_upsample(info); + else + stbir__buffer_loop_downsample(info); + + STBIR_PROGRESS_REPORT(1); + +#ifdef STBIR_DEBUG_OVERWRITE_TEST + STBIR_ASSERT(memcmp(overwrite_output_before_pre, &((unsigned char*)output_data)[-OVERWRITE_ARRAY_SIZE], OVERWRITE_ARRAY_SIZE) == 0); + STBIR_ASSERT(memcmp(overwrite_output_after_pre, &((unsigned char*)output_data)[begin_forbidden], OVERWRITE_ARRAY_SIZE) == 0); + STBIR_ASSERT(memcmp(overwrite_tempmem_before_pre, &((unsigned char*)tempmem)[-OVERWRITE_ARRAY_SIZE], OVERWRITE_ARRAY_SIZE) == 0); + STBIR_ASSERT(memcmp(overwrite_tempmem_after_pre, &((unsigned char*)tempmem)[tempmem_size_in_bytes], OVERWRITE_ARRAY_SIZE) == 0); +#endif + + return 1; +} + + +static int stbir__resize_arbitrary( + void *alloc_context, + const void* input_data, int input_w, int input_h, int input_stride_in_bytes, + void* output_data, int output_w, int output_h, int output_stride_in_bytes, + float s0, float t0, float s1, float t1, float *transform, + int channels, int alpha_channel, stbir_uint32 flags, stbir_datatype type, + stbir_filter h_filter, stbir_filter v_filter, + stbir_edge edge_horizontal, stbir_edge edge_vertical, stbir_colorspace colorspace) +{ + stbir__info info; + int result; + size_t memory_required; + void* extra_memory; + + stbir__setup(&info, input_w, input_h, output_w, output_h, channels); + stbir__calculate_transform(&info, s0,t0,s1,t1,transform); + stbir__choose_filter(&info, h_filter, v_filter); + memory_required = stbir__calculate_memory(&info); + extra_memory = STBIR_MALLOC(memory_required, alloc_context); + + if (!extra_memory) + return 0; + + result = stbir__resize_allocated(&info, input_data, input_stride_in_bytes, + output_data, output_stride_in_bytes, + alpha_channel, flags, type, + edge_horizontal, edge_vertical, + colorspace, extra_memory, memory_required); + + STBIR_FREE(extra_memory, alloc_context); + + return result; +} + +STBIRDEF int stbir_resize_uint8( const unsigned char *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + unsigned char *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + int num_channels) +{ + return stbir__resize_arbitrary(NULL, input_pixels, input_w, input_h, input_stride_in_bytes, + output_pixels, output_w, output_h, output_stride_in_bytes, + 0,0,1,1,NULL,num_channels,-1,0, STBIR_TYPE_UINT8, STBIR_FILTER_DEFAULT, STBIR_FILTER_DEFAULT, + STBIR_EDGE_CLAMP, STBIR_EDGE_CLAMP, STBIR_COLORSPACE_LINEAR); +} + +STBIRDEF int stbir_resize_float( const float *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + float *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + int num_channels) +{ + return stbir__resize_arbitrary(NULL, input_pixels, input_w, input_h, input_stride_in_bytes, + output_pixels, output_w, output_h, output_stride_in_bytes, + 0,0,1,1,NULL,num_channels,-1,0, STBIR_TYPE_FLOAT, STBIR_FILTER_DEFAULT, STBIR_FILTER_DEFAULT, + STBIR_EDGE_CLAMP, STBIR_EDGE_CLAMP, STBIR_COLORSPACE_LINEAR); +} + +STBIRDEF int stbir_resize_uint8_srgb(const unsigned char *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + unsigned char *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + int num_channels, int alpha_channel, int flags) +{ + return stbir__resize_arbitrary(NULL, input_pixels, input_w, input_h, input_stride_in_bytes, + output_pixels, output_w, output_h, output_stride_in_bytes, + 0,0,1,1,NULL,num_channels,alpha_channel,flags, STBIR_TYPE_UINT8, STBIR_FILTER_DEFAULT, STBIR_FILTER_DEFAULT, + STBIR_EDGE_CLAMP, STBIR_EDGE_CLAMP, STBIR_COLORSPACE_SRGB); +} + +STBIRDEF int stbir_resize_uint8_srgb_edgemode(const unsigned char *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + unsigned char *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + int num_channels, int alpha_channel, int flags, + stbir_edge edge_wrap_mode) +{ + return stbir__resize_arbitrary(NULL, input_pixels, input_w, input_h, input_stride_in_bytes, + output_pixels, output_w, output_h, output_stride_in_bytes, + 0,0,1,1,NULL,num_channels,alpha_channel,flags, STBIR_TYPE_UINT8, STBIR_FILTER_DEFAULT, STBIR_FILTER_DEFAULT, + edge_wrap_mode, edge_wrap_mode, STBIR_COLORSPACE_SRGB); +} + +STBIRDEF int stbir_resize_uint8_generic( const unsigned char *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + unsigned char *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + int num_channels, int alpha_channel, int flags, + stbir_edge edge_wrap_mode, stbir_filter filter, stbir_colorspace space, + void *alloc_context) +{ + return stbir__resize_arbitrary(alloc_context, input_pixels, input_w, input_h, input_stride_in_bytes, + output_pixels, output_w, output_h, output_stride_in_bytes, + 0,0,1,1,NULL,num_channels,alpha_channel,flags, STBIR_TYPE_UINT8, filter, filter, + edge_wrap_mode, edge_wrap_mode, space); +} + +STBIRDEF int stbir_resize_uint16_generic(const stbir_uint16 *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + stbir_uint16 *output_pixels , int output_w, int output_h, int output_stride_in_bytes, + int num_channels, int alpha_channel, int flags, + stbir_edge edge_wrap_mode, stbir_filter filter, stbir_colorspace space, + void *alloc_context) +{ + return stbir__resize_arbitrary(alloc_context, input_pixels, input_w, input_h, input_stride_in_bytes, + output_pixels, output_w, output_h, output_stride_in_bytes, + 0,0,1,1,NULL,num_channels,alpha_channel,flags, STBIR_TYPE_UINT16, filter, filter, + edge_wrap_mode, edge_wrap_mode, space); +} + + +STBIRDEF int stbir_resize_float_generic( const float *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + float *output_pixels , int output_w, int output_h, int output_stride_in_bytes, + int num_channels, int alpha_channel, int flags, + stbir_edge edge_wrap_mode, stbir_filter filter, stbir_colorspace space, + void *alloc_context) +{ + return stbir__resize_arbitrary(alloc_context, input_pixels, input_w, input_h, input_stride_in_bytes, + output_pixels, output_w, output_h, output_stride_in_bytes, + 0,0,1,1,NULL,num_channels,alpha_channel,flags, STBIR_TYPE_FLOAT, filter, filter, + edge_wrap_mode, edge_wrap_mode, space); +} + + +STBIRDEF int stbir_resize( const void *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + void *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + stbir_datatype datatype, + int num_channels, int alpha_channel, int flags, + stbir_edge edge_mode_horizontal, stbir_edge edge_mode_vertical, + stbir_filter filter_horizontal, stbir_filter filter_vertical, + stbir_colorspace space, void *alloc_context) +{ + return stbir__resize_arbitrary(alloc_context, input_pixels, input_w, input_h, input_stride_in_bytes, + output_pixels, output_w, output_h, output_stride_in_bytes, + 0,0,1,1,NULL,num_channels,alpha_channel,flags, datatype, filter_horizontal, filter_vertical, + edge_mode_horizontal, edge_mode_vertical, space); +} + + +STBIRDEF int stbir_resize_subpixel(const void *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + void *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + stbir_datatype datatype, + int num_channels, int alpha_channel, int flags, + stbir_edge edge_mode_horizontal, stbir_edge edge_mode_vertical, + stbir_filter filter_horizontal, stbir_filter filter_vertical, + stbir_colorspace space, void *alloc_context, + float x_scale, float y_scale, + float x_offset, float y_offset) +{ + float transform[4]; + transform[0] = x_scale; + transform[1] = y_scale; + transform[2] = x_offset; + transform[3] = y_offset; + return stbir__resize_arbitrary(alloc_context, input_pixels, input_w, input_h, input_stride_in_bytes, + output_pixels, output_w, output_h, output_stride_in_bytes, + 0,0,1,1,transform,num_channels,alpha_channel,flags, datatype, filter_horizontal, filter_vertical, + edge_mode_horizontal, edge_mode_vertical, space); +} + +STBIRDEF int stbir_resize_region( const void *input_pixels , int input_w , int input_h , int input_stride_in_bytes, + void *output_pixels, int output_w, int output_h, int output_stride_in_bytes, + stbir_datatype datatype, + int num_channels, int alpha_channel, int flags, + stbir_edge edge_mode_horizontal, stbir_edge edge_mode_vertical, + stbir_filter filter_horizontal, stbir_filter filter_vertical, + stbir_colorspace space, void *alloc_context, + float s0, float t0, float s1, float t1) +{ + return stbir__resize_arbitrary(alloc_context, input_pixels, input_w, input_h, input_stride_in_bytes, + output_pixels, output_w, output_h, output_stride_in_bytes, + s0,t0,s1,t1,NULL,num_channels,alpha_channel,flags, datatype, filter_horizontal, filter_vertical, + edge_mode_horizontal, edge_mode_vertical, space); +} + +#endif // STB_IMAGE_RESIZE_IMPLEMENTATION + +/* +------------------------------------------------------------------------------ +This software is available under 2 licenses -- choose whichever you prefer. +------------------------------------------------------------------------------ +ALTERNATIVE A - MIT License +Copyright (c) 2017 Sean Barrett +Permission is hereby granted, free of charge, to any person obtaining a copy of +this software and associated documentation files (the "Software"), to deal in +the Software without restriction, including without limitation the rights to +use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies +of the Software, and to permit persons to whom the Software is furnished to do +so, subject to the following conditions: +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. +------------------------------------------------------------------------------ +ALTERNATIVE B - Public Domain (www.unlicense.org) +This is free and unencumbered software released into the public domain. +Anyone is free to copy, modify, publish, use, compile, sell, or distribute this +software, either in source code form or as a compiled binary, for any purpose, +commercial or non-commercial, and by any means. +In jurisdictions that recognize copyright laws, the author or authors of this +software dedicate any and all copyright interest in the software to the public +domain. We make this dedication for the benefit of the public at large and to +the detriment of our heirs and successors. We intend this dedication to be an +overt act of relinquishment in perpetuity of all present and future rights to +this software under copyright law. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN +ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION +WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. +------------------------------------------------------------------------------ +*/ diff --git a/src/files/wal.c b/src/files/wal.c new file mode 100644 index 0000000..ddc184c --- /dev/null +++ b/src/files/wal.c @@ -0,0 +1,86 @@ +/* + * Copyright (C) 1997-2001 Id Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + * + * ======================================================================= + * + * The Wal image format + * + * ======================================================================= + */ + +#include "../common/header/ref_shared.h" + +/* + * NOTE: LoadWal() is in *_image.c because it needs the renderer-specific image_t + */ + +void +GetWalInfo(char *name, int *width, int *height) +{ + miptex_t *mt; + int size; + + size = ri.FS_LoadFile(name, (void **)&mt); + + if (!mt) + { + return; + } + + if (size < sizeof(miptex_t)) + { + ri.FS_FreeFile((void *)mt); + return; + } + + *width = LittleLong(mt->width); + *height = LittleLong(mt->height); + + ri.FS_FreeFile((void *)mt); + + return; +} + +void +GetM8Info(char *name, int *width, int *height) +{ + m8tex_t *mt; + int size; + + size = ri.FS_LoadFile(name, (void **)&mt); + + if (!mt) + { + return; + } + + + if (size < sizeof(m8tex_t) || LittleLong (mt->version) != M8_VERSION) + { + ri.FS_FreeFile((void *)mt); + return; + } + + *width = LittleLong(mt->width[0]); + *height = LittleLong(mt->height[0]); + + ri.FS_FreeFile((void *)mt); + + return; +} diff --git a/src/vk/header/local.h b/src/vk/header/local.h new file mode 100644 index 0000000..c078961 --- /dev/null +++ b/src/vk/header/local.h @@ -0,0 +1,331 @@ +/* +Copyright (C) 1997-2001 Id Software, Inc. +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. +*/ + +#ifndef __VK_LOCAL_H__ +#define __VK_LOCAL_H__ + +#include +#include + +#include +#include + + +#include "../../common/header/ref_shared.h" +#include "../volk/volk.h" +#include "qvk.h" + +// verify if VkResult is VK_SUCCESS +#define VK_VERIFY(x) { \ + VkResult res = (x); \ + if(res != VK_SUCCESS) { \ + R_Printf(PRINT_ALL, "%s:%d: VkResult verification failed: %s\n", \ + __func__, __LINE__, QVk_GetError(res)); \ + } \ +} + +// up / down +#define PITCH 0 + +// left / right +#define YAW 1 + +// fall over +#define ROLL 2 + +extern viddef_t vid; + +typedef struct image_s +{ + char name[MAX_QPATH]; // game path, including extension + imagetype_t type; + int width, height; // source image + int upload_width, upload_height; // after power of two and picmip + int registration_sequence; // 0 = free + struct msurface_s *texturechain; // for sort-by-texture world drawing + qvktexture_t vk_texture; // Vulkan texture handle +} image_t; + +#define MAX_VKTEXTURES 1024 + +//=================================================================== + +typedef enum +{ + rserr_ok, + + rserr_invalid_fullscreen, + rserr_invalid_mode, + + rserr_unknown +} rserr_t; + +#include "model.h" + +#define MAX_LBM_HEIGHT 480 + +#define BACKFACE_EPSILON 0.01 + + +//==================================================== + +extern image_t vktextures[MAX_VKTEXTURES]; +extern int numvktextures; + +extern image_t *r_notexture; +extern image_t *r_particletexture; +extern image_t *r_squaretexture; +extern int r_visframecount; +extern int r_framecount; +extern cplane_t frustum[4]; +extern int c_brush_polys, c_alias_polys; + +// +// view origin +// +extern vec3_t vup; +extern vec3_t vpn; +extern vec3_t vright; +extern vec3_t r_origin; + +// +// screen size info +// +extern refdef_t r_newrefdef; +extern int r_viewcluster, r_viewcluster2, r_oldviewcluster, r_oldviewcluster2; + +extern cvar_t *r_lefthand; +extern cvar_t *r_drawworld; +extern cvar_t *r_novis; +extern cvar_t *r_lerpmodels; +extern cvar_t *r_lockpvs; +extern cvar_t *r_modulate; +extern cvar_t *r_vsync; +extern cvar_t *r_clear; +extern cvar_t *r_lightlevel; // FIXME: This is a HACK to get the client's light level +extern cvar_t *r_gunfov; +extern cvar_t *r_farsee; + +extern cvar_t *vk_overbrightbits; +extern cvar_t *vk_validation; +extern cvar_t *vk_picmip; +extern cvar_t *vk_skymip; +extern cvar_t *vk_flashblend; +extern cvar_t *vk_finish; +extern cvar_t *vk_shadows; +extern cvar_t *vk_dynamic; +extern cvar_t *vk_msaa; +extern cvar_t *vk_showtris; +extern cvar_t *vk_lightmap; +extern cvar_t *vk_texturemode; +extern cvar_t *vk_lmaptexturemode; +extern cvar_t *vk_aniso; +extern cvar_t *vk_sampleshading; +extern cvar_t *vk_device_idx; +extern cvar_t *vk_retexturing; +extern cvar_t *vk_nolerp_list; +extern cvar_t *vk_pixel_size; +extern cvar_t *r_fixsurfsky; + +extern cvar_t *vid_fullscreen; +extern cvar_t *vid_gamma; + +extern int c_visible_lightmaps; +extern int c_visible_textures; + +extern float r_viewproj_matrix[16]; + +void R_LightPoint (vec3_t p, vec3_t color, entity_t *currententity); +void R_PushDlights (void); + +//==================================================================== + +extern model_t *r_worldmodel; + +extern unsigned d_8to24table[256]; + +extern int registration_sequence; +extern qvksampler_t vk_current_sampler; +extern qvksampler_t vk_current_lmap_sampler; + +void RE_Shutdown( void ); + +void Vk_ScreenShot_f (void); +void R_DrawAliasModel (entity_t *currententity, model_t *currentmodel); +void R_DrawBrushModel (entity_t *currententity, model_t *currentmodel); +void R_DrawSpriteModel (entity_t *currententity, model_t *currentmodel); +void R_DrawBeam (entity_t *currententity); +void R_DrawWorld (void); +void R_RenderDlights (void); +void R_DrawAlphaSurfaces (void); +void RE_InitParticleTexture (void); +void Draw_InitLocal (void); +void Vk_SubdivideSurface (msurface_t *fa, model_t *loadmodel); +qboolean R_CullBox (vec3_t mins, vec3_t maxs); +void R_RotateForEntity (entity_t *e, float *mvMatrix); +void R_MarkLeaves (void); + +void EmitWaterPolys (msurface_t *fa, image_t *texture, + float *modelMatrix, float *color, + qboolean solid_surface); +void R_AddSkySurface (msurface_t *fa); +void R_ClearSkyBox (void); +void R_DrawSkyBox (void); +void R_MarkLights (dlight_t *light, int bit, mnode_t *node); + +struct image_s *RE_Draw_FindPic (char *name); + +void RE_Draw_GetPicSize (int *w, int *h, char *name); +void RE_Draw_PicScaled (int x, int y, char *name, float scale); +void RE_Draw_StretchPic (int x, int y, int w, int h, char *name); +void RE_Draw_CharScaled (int x, int y, int num, float scale); +void RE_Draw_TileClear (int x, int y, int w, int h, char *name); +void RE_Draw_Fill (int x, int y, int w, int h, int c); +void RE_Draw_FadeScreen (void); +void RE_Draw_StretchRaw (int x, int y, int w, int h, int cols, int rows, byte *data); + +qboolean RE_EndWorldRenderpass( void ); + +struct image_s *RE_RegisterSkin (char *name); + +void LoadPCX (char *filename, byte **pic, byte **palette, int *width, int *height); +image_t *Vk_LoadPic(char *name, byte *pic, int width, int realwidth, + int height, int realheight, imagetype_t type, + int bits); +image_t *Vk_FindImage (char *name, imagetype_t type); +void Vk_TextureMode( char *string ); +void Vk_LmapTextureMode( char *string ); +void Vk_ImageList_f (void); + +void Vk_BuildPolygonFromSurface(msurface_t *fa, model_t *currentmodel); +void Vk_CreateSurfaceLightmap (msurface_t *surf); +void Vk_EndBuildingLightmaps (void); +void Vk_BeginBuildingLightmaps (model_t *m); + +void Vk_InitImages (void); +void Vk_ShutdownImages (void); +void Vk_FreeUnusedImages (void); +qboolean Vk_ImageHasFreeSpace(void); + +void RE_BeginRegistration (char *model); +struct model_s *RE_RegisterModel (char *name); +struct image_s *RE_RegisterSkin (char *name); +void RE_SetSky (char *name, float rotate, vec3_t axis); +void RE_EndRegistration (void); + +void Mat_Identity(float *matrix); +void Mat_Mul(float *m1, float *m2, float *res); +void Mat_Translate(float *matrix, float x, float y, float z); +void Mat_Rotate(float *matrix, float deg, float x, float y, float z); +void Mat_Scale(float *matrix, float x, float y, float z); +void Mat_Perspective(float *matrix, float *correction_matrix, float fovy, float aspect, float zNear, float zFar); +void Mat_Ortho(float *matrix, float left, float right, float bottom, float top, float zNear, float zFar); + +typedef struct +{ + uint32_t vk_version; + const char *vendor_name; + const char *device_type; + const char *present_mode; + const char *supported_present_modes[256]; + const char *extensions[256]; + const char *layers[256]; + uint32_t vertex_buffer_usage; + uint32_t vertex_buffer_max_usage; + uint32_t vertex_buffer_size; + uint32_t index_buffer_usage; + uint32_t index_buffer_max_usage; + uint32_t index_buffer_size; + uint32_t uniform_buffer_usage; + uint32_t uniform_buffer_max_usage; + uint32_t uniform_buffer_size; + uint32_t triangle_index_usage; + uint32_t triangle_index_max_usage; + uint32_t triangle_index_count; +} vkconfig_t; + +#define MAX_LIGHTMAPS 128 +#define DYNLIGHTMAP_OFFSET MAX_LIGHTMAPS + +typedef struct +{ + float inverse_intensity; + qboolean fullscreen; + + int prev_mode; + + unsigned char *d_16to8table; + + qvktexture_t lightmap_textures[MAX_LIGHTMAPS*2]; + + int currenttextures[2]; + int currenttmu; + + float camera_separation; + qboolean stereo_enabled; + + VkPipeline current_pipeline; + qvkrenderpasstype_t current_renderpass; +} vkstate_t; + +extern vkconfig_t vk_config; +extern vkstate_t vk_state; + +/* +==================================================================== + +IMPORTED FUNCTIONS + +==================================================================== +*/ + +extern refimport_t ri; + + +/* +==================================================================== + +IMPLEMENTATION SPECIFIC FUNCTIONS + +==================================================================== +*/ + +qboolean Vkimp_CreateSurface(SDL_Window *window); + +// buffers reallocate +typedef struct { + float vertex[3]; + float texCoord[2]; +} polyvert_t; + +typedef struct { + float vertex[3]; + float texCoord[2]; + float texCoordLmap[2]; +} lmappolyvert_t; + +extern polyvert_t *verts_buffer; +extern lmappolyvert_t *lmappolyverts_buffer; + +void Mesh_Init (void); +void Mesh_Free (void); +int Mesh_VertsRealloc(int count); + +#endif diff --git a/src/vk/header/model.h b/src/vk/header/model.h new file mode 100644 index 0000000..4821f6c --- /dev/null +++ b/src/vk/header/model.h @@ -0,0 +1,254 @@ +/* +Copyright (C) 1997-2001 Id Software, Inc. +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +*/ + +#ifndef __VK_MODEL_H__ +#define __VK_MODEL_H__ + +/* + +d*_t structures are on-disk representations +m*_t structures are in-memory + +*/ + +/* +============================================================================== + +BRUSH MODELS + +============================================================================== +*/ + + +// +// in memory representation +// +typedef struct +{ + vec3_t position; +} mvertex_t; + + +#define SIDE_FRONT 0 +#define SIDE_BACK 1 +#define SIDE_ON 2 + + +#define SURF_PLANEBACK 2 +#define SURF_DRAWSKY 4 +#define SURF_DRAWTURB 0x10 +#define SURF_DRAWBACKGROUND 0x40 +#define SURF_UNDERWATER 0x80 + +typedef struct +{ + unsigned short v[2]; + unsigned int cachededgeoffset; +} medge_t; + +typedef struct mtexinfo_s +{ + float vecs[2][4]; + int flags; + int numframes; + struct mtexinfo_s *next; // animation chain + image_t *image; +} mtexinfo_t; + +#define VERTEXSIZE 7 + +typedef struct vkpoly_s +{ + struct vkpoly_s *next; + struct vkpoly_s *chain; + int numverts; + int flags; // for SURF_UNDERWATER (not needed anymore?) + float verts[4][VERTEXSIZE]; // variable sized (xyz s1t1 s2t2) +} vkpoly_t; + +typedef struct msurface_s +{ + int visframe; // should be drawn when node is crossed + + cplane_t *plane; + int flags; + + int firstedge; // look up in model->surfedges[], negative numbers + int numedges; // are backwards edges + + short texturemins[2]; + short extents[2]; + + int light_s, light_t; // gl lightmap coordinates + int dlight_s, dlight_t; // gl lightmap coordinates for dynamic lightmaps + + vkpoly_t *polys; // multiple if warped + struct msurface_s *texturechain; + struct msurface_s *lightmapchain; + + mtexinfo_t *texinfo; + +// lighting info + int dlightframe; + int dlightbits; + + int lightmaptexturenum; + byte styles[MAXLIGHTMAPS]; + float cached_light[MAXLIGHTMAPS]; // values currently used in lightmap + byte *samples; // [numstyles*surfsize] +} msurface_t; + +typedef struct mnode_s +{ +// common with leaf + int contents; // -1, to differentiate from leafs + int visframe; // node needs to be traversed if current + + float minmaxs[6]; // for bounding box culling + + struct mnode_s *parent; + +// node specific + cplane_t *plane; + struct mnode_s *children[2]; + + unsigned short firstsurface; + unsigned short numsurfaces; +} mnode_t; + + + +typedef struct mleaf_s +{ +// common with node + int contents; // wil be a negative contents number + int visframe; // node needs to be traversed if current + + float minmaxs[6]; // for bounding box culling + + struct mnode_s *parent; + +// leaf specific + int cluster; + int area; + + msurface_t **firstmarksurface; + int nummarksurfaces; +} mleaf_t; + + +//=================================================================== + +// +// Whole model +// + +typedef struct model_s +{ + char name[MAX_QPATH]; + + int registration_sequence; + + modtype_t type; + int numframes; + + int flags; + +// +// volume occupied by the model graphics +// + vec3_t mins, maxs; + float radius; + +// +// solid volume for clipping +// + qboolean clipbox; + vec3_t clipmins, clipmaxs; + +// +// brush model +// + int firstmodelsurface, nummodelsurfaces; + int lightmap; // only for submodels + + int numsubmodels; + struct model_s *submodels; + + int numplanes; + cplane_t *planes; + + int numleafs; // number of visible leafs, not counting 0 + mleaf_t *leafs; + + int numvertexes; + mvertex_t *vertexes; + + int numedges; + medge_t *edges; + + int numnodes; + int firstnode; + mnode_t *nodes; + + int numtexinfo; + mtexinfo_t *texinfo; + + int numsurfaces; + msurface_t *surfaces; + + int numsurfedges; + int *surfedges; + + int nummarksurfaces; + msurface_t **marksurfaces; + + dvis_t *vis; + + byte *lightdata; + + // for alias models and skins + image_t *skins[MAX_MD2SKINS]; + + int extradatasize; + void *extradata; + + // submodules + vec3_t origin; // for sounds or lights +} model_t; + +//============================================================================ + +void Mod_Init (void); +mleaf_t *Mod_PointInLeaf (float *p, model_t *model); +byte *Mod_ClusterPVS (int cluster, model_t *model); + +void Mod_Modellist_f (void); + +void *Hunk_Begin (int maxsize); +void *Hunk_Alloc (int size); +int Hunk_End (void); +void Hunk_Free (void *base); + +void Mod_FreeAll (void); +void Mod_FreeModelsKnown (void); + +#endif diff --git a/src/vk/header/qvk.h b/src/vk/header/qvk.h new file mode 100644 index 0000000..fb58d28 --- /dev/null +++ b/src/vk/header/qvk.h @@ -0,0 +1,326 @@ +/* +Copyright (C) 1997-2001 Id Software, Inc. +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +*/ +/* +** QVK.H +*/ + +#ifndef __QVK_H__ +#define __QVK_H__ + +#ifdef _WIN32 +# include +#endif + +#include "local.h" +#include "util.h" +#include "shaders.h" + +// Vulkan device +typedef struct +{ + VkPhysicalDevice physical; + VkDevice logical; + VkPhysicalDeviceMemoryProperties mem_properties; + VkPhysicalDeviceProperties properties; + VkPhysicalDeviceFeatures features; + VkQueue gfxQueue; + VkQueue presentQueue; + VkQueue transferQueue; + int gfxFamilyIndex; + int presentFamilyIndex; + int transferFamilyIndex; + qboolean screenshotSupported; +} qvkdevice_t; + +// Vulkan swapchain +typedef struct +{ + VkSwapchainKHR sc; + VkFormat format; + VkPresentModeKHR presentMode; + VkExtent2D extent; + VkImage *images; + int imageCount; +} qvkswapchain_t; + +// available sampler types +typedef enum +{ + S_NEAREST = 0, + S_LINEAR = 1, + S_MIPMAP_NEAREST = 2, + S_MIPMAP_LINEAR = 3, + S_NEAREST_UNNORMALIZED = 4, + S_SAMPLER_CNT = 5 +} qvksampler_t; + +#define NUM_SAMPLERS (S_SAMPLER_CNT * 2) + +// texture object +typedef struct +{ + ImageResource_t resource; + + VkImageView imageView; + VkSharingMode sharingMode; + VkSampleCountFlagBits sampleCount; + VkFormat format; + VkDescriptorSet descriptorSet; + uint32_t mipLevels; + qboolean clampToEdge; +} qvktexture_t; + +#define QVVKTEXTURE_INIT { \ + .resource = { \ + .image = VK_NULL_HANDLE, \ + .memory = VK_NULL_HANDLE, \ + .size = 0, \ + }, \ + .imageView = VK_NULL_HANDLE, \ + .sharingMode = VK_SHARING_MODE_MAX_ENUM, \ + .sampleCount = VK_SAMPLE_COUNT_1_BIT, \ + .format = VK_FORMAT_R8G8B8A8_UNORM, \ + .descriptorSet = VK_NULL_HANDLE, \ + .mipLevels = 1, \ +} + +#define QVVKTEXTURE_CLEAR(i) { \ + (i).resource.image = VK_NULL_HANDLE; \ + (i).resource.memory = VK_NULL_HANDLE; \ + (i).resource.size = 0; \ + (i).imageView = VK_NULL_HANDLE; \ + (i).sharingMode = VK_SHARING_MODE_MAX_ENUM; \ + (i).sampleCount = VK_SAMPLE_COUNT_1_BIT; \ + (i).format = VK_FORMAT_R8G8B8A8_UNORM; \ + (i).mipLevels = 1; \ +} + +// Vulkan renderpass +typedef struct +{ + VkRenderPass rp; + VkAttachmentLoadOp colorLoadOp; + VkSampleCountFlagBits sampleCount; +} qvkrenderpass_t; + +// Vulkan buffer +typedef struct +{ + VkDeviceSize currentOffset; + + BufferResource_t resource; + void *pMappedData; +} qvkbuffer_t; + +// Vulkan staging buffer +typedef struct +{ + VkDeviceSize currentOffset; + VkCommandBuffer cmdBuffer; + VkFence fence; + qboolean submitted; + + BufferResource_t resource; + void *pMappedData; +} qvkstagingbuffer_t; + +// Vulkan buffer options +typedef struct +{ + VkBufferUsageFlags usage; + VkMemoryPropertyFlags reqMemFlags; + VkMemoryPropertyFlags prefMemFlags; +} qvkbufferopts_t; + +// Vulkan pipeline +typedef struct +{ + VkPipelineLayout layout; + VkPipeline pl; + VkPipelineCreateFlags flags; + VkCullModeFlags cullMode; + VkPrimitiveTopology topology; + VkPipelineColorBlendAttachmentState blendOpts; + VkBool32 depthTestEnable; + VkBool32 depthWriteEnable; +} qvkpipeline_t; + +// Vulkan shader +typedef struct +{ + VkPipelineShaderStageCreateInfo createInfo; + VkShaderModule module; +} qvkshader_t; + +#define QVKPIPELINE_INIT { \ + .layout = VK_NULL_HANDLE, \ + .pl = VK_NULL_HANDLE, \ + .flags = 0, \ + .cullMode = VK_CULL_MODE_BACK_BIT, \ + .topology = VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST, \ + .blendOpts = { \ + .blendEnable = VK_FALSE, \ + .srcColorBlendFactor = VK_BLEND_FACTOR_SRC_ALPHA, \ + .dstColorBlendFactor = VK_BLEND_FACTOR_ONE_MINUS_SRC_ALPHA, \ + .colorBlendOp = VK_BLEND_OP_ADD, \ + .srcAlphaBlendFactor = VK_BLEND_FACTOR_SRC_ALPHA, \ + .dstAlphaBlendFactor = VK_BLEND_FACTOR_ONE_MINUS_SRC_ALPHA, \ + .alphaBlendOp = VK_BLEND_OP_ADD, \ + .colorWriteMask = VK_COLOR_COMPONENT_R_BIT | VK_COLOR_COMPONENT_G_BIT | VK_COLOR_COMPONENT_B_BIT | VK_COLOR_COMPONENT_A_BIT \ + }, \ + .depthTestEnable = VK_TRUE, \ + .depthWriteEnable = VK_TRUE \ +} + +// renderpass type +typedef enum +{ + RP_WORLD = 0, // renders game world to offscreen buffer + RP_UI = 1, // render UI elements and game console + RP_WORLD_WARP = 2, // perform postprocessing on RP_WORLD (underwater screen warp) + RP_COUNT = 3 +} qvkrenderpasstype_t; + +// Vulkan constants: command and dynamic buffer count +#define NUM_CMDBUFFERS 2 +#define NUM_DYNBUFFERS 2 + +// Vulkan instance +extern VkInstance vk_instance; +// Vulkan surface +extern VkSurfaceKHR vk_surface; +// Vulkan device +extern qvkdevice_t vk_device; +// Vulkan swapchain +extern qvkswapchain_t vk_swapchain; +// Vulkan command buffer currently in use +extern VkCommandBuffer vk_activeCmdbuffer; +// Vulkan command pools +extern VkCommandPool vk_commandPool[NUM_CMDBUFFERS]; +extern VkCommandPool vk_transferCommandPool; +// Vulkan descriptor pool +extern VkDescriptorPool vk_descriptorPool; +// viewport/scissor +extern VkViewport vk_viewport; +extern VkRect2D vk_scissor; + +// Vulkan descriptor sets +extern VkDescriptorSetLayout vk_samplerDescSetLayout; + +// *** pipelines *** +extern qvkpipeline_t vk_drawTexQuadPipeline[RP_COUNT]; +extern qvkpipeline_t vk_drawColorQuadPipeline[RP_COUNT]; +extern qvkpipeline_t vk_drawModelPipelineFan[RP_COUNT]; +extern qvkpipeline_t vk_drawNoDepthModelPipelineFan; +extern qvkpipeline_t vk_drawLefthandModelPipelineFan; +extern qvkpipeline_t vk_drawNullModelPipeline; +extern qvkpipeline_t vk_drawParticlesPipeline; +extern qvkpipeline_t vk_drawPointParticlesPipeline; +extern qvkpipeline_t vk_drawSpritePipeline; +extern qvkpipeline_t vk_drawPolyPipeline; +extern qvkpipeline_t vk_drawPolyLmapPipeline; +extern qvkpipeline_t vk_drawPolyWarpPipeline; +extern qvkpipeline_t vk_drawPolySolidWarpPipeline; +extern qvkpipeline_t vk_drawBeamPipeline; +extern qvkpipeline_t vk_drawSkyboxPipeline; +extern qvkpipeline_t vk_drawDLightPipeline; +extern qvkpipeline_t vk_showTrisPipeline; +extern qvkpipeline_t vk_shadowsPipelineFan; +extern qvkpipeline_t vk_worldWarpPipeline; +extern qvkpipeline_t vk_postprocessPipeline; + +// color buffer containing main game/world view +extern qvktexture_t vk_colorbuffer; +// color buffer with postprocessed game view +extern qvktexture_t vk_colorbufferWarp; +// indicator if the frame is currently being rendered +extern qboolean vk_frameStarted; +// Indicates if the renderer needs to be restarted. +extern qboolean vk_restartNeeded; +// is QVk initialized? +extern qboolean vk_initialized; + +// function pointers +extern PFN_vkCreateDebugUtilsMessengerEXT qvkCreateDebugUtilsMessengerEXT; +extern PFN_vkDestroyDebugUtilsMessengerEXT qvkDestroyDebugUtilsMessengerEXT; +extern PFN_vkSetDebugUtilsObjectNameEXT qvkSetDebugUtilsObjectNameEXT; +extern PFN_vkSetDebugUtilsObjectTagEXT qvkSetDebugUtilsObjectTagEXT; +extern PFN_vkCmdBeginDebugUtilsLabelEXT qvkCmdBeginDebugUtilsLabelEXT; +extern PFN_vkCmdEndDebugUtilsLabelEXT qvkCmdEndDebugUtilsLabelEXT; +extern PFN_vkCmdInsertDebugUtilsLabelEXT qvkInsertDebugUtilsLabelEXT; + +// The Interface Functions (tm) +void QVk_SetWindow(SDL_Window*); +qboolean QVk_Init(void); +void QVk_PostInit(void); +void QVk_WaitAndShutdownAll(void); +void QVk_Shutdown(void); +void QVk_Restart(void); +void QVk_CreateValidationLayers(void); +void QVk_DestroyValidationLayers(void); +qboolean QVk_CreateDevice(int preferredDeviceIdx); +VkResult QVk_CreateSwapchain(void); +VkFormat QVk_FindDepthFormat(void); +VkResult QVk_CreateCommandPool(VkCommandPool *commandPool, uint32_t queueFamilyIndex); +VkResult QVk_CreateImageView(const VkImage *image, VkImageAspectFlags aspectFlags, VkImageView *imageView, VkFormat format, uint32_t mipLevels); +VkResult QVk_CreateImage(uint32_t width, uint32_t height, VkFormat format, VkImageTiling tiling, VkImageUsageFlags usage, qvktexture_t *texture); +void QVk_CreateDepthBuffer(VkSampleCountFlagBits sampleCount, qvktexture_t *depthBuffer); +void QVk_CreateColorBuffer(VkSampleCountFlagBits sampleCount, qvktexture_t *colorBuffer, int extraFlags); +void QVk_CreateTexture(qvktexture_t *texture, const unsigned char *data, uint32_t width, uint32_t height, qvksampler_t samplerType, qboolean clampToEdge); +void QVk_UpdateTextureData(qvktexture_t *texture, const unsigned char *data, uint32_t offset_x, uint32_t offset_y, uint32_t width, uint32_t height); +VkSampler QVk_UpdateTextureSampler(qvktexture_t *texture, qvksampler_t samplerType, qboolean clampToEdge); +void QVk_ReadPixels(uint8_t *dstBuffer, const VkOffset2D *offset, const VkExtent2D *extent); +VkResult QVk_BeginCommand(const VkCommandBuffer *commandBuffer); +void QVk_SubmitCommand(const VkCommandBuffer *commandBuffer, const VkQueue *queue); +VkCommandBuffer QVk_CreateCommandBuffer(const VkCommandPool *commandPool, VkCommandBufferLevel level); +const char* QVk_GetError(VkResult errorCode); +VkResult QVk_BeginFrame(const VkViewport* viewport, const VkRect2D* scissor); +VkResult QVk_EndFrame(qboolean force); +void QVk_BeginRenderpass(qvkrenderpasstype_t rpType); +void QVk_FreeStagingBuffer(qvkstagingbuffer_t *buffer); +VkResult QVk_CreateBuffer(VkDeviceSize size, qvkbuffer_t *dstBuffer, const qvkbufferopts_t options); +void QVk_FreeBuffer(qvkbuffer_t *buffer); +VkResult QVk_CreateStagingBuffer(VkDeviceSize size, qvkstagingbuffer_t *dstBuffer, VkMemoryPropertyFlags reqMemFlags, VkMemoryPropertyFlags prefMemFlags); +VkResult QVk_CreateUniformBuffer(VkDeviceSize size, qvkbuffer_t *dstBuffer, VkMemoryPropertyFlags reqMemFlags, VkMemoryPropertyFlags prefMemFlags); +void QVk_CreateVertexBuffer(const void *data, VkDeviceSize size, qvkbuffer_t *dstBuffer, VkMemoryPropertyFlags reqMemFlags, VkMemoryPropertyFlags prefMemFlags); +void QVk_CreateIndexBuffer(const void *data, VkDeviceSize size, qvkbuffer_t *dstBuffer, VkMemoryPropertyFlags reqMemFlags, VkMemoryPropertyFlags prefMemFlags); +qvkshader_t QVk_CreateShader(const uint32_t *shaderSrc, size_t shaderCodeSize, VkShaderStageFlagBits shaderStage); +void QVk_CreatePipeline(const VkDescriptorSetLayout *descriptorLayout, const uint32_t descLayoutCount, const VkPipelineVertexInputStateCreateInfo *vertexInputInfo, qvkpipeline_t *pipeline, const qvkrenderpass_t *renderpass, const qvkshader_t *shaders, uint32_t shaderCount); +void QVk_DestroyPipeline(qvkpipeline_t *pipeline); +uint8_t* QVk_GetVertexBuffer(VkDeviceSize size, VkBuffer *dstBuffer, VkDeviceSize *dstOffset); +uint8_t* QVk_GetUniformBuffer(VkDeviceSize size, uint32_t *dstOffset, VkDescriptorSet *dstUboDescriptorSet); +uint8_t* QVk_GetStagingBuffer(VkDeviceSize size, int alignment, VkCommandBuffer *cmdBuffer, VkBuffer *buffer, uint32_t *dstOffset); +VkBuffer QVk_GetTriangleFanIbo(VkDeviceSize indexCount); +VkBuffer QVk_GetTriangleStripIbo(VkDeviceSize indexCount); +void QVk_DrawColorRect(float *ubo, VkDeviceSize uboSize, qvkrenderpasstype_t rpType); +void QVk_DrawTexRect(const float *ubo, VkDeviceSize uboSize, qvktexture_t *texture); +void QVk_BindPipeline(qvkpipeline_t *pipeline); +void QVk_SubmitStagingBuffers(void); +void Qvk_MemoryBarrier(VkCommandBuffer cmdBuffer, VkPipelineStageFlags srcStageMask, VkPipelineStageFlags dstStageMask, VkAccessFlags srcAccessMask, VkAccessFlags dstAccessMask); +qboolean QVk_CheckExtent(void); + +// debug label related functions +void QVk_DebugSetObjectName(uint64_t obj, VkObjectType objType, const char *objName); +void QVk_DebugSetObjectTag(uint64_t obj, VkObjectType objType, uint64_t tagName, size_t tagSize, const void *tagData); +void QVk_DebugLabelBegin(const VkCommandBuffer *cmdBuffer, const char *labelName, const float r, const float g, const float b); +void QVk_DebugLabelEnd(const VkCommandBuffer *cmdBuffer); +void QVk_DebugLabelInsert(const VkCommandBuffer *cmdBuffer, const char *labelName, const float r, const float g, const float b); +#endif diff --git a/src/vk/header/shaders.h b/src/vk/header/shaders.h new file mode 100644 index 0000000..64f6e67 --- /dev/null +++ b/src/vk/header/shaders.h @@ -0,0 +1,115 @@ +/* +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +*/ + +// +// game shaders, compiled offline +// + +#ifndef __VK_SHADERS_H__ +#define __VK_SHADERS_H__ +#include +#include + +// textured quad (alpha) +extern const uint32_t basic_vert_spv[]; +extern const size_t basic_vert_size; + +extern const uint32_t basic_frag_spv[]; +extern const size_t basic_frag_size; + +// colored quad (alpha) +extern const uint32_t basic_color_quad_vert_spv[]; +extern const size_t basic_color_quad_vert_size; + +extern const uint32_t basic_color_quad_frag_spv[]; +extern const size_t basic_color_quad_frag_size; + +// textured model +extern const uint32_t model_vert_spv[]; +extern const size_t model_vert_size; + +extern const uint32_t model_frag_spv[]; +extern const size_t model_frag_size; + +// null model +extern const uint32_t nullmodel_vert_spv[]; +extern const size_t nullmodel_vert_size; + +// particle (texture) +extern const uint32_t particle_vert_spv[]; +extern const size_t particle_vert_size; + +// particle (point) +extern const uint32_t point_particle_vert_spv[]; +extern const size_t point_particle_vert_size; + +extern const uint32_t point_particle_frag_spv[]; +extern const size_t point_particle_frag_size; + +// sprite model +extern const uint32_t sprite_vert_spv[]; +extern const size_t sprite_vert_size; + +// beam +extern const uint32_t beam_vert_spv[]; +extern const size_t beam_vert_size; + +// skybox +extern const uint32_t skybox_vert_spv[]; +extern const size_t skybox_vert_size; + +// dynamic lights +extern const uint32_t d_light_vert_spv[]; +extern const size_t d_light_vert_size; + +// textured, alpha blended polygon +extern const uint32_t polygon_vert_spv[]; +extern const size_t polygon_vert_size; + +// textured, lightmapped polygon +extern const uint32_t polygon_lmap_vert_spv[]; +extern const size_t polygon_lmap_vert_size; + +extern const uint32_t polygon_lmap_frag_spv[]; +extern const size_t polygon_lmap_frag_size; + +// warped polygon (liquids) +extern const uint32_t polygon_warp_vert_spv[]; +extern const size_t polygon_warp_vert_size; + +// entity shadows +extern const uint32_t shadows_vert_spv[]; +extern const size_t shadows_vert_size; + +// postprocess +extern const uint32_t postprocess_vert_spv[]; +extern const size_t postprocess_vert_size; + +extern const uint32_t postprocess_frag_spv[]; +extern const size_t postprocess_frag_size; + +// underwater vision warp +extern const uint32_t world_warp_vert_spv[]; +extern const size_t world_warp_vert_size; + +extern const uint32_t world_warp_frag_spv[]; +extern const size_t world_warp_frag_size; + +#endif diff --git a/src/vk/header/util.h b/src/vk/header/util.h new file mode 100644 index 0000000..92ffc1a --- /dev/null +++ b/src/vk/header/util.h @@ -0,0 +1,71 @@ +/* +Copyright (C) 2018 Christoph Schied +Copyright (C) 2020 Denis Pauk + +This program is free software; you can redistribute it and/or modify +it under the terms of the GNU General Public License as published by +the Free Software Foundation; either version 2 of the License, or +(at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +GNU General Public License for more details. + +You should have received a copy of the GNU General Public License along +with this program; if not, write to the Free Software Foundation, Inc., +51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. +*/ + +#ifndef __VK_UTIL_H__ +#define __VK_UTIL_H__ + +#include "../volk/volk.h" + +#define ROUNDUP(a, b) (((a) + ((b)-1)) & ~((b)-1)) + +typedef struct BufferResource_s { + VkBuffer buffer; + // shared memory used for buffer + VkDeviceMemory memory; + // image size + VkDeviceSize size; + // posision in shared memory + VkDeviceSize offset; + // is mapped? + VkBool32 is_mapped; +} BufferResource_t; + +typedef struct ImageResource_s { + VkImage image; + // shared memory used for image + VkDeviceMemory memory; + // image size + VkDeviceSize size; + // posision in shared memory + VkDeviceSize offset; +} ImageResource_t; + +VkResult buffer_create(BufferResource_t *buf, + VkBufferCreateInfo buf_create_info, + VkMemoryPropertyFlags mem_properties, + VkMemoryPropertyFlags mem_preferences); + +VkResult buffer_destroy(BufferResource_t *buf); +void buffer_unmap(BufferResource_t *buf); +void *buffer_map(BufferResource_t *buf); +VkResult buffer_flush(BufferResource_t *buf); +VkResult buffer_invalidate(BufferResource_t *buf); + +VkResult image_create(ImageResource_t *img, + VkImageCreateInfo img_create_info, + VkMemoryPropertyFlags mem_properties, + VkMemoryPropertyFlags mem_preferences); +VkResult image_destroy(ImageResource_t *img); + +void vulkan_memory_init(void); +void vulkan_memory_types_show(void); +void vulkan_memory_free_unused(void); +void vulkan_memory_delete(void); + +#endif /*__VK_UTIL_H__*/ diff --git a/src/vk/spirv/basic_color_quad_frag.c b/src/vk/spirv/basic_color_quad_frag.c new file mode 100644 index 0000000..4fb19bf --- /dev/null +++ b/src/vk/spirv/basic_color_quad_frag.c @@ -0,0 +1,16 @@ + // 8.13.3559 + #pragma once +const uint32_t basic_color_quad_frag_spv[] = { + 0x07230203,0x00010000,0x00080008,0x0000000d,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0007000f,0x00000004,0x00000004,0x6e69616d,0x00000000,0x00000009,0x0000000b,0x00030010, + 0x00000004,0x00000007,0x00030003,0x00000002,0x000001c2,0x00040005,0x00000004,0x6e69616d, + 0x00000000,0x00060005,0x00000009,0x67617266,0x746e656d,0x6f6c6f43,0x00000072,0x00040005, + 0x0000000b,0x6f6c6f63,0x00000072,0x00040047,0x00000009,0x0000001e,0x00000000,0x00040047, + 0x0000000b,0x0000001e,0x00000000,0x00020013,0x00000002,0x00030021,0x00000003,0x00000002, + 0x00030016,0x00000006,0x00000020,0x00040017,0x00000007,0x00000006,0x00000004,0x00040020, + 0x00000008,0x00000003,0x00000007,0x0004003b,0x00000008,0x00000009,0x00000003,0x00040020, + 0x0000000a,0x00000001,0x00000007,0x0004003b,0x0000000a,0x0000000b,0x00000001,0x00050036, + 0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,0x0004003d,0x00000007, + 0x0000000c,0x0000000b,0x0003003e,0x00000009,0x0000000c,0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/basic_color_quad_vert.c b/src/vk/spirv/basic_color_quad_vert.c new file mode 100644 index 0000000..c884bef --- /dev/null +++ b/src/vk/spirv/basic_color_quad_vert.c @@ -0,0 +1,48 @@ + // 8.13.3559 + #pragma once +const uint32_t basic_color_quad_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000032,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0008000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000b,0x0000001f,0x0000002d, + 0x00030003,0x00000002,0x000001c2,0x00090004,0x415f4c47,0x735f4252,0x72617065,0x5f657461, + 0x64616873,0x6f5f7265,0x63656a62,0x00007374,0x00040005,0x00000004,0x6e69616d,0x00000000, + 0x00040005,0x00000009,0x736f5076,0x00000000,0x00050005,0x0000000b,0x65566e69,0x78657472, + 0x00000000,0x00060005,0x0000000e,0x67616d69,0x61725465,0x6f66736e,0x00006d72,0x00050006, + 0x0000000e,0x00000000,0x7366666f,0x00007465,0x00050006,0x0000000e,0x00000001,0x6c616373, + 0x00000065,0x00050006,0x0000000e,0x00000002,0x6f6c6f63,0x00000072,0x00030005,0x00000010, + 0x00007469,0x00060005,0x0000001d,0x505f6c67,0x65567265,0x78657472,0x00000000,0x00060006, + 0x0000001d,0x00000000,0x505f6c67,0x7469736f,0x006e6f69,0x00030005,0x0000001f,0x00000000, + 0x00040005,0x0000002d,0x6f6c6f63,0x00000072,0x00040047,0x0000000b,0x0000001e,0x00000000, + 0x00050048,0x0000000e,0x00000000,0x00000023,0x00000000,0x00050048,0x0000000e,0x00000001, + 0x00000023,0x00000008,0x00050048,0x0000000e,0x00000002,0x00000023,0x00000010,0x00030047, + 0x0000000e,0x00000002,0x00040047,0x00000010,0x00000022,0x00000000,0x00040047,0x00000010, + 0x00000021,0x00000000,0x00050048,0x0000001d,0x00000000,0x0000000b,0x00000000,0x00030047, + 0x0000001d,0x00000002,0x00040047,0x0000002d,0x0000001e,0x00000000,0x00020013,0x00000002, + 0x00030021,0x00000003,0x00000002,0x00030016,0x00000006,0x00000020,0x00040017,0x00000007, + 0x00000006,0x00000002,0x00040020,0x00000008,0x00000007,0x00000007,0x00040020,0x0000000a, + 0x00000001,0x00000007,0x0004003b,0x0000000a,0x0000000b,0x00000001,0x00040017,0x0000000d, + 0x00000006,0x00000004,0x0005001e,0x0000000e,0x00000007,0x00000007,0x0000000d,0x00040020, + 0x0000000f,0x00000002,0x0000000e,0x0004003b,0x0000000f,0x00000010,0x00000002,0x00040015, + 0x00000011,0x00000020,0x00000001,0x0004002b,0x00000011,0x00000012,0x00000001,0x00040020, + 0x00000013,0x00000002,0x00000007,0x0004002b,0x00000006,0x00000017,0x3f800000,0x0005002c, + 0x00000007,0x00000018,0x00000017,0x00000017,0x0003001e,0x0000001d,0x0000000d,0x00040020, + 0x0000001e,0x00000003,0x0000001d,0x0004003b,0x0000001e,0x0000001f,0x00000003,0x0004002b, + 0x00000011,0x00000020,0x00000000,0x0004002b,0x00000006,0x00000024,0x40000000,0x0004002b, + 0x00000006,0x00000027,0x00000000,0x00040020,0x0000002b,0x00000003,0x0000000d,0x0004003b, + 0x0000002b,0x0000002d,0x00000003,0x0004002b,0x00000011,0x0000002e,0x00000002,0x00040020, + 0x0000002f,0x00000002,0x0000000d,0x00050036,0x00000002,0x00000004,0x00000000,0x00000003, + 0x000200f8,0x00000005,0x0004003b,0x00000008,0x00000009,0x00000007,0x0004003d,0x00000007, + 0x0000000c,0x0000000b,0x00050041,0x00000013,0x00000014,0x00000010,0x00000012,0x0004003d, + 0x00000007,0x00000015,0x00000014,0x00050085,0x00000007,0x00000016,0x0000000c,0x00000015, + 0x00050041,0x00000013,0x00000019,0x00000010,0x00000012,0x0004003d,0x00000007,0x0000001a, + 0x00000019,0x00050083,0x00000007,0x0000001b,0x00000018,0x0000001a,0x00050083,0x00000007, + 0x0000001c,0x00000016,0x0000001b,0x0003003e,0x00000009,0x0000001c,0x0004003d,0x00000007, + 0x00000021,0x00000009,0x00050041,0x00000013,0x00000022,0x00000010,0x00000020,0x0004003d, + 0x00000007,0x00000023,0x00000022,0x0005008e,0x00000007,0x00000025,0x00000023,0x00000024, + 0x00050081,0x00000007,0x00000026,0x00000021,0x00000025,0x00050051,0x00000006,0x00000028, + 0x00000026,0x00000000,0x00050051,0x00000006,0x00000029,0x00000026,0x00000001,0x00070050, + 0x0000000d,0x0000002a,0x00000028,0x00000029,0x00000027,0x00000017,0x00050041,0x0000002b, + 0x0000002c,0x0000001f,0x00000020,0x0003003e,0x0000002c,0x0000002a,0x00050041,0x0000002f, + 0x00000030,0x00000010,0x0000002e,0x0004003d,0x0000000d,0x00000031,0x00000030,0x0003003e, + 0x0000002d,0x00000031,0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/basic_frag.c b/src/vk/spirv/basic_frag.c new file mode 100644 index 0000000..9885101 --- /dev/null +++ b/src/vk/spirv/basic_frag.c @@ -0,0 +1,48 @@ + // 8.13.3559 + #pragma once +const uint32_t basic_frag_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000038,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0009000f,0x00000004,0x00000004,0x6e69616d,0x00000000,0x00000009,0x00000011,0x00000015, + 0x0000001e,0x00030010,0x00000004,0x00000007,0x00030003,0x00000002,0x000001c2,0x00040005, + 0x00000004,0x6e69616d,0x00000000,0x00060005,0x00000009,0x67617266,0x746e656d,0x6f6c6f43, + 0x00000072,0x00050005,0x0000000d,0x78655473,0x65727574,0x00000000,0x00050005,0x00000011, + 0x43786574,0x64726f6f,0x00000000,0x00040005,0x00000015,0x6f6c6f63,0x00000072,0x00050005, + 0x0000001e,0x65725461,0x6c6f6873,0x00000064,0x00060005,0x00000028,0x68737550,0x736e6f43, + 0x746e6174,0x00000000,0x00050006,0x00000028,0x00000000,0x6d6d6167,0x00000061,0x00030005, + 0x0000002a,0x00006370,0x00040047,0x00000009,0x0000001e,0x00000000,0x00040047,0x0000000d, + 0x00000022,0x00000000,0x00040047,0x0000000d,0x00000021,0x00000000,0x00040047,0x00000011, + 0x0000001e,0x00000000,0x00040047,0x00000015,0x0000001e,0x00000001,0x00040047,0x0000001e, + 0x0000001e,0x00000002,0x00050048,0x00000028,0x00000000,0x00000023,0x00000044,0x00030047, + 0x00000028,0x00000002,0x00020013,0x00000002,0x00030021,0x00000003,0x00000002,0x00030016, + 0x00000006,0x00000020,0x00040017,0x00000007,0x00000006,0x00000004,0x00040020,0x00000008, + 0x00000003,0x00000007,0x0004003b,0x00000008,0x00000009,0x00000003,0x00090019,0x0000000a, + 0x00000006,0x00000001,0x00000000,0x00000000,0x00000000,0x00000001,0x00000000,0x0003001b, + 0x0000000b,0x0000000a,0x00040020,0x0000000c,0x00000000,0x0000000b,0x0004003b,0x0000000c, + 0x0000000d,0x00000000,0x00040017,0x0000000f,0x00000006,0x00000002,0x00040020,0x00000010, + 0x00000001,0x0000000f,0x0004003b,0x00000010,0x00000011,0x00000001,0x00040020,0x00000014, + 0x00000001,0x00000007,0x0004003b,0x00000014,0x00000015,0x00000001,0x00040015,0x00000018, + 0x00000020,0x00000000,0x0004002b,0x00000018,0x00000019,0x00000003,0x00040020,0x0000001a, + 0x00000003,0x00000006,0x00040020,0x0000001d,0x00000001,0x00000006,0x0004003b,0x0000001d, + 0x0000001e,0x00000001,0x00020014,0x00000020,0x00040017,0x00000025,0x00000006,0x00000003, + 0x0003001e,0x00000028,0x00000006,0x00040020,0x00000029,0x00000009,0x00000028,0x0004003b, + 0x00000029,0x0000002a,0x00000009,0x00040015,0x0000002b,0x00000020,0x00000001,0x0004002b, + 0x0000002b,0x0000002c,0x00000000,0x00040020,0x0000002d,0x00000009,0x00000006,0x00050036, + 0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,0x0004003d,0x0000000b, + 0x0000000e,0x0000000d,0x0004003d,0x0000000f,0x00000012,0x00000011,0x00050057,0x00000007, + 0x00000013,0x0000000e,0x00000012,0x0004003d,0x00000007,0x00000016,0x00000015,0x00050085, + 0x00000007,0x00000017,0x00000013,0x00000016,0x0003003e,0x00000009,0x00000017,0x00050041, + 0x0000001a,0x0000001b,0x00000009,0x00000019,0x0004003d,0x00000006,0x0000001c,0x0000001b, + 0x0004003d,0x00000006,0x0000001f,0x0000001e,0x000500b8,0x00000020,0x00000021,0x0000001c, + 0x0000001f,0x000300f7,0x00000023,0x00000000,0x000400fa,0x00000021,0x00000022,0x00000023, + 0x000200f8,0x00000022,0x000100fc,0x000200f8,0x00000023,0x0004003d,0x00000007,0x00000026, + 0x00000009,0x0008004f,0x00000025,0x00000027,0x00000026,0x00000026,0x00000000,0x00000001, + 0x00000002,0x00050041,0x0000002d,0x0000002e,0x0000002a,0x0000002c,0x0004003d,0x00000006, + 0x0000002f,0x0000002e,0x00060050,0x00000025,0x00000030,0x0000002f,0x0000002f,0x0000002f, + 0x0007000c,0x00000025,0x00000031,0x00000001,0x0000001a,0x00000027,0x00000030,0x00050041, + 0x0000001a,0x00000032,0x00000009,0x00000019,0x0004003d,0x00000006,0x00000033,0x00000032, + 0x00050051,0x00000006,0x00000034,0x00000031,0x00000000,0x00050051,0x00000006,0x00000035, + 0x00000031,0x00000001,0x00050051,0x00000006,0x00000036,0x00000031,0x00000002,0x00070050, + 0x00000007,0x00000037,0x00000034,0x00000035,0x00000036,0x00000033,0x0003003e,0x00000009, + 0x00000037,0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/basic_vert.c b/src/vk/spirv/basic_vert.c new file mode 100644 index 0000000..3e15183 --- /dev/null +++ b/src/vk/spirv/basic_vert.c @@ -0,0 +1,61 @@ + // 8.13.3559 + #pragma once +const uint32_t basic_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x0000003e,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x000b000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000b,0x0000001f,0x0000002e, + 0x0000002f,0x00000039,0x0000003c,0x00030003,0x00000002,0x000001c2,0x00090004,0x415f4c47, + 0x735f4252,0x72617065,0x5f657461,0x64616873,0x6f5f7265,0x63656a62,0x00007374,0x00040005, + 0x00000004,0x6e69616d,0x00000000,0x00040005,0x00000009,0x736f5076,0x00000000,0x00050005, + 0x0000000b,0x65566e69,0x78657472,0x00000000,0x00060005,0x0000000d,0x67616d69,0x61725465, + 0x6f66736e,0x00006d72,0x00050006,0x0000000d,0x00000000,0x7366666f,0x00007465,0x00050006, + 0x0000000d,0x00000001,0x6c616373,0x00000065,0x00060006,0x0000000d,0x00000002,0x664f7675, + 0x74657366,0x00000000,0x00050006,0x0000000d,0x00000003,0x63537675,0x00656c61,0x00030005, + 0x0000000f,0x00007469,0x00060005,0x0000001d,0x505f6c67,0x65567265,0x78657472,0x00000000, + 0x00060006,0x0000001d,0x00000000,0x505f6c67,0x7469736f,0x006e6f69,0x00030005,0x0000001f, + 0x00000000,0x00050005,0x0000002e,0x43786574,0x64726f6f,0x00000000,0x00050005,0x0000002f, + 0x65546e69,0x6f6f4378,0x00006472,0x00040005,0x00000039,0x6f6c6f63,0x00000072,0x00050005, + 0x0000003c,0x65725461,0x6c6f6873,0x00000064,0x00040047,0x0000000b,0x0000001e,0x00000000, + 0x00050048,0x0000000d,0x00000000,0x00000023,0x00000000,0x00050048,0x0000000d,0x00000001, + 0x00000023,0x00000008,0x00050048,0x0000000d,0x00000002,0x00000023,0x00000010,0x00050048, + 0x0000000d,0x00000003,0x00000023,0x00000018,0x00030047,0x0000000d,0x00000002,0x00040047, + 0x0000000f,0x00000022,0x00000001,0x00040047,0x0000000f,0x00000021,0x00000000,0x00050048, + 0x0000001d,0x00000000,0x0000000b,0x00000000,0x00030047,0x0000001d,0x00000002,0x00040047, + 0x0000002e,0x0000001e,0x00000000,0x00040047,0x0000002f,0x0000001e,0x00000001,0x00040047, + 0x00000039,0x0000001e,0x00000001,0x00040047,0x0000003c,0x0000001e,0x00000002,0x00020013, + 0x00000002,0x00030021,0x00000003,0x00000002,0x00030016,0x00000006,0x00000020,0x00040017, + 0x00000007,0x00000006,0x00000002,0x00040020,0x00000008,0x00000007,0x00000007,0x00040020, + 0x0000000a,0x00000001,0x00000007,0x0004003b,0x0000000a,0x0000000b,0x00000001,0x0006001e, + 0x0000000d,0x00000007,0x00000007,0x00000007,0x00000007,0x00040020,0x0000000e,0x00000002, + 0x0000000d,0x0004003b,0x0000000e,0x0000000f,0x00000002,0x00040015,0x00000010,0x00000020, + 0x00000001,0x0004002b,0x00000010,0x00000011,0x00000001,0x00040020,0x00000012,0x00000002, + 0x00000007,0x0004002b,0x00000006,0x00000016,0x3f800000,0x0005002c,0x00000007,0x00000017, + 0x00000016,0x00000016,0x00040017,0x0000001c,0x00000006,0x00000004,0x0003001e,0x0000001d, + 0x0000001c,0x00040020,0x0000001e,0x00000003,0x0000001d,0x0004003b,0x0000001e,0x0000001f, + 0x00000003,0x0004002b,0x00000010,0x00000020,0x00000000,0x0004002b,0x00000006,0x00000024, + 0x40000000,0x0004002b,0x00000006,0x00000027,0x00000000,0x00040020,0x0000002b,0x00000003, + 0x0000001c,0x00040020,0x0000002d,0x00000003,0x00000007,0x0004003b,0x0000002d,0x0000002e, + 0x00000003,0x0004003b,0x0000000a,0x0000002f,0x00000001,0x0004002b,0x00000010,0x00000031, + 0x00000003,0x0004002b,0x00000010,0x00000035,0x00000002,0x0004003b,0x0000002b,0x00000039, + 0x00000003,0x0007002c,0x0000001c,0x0000003a,0x00000016,0x00000016,0x00000016,0x00000016, + 0x00040020,0x0000003b,0x00000003,0x00000006,0x0004003b,0x0000003b,0x0000003c,0x00000003, + 0x0004002b,0x00000006,0x0000003d,0x3f2a7efa,0x00050036,0x00000002,0x00000004,0x00000000, + 0x00000003,0x000200f8,0x00000005,0x0004003b,0x00000008,0x00000009,0x00000007,0x0004003d, + 0x00000007,0x0000000c,0x0000000b,0x00050041,0x00000012,0x00000013,0x0000000f,0x00000011, + 0x0004003d,0x00000007,0x00000014,0x00000013,0x00050085,0x00000007,0x00000015,0x0000000c, + 0x00000014,0x00050041,0x00000012,0x00000018,0x0000000f,0x00000011,0x0004003d,0x00000007, + 0x00000019,0x00000018,0x00050083,0x00000007,0x0000001a,0x00000017,0x00000019,0x00050083, + 0x00000007,0x0000001b,0x00000015,0x0000001a,0x0003003e,0x00000009,0x0000001b,0x0004003d, + 0x00000007,0x00000021,0x00000009,0x00050041,0x00000012,0x00000022,0x0000000f,0x00000020, + 0x0004003d,0x00000007,0x00000023,0x00000022,0x0005008e,0x00000007,0x00000025,0x00000023, + 0x00000024,0x00050081,0x00000007,0x00000026,0x00000021,0x00000025,0x00050051,0x00000006, + 0x00000028,0x00000026,0x00000000,0x00050051,0x00000006,0x00000029,0x00000026,0x00000001, + 0x00070050,0x0000001c,0x0000002a,0x00000028,0x00000029,0x00000027,0x00000016,0x00050041, + 0x0000002b,0x0000002c,0x0000001f,0x00000020,0x0003003e,0x0000002c,0x0000002a,0x0004003d, + 0x00000007,0x00000030,0x0000002f,0x00050041,0x00000012,0x00000032,0x0000000f,0x00000031, + 0x0004003d,0x00000007,0x00000033,0x00000032,0x00050085,0x00000007,0x00000034,0x00000030, + 0x00000033,0x00050041,0x00000012,0x00000036,0x0000000f,0x00000035,0x0004003d,0x00000007, + 0x00000037,0x00000036,0x00050081,0x00000007,0x00000038,0x00000034,0x00000037,0x0003003e, + 0x0000002e,0x00000038,0x0003003e,0x00000039,0x0000003a,0x0003003e,0x0000003c,0x0000003d, + 0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/beam_vert.c b/src/vk/spirv/beam_vert.c new file mode 100644 index 0000000..9dc6b07 --- /dev/null +++ b/src/vk/spirv/beam_vert.c @@ -0,0 +1,42 @@ + // 8.13.3559 + #pragma once +const uint32_t beam_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000027,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0008000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000a,0x00000016,0x00000020, + 0x00030003,0x00000002,0x000001c2,0x00090004,0x415f4c47,0x735f4252,0x72617065,0x5f657461, + 0x64616873,0x6f5f7265,0x63656a62,0x00007374,0x00040005,0x00000004,0x6e69616d,0x00000000, + 0x00060005,0x00000008,0x505f6c67,0x65567265,0x78657472,0x00000000,0x00060006,0x00000008, + 0x00000000,0x505f6c67,0x7469736f,0x006e6f69,0x00030005,0x0000000a,0x00000000,0x00060005, + 0x0000000e,0x68737550,0x736e6f43,0x746e6174,0x00000000,0x00060006,0x0000000e,0x00000000, + 0x4d70766d,0x69727461,0x00000078,0x00030005,0x00000010,0x00006370,0x00050005,0x00000016, + 0x65566e69,0x78657472,0x00000000,0x00040005,0x00000020,0x6f6c6f63,0x00000072,0x00070005, + 0x00000021,0x66696e55,0x426d726f,0x65666675,0x6a624f72,0x00746365,0x00050006,0x00000021, + 0x00000000,0x6f6c6f63,0x00000072,0x00030005,0x00000023,0x006f6275,0x00050048,0x00000008, + 0x00000000,0x0000000b,0x00000000,0x00030047,0x00000008,0x00000002,0x00040048,0x0000000e, + 0x00000000,0x00000005,0x00050048,0x0000000e,0x00000000,0x00000023,0x00000000,0x00050048, + 0x0000000e,0x00000000,0x00000007,0x00000010,0x00030047,0x0000000e,0x00000002,0x00040047, + 0x00000016,0x0000001e,0x00000000,0x00040047,0x00000020,0x0000001e,0x00000000,0x00050048, + 0x00000021,0x00000000,0x00000023,0x00000000,0x00030047,0x00000021,0x00000002,0x00040047, + 0x00000023,0x00000022,0x00000000,0x00040047,0x00000023,0x00000021,0x00000000,0x00020013, + 0x00000002,0x00030021,0x00000003,0x00000002,0x00030016,0x00000006,0x00000020,0x00040017, + 0x00000007,0x00000006,0x00000004,0x0003001e,0x00000008,0x00000007,0x00040020,0x00000009, + 0x00000003,0x00000008,0x0004003b,0x00000009,0x0000000a,0x00000003,0x00040015,0x0000000b, + 0x00000020,0x00000001,0x0004002b,0x0000000b,0x0000000c,0x00000000,0x00040018,0x0000000d, + 0x00000007,0x00000004,0x0003001e,0x0000000e,0x0000000d,0x00040020,0x0000000f,0x00000009, + 0x0000000e,0x0004003b,0x0000000f,0x00000010,0x00000009,0x00040020,0x00000011,0x00000009, + 0x0000000d,0x00040017,0x00000014,0x00000006,0x00000003,0x00040020,0x00000015,0x00000001, + 0x00000014,0x0004003b,0x00000015,0x00000016,0x00000001,0x0004002b,0x00000006,0x00000018, + 0x3f800000,0x00040020,0x0000001e,0x00000003,0x00000007,0x0004003b,0x0000001e,0x00000020, + 0x00000003,0x0003001e,0x00000021,0x00000007,0x00040020,0x00000022,0x00000002,0x00000021, + 0x0004003b,0x00000022,0x00000023,0x00000002,0x00040020,0x00000024,0x00000002,0x00000007, + 0x00050036,0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,0x00050041, + 0x00000011,0x00000012,0x00000010,0x0000000c,0x0004003d,0x0000000d,0x00000013,0x00000012, + 0x0004003d,0x00000014,0x00000017,0x00000016,0x00050051,0x00000006,0x00000019,0x00000017, + 0x00000000,0x00050051,0x00000006,0x0000001a,0x00000017,0x00000001,0x00050051,0x00000006, + 0x0000001b,0x00000017,0x00000002,0x00070050,0x00000007,0x0000001c,0x00000019,0x0000001a, + 0x0000001b,0x00000018,0x00050091,0x00000007,0x0000001d,0x00000013,0x0000001c,0x00050041, + 0x0000001e,0x0000001f,0x0000000a,0x0000000c,0x0003003e,0x0000001f,0x0000001d,0x00050041, + 0x00000024,0x00000025,0x00000023,0x0000000c,0x0004003d,0x00000007,0x00000026,0x00000025, + 0x0003003e,0x00000020,0x00000026,0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/d_light_vert.c b/src/vk/spirv/d_light_vert.c new file mode 100644 index 0000000..8f8e04b --- /dev/null +++ b/src/vk/spirv/d_light_vert.c @@ -0,0 +1,41 @@ + // 8.13.3559 + #pragma once +const uint32_t d_light_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000027,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0009000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000a,0x00000016,0x00000020, + 0x00000021,0x00030003,0x00000002,0x000001c2,0x00090004,0x415f4c47,0x735f4252,0x72617065, + 0x5f657461,0x64616873,0x6f5f7265,0x63656a62,0x00007374,0x00040005,0x00000004,0x6e69616d, + 0x00000000,0x00060005,0x00000008,0x505f6c67,0x65567265,0x78657472,0x00000000,0x00060006, + 0x00000008,0x00000000,0x505f6c67,0x7469736f,0x006e6f69,0x00030005,0x0000000a,0x00000000, + 0x00070005,0x0000000e,0x66696e55,0x426d726f,0x65666675,0x6a624f72,0x00746365,0x00060006, + 0x0000000e,0x00000000,0x4d70766d,0x69727461,0x00000078,0x00030005,0x00000010,0x006f6275, + 0x00050005,0x00000016,0x65566e69,0x78657472,0x00000000,0x00040005,0x00000020,0x6f6c6f63, + 0x00000072,0x00040005,0x00000021,0x6f436e69,0x00726f6c,0x00050048,0x00000008,0x00000000, + 0x0000000b,0x00000000,0x00030047,0x00000008,0x00000002,0x00040048,0x0000000e,0x00000000, + 0x00000005,0x00050048,0x0000000e,0x00000000,0x00000023,0x00000000,0x00050048,0x0000000e, + 0x00000000,0x00000007,0x00000010,0x00030047,0x0000000e,0x00000002,0x00040047,0x00000010, + 0x00000022,0x00000000,0x00040047,0x00000010,0x00000021,0x00000000,0x00040047,0x00000016, + 0x0000001e,0x00000000,0x00040047,0x00000020,0x0000001e,0x00000000,0x00040047,0x00000021, + 0x0000001e,0x00000001,0x00020013,0x00000002,0x00030021,0x00000003,0x00000002,0x00030016, + 0x00000006,0x00000020,0x00040017,0x00000007,0x00000006,0x00000004,0x0003001e,0x00000008, + 0x00000007,0x00040020,0x00000009,0x00000003,0x00000008,0x0004003b,0x00000009,0x0000000a, + 0x00000003,0x00040015,0x0000000b,0x00000020,0x00000001,0x0004002b,0x0000000b,0x0000000c, + 0x00000000,0x00040018,0x0000000d,0x00000007,0x00000004,0x0003001e,0x0000000e,0x0000000d, + 0x00040020,0x0000000f,0x00000002,0x0000000e,0x0004003b,0x0000000f,0x00000010,0x00000002, + 0x00040020,0x00000011,0x00000002,0x0000000d,0x00040017,0x00000014,0x00000006,0x00000003, + 0x00040020,0x00000015,0x00000001,0x00000014,0x0004003b,0x00000015,0x00000016,0x00000001, + 0x0004002b,0x00000006,0x00000018,0x3f800000,0x00040020,0x0000001e,0x00000003,0x00000007, + 0x0004003b,0x0000001e,0x00000020,0x00000003,0x0004003b,0x00000015,0x00000021,0x00000001, + 0x00050036,0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,0x00050041, + 0x00000011,0x00000012,0x00000010,0x0000000c,0x0004003d,0x0000000d,0x00000013,0x00000012, + 0x0004003d,0x00000014,0x00000017,0x00000016,0x00050051,0x00000006,0x00000019,0x00000017, + 0x00000000,0x00050051,0x00000006,0x0000001a,0x00000017,0x00000001,0x00050051,0x00000006, + 0x0000001b,0x00000017,0x00000002,0x00070050,0x00000007,0x0000001c,0x00000019,0x0000001a, + 0x0000001b,0x00000018,0x00050091,0x00000007,0x0000001d,0x00000013,0x0000001c,0x00050041, + 0x0000001e,0x0000001f,0x0000000a,0x0000000c,0x0003003e,0x0000001f,0x0000001d,0x0004003d, + 0x00000014,0x00000022,0x00000021,0x00050051,0x00000006,0x00000023,0x00000022,0x00000000, + 0x00050051,0x00000006,0x00000024,0x00000022,0x00000001,0x00050051,0x00000006,0x00000025, + 0x00000022,0x00000002,0x00070050,0x00000007,0x00000026,0x00000023,0x00000024,0x00000025, + 0x00000018,0x0003003e,0x00000020,0x00000026,0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/model_frag.c b/src/vk/spirv/model_frag.c new file mode 100644 index 0000000..953ca57 --- /dev/null +++ b/src/vk/spirv/model_frag.c @@ -0,0 +1,37 @@ + // 8.13.3559 + #pragma once +const uint32_t model_frag_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000028,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0009000f,0x00000004,0x00000004,0x6e69616d,0x00000000,0x00000008,0x00000012,0x0000001a, + 0x0000001e,0x00030010,0x00000004,0x00000007,0x00030003,0x00000002,0x000001c2,0x00040005, + 0x00000004,0x6e69616d,0x00000000,0x00050005,0x00000008,0x74786574,0x64657275,0x00000000, + 0x00060005,0x00000012,0x67617266,0x746e656d,0x6f6c6f43,0x00000072,0x00050005,0x00000016, + 0x78655473,0x65727574,0x00000000,0x00050005,0x0000001a,0x43786574,0x64726f6f,0x00000000, + 0x00040005,0x0000001e,0x6f6c6f63,0x00000072,0x00030047,0x00000008,0x0000000e,0x00040047, + 0x00000008,0x0000001e,0x00000002,0x00040047,0x00000012,0x0000001e,0x00000000,0x00040047, + 0x00000016,0x00000022,0x00000000,0x00040047,0x00000016,0x00000021,0x00000000,0x00040047, + 0x0000001a,0x0000001e,0x00000001,0x00040047,0x0000001e,0x0000001e,0x00000000,0x00020013, + 0x00000002,0x00030021,0x00000003,0x00000002,0x00040015,0x00000006,0x00000020,0x00000001, + 0x00040020,0x00000007,0x00000001,0x00000006,0x0004003b,0x00000007,0x00000008,0x00000001, + 0x0004002b,0x00000006,0x0000000a,0x00000000,0x00020014,0x0000000b,0x00030016,0x0000000f, + 0x00000020,0x00040017,0x00000010,0x0000000f,0x00000004,0x00040020,0x00000011,0x00000003, + 0x00000010,0x0004003b,0x00000011,0x00000012,0x00000003,0x00090019,0x00000013,0x0000000f, + 0x00000001,0x00000000,0x00000000,0x00000000,0x00000001,0x00000000,0x0003001b,0x00000014, + 0x00000013,0x00040020,0x00000015,0x00000000,0x00000014,0x0004003b,0x00000015,0x00000016, + 0x00000000,0x00040017,0x00000018,0x0000000f,0x00000002,0x00040020,0x00000019,0x00000001, + 0x00000018,0x0004003b,0x00000019,0x0000001a,0x00000001,0x00040020,0x0000001d,0x00000001, + 0x00000010,0x0004003b,0x0000001d,0x0000001e,0x00000001,0x0004002b,0x0000000f,0x00000020, + 0x00000000,0x0004002b,0x0000000f,0x00000021,0x3f800000,0x00050036,0x00000002,0x00000004, + 0x00000000,0x00000003,0x000200f8,0x00000005,0x0004003d,0x00000006,0x00000009,0x00000008, + 0x000500ab,0x0000000b,0x0000000c,0x00000009,0x0000000a,0x000300f7,0x0000000e,0x00000000, + 0x000400fa,0x0000000c,0x0000000d,0x00000026,0x000200f8,0x0000000d,0x0004003d,0x00000014, + 0x00000017,0x00000016,0x0004003d,0x00000018,0x0000001b,0x0000001a,0x00050057,0x00000010, + 0x0000001c,0x00000017,0x0000001b,0x0004003d,0x00000010,0x0000001f,0x0000001e,0x00070050, + 0x00000010,0x00000022,0x00000020,0x00000020,0x00000020,0x00000020,0x00070050,0x00000010, + 0x00000023,0x00000021,0x00000021,0x00000021,0x00000021,0x0008000c,0x00000010,0x00000024, + 0x00000001,0x0000002b,0x0000001f,0x00000022,0x00000023,0x00050085,0x00000010,0x00000025, + 0x0000001c,0x00000024,0x0003003e,0x00000012,0x00000025,0x000200f9,0x0000000e,0x000200f8, + 0x00000026,0x0004003d,0x00000010,0x00000027,0x0000001e,0x0003003e,0x00000012,0x00000027, + 0x000200f9,0x0000000e,0x000200f8,0x0000000e,0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/model_vert.c b/src/vk/spirv/model_vert.c new file mode 100644 index 0000000..86462f5 --- /dev/null +++ b/src/vk/spirv/model_vert.c @@ -0,0 +1,59 @@ + // 8.13.3559 + #pragma once +const uint32_t model_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000037,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x000c000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000a,0x0000001d,0x00000027, + 0x00000029,0x0000002d,0x0000002f,0x00000032,0x00030003,0x00000002,0x000001c2,0x00090004, + 0x415f4c47,0x735f4252,0x72617065,0x5f657461,0x64616873,0x6f5f7265,0x63656a62,0x00007374, + 0x00040005,0x00000004,0x6e69616d,0x00000000,0x00060005,0x00000008,0x505f6c67,0x65567265, + 0x78657472,0x00000000,0x00060006,0x00000008,0x00000000,0x505f6c67,0x7469736f,0x006e6f69, + 0x00030005,0x0000000a,0x00000000,0x00060005,0x0000000e,0x68737550,0x736e6f43,0x746e6174, + 0x00000000,0x00060006,0x0000000e,0x00000000,0x614d7076,0x78697274,0x00000000,0x00030005, + 0x00000010,0x00006370,0x00070005,0x00000014,0x66696e55,0x426d726f,0x65666675,0x6a624f72, + 0x00746365,0x00050006,0x00000014,0x00000000,0x65646f6d,0x0000006c,0x00060006,0x00000014, + 0x00000001,0x74786574,0x64657275,0x00000000,0x00030005,0x00000016,0x006f6275,0x00050005, + 0x0000001d,0x65566e69,0x78657472,0x00000000,0x00040005,0x00000027,0x6f6c6f63,0x00000072, + 0x00040005,0x00000029,0x6f436e69,0x00726f6c,0x00050005,0x0000002d,0x43786574,0x64726f6f, + 0x00000000,0x00050005,0x0000002f,0x65546e69,0x6f6f4378,0x00006472,0x00050005,0x00000032, + 0x74786574,0x64657275,0x00000000,0x00050048,0x00000008,0x00000000,0x0000000b,0x00000000, + 0x00030047,0x00000008,0x00000002,0x00040048,0x0000000e,0x00000000,0x00000005,0x00050048, + 0x0000000e,0x00000000,0x00000023,0x00000000,0x00050048,0x0000000e,0x00000000,0x00000007, + 0x00000010,0x00030047,0x0000000e,0x00000002,0x00040048,0x00000014,0x00000000,0x00000005, + 0x00050048,0x00000014,0x00000000,0x00000023,0x00000000,0x00050048,0x00000014,0x00000000, + 0x00000007,0x00000010,0x00050048,0x00000014,0x00000001,0x00000023,0x00000040,0x00030047, + 0x00000014,0x00000002,0x00040047,0x00000016,0x00000022,0x00000001,0x00040047,0x00000016, + 0x00000021,0x00000000,0x00040047,0x0000001d,0x0000001e,0x00000000,0x00040047,0x00000027, + 0x0000001e,0x00000000,0x00040047,0x00000029,0x0000001e,0x00000001,0x00040047,0x0000002d, + 0x0000001e,0x00000001,0x00040047,0x0000002f,0x0000001e,0x00000002,0x00040047,0x00000032, + 0x0000001e,0x00000002,0x00020013,0x00000002,0x00030021,0x00000003,0x00000002,0x00030016, + 0x00000006,0x00000020,0x00040017,0x00000007,0x00000006,0x00000004,0x0003001e,0x00000008, + 0x00000007,0x00040020,0x00000009,0x00000003,0x00000008,0x0004003b,0x00000009,0x0000000a, + 0x00000003,0x00040015,0x0000000b,0x00000020,0x00000001,0x0004002b,0x0000000b,0x0000000c, + 0x00000000,0x00040018,0x0000000d,0x00000007,0x00000004,0x0003001e,0x0000000e,0x0000000d, + 0x00040020,0x0000000f,0x00000009,0x0000000e,0x0004003b,0x0000000f,0x00000010,0x00000009, + 0x00040020,0x00000011,0x00000009,0x0000000d,0x0004001e,0x00000014,0x0000000d,0x0000000b, + 0x00040020,0x00000015,0x00000002,0x00000014,0x0004003b,0x00000015,0x00000016,0x00000002, + 0x00040020,0x00000017,0x00000002,0x0000000d,0x00040017,0x0000001b,0x00000006,0x00000003, + 0x00040020,0x0000001c,0x00000001,0x0000001b,0x0004003b,0x0000001c,0x0000001d,0x00000001, + 0x0004002b,0x00000006,0x0000001f,0x3f800000,0x00040020,0x00000025,0x00000003,0x00000007, + 0x0004003b,0x00000025,0x00000027,0x00000003,0x00040020,0x00000028,0x00000001,0x00000007, + 0x0004003b,0x00000028,0x00000029,0x00000001,0x00040017,0x0000002b,0x00000006,0x00000002, + 0x00040020,0x0000002c,0x00000003,0x0000002b,0x0004003b,0x0000002c,0x0000002d,0x00000003, + 0x00040020,0x0000002e,0x00000001,0x0000002b,0x0004003b,0x0000002e,0x0000002f,0x00000001, + 0x00040020,0x00000031,0x00000003,0x0000000b,0x0004003b,0x00000031,0x00000032,0x00000003, + 0x0004002b,0x0000000b,0x00000033,0x00000001,0x00040020,0x00000034,0x00000002,0x0000000b, + 0x00050036,0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,0x00050041, + 0x00000011,0x00000012,0x00000010,0x0000000c,0x0004003d,0x0000000d,0x00000013,0x00000012, + 0x00050041,0x00000017,0x00000018,0x00000016,0x0000000c,0x0004003d,0x0000000d,0x00000019, + 0x00000018,0x00050092,0x0000000d,0x0000001a,0x00000013,0x00000019,0x0004003d,0x0000001b, + 0x0000001e,0x0000001d,0x00050051,0x00000006,0x00000020,0x0000001e,0x00000000,0x00050051, + 0x00000006,0x00000021,0x0000001e,0x00000001,0x00050051,0x00000006,0x00000022,0x0000001e, + 0x00000002,0x00070050,0x00000007,0x00000023,0x00000020,0x00000021,0x00000022,0x0000001f, + 0x00050091,0x00000007,0x00000024,0x0000001a,0x00000023,0x00050041,0x00000025,0x00000026, + 0x0000000a,0x0000000c,0x0003003e,0x00000026,0x00000024,0x0004003d,0x00000007,0x0000002a, + 0x00000029,0x0003003e,0x00000027,0x0000002a,0x0004003d,0x0000002b,0x00000030,0x0000002f, + 0x0003003e,0x0000002d,0x00000030,0x00050041,0x00000034,0x00000035,0x00000016,0x00000033, + 0x0004003d,0x0000000b,0x00000036,0x00000035,0x0003003e,0x00000032,0x00000036,0x000100fd, + 0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/nullmodel_vert.c b/src/vk/spirv/nullmodel_vert.c new file mode 100644 index 0000000..b86cff4 --- /dev/null +++ b/src/vk/spirv/nullmodel_vert.c @@ -0,0 +1,49 @@ + // 8.13.3559 + #pragma once +const uint32_t nullmodel_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x0000002e,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0009000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000a,0x0000001d,0x00000027, + 0x00000028,0x00030003,0x00000002,0x000001c2,0x00090004,0x415f4c47,0x735f4252,0x72617065, + 0x5f657461,0x64616873,0x6f5f7265,0x63656a62,0x00007374,0x00040005,0x00000004,0x6e69616d, + 0x00000000,0x00060005,0x00000008,0x505f6c67,0x65567265,0x78657472,0x00000000,0x00060006, + 0x00000008,0x00000000,0x505f6c67,0x7469736f,0x006e6f69,0x00030005,0x0000000a,0x00000000, + 0x00060005,0x0000000e,0x68737550,0x736e6f43,0x746e6174,0x00000000,0x00060006,0x0000000e, + 0x00000000,0x614d7076,0x78697274,0x00000000,0x00030005,0x00000010,0x00006370,0x00070005, + 0x00000014,0x66696e55,0x426d726f,0x65666675,0x6a624f72,0x00746365,0x00050006,0x00000014, + 0x00000000,0x65646f6d,0x0000006c,0x00030005,0x00000016,0x006f6275,0x00050005,0x0000001d, + 0x65566e69,0x78657472,0x00000000,0x00040005,0x00000027,0x6f6c6f63,0x00000072,0x00040005, + 0x00000028,0x6f436e69,0x00726f6c,0x00050048,0x00000008,0x00000000,0x0000000b,0x00000000, + 0x00030047,0x00000008,0x00000002,0x00040048,0x0000000e,0x00000000,0x00000005,0x00050048, + 0x0000000e,0x00000000,0x00000023,0x00000000,0x00050048,0x0000000e,0x00000000,0x00000007, + 0x00000010,0x00030047,0x0000000e,0x00000002,0x00040048,0x00000014,0x00000000,0x00000005, + 0x00050048,0x00000014,0x00000000,0x00000023,0x00000000,0x00050048,0x00000014,0x00000000, + 0x00000007,0x00000010,0x00030047,0x00000014,0x00000002,0x00040047,0x00000016,0x00000022, + 0x00000000,0x00040047,0x00000016,0x00000021,0x00000000,0x00040047,0x0000001d,0x0000001e, + 0x00000000,0x00040047,0x00000027,0x0000001e,0x00000000,0x00040047,0x00000028,0x0000001e, + 0x00000001,0x00020013,0x00000002,0x00030021,0x00000003,0x00000002,0x00030016,0x00000006, + 0x00000020,0x00040017,0x00000007,0x00000006,0x00000004,0x0003001e,0x00000008,0x00000007, + 0x00040020,0x00000009,0x00000003,0x00000008,0x0004003b,0x00000009,0x0000000a,0x00000003, + 0x00040015,0x0000000b,0x00000020,0x00000001,0x0004002b,0x0000000b,0x0000000c,0x00000000, + 0x00040018,0x0000000d,0x00000007,0x00000004,0x0003001e,0x0000000e,0x0000000d,0x00040020, + 0x0000000f,0x00000009,0x0000000e,0x0004003b,0x0000000f,0x00000010,0x00000009,0x00040020, + 0x00000011,0x00000009,0x0000000d,0x0003001e,0x00000014,0x0000000d,0x00040020,0x00000015, + 0x00000002,0x00000014,0x0004003b,0x00000015,0x00000016,0x00000002,0x00040020,0x00000017, + 0x00000002,0x0000000d,0x00040017,0x0000001b,0x00000006,0x00000003,0x00040020,0x0000001c, + 0x00000001,0x0000001b,0x0004003b,0x0000001c,0x0000001d,0x00000001,0x0004002b,0x00000006, + 0x0000001f,0x3f800000,0x00040020,0x00000025,0x00000003,0x00000007,0x0004003b,0x00000025, + 0x00000027,0x00000003,0x0004003b,0x0000001c,0x00000028,0x00000001,0x00050036,0x00000002, + 0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,0x00050041,0x00000011,0x00000012, + 0x00000010,0x0000000c,0x0004003d,0x0000000d,0x00000013,0x00000012,0x00050041,0x00000017, + 0x00000018,0x00000016,0x0000000c,0x0004003d,0x0000000d,0x00000019,0x00000018,0x00050092, + 0x0000000d,0x0000001a,0x00000013,0x00000019,0x0004003d,0x0000001b,0x0000001e,0x0000001d, + 0x00050051,0x00000006,0x00000020,0x0000001e,0x00000000,0x00050051,0x00000006,0x00000021, + 0x0000001e,0x00000001,0x00050051,0x00000006,0x00000022,0x0000001e,0x00000002,0x00070050, + 0x00000007,0x00000023,0x00000020,0x00000021,0x00000022,0x0000001f,0x00050091,0x00000007, + 0x00000024,0x0000001a,0x00000023,0x00050041,0x00000025,0x00000026,0x0000000a,0x0000000c, + 0x0003003e,0x00000026,0x00000024,0x0004003d,0x0000001b,0x00000029,0x00000028,0x00050051, + 0x00000006,0x0000002a,0x00000029,0x00000000,0x00050051,0x00000006,0x0000002b,0x00000029, + 0x00000001,0x00050051,0x00000006,0x0000002c,0x00000029,0x00000002,0x00070050,0x00000007, + 0x0000002d,0x0000002a,0x0000002b,0x0000002c,0x0000001f,0x0003003e,0x00000027,0x0000002d, + 0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/particle_vert.c b/src/vk/spirv/particle_vert.c new file mode 100644 index 0000000..a0206ec --- /dev/null +++ b/src/vk/spirv/particle_vert.c @@ -0,0 +1,47 @@ + // 8.13.3743 + #pragma once +const uint32_t particle_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x0000002d,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x000c000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000a,0x00000016,0x00000022, + 0x00000024,0x00000026,0x00000028,0x0000002b,0x00030003,0x00000002,0x000001c2,0x00090004, + 0x415f4c47,0x735f4252,0x72617065,0x5f657461,0x64616873,0x6f5f7265,0x63656a62,0x00007374, + 0x00040005,0x00000004,0x6e69616d,0x00000000,0x00060005,0x00000008,0x505f6c67,0x65567265, + 0x78657472,0x00000000,0x00060006,0x00000008,0x00000000,0x505f6c67,0x7469736f,0x006e6f69, + 0x00030005,0x0000000a,0x00000000,0x00060005,0x0000000e,0x68737550,0x736e6f43,0x746e6174, + 0x00000000,0x00060006,0x0000000e,0x00000000,0x4d70766d,0x69727461,0x00000078,0x00030005, + 0x00000010,0x00006370,0x00050005,0x00000016,0x65566e69,0x78657472,0x00000000,0x00050005, + 0x00000022,0x43786574,0x64726f6f,0x00000000,0x00050005,0x00000024,0x65546e69,0x6f6f4378, + 0x00006472,0x00040005,0x00000026,0x6f6c6f63,0x00000072,0x00040005,0x00000028,0x6f436e69, + 0x00726f6c,0x00050005,0x0000002b,0x65725461,0x6c6f6873,0x00000064,0x00050048,0x00000008, + 0x00000000,0x0000000b,0x00000000,0x00030047,0x00000008,0x00000002,0x00040048,0x0000000e, + 0x00000000,0x00000005,0x00050048,0x0000000e,0x00000000,0x00000023,0x00000000,0x00050048, + 0x0000000e,0x00000000,0x00000007,0x00000010,0x00030047,0x0000000e,0x00000002,0x00040047, + 0x00000016,0x0000001e,0x00000000,0x00040047,0x00000022,0x0000001e,0x00000000,0x00040047, + 0x00000024,0x0000001e,0x00000002,0x00040047,0x00000026,0x0000001e,0x00000001,0x00040047, + 0x00000028,0x0000001e,0x00000001,0x00040047,0x0000002b,0x0000001e,0x00000002,0x00020013, + 0x00000002,0x00030021,0x00000003,0x00000002,0x00030016,0x00000006,0x00000020,0x00040017, + 0x00000007,0x00000006,0x00000004,0x0003001e,0x00000008,0x00000007,0x00040020,0x00000009, + 0x00000003,0x00000008,0x0004003b,0x00000009,0x0000000a,0x00000003,0x00040015,0x0000000b, + 0x00000020,0x00000001,0x0004002b,0x0000000b,0x0000000c,0x00000000,0x00040018,0x0000000d, + 0x00000007,0x00000004,0x0003001e,0x0000000e,0x0000000d,0x00040020,0x0000000f,0x00000009, + 0x0000000e,0x0004003b,0x0000000f,0x00000010,0x00000009,0x00040020,0x00000011,0x00000009, + 0x0000000d,0x00040017,0x00000014,0x00000006,0x00000003,0x00040020,0x00000015,0x00000001, + 0x00000014,0x0004003b,0x00000015,0x00000016,0x00000001,0x0004002b,0x00000006,0x00000018, + 0x3f800000,0x00040020,0x0000001e,0x00000003,0x00000007,0x00040017,0x00000020,0x00000006, + 0x00000002,0x00040020,0x00000021,0x00000003,0x00000020,0x0004003b,0x00000021,0x00000022, + 0x00000003,0x00040020,0x00000023,0x00000001,0x00000020,0x0004003b,0x00000023,0x00000024, + 0x00000001,0x0004003b,0x0000001e,0x00000026,0x00000003,0x00040020,0x00000027,0x00000001, + 0x00000007,0x0004003b,0x00000027,0x00000028,0x00000001,0x00040020,0x0000002a,0x00000003, + 0x00000006,0x0004003b,0x0000002a,0x0000002b,0x00000003,0x0004002b,0x00000006,0x0000002c, + 0x3f000000,0x00050036,0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005, + 0x00050041,0x00000011,0x00000012,0x00000010,0x0000000c,0x0004003d,0x0000000d,0x00000013, + 0x00000012,0x0004003d,0x00000014,0x00000017,0x00000016,0x00050051,0x00000006,0x00000019, + 0x00000017,0x00000000,0x00050051,0x00000006,0x0000001a,0x00000017,0x00000001,0x00050051, + 0x00000006,0x0000001b,0x00000017,0x00000002,0x00070050,0x00000007,0x0000001c,0x00000019, + 0x0000001a,0x0000001b,0x00000018,0x00050091,0x00000007,0x0000001d,0x00000013,0x0000001c, + 0x00050041,0x0000001e,0x0000001f,0x0000000a,0x0000000c,0x0003003e,0x0000001f,0x0000001d, + 0x0004003d,0x00000020,0x00000025,0x00000024,0x0003003e,0x00000022,0x00000025,0x0004003d, + 0x00000007,0x00000029,0x00000028,0x0003003e,0x00000026,0x00000029,0x0003003e,0x0000002b, + 0x0000002c,0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/point_particle_frag.c b/src/vk/spirv/point_particle_frag.c new file mode 100644 index 0000000..3dc1bc3 --- /dev/null +++ b/src/vk/spirv/point_particle_frag.c @@ -0,0 +1,28 @@ + // 8.13.3559 + #pragma once +const uint32_t point_particle_frag_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000020,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0008000f,0x00000004,0x00000004,0x6e69616d,0x00000000,0x0000000c,0x0000001c,0x0000001e, + 0x00030010,0x00000004,0x00000007,0x00030003,0x00000002,0x000001c2,0x00040005,0x00000004, + 0x6e69616d,0x00000000,0x00030005,0x00000009,0x00797863,0x00060005,0x0000000c,0x505f6c67, + 0x746e696f,0x726f6f43,0x00000064,0x00060005,0x0000001c,0x67617266,0x746e656d,0x6f6c6f43, + 0x00000072,0x00040005,0x0000001e,0x6f6c6f63,0x00000072,0x00040047,0x0000000c,0x0000000b, + 0x00000010,0x00040047,0x0000001c,0x0000001e,0x00000000,0x00040047,0x0000001e,0x0000001e, + 0x00000000,0x00020013,0x00000002,0x00030021,0x00000003,0x00000002,0x00030016,0x00000006, + 0x00000020,0x00040017,0x00000007,0x00000006,0x00000002,0x00040020,0x00000008,0x00000007, + 0x00000007,0x0004002b,0x00000006,0x0000000a,0x40000000,0x00040020,0x0000000b,0x00000001, + 0x00000007,0x0004003b,0x0000000b,0x0000000c,0x00000001,0x0004002b,0x00000006,0x0000000f, + 0x3f800000,0x00020014,0x00000015,0x00040017,0x0000001a,0x00000006,0x00000004,0x00040020, + 0x0000001b,0x00000003,0x0000001a,0x0004003b,0x0000001b,0x0000001c,0x00000003,0x00040020, + 0x0000001d,0x00000001,0x0000001a,0x0004003b,0x0000001d,0x0000001e,0x00000001,0x00050036, + 0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,0x0004003b,0x00000008, + 0x00000009,0x00000007,0x0004003d,0x00000007,0x0000000d,0x0000000c,0x0005008e,0x00000007, + 0x0000000e,0x0000000d,0x0000000a,0x00050050,0x00000007,0x00000010,0x0000000f,0x0000000f, + 0x00050083,0x00000007,0x00000011,0x0000000e,0x00000010,0x0003003e,0x00000009,0x00000011, + 0x0004003d,0x00000007,0x00000012,0x00000009,0x0004003d,0x00000007,0x00000013,0x00000009, + 0x00050094,0x00000006,0x00000014,0x00000012,0x00000013,0x000500ba,0x00000015,0x00000016, + 0x00000014,0x0000000f,0x000300f7,0x00000018,0x00000000,0x000400fa,0x00000016,0x00000017, + 0x00000018,0x000200f8,0x00000017,0x000100fc,0x000200f8,0x00000018,0x0004003d,0x0000001a, + 0x0000001f,0x0000001e,0x0003003e,0x0000001c,0x0000001f,0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/point_particle_vert.c b/src/vk/spirv/point_particle_vert.c new file mode 100644 index 0000000..d257aeb --- /dev/null +++ b/src/vk/spirv/point_particle_vert.c @@ -0,0 +1,82 @@ + // 8.13.3559 + #pragma once +const uint32_t point_particle_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000055,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0009000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000a,0x00000016,0x00000051, + 0x00000053,0x00030003,0x00000002,0x000001c2,0x00090004,0x415f4c47,0x735f4252,0x72617065, + 0x5f657461,0x64616873,0x6f5f7265,0x63656a62,0x00007374,0x00040005,0x00000004,0x6e69616d, + 0x00000000,0x00060005,0x00000008,0x505f6c67,0x65567265,0x78657472,0x00000000,0x00060006, + 0x00000008,0x00000000,0x505f6c67,0x7469736f,0x006e6f69,0x00070006,0x00000008,0x00000001, + 0x505f6c67,0x746e696f,0x657a6953,0x00000000,0x00030005,0x0000000a,0x00000000,0x00060005, + 0x0000000e,0x68737550,0x736e6f43,0x746e6174,0x00000000,0x00060006,0x0000000e,0x00000000, + 0x4d70766d,0x69727461,0x00000078,0x00030005,0x00000010,0x00006370,0x00050005,0x00000016, + 0x65566e69,0x78657472,0x00000000,0x00050005,0x00000021,0x74736964,0x7474615f,0x00006e65, + 0x00070005,0x00000022,0x66696e55,0x426d726f,0x65666675,0x6a624f72,0x00746365,0x00060006, + 0x00000022,0x00000000,0x6e696f70,0x7a695374,0x00000065,0x00060006,0x00000022,0x00000001, + 0x6e696f70,0x61635374,0x0000656c,0x00070006,0x00000022,0x00000002,0x506e696d,0x746e696f, + 0x657a6953,0x00000000,0x00070006,0x00000022,0x00000003,0x5078616d,0x746e696f,0x657a6953, + 0x00000000,0x00050006,0x00000022,0x00000004,0x5f747461,0x00000061,0x00050006,0x00000022, + 0x00000005,0x5f747461,0x00000062,0x00050006,0x00000022,0x00000006,0x5f747461,0x00000063, + 0x00030005,0x00000024,0x006f6275,0x00040005,0x00000051,0x6f6c6f63,0x00000072,0x00040005, + 0x00000053,0x6f436e69,0x00726f6c,0x00050048,0x00000008,0x00000000,0x0000000b,0x00000000, + 0x00050048,0x00000008,0x00000001,0x0000000b,0x00000001,0x00030047,0x00000008,0x00000002, + 0x00040048,0x0000000e,0x00000000,0x00000005,0x00050048,0x0000000e,0x00000000,0x00000023, + 0x00000000,0x00050048,0x0000000e,0x00000000,0x00000007,0x00000010,0x00030047,0x0000000e, + 0x00000002,0x00040047,0x00000016,0x0000001e,0x00000000,0x00050048,0x00000022,0x00000000, + 0x00000023,0x00000000,0x00050048,0x00000022,0x00000001,0x00000023,0x00000004,0x00050048, + 0x00000022,0x00000002,0x00000023,0x00000008,0x00050048,0x00000022,0x00000003,0x00000023, + 0x0000000c,0x00050048,0x00000022,0x00000004,0x00000023,0x00000010,0x00050048,0x00000022, + 0x00000005,0x00000023,0x00000014,0x00050048,0x00000022,0x00000006,0x00000023,0x00000018, + 0x00030047,0x00000022,0x00000002,0x00040047,0x00000024,0x00000022,0x00000000,0x00040047, + 0x00000024,0x00000021,0x00000000,0x00040047,0x00000051,0x0000001e,0x00000000,0x00040047, + 0x00000053,0x0000001e,0x00000001,0x00020013,0x00000002,0x00030021,0x00000003,0x00000002, + 0x00030016,0x00000006,0x00000020,0x00040017,0x00000007,0x00000006,0x00000004,0x0004001e, + 0x00000008,0x00000007,0x00000006,0x00040020,0x00000009,0x00000003,0x00000008,0x0004003b, + 0x00000009,0x0000000a,0x00000003,0x00040015,0x0000000b,0x00000020,0x00000001,0x0004002b, + 0x0000000b,0x0000000c,0x00000000,0x00040018,0x0000000d,0x00000007,0x00000004,0x0003001e, + 0x0000000e,0x0000000d,0x00040020,0x0000000f,0x00000009,0x0000000e,0x0004003b,0x0000000f, + 0x00000010,0x00000009,0x00040020,0x00000011,0x00000009,0x0000000d,0x00040017,0x00000014, + 0x00000006,0x00000003,0x00040020,0x00000015,0x00000001,0x00000014,0x0004003b,0x00000015, + 0x00000016,0x00000001,0x0004002b,0x00000006,0x00000018,0x3f800000,0x00040020,0x0000001e, + 0x00000003,0x00000007,0x00040020,0x00000020,0x00000007,0x00000006,0x0009001e,0x00000022, + 0x00000006,0x00000006,0x00000006,0x00000006,0x00000006,0x00000006,0x00000006,0x00040020, + 0x00000023,0x00000002,0x00000022,0x0004003b,0x00000023,0x00000024,0x00000002,0x0004002b, + 0x0000000b,0x00000025,0x00000001,0x00040020,0x00000026,0x00000002,0x00000006,0x0004002b, + 0x0000000b,0x00000029,0x00000004,0x0004002b,0x0000000b,0x0000002c,0x00000005,0x00040015, + 0x0000002f,0x00000020,0x00000000,0x0004002b,0x0000002f,0x00000030,0x00000003,0x00040020, + 0x00000031,0x00000003,0x00000006,0x0004002b,0x0000000b,0x00000036,0x00000006,0x0004002b, + 0x0000000b,0x00000049,0x00000002,0x0004002b,0x0000000b,0x0000004c,0x00000003,0x0004003b, + 0x0000001e,0x00000051,0x00000003,0x00040020,0x00000052,0x00000001,0x00000007,0x0004003b, + 0x00000052,0x00000053,0x00000001,0x00050036,0x00000002,0x00000004,0x00000000,0x00000003, + 0x000200f8,0x00000005,0x0004003b,0x00000020,0x00000021,0x00000007,0x00050041,0x00000011, + 0x00000012,0x00000010,0x0000000c,0x0004003d,0x0000000d,0x00000013,0x00000012,0x0004003d, + 0x00000014,0x00000017,0x00000016,0x00050051,0x00000006,0x00000019,0x00000017,0x00000000, + 0x00050051,0x00000006,0x0000001a,0x00000017,0x00000001,0x00050051,0x00000006,0x0000001b, + 0x00000017,0x00000002,0x00070050,0x00000007,0x0000001c,0x00000019,0x0000001a,0x0000001b, + 0x00000018,0x00050091,0x00000007,0x0000001d,0x00000013,0x0000001c,0x00050041,0x0000001e, + 0x0000001f,0x0000000a,0x0000000c,0x0003003e,0x0000001f,0x0000001d,0x00050041,0x00000026, + 0x00000027,0x00000024,0x00000025,0x0004003d,0x00000006,0x00000028,0x00000027,0x00050041, + 0x00000026,0x0000002a,0x00000024,0x00000029,0x0004003d,0x00000006,0x0000002b,0x0000002a, + 0x00050041,0x00000026,0x0000002d,0x00000024,0x0000002c,0x0004003d,0x00000006,0x0000002e, + 0x0000002d,0x00060041,0x00000031,0x00000032,0x0000000a,0x0000000c,0x00000030,0x0004003d, + 0x00000006,0x00000033,0x00000032,0x00050085,0x00000006,0x00000034,0x0000002e,0x00000033, + 0x00050081,0x00000006,0x00000035,0x0000002b,0x00000034,0x00050041,0x00000026,0x00000037, + 0x00000024,0x00000036,0x0004003d,0x00000006,0x00000038,0x00000037,0x00060041,0x00000031, + 0x00000039,0x0000000a,0x0000000c,0x00000030,0x0004003d,0x00000006,0x0000003a,0x00000039, + 0x00050085,0x00000006,0x0000003b,0x00000038,0x0000003a,0x00060041,0x00000031,0x0000003c, + 0x0000000a,0x0000000c,0x00000030,0x0004003d,0x00000006,0x0000003d,0x0000003c,0x00050085, + 0x00000006,0x0000003e,0x0000003b,0x0000003d,0x00050081,0x00000006,0x0000003f,0x00000035, + 0x0000003e,0x00050088,0x00000006,0x00000040,0x00000028,0x0000003f,0x0003003e,0x00000021, + 0x00000040,0x00050041,0x00000026,0x00000041,0x00000024,0x00000025,0x0004003d,0x00000006, + 0x00000042,0x00000041,0x00050041,0x00000026,0x00000043,0x00000024,0x0000000c,0x0004003d, + 0x00000006,0x00000044,0x00000043,0x00050085,0x00000006,0x00000045,0x00000042,0x00000044, + 0x0004003d,0x00000006,0x00000046,0x00000021,0x0006000c,0x00000006,0x00000047,0x00000001, + 0x0000001f,0x00000046,0x00050085,0x00000006,0x00000048,0x00000045,0x00000047,0x00050041, + 0x00000026,0x0000004a,0x00000024,0x00000049,0x0004003d,0x00000006,0x0000004b,0x0000004a, + 0x00050041,0x00000026,0x0000004d,0x00000024,0x0000004c,0x0004003d,0x00000006,0x0000004e, + 0x0000004d,0x0008000c,0x00000006,0x0000004f,0x00000001,0x0000002b,0x00000048,0x0000004b, + 0x0000004e,0x00050041,0x00000031,0x00000050,0x0000000a,0x00000025,0x0003003e,0x00000050, + 0x0000004f,0x0004003d,0x00000007,0x00000054,0x00000053,0x0003003e,0x00000051,0x00000054, + 0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/polygon_lmap_frag.c b/src/vk/spirv/polygon_lmap_frag.c new file mode 100644 index 0000000..5673000 --- /dev/null +++ b/src/vk/spirv/polygon_lmap_frag.c @@ -0,0 +1,40 @@ + // 8.13.3559 + #pragma once +const uint32_t polygon_lmap_frag_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000029,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0009000f,0x00000004,0x00000004,0x6e69616d,0x00000000,0x00000011,0x00000017,0x0000001b, + 0x0000001e,0x00030010,0x00000004,0x00000007,0x00030003,0x00000002,0x000001c2,0x00040005, + 0x00000004,0x6e69616d,0x00000000,0x00040005,0x00000009,0x6f6c6f63,0x00000072,0x00050005, + 0x0000000d,0x78655473,0x65727574,0x00000000,0x00050005,0x00000011,0x43786574,0x64726f6f, + 0x00000000,0x00040005,0x00000014,0x6867696c,0x00000074,0x00050005,0x00000015,0x67694c73, + 0x616d7468,0x00000070,0x00060005,0x00000017,0x43786574,0x64726f6f,0x70616d4c,0x00000000, + 0x00060005,0x0000001b,0x67617266,0x746e656d,0x6f6c6f43,0x00000072,0x00060005,0x0000001e, + 0x77656976,0x6867694c,0x70616d74,0x00000073,0x00040047,0x0000000d,0x00000022,0x00000000, + 0x00040047,0x0000000d,0x00000021,0x00000000,0x00040047,0x00000011,0x0000001e,0x00000000, + 0x00040047,0x00000015,0x00000022,0x00000002,0x00040047,0x00000015,0x00000021,0x00000000, + 0x00040047,0x00000017,0x0000001e,0x00000001,0x00040047,0x0000001b,0x0000001e,0x00000000, + 0x00040047,0x0000001e,0x0000001e,0x00000002,0x00020013,0x00000002,0x00030021,0x00000003, + 0x00000002,0x00030016,0x00000006,0x00000020,0x00040017,0x00000007,0x00000006,0x00000004, + 0x00040020,0x00000008,0x00000007,0x00000007,0x00090019,0x0000000a,0x00000006,0x00000001, + 0x00000000,0x00000000,0x00000000,0x00000001,0x00000000,0x0003001b,0x0000000b,0x0000000a, + 0x00040020,0x0000000c,0x00000000,0x0000000b,0x0004003b,0x0000000c,0x0000000d,0x00000000, + 0x00040017,0x0000000f,0x00000006,0x00000002,0x00040020,0x00000010,0x00000001,0x0000000f, + 0x0004003b,0x00000010,0x00000011,0x00000001,0x0004003b,0x0000000c,0x00000015,0x00000000, + 0x0004003b,0x00000010,0x00000017,0x00000001,0x00040020,0x0000001a,0x00000003,0x00000007, + 0x0004003b,0x0000001a,0x0000001b,0x00000003,0x0004002b,0x00000006,0x0000001c,0x3f800000, + 0x00040020,0x0000001d,0x00000001,0x00000006,0x0004003b,0x0000001d,0x0000001e,0x00000001, + 0x00050036,0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,0x0004003b, + 0x00000008,0x00000009,0x00000007,0x0004003b,0x00000008,0x00000014,0x00000007,0x0004003d, + 0x0000000b,0x0000000e,0x0000000d,0x0004003d,0x0000000f,0x00000012,0x00000011,0x00050057, + 0x00000007,0x00000013,0x0000000e,0x00000012,0x0003003e,0x00000009,0x00000013,0x0004003d, + 0x0000000b,0x00000016,0x00000015,0x0004003d,0x0000000f,0x00000018,0x00000017,0x00050057, + 0x00000007,0x00000019,0x00000016,0x00000018,0x0003003e,0x00000014,0x00000019,0x0004003d, + 0x00000006,0x0000001f,0x0000001e,0x00050083,0x00000006,0x00000020,0x0000001c,0x0000001f, + 0x0004003d,0x00000007,0x00000021,0x00000009,0x0005008e,0x00000007,0x00000022,0x00000021, + 0x00000020,0x0004003d,0x00000007,0x00000023,0x00000014,0x00050085,0x00000007,0x00000024, + 0x00000022,0x00000023,0x0004003d,0x00000006,0x00000025,0x0000001e,0x0004003d,0x00000007, + 0x00000026,0x00000014,0x0005008e,0x00000007,0x00000027,0x00000026,0x00000025,0x00050081, + 0x00000007,0x00000028,0x00000024,0x00000027,0x0003003e,0x0000001b,0x00000028,0x000100fd, + 0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/polygon_lmap_vert.c b/src/vk/spirv/polygon_lmap_vert.c new file mode 100644 index 0000000..4e66eeb --- /dev/null +++ b/src/vk/spirv/polygon_lmap_vert.c @@ -0,0 +1,59 @@ + // 8.13.3559 + #pragma once +const uint32_t polygon_lmap_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000036,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x000c000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000a,0x0000001d,0x00000029, + 0x0000002b,0x0000002d,0x0000002e,0x00000031,0x00030003,0x00000002,0x000001c2,0x00090004, + 0x415f4c47,0x735f4252,0x72617065,0x5f657461,0x64616873,0x6f5f7265,0x63656a62,0x00007374, + 0x00040005,0x00000004,0x6e69616d,0x00000000,0x00060005,0x00000008,0x505f6c67,0x65567265, + 0x78657472,0x00000000,0x00060006,0x00000008,0x00000000,0x505f6c67,0x7469736f,0x006e6f69, + 0x00030005,0x0000000a,0x00000000,0x00060005,0x0000000e,0x68737550,0x736e6f43,0x746e6174, + 0x00000000,0x00060006,0x0000000e,0x00000000,0x614d7076,0x78697274,0x00000000,0x00030005, + 0x00000010,0x00006370,0x00070005,0x00000014,0x66696e55,0x426d726f,0x65666675,0x6a624f72, + 0x00746365,0x00050006,0x00000014,0x00000000,0x65646f6d,0x0000006c,0x00070006,0x00000014, + 0x00000001,0x77656976,0x6867694c,0x70616d74,0x00000073,0x00030005,0x00000016,0x006f6275, + 0x00050005,0x0000001d,0x65566e69,0x78657472,0x00000000,0x00050005,0x00000029,0x43786574, + 0x64726f6f,0x00000000,0x00050005,0x0000002b,0x65546e69,0x6f6f4378,0x00006472,0x00060005, + 0x0000002d,0x43786574,0x64726f6f,0x70616d4c,0x00000000,0x00060005,0x0000002e,0x65546e69, + 0x6f6f4378,0x6d4c6472,0x00007061,0x00060005,0x00000031,0x77656976,0x6867694c,0x70616d74, + 0x00000073,0x00050048,0x00000008,0x00000000,0x0000000b,0x00000000,0x00030047,0x00000008, + 0x00000002,0x00040048,0x0000000e,0x00000000,0x00000005,0x00050048,0x0000000e,0x00000000, + 0x00000023,0x00000000,0x00050048,0x0000000e,0x00000000,0x00000007,0x00000010,0x00030047, + 0x0000000e,0x00000002,0x00040048,0x00000014,0x00000000,0x00000005,0x00050048,0x00000014, + 0x00000000,0x00000023,0x00000000,0x00050048,0x00000014,0x00000000,0x00000007,0x00000010, + 0x00050048,0x00000014,0x00000001,0x00000023,0x00000040,0x00030047,0x00000014,0x00000002, + 0x00040047,0x00000016,0x00000022,0x00000001,0x00040047,0x00000016,0x00000021,0x00000000, + 0x00040047,0x0000001d,0x0000001e,0x00000000,0x00040047,0x00000029,0x0000001e,0x00000000, + 0x00040047,0x0000002b,0x0000001e,0x00000001,0x00040047,0x0000002d,0x0000001e,0x00000001, + 0x00040047,0x0000002e,0x0000001e,0x00000002,0x00040047,0x00000031,0x0000001e,0x00000002, + 0x00020013,0x00000002,0x00030021,0x00000003,0x00000002,0x00030016,0x00000006,0x00000020, + 0x00040017,0x00000007,0x00000006,0x00000004,0x0003001e,0x00000008,0x00000007,0x00040020, + 0x00000009,0x00000003,0x00000008,0x0004003b,0x00000009,0x0000000a,0x00000003,0x00040015, + 0x0000000b,0x00000020,0x00000001,0x0004002b,0x0000000b,0x0000000c,0x00000000,0x00040018, + 0x0000000d,0x00000007,0x00000004,0x0003001e,0x0000000e,0x0000000d,0x00040020,0x0000000f, + 0x00000009,0x0000000e,0x0004003b,0x0000000f,0x00000010,0x00000009,0x00040020,0x00000011, + 0x00000009,0x0000000d,0x0004001e,0x00000014,0x0000000d,0x00000006,0x00040020,0x00000015, + 0x00000002,0x00000014,0x0004003b,0x00000015,0x00000016,0x00000002,0x00040020,0x00000017, + 0x00000002,0x0000000d,0x00040017,0x0000001b,0x00000006,0x00000003,0x00040020,0x0000001c, + 0x00000001,0x0000001b,0x0004003b,0x0000001c,0x0000001d,0x00000001,0x0004002b,0x00000006, + 0x0000001f,0x3f800000,0x00040020,0x00000025,0x00000003,0x00000007,0x00040017,0x00000027, + 0x00000006,0x00000002,0x00040020,0x00000028,0x00000003,0x00000027,0x0004003b,0x00000028, + 0x00000029,0x00000003,0x00040020,0x0000002a,0x00000001,0x00000027,0x0004003b,0x0000002a, + 0x0000002b,0x00000001,0x0004003b,0x00000028,0x0000002d,0x00000003,0x0004003b,0x0000002a, + 0x0000002e,0x00000001,0x00040020,0x00000030,0x00000003,0x00000006,0x0004003b,0x00000030, + 0x00000031,0x00000003,0x0004002b,0x0000000b,0x00000032,0x00000001,0x00040020,0x00000033, + 0x00000002,0x00000006,0x00050036,0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8, + 0x00000005,0x00050041,0x00000011,0x00000012,0x00000010,0x0000000c,0x0004003d,0x0000000d, + 0x00000013,0x00000012,0x00050041,0x00000017,0x00000018,0x00000016,0x0000000c,0x0004003d, + 0x0000000d,0x00000019,0x00000018,0x00050092,0x0000000d,0x0000001a,0x00000013,0x00000019, + 0x0004003d,0x0000001b,0x0000001e,0x0000001d,0x00050051,0x00000006,0x00000020,0x0000001e, + 0x00000000,0x00050051,0x00000006,0x00000021,0x0000001e,0x00000001,0x00050051,0x00000006, + 0x00000022,0x0000001e,0x00000002,0x00070050,0x00000007,0x00000023,0x00000020,0x00000021, + 0x00000022,0x0000001f,0x00050091,0x00000007,0x00000024,0x0000001a,0x00000023,0x00050041, + 0x00000025,0x00000026,0x0000000a,0x0000000c,0x0003003e,0x00000026,0x00000024,0x0004003d, + 0x00000027,0x0000002c,0x0000002b,0x0003003e,0x00000029,0x0000002c,0x0004003d,0x00000027, + 0x0000002f,0x0000002e,0x0003003e,0x0000002d,0x0000002f,0x00050041,0x00000033,0x00000034, + 0x00000016,0x00000032,0x0004003d,0x00000006,0x00000035,0x00000034,0x0003003e,0x00000031, + 0x00000035,0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/polygon_vert.c b/src/vk/spirv/polygon_vert.c new file mode 100644 index 0000000..e7bc662 --- /dev/null +++ b/src/vk/spirv/polygon_vert.c @@ -0,0 +1,51 @@ + // 8.13.3559 + #pragma once +const uint32_t polygon_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000030,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x000b000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000a,0x00000016,0x00000022, + 0x00000024,0x00000026,0x0000002e,0x00030003,0x00000002,0x000001c2,0x00090004,0x415f4c47, + 0x735f4252,0x72617065,0x5f657461,0x64616873,0x6f5f7265,0x63656a62,0x00007374,0x00040005, + 0x00000004,0x6e69616d,0x00000000,0x00060005,0x00000008,0x505f6c67,0x65567265,0x78657472, + 0x00000000,0x00060006,0x00000008,0x00000000,0x505f6c67,0x7469736f,0x006e6f69,0x00030005, + 0x0000000a,0x00000000,0x00060005,0x0000000e,0x68737550,0x736e6f43,0x746e6174,0x00000000, + 0x00060006,0x0000000e,0x00000000,0x4d70766d,0x69727461,0x00000078,0x00030005,0x00000010, + 0x00006370,0x00050005,0x00000016,0x65566e69,0x78657472,0x00000000,0x00050005,0x00000022, + 0x43786574,0x64726f6f,0x00000000,0x00050005,0x00000024,0x65546e69,0x6f6f4378,0x00006472, + 0x00040005,0x00000026,0x6f6c6f63,0x00000072,0x00070005,0x00000027,0x66696e55,0x426d726f, + 0x65666675,0x6a624f72,0x00746365,0x00050006,0x00000027,0x00000000,0x6f6c6f63,0x00000072, + 0x00030005,0x00000029,0x006f6275,0x00050005,0x0000002e,0x65725461,0x6c6f6873,0x00000064, + 0x00050048,0x00000008,0x00000000,0x0000000b,0x00000000,0x00030047,0x00000008,0x00000002, + 0x00040048,0x0000000e,0x00000000,0x00000005,0x00050048,0x0000000e,0x00000000,0x00000023, + 0x00000000,0x00050048,0x0000000e,0x00000000,0x00000007,0x00000010,0x00030047,0x0000000e, + 0x00000002,0x00040047,0x00000016,0x0000001e,0x00000000,0x00040047,0x00000022,0x0000001e, + 0x00000000,0x00040047,0x00000024,0x0000001e,0x00000001,0x00040047,0x00000026,0x0000001e, + 0x00000001,0x00050048,0x00000027,0x00000000,0x00000023,0x00000000,0x00030047,0x00000027, + 0x00000002,0x00040047,0x00000029,0x00000022,0x00000001,0x00040047,0x00000029,0x00000021, + 0x00000000,0x00040047,0x0000002e,0x0000001e,0x00000002,0x00020013,0x00000002,0x00030021, + 0x00000003,0x00000002,0x00030016,0x00000006,0x00000020,0x00040017,0x00000007,0x00000006, + 0x00000004,0x0003001e,0x00000008,0x00000007,0x00040020,0x00000009,0x00000003,0x00000008, + 0x0004003b,0x00000009,0x0000000a,0x00000003,0x00040015,0x0000000b,0x00000020,0x00000001, + 0x0004002b,0x0000000b,0x0000000c,0x00000000,0x00040018,0x0000000d,0x00000007,0x00000004, + 0x0003001e,0x0000000e,0x0000000d,0x00040020,0x0000000f,0x00000009,0x0000000e,0x0004003b, + 0x0000000f,0x00000010,0x00000009,0x00040020,0x00000011,0x00000009,0x0000000d,0x00040017, + 0x00000014,0x00000006,0x00000003,0x00040020,0x00000015,0x00000001,0x00000014,0x0004003b, + 0x00000015,0x00000016,0x00000001,0x0004002b,0x00000006,0x00000018,0x3f800000,0x00040020, + 0x0000001e,0x00000003,0x00000007,0x00040017,0x00000020,0x00000006,0x00000002,0x00040020, + 0x00000021,0x00000003,0x00000020,0x0004003b,0x00000021,0x00000022,0x00000003,0x00040020, + 0x00000023,0x00000001,0x00000020,0x0004003b,0x00000023,0x00000024,0x00000001,0x0004003b, + 0x0000001e,0x00000026,0x00000003,0x0003001e,0x00000027,0x00000007,0x00040020,0x00000028, + 0x00000002,0x00000027,0x0004003b,0x00000028,0x00000029,0x00000002,0x00040020,0x0000002a, + 0x00000002,0x00000007,0x00040020,0x0000002d,0x00000003,0x00000006,0x0004003b,0x0000002d, + 0x0000002e,0x00000003,0x0004002b,0x00000006,0x0000002f,0x00000000,0x00050036,0x00000002, + 0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,0x00050041,0x00000011,0x00000012, + 0x00000010,0x0000000c,0x0004003d,0x0000000d,0x00000013,0x00000012,0x0004003d,0x00000014, + 0x00000017,0x00000016,0x00050051,0x00000006,0x00000019,0x00000017,0x00000000,0x00050051, + 0x00000006,0x0000001a,0x00000017,0x00000001,0x00050051,0x00000006,0x0000001b,0x00000017, + 0x00000002,0x00070050,0x00000007,0x0000001c,0x00000019,0x0000001a,0x0000001b,0x00000018, + 0x00050091,0x00000007,0x0000001d,0x00000013,0x0000001c,0x00050041,0x0000001e,0x0000001f, + 0x0000000a,0x0000000c,0x0003003e,0x0000001f,0x0000001d,0x0004003d,0x00000020,0x00000025, + 0x00000024,0x0003003e,0x00000022,0x00000025,0x00050041,0x0000002a,0x0000002b,0x00000029, + 0x0000000c,0x0004003d,0x00000007,0x0000002c,0x0000002b,0x0003003e,0x00000026,0x0000002c, + 0x0003003e,0x0000002e,0x0000002f,0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/polygon_warp_vert.c b/src/vk/spirv/polygon_warp_vert.c new file mode 100644 index 0000000..b1127ae --- /dev/null +++ b/src/vk/spirv/polygon_warp_vert.c @@ -0,0 +1,80 @@ + // 8.13.3559 + #pragma once +const uint32_t polygon_warp_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000058,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x000b000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000a,0x0000001d,0x00000029, + 0x0000002b,0x00000051,0x00000056,0x00030003,0x00000002,0x000001c2,0x00090004,0x415f4c47, + 0x735f4252,0x72617065,0x5f657461,0x64616873,0x6f5f7265,0x63656a62,0x00007374,0x00040005, + 0x00000004,0x6e69616d,0x00000000,0x00060005,0x00000008,0x505f6c67,0x65567265,0x78657472, + 0x00000000,0x00060006,0x00000008,0x00000000,0x505f6c67,0x7469736f,0x006e6f69,0x00030005, + 0x0000000a,0x00000000,0x00060005,0x0000000e,0x68737550,0x736e6f43,0x746e6174,0x00000000, + 0x00060006,0x0000000e,0x00000000,0x614d7076,0x78697274,0x00000000,0x00030005,0x00000010, + 0x00006370,0x00070005,0x00000014,0x66696e55,0x426d726f,0x65666675,0x6a624f72,0x00746365, + 0x00050006,0x00000014,0x00000000,0x65646f6d,0x0000006c,0x00050006,0x00000014,0x00000001, + 0x6f6c6f63,0x00000072,0x00050006,0x00000014,0x00000002,0x656d6974,0x00000000,0x00050006, + 0x00000014,0x00000003,0x6f726373,0x00006c6c,0x00030005,0x00000016,0x006f6275,0x00050005, + 0x0000001d,0x65566e69,0x78657472,0x00000000,0x00050005,0x00000029,0x43786574,0x64726f6f, + 0x00000000,0x00050005,0x0000002b,0x65546e69,0x6f6f4378,0x00006472,0x00040005,0x00000051, + 0x6f6c6f63,0x00000072,0x00050005,0x00000056,0x65725461,0x6c6f6873,0x00000064,0x00050048, + 0x00000008,0x00000000,0x0000000b,0x00000000,0x00030047,0x00000008,0x00000002,0x00040048, + 0x0000000e,0x00000000,0x00000005,0x00050048,0x0000000e,0x00000000,0x00000023,0x00000000, + 0x00050048,0x0000000e,0x00000000,0x00000007,0x00000010,0x00030047,0x0000000e,0x00000002, + 0x00040048,0x00000014,0x00000000,0x00000005,0x00050048,0x00000014,0x00000000,0x00000023, + 0x00000000,0x00050048,0x00000014,0x00000000,0x00000007,0x00000010,0x00050048,0x00000014, + 0x00000001,0x00000023,0x00000040,0x00050048,0x00000014,0x00000002,0x00000023,0x00000050, + 0x00050048,0x00000014,0x00000003,0x00000023,0x00000054,0x00030047,0x00000014,0x00000002, + 0x00040047,0x00000016,0x00000022,0x00000001,0x00040047,0x00000016,0x00000021,0x00000000, + 0x00040047,0x0000001d,0x0000001e,0x00000000,0x00040047,0x00000029,0x0000001e,0x00000000, + 0x00040047,0x0000002b,0x0000001e,0x00000001,0x00040047,0x00000051,0x0000001e,0x00000001, + 0x00040047,0x00000056,0x0000001e,0x00000002,0x00020013,0x00000002,0x00030021,0x00000003, + 0x00000002,0x00030016,0x00000006,0x00000020,0x00040017,0x00000007,0x00000006,0x00000004, + 0x0003001e,0x00000008,0x00000007,0x00040020,0x00000009,0x00000003,0x00000008,0x0004003b, + 0x00000009,0x0000000a,0x00000003,0x00040015,0x0000000b,0x00000020,0x00000001,0x0004002b, + 0x0000000b,0x0000000c,0x00000000,0x00040018,0x0000000d,0x00000007,0x00000004,0x0003001e, + 0x0000000e,0x0000000d,0x00040020,0x0000000f,0x00000009,0x0000000e,0x0004003b,0x0000000f, + 0x00000010,0x00000009,0x00040020,0x00000011,0x00000009,0x0000000d,0x0006001e,0x00000014, + 0x0000000d,0x00000007,0x00000006,0x00000006,0x00040020,0x00000015,0x00000002,0x00000014, + 0x0004003b,0x00000015,0x00000016,0x00000002,0x00040020,0x00000017,0x00000002,0x0000000d, + 0x00040017,0x0000001b,0x00000006,0x00000003,0x00040020,0x0000001c,0x00000001,0x0000001b, + 0x0004003b,0x0000001c,0x0000001d,0x00000001,0x0004002b,0x00000006,0x0000001f,0x3f800000, + 0x00040020,0x00000025,0x00000003,0x00000007,0x00040017,0x00000027,0x00000006,0x00000002, + 0x00040020,0x00000028,0x00000003,0x00000027,0x0004003b,0x00000028,0x00000029,0x00000003, + 0x00040020,0x0000002a,0x00000001,0x00000027,0x0004003b,0x0000002a,0x0000002b,0x00000001, + 0x0004002b,0x00000006,0x0000002d,0x40000000,0x0004002b,0x0000000b,0x0000002e,0x00000002, + 0x00040020,0x0000002f,0x00000002,0x00000006,0x00040015,0x00000033,0x00000020,0x00000000, + 0x0004002b,0x00000033,0x00000034,0x00000001,0x00040020,0x00000035,0x00000001,0x00000006, + 0x0004002b,0x00000006,0x00000038,0x4051eb85,0x0004002b,0x00000033,0x0000003f,0x00000000, + 0x0004002b,0x00000006,0x00000046,0x3d4ccccd,0x0004002b,0x0000000b,0x00000049,0x00000003, + 0x00040020,0x0000004c,0x00000003,0x00000006,0x0004003b,0x00000025,0x00000051,0x00000003, + 0x0004002b,0x0000000b,0x00000052,0x00000001,0x00040020,0x00000053,0x00000002,0x00000007, + 0x0004003b,0x0000004c,0x00000056,0x00000003,0x0004002b,0x00000006,0x00000057,0x00000000, + 0x00050036,0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,0x00050041, + 0x00000011,0x00000012,0x00000010,0x0000000c,0x0004003d,0x0000000d,0x00000013,0x00000012, + 0x00050041,0x00000017,0x00000018,0x00000016,0x0000000c,0x0004003d,0x0000000d,0x00000019, + 0x00000018,0x00050092,0x0000000d,0x0000001a,0x00000013,0x00000019,0x0004003d,0x0000001b, + 0x0000001e,0x0000001d,0x00050051,0x00000006,0x00000020,0x0000001e,0x00000000,0x00050051, + 0x00000006,0x00000021,0x0000001e,0x00000001,0x00050051,0x00000006,0x00000022,0x0000001e, + 0x00000002,0x00070050,0x00000007,0x00000023,0x00000020,0x00000021,0x00000022,0x0000001f, + 0x00050091,0x00000007,0x00000024,0x0000001a,0x00000023,0x00050041,0x00000025,0x00000026, + 0x0000000a,0x0000000c,0x0003003e,0x00000026,0x00000024,0x0004003d,0x00000027,0x0000002c, + 0x0000002b,0x00050041,0x0000002f,0x00000030,0x00000016,0x0000002e,0x0004003d,0x00000006, + 0x00000031,0x00000030,0x00050085,0x00000006,0x00000032,0x0000002d,0x00000031,0x00050041, + 0x00000035,0x00000036,0x0000002b,0x00000034,0x0004003d,0x00000006,0x00000037,0x00000036, + 0x00050085,0x00000006,0x00000039,0x00000037,0x00000038,0x00050081,0x00000006,0x0000003a, + 0x00000032,0x00000039,0x0006000c,0x00000006,0x0000003b,0x00000001,0x0000000d,0x0000003a, + 0x00050041,0x0000002f,0x0000003c,0x00000016,0x0000002e,0x0004003d,0x00000006,0x0000003d, + 0x0000003c,0x00050085,0x00000006,0x0000003e,0x0000002d,0x0000003d,0x00050041,0x00000035, + 0x00000040,0x0000002b,0x0000003f,0x0004003d,0x00000006,0x00000041,0x00000040,0x00050085, + 0x00000006,0x00000042,0x00000041,0x00000038,0x00050081,0x00000006,0x00000043,0x0000003e, + 0x00000042,0x0006000c,0x00000006,0x00000044,0x00000001,0x0000000d,0x00000043,0x00050050, + 0x00000027,0x00000045,0x0000003b,0x00000044,0x0005008e,0x00000027,0x00000047,0x00000045, + 0x00000046,0x00050081,0x00000027,0x00000048,0x0000002c,0x00000047,0x0003003e,0x00000029, + 0x00000048,0x00050041,0x0000002f,0x0000004a,0x00000016,0x00000049,0x0004003d,0x00000006, + 0x0000004b,0x0000004a,0x00050041,0x0000004c,0x0000004d,0x00000029,0x0000003f,0x0004003d, + 0x00000006,0x0000004e,0x0000004d,0x00050081,0x00000006,0x0000004f,0x0000004e,0x0000004b, + 0x00050041,0x0000004c,0x00000050,0x00000029,0x0000003f,0x0003003e,0x00000050,0x0000004f, + 0x00050041,0x00000053,0x00000054,0x00000016,0x00000052,0x0004003d,0x00000007,0x00000055, + 0x00000054,0x0003003e,0x00000051,0x00000055,0x0003003e,0x00000056,0x00000057,0x000100fd, + 0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/postprocess_frag.c b/src/vk/spirv/postprocess_frag.c new file mode 100644 index 0000000..7196d32 --- /dev/null +++ b/src/vk/spirv/postprocess_frag.c @@ -0,0 +1,64 @@ + // 1011.0.0 + #pragma once +const uint32_t postprocess_frag_spv[] = { + 0x07230203,0x00010000,0x0008000a,0x00000046,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0007000f,0x00000004,0x00000004,0x6e69616d,0x00000000,0x0000000b,0x0000002c,0x00030010, + 0x00000004,0x00000007,0x00030003,0x00000002,0x000001c2,0x00090004,0x415f4c47,0x735f4252, + 0x72617065,0x5f657461,0x64616873,0x6f5f7265,0x63656a62,0x00007374,0x00040005,0x00000004, + 0x6e69616d,0x00000000,0x00060005,0x00000009,0x6f6e6e75,0x65546d72,0x6f6f4378,0x00006472, + 0x00050005,0x0000000b,0x43786574,0x64726f6f,0x00000000,0x00060005,0x0000000d,0x68737550, + 0x736e6f43,0x746e6174,0x00000000,0x00060006,0x0000000d,0x00000000,0x74736f70,0x636f7270, + 0x00737365,0x00050006,0x0000000d,0x00000001,0x6d6d6167,0x00000061,0x00060006,0x0000000d, + 0x00000002,0x57726373,0x68746469,0x00000000,0x00060006,0x0000000d,0x00000003,0x48726373, + 0x68676965,0x00000074,0x00050006,0x0000000d,0x00000004,0x7366666f,0x00587465,0x00050006, + 0x0000000d,0x00000005,0x7366666f,0x00597465,0x00030005,0x0000000f,0x00006370,0x00060005, + 0x0000002c,0x67617266,0x746e656d,0x6f6c6f43,0x00000072,0x00050005,0x00000030,0x78655473, + 0x65727574,0x00000000,0x00040047,0x0000000b,0x0000001e,0x00000000,0x00050048,0x0000000d, + 0x00000000,0x00000023,0x00000044,0x00050048,0x0000000d,0x00000001,0x00000023,0x00000048, + 0x00050048,0x0000000d,0x00000002,0x00000023,0x0000004c,0x00050048,0x0000000d,0x00000003, + 0x00000023,0x00000050,0x00050048,0x0000000d,0x00000004,0x00000023,0x00000054,0x00050048, + 0x0000000d,0x00000005,0x00000023,0x00000058,0x00030047,0x0000000d,0x00000002,0x00040047, + 0x0000002c,0x0000001e,0x00000000,0x00040047,0x00000030,0x00000022,0x00000000,0x00040047, + 0x00000030,0x00000021,0x00000000,0x00020013,0x00000002,0x00030021,0x00000003,0x00000002, + 0x00030016,0x00000006,0x00000020,0x00040017,0x00000007,0x00000006,0x00000002,0x00040020, + 0x00000008,0x00000007,0x00000007,0x00040020,0x0000000a,0x00000001,0x00000007,0x0004003b, + 0x0000000a,0x0000000b,0x00000001,0x0008001e,0x0000000d,0x00000006,0x00000006,0x00000006, + 0x00000006,0x00000006,0x00000006,0x00040020,0x0000000e,0x00000009,0x0000000d,0x0004003b, + 0x0000000e,0x0000000f,0x00000009,0x00040015,0x00000010,0x00000020,0x00000001,0x0004002b, + 0x00000010,0x00000011,0x00000002,0x00040020,0x00000012,0x00000009,0x00000006,0x0004002b, + 0x00000010,0x00000015,0x00000003,0x0004002b,0x00000010,0x0000001a,0x00000004,0x0004002b, + 0x00000010,0x0000001d,0x00000005,0x0004002b,0x00000010,0x00000022,0x00000000,0x0004002b, + 0x00000006,0x00000025,0x00000000,0x00020014,0x00000026,0x00040017,0x0000002a,0x00000006, + 0x00000004,0x00040020,0x0000002b,0x00000003,0x0000002a,0x0004003b,0x0000002b,0x0000002c, + 0x00000003,0x00090019,0x0000002d,0x00000006,0x00000001,0x00000000,0x00000000,0x00000000, + 0x00000001,0x00000000,0x0003001b,0x0000002e,0x0000002d,0x00040020,0x0000002f,0x00000000, + 0x0000002e,0x0004003b,0x0000002f,0x00000030,0x00000000,0x00040017,0x00000034,0x00000006, + 0x00000003,0x0004002b,0x00000006,0x00000036,0x3fc00000,0x0004002b,0x00000010,0x00000038, + 0x00000001,0x0004002b,0x00000006,0x0000003d,0x3f800000,0x00050036,0x00000002,0x00000004, + 0x00000000,0x00000003,0x000200f8,0x00000005,0x0004003b,0x00000008,0x00000009,0x00000007, + 0x0004003d,0x00000007,0x0000000c,0x0000000b,0x00050041,0x00000012,0x00000013,0x0000000f, + 0x00000011,0x0004003d,0x00000006,0x00000014,0x00000013,0x00050041,0x00000012,0x00000016, + 0x0000000f,0x00000015,0x0004003d,0x00000006,0x00000017,0x00000016,0x00050050,0x00000007, + 0x00000018,0x00000014,0x00000017,0x00050085,0x00000007,0x00000019,0x0000000c,0x00000018, + 0x00050041,0x00000012,0x0000001b,0x0000000f,0x0000001a,0x0004003d,0x00000006,0x0000001c, + 0x0000001b,0x00050041,0x00000012,0x0000001e,0x0000000f,0x0000001d,0x0004003d,0x00000006, + 0x0000001f,0x0000001e,0x00050050,0x00000007,0x00000020,0x0000001c,0x0000001f,0x00050081, + 0x00000007,0x00000021,0x00000019,0x00000020,0x0003003e,0x00000009,0x00000021,0x00050041, + 0x00000012,0x00000023,0x0000000f,0x00000022,0x0004003d,0x00000006,0x00000024,0x00000023, + 0x000500ba,0x00000026,0x00000027,0x00000024,0x00000025,0x000300f7,0x00000029,0x00000000, + 0x000400fa,0x00000027,0x00000028,0x00000042,0x000200f8,0x00000028,0x0004003d,0x0000002e, + 0x00000031,0x00000030,0x0004003d,0x00000007,0x00000032,0x00000009,0x00070058,0x0000002a, + 0x00000033,0x00000031,0x00000032,0x00000002,0x00000025,0x0008004f,0x00000034,0x00000035, + 0x00000033,0x00000033,0x00000000,0x00000001,0x00000002,0x0005008e,0x00000034,0x00000037, + 0x00000035,0x00000036,0x00050041,0x00000012,0x00000039,0x0000000f,0x00000038,0x0004003d, + 0x00000006,0x0000003a,0x00000039,0x00060050,0x00000034,0x0000003b,0x0000003a,0x0000003a, + 0x0000003a,0x0007000c,0x00000034,0x0000003c,0x00000001,0x0000001a,0x00000037,0x0000003b, + 0x00050051,0x00000006,0x0000003e,0x0000003c,0x00000000,0x00050051,0x00000006,0x0000003f, + 0x0000003c,0x00000001,0x00050051,0x00000006,0x00000040,0x0000003c,0x00000002,0x00070050, + 0x0000002a,0x00000041,0x0000003e,0x0000003f,0x00000040,0x0000003d,0x0003003e,0x0000002c, + 0x00000041,0x000200f9,0x00000029,0x000200f8,0x00000042,0x0004003d,0x0000002e,0x00000043, + 0x00000030,0x0004003d,0x00000007,0x00000044,0x00000009,0x00070058,0x0000002a,0x00000045, + 0x00000043,0x00000044,0x00000002,0x00000025,0x0003003e,0x0000002c,0x00000045,0x000200f9, + 0x00000029,0x000200f8,0x00000029,0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/postprocess_vert.c b/src/vk/spirv/postprocess_vert.c new file mode 100644 index 0000000..7ce9dbe --- /dev/null +++ b/src/vk/spirv/postprocess_vert.c @@ -0,0 +1,36 @@ + // 8.13.3559 + #pragma once +const uint32_t postprocess_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000029,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0008000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x00000009,0x0000000c,0x0000001a, + 0x00030003,0x00000002,0x000001c2,0x00090004,0x415f4c47,0x735f4252,0x72617065,0x5f657461, + 0x64616873,0x6f5f7265,0x63656a62,0x00007374,0x00040005,0x00000004,0x6e69616d,0x00000000, + 0x00050005,0x00000009,0x43786574,0x64726f6f,0x00000000,0x00060005,0x0000000c,0x565f6c67, + 0x65747265,0x646e4978,0x00007865,0x00060005,0x00000018,0x505f6c67,0x65567265,0x78657472, + 0x00000000,0x00060006,0x00000018,0x00000000,0x505f6c67,0x7469736f,0x006e6f69,0x00030005, + 0x0000001a,0x00000000,0x00040047,0x00000009,0x0000001e,0x00000000,0x00040047,0x0000000c, + 0x0000000b,0x0000002a,0x00050048,0x00000018,0x00000000,0x0000000b,0x00000000,0x00030047, + 0x00000018,0x00000002,0x00020013,0x00000002,0x00030021,0x00000003,0x00000002,0x00030016, + 0x00000006,0x00000020,0x00040017,0x00000007,0x00000006,0x00000002,0x00040020,0x00000008, + 0x00000003,0x00000007,0x0004003b,0x00000008,0x00000009,0x00000003,0x00040015,0x0000000a, + 0x00000020,0x00000001,0x00040020,0x0000000b,0x00000001,0x0000000a,0x0004003b,0x0000000b, + 0x0000000c,0x00000001,0x0004002b,0x0000000a,0x0000000e,0x00000001,0x0004002b,0x0000000a, + 0x00000010,0x00000002,0x00040017,0x00000017,0x00000006,0x00000004,0x0003001e,0x00000018, + 0x00000017,0x00040020,0x00000019,0x00000003,0x00000018,0x0004003b,0x00000019,0x0000001a, + 0x00000003,0x0004002b,0x0000000a,0x0000001b,0x00000000,0x0004002b,0x00000006,0x0000001d, + 0x40000000,0x0004002b,0x00000006,0x0000001f,0xbf800000,0x0004002b,0x00000006,0x00000022, + 0x00000000,0x0004002b,0x00000006,0x00000023,0x3f800000,0x00040020,0x00000027,0x00000003, + 0x00000017,0x00050036,0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005, + 0x0004003d,0x0000000a,0x0000000d,0x0000000c,0x000500c4,0x0000000a,0x0000000f,0x0000000d, + 0x0000000e,0x000500c7,0x0000000a,0x00000011,0x0000000f,0x00000010,0x0004006f,0x00000006, + 0x00000012,0x00000011,0x0004003d,0x0000000a,0x00000013,0x0000000c,0x000500c7,0x0000000a, + 0x00000014,0x00000013,0x00000010,0x0004006f,0x00000006,0x00000015,0x00000014,0x00050050, + 0x00000007,0x00000016,0x00000012,0x00000015,0x0003003e,0x00000009,0x00000016,0x0004003d, + 0x00000007,0x0000001c,0x00000009,0x0005008e,0x00000007,0x0000001e,0x0000001c,0x0000001d, + 0x00050050,0x00000007,0x00000020,0x0000001f,0x0000001f,0x00050081,0x00000007,0x00000021, + 0x0000001e,0x00000020,0x00050051,0x00000006,0x00000024,0x00000021,0x00000000,0x00050051, + 0x00000006,0x00000025,0x00000021,0x00000001,0x00070050,0x00000017,0x00000026,0x00000024, + 0x00000025,0x00000022,0x00000023,0x00050041,0x00000027,0x00000028,0x0000001a,0x0000001b, + 0x0003003e,0x00000028,0x00000026,0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/shadows_vert.c b/src/vk/spirv/shadows_vert.c new file mode 100644 index 0000000..5114b8a --- /dev/null +++ b/src/vk/spirv/shadows_vert.c @@ -0,0 +1,46 @@ + // 8.13.3559 + #pragma once +const uint32_t shadows_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x0000002b,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0008000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000a,0x0000001d,0x00000027, + 0x00030003,0x00000002,0x000001c2,0x00090004,0x415f4c47,0x735f4252,0x72617065,0x5f657461, + 0x64616873,0x6f5f7265,0x63656a62,0x00007374,0x00040005,0x00000004,0x6e69616d,0x00000000, + 0x00060005,0x00000008,0x505f6c67,0x65567265,0x78657472,0x00000000,0x00060006,0x00000008, + 0x00000000,0x505f6c67,0x7469736f,0x006e6f69,0x00030005,0x0000000a,0x00000000,0x00060005, + 0x0000000e,0x68737550,0x736e6f43,0x746e6174,0x00000000,0x00060006,0x0000000e,0x00000000, + 0x614d7076,0x78697274,0x00000000,0x00030005,0x00000010,0x00006370,0x00070005,0x00000014, + 0x66696e55,0x426d726f,0x65666675,0x6a624f72,0x00746365,0x00050006,0x00000014,0x00000000, + 0x65646f6d,0x0000006c,0x00030005,0x00000016,0x006f6275,0x00050005,0x0000001d,0x65566e69, + 0x78657472,0x00000000,0x00040005,0x00000027,0x6f6c6f63,0x00000072,0x00050048,0x00000008, + 0x00000000,0x0000000b,0x00000000,0x00030047,0x00000008,0x00000002,0x00040048,0x0000000e, + 0x00000000,0x00000005,0x00050048,0x0000000e,0x00000000,0x00000023,0x00000000,0x00050048, + 0x0000000e,0x00000000,0x00000007,0x00000010,0x00030047,0x0000000e,0x00000002,0x00040048, + 0x00000014,0x00000000,0x00000005,0x00050048,0x00000014,0x00000000,0x00000023,0x00000000, + 0x00050048,0x00000014,0x00000000,0x00000007,0x00000010,0x00030047,0x00000014,0x00000002, + 0x00040047,0x00000016,0x00000022,0x00000000,0x00040047,0x00000016,0x00000021,0x00000000, + 0x00040047,0x0000001d,0x0000001e,0x00000000,0x00040047,0x00000027,0x0000001e,0x00000000, + 0x00020013,0x00000002,0x00030021,0x00000003,0x00000002,0x00030016,0x00000006,0x00000020, + 0x00040017,0x00000007,0x00000006,0x00000004,0x0003001e,0x00000008,0x00000007,0x00040020, + 0x00000009,0x00000003,0x00000008,0x0004003b,0x00000009,0x0000000a,0x00000003,0x00040015, + 0x0000000b,0x00000020,0x00000001,0x0004002b,0x0000000b,0x0000000c,0x00000000,0x00040018, + 0x0000000d,0x00000007,0x00000004,0x0003001e,0x0000000e,0x0000000d,0x00040020,0x0000000f, + 0x00000009,0x0000000e,0x0004003b,0x0000000f,0x00000010,0x00000009,0x00040020,0x00000011, + 0x00000009,0x0000000d,0x0003001e,0x00000014,0x0000000d,0x00040020,0x00000015,0x00000002, + 0x00000014,0x0004003b,0x00000015,0x00000016,0x00000002,0x00040020,0x00000017,0x00000002, + 0x0000000d,0x00040017,0x0000001b,0x00000006,0x00000003,0x00040020,0x0000001c,0x00000001, + 0x0000001b,0x0004003b,0x0000001c,0x0000001d,0x00000001,0x0004002b,0x00000006,0x0000001f, + 0x3f800000,0x00040020,0x00000025,0x00000003,0x00000007,0x0004003b,0x00000025,0x00000027, + 0x00000003,0x0004002b,0x00000006,0x00000028,0x00000000,0x0004002b,0x00000006,0x00000029, + 0x3f000000,0x0007002c,0x00000007,0x0000002a,0x00000028,0x00000028,0x00000028,0x00000029, + 0x00050036,0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,0x00050041, + 0x00000011,0x00000012,0x00000010,0x0000000c,0x0004003d,0x0000000d,0x00000013,0x00000012, + 0x00050041,0x00000017,0x00000018,0x00000016,0x0000000c,0x0004003d,0x0000000d,0x00000019, + 0x00000018,0x00050092,0x0000000d,0x0000001a,0x00000013,0x00000019,0x0004003d,0x0000001b, + 0x0000001e,0x0000001d,0x00050051,0x00000006,0x00000020,0x0000001e,0x00000000,0x00050051, + 0x00000006,0x00000021,0x0000001e,0x00000001,0x00050051,0x00000006,0x00000022,0x0000001e, + 0x00000002,0x00070050,0x00000007,0x00000023,0x00000020,0x00000021,0x00000022,0x0000001f, + 0x00050091,0x00000007,0x00000024,0x0000001a,0x00000023,0x00050041,0x00000025,0x00000026, + 0x0000000a,0x0000000c,0x0003003e,0x00000026,0x00000024,0x0003003e,0x00000027,0x0000002a, + 0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/skybox_vert.c b/src/vk/spirv/skybox_vert.c new file mode 100644 index 0000000..2797070 --- /dev/null +++ b/src/vk/spirv/skybox_vert.c @@ -0,0 +1,54 @@ + // 8.13.3559 + #pragma once +const uint32_t skybox_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000032,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x000b000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000a,0x0000001d,0x00000029, + 0x0000002b,0x0000002d,0x00000030,0x00030003,0x00000002,0x000001c2,0x00090004,0x415f4c47, + 0x735f4252,0x72617065,0x5f657461,0x64616873,0x6f5f7265,0x63656a62,0x00007374,0x00040005, + 0x00000004,0x6e69616d,0x00000000,0x00060005,0x00000008,0x505f6c67,0x65567265,0x78657472, + 0x00000000,0x00060006,0x00000008,0x00000000,0x505f6c67,0x7469736f,0x006e6f69,0x00030005, + 0x0000000a,0x00000000,0x00060005,0x0000000e,0x68737550,0x736e6f43,0x746e6174,0x00000000, + 0x00060006,0x0000000e,0x00000000,0x614d7076,0x78697274,0x00000000,0x00030005,0x00000010, + 0x00006370,0x00070005,0x00000014,0x66696e55,0x426d726f,0x65666675,0x6a624f72,0x00746365, + 0x00050006,0x00000014,0x00000000,0x65646f6d,0x0000006c,0x00030005,0x00000016,0x006f6275, + 0x00050005,0x0000001d,0x65566e69,0x78657472,0x00000000,0x00050005,0x00000029,0x43786574, + 0x64726f6f,0x00000000,0x00050005,0x0000002b,0x65546e69,0x6f6f4378,0x00006472,0x00040005, + 0x0000002d,0x6f6c6f63,0x00000072,0x00050005,0x00000030,0x65725461,0x6c6f6873,0x00000064, + 0x00050048,0x00000008,0x00000000,0x0000000b,0x00000000,0x00030047,0x00000008,0x00000002, + 0x00040048,0x0000000e,0x00000000,0x00000005,0x00050048,0x0000000e,0x00000000,0x00000023, + 0x00000000,0x00050048,0x0000000e,0x00000000,0x00000007,0x00000010,0x00030047,0x0000000e, + 0x00000002,0x00040048,0x00000014,0x00000000,0x00000005,0x00050048,0x00000014,0x00000000, + 0x00000023,0x00000000,0x00050048,0x00000014,0x00000000,0x00000007,0x00000010,0x00030047, + 0x00000014,0x00000002,0x00040047,0x00000016,0x00000022,0x00000001,0x00040047,0x00000016, + 0x00000021,0x00000000,0x00040047,0x0000001d,0x0000001e,0x00000000,0x00040047,0x00000029, + 0x0000001e,0x00000000,0x00040047,0x0000002b,0x0000001e,0x00000001,0x00040047,0x0000002d, + 0x0000001e,0x00000001,0x00040047,0x00000030,0x0000001e,0x00000002,0x00020013,0x00000002, + 0x00030021,0x00000003,0x00000002,0x00030016,0x00000006,0x00000020,0x00040017,0x00000007, + 0x00000006,0x00000004,0x0003001e,0x00000008,0x00000007,0x00040020,0x00000009,0x00000003, + 0x00000008,0x0004003b,0x00000009,0x0000000a,0x00000003,0x00040015,0x0000000b,0x00000020, + 0x00000001,0x0004002b,0x0000000b,0x0000000c,0x00000000,0x00040018,0x0000000d,0x00000007, + 0x00000004,0x0003001e,0x0000000e,0x0000000d,0x00040020,0x0000000f,0x00000009,0x0000000e, + 0x0004003b,0x0000000f,0x00000010,0x00000009,0x00040020,0x00000011,0x00000009,0x0000000d, + 0x0003001e,0x00000014,0x0000000d,0x00040020,0x00000015,0x00000002,0x00000014,0x0004003b, + 0x00000015,0x00000016,0x00000002,0x00040020,0x00000017,0x00000002,0x0000000d,0x00040017, + 0x0000001b,0x00000006,0x00000003,0x00040020,0x0000001c,0x00000001,0x0000001b,0x0004003b, + 0x0000001c,0x0000001d,0x00000001,0x0004002b,0x00000006,0x0000001f,0x3f800000,0x00040020, + 0x00000025,0x00000003,0x00000007,0x00040017,0x00000027,0x00000006,0x00000002,0x00040020, + 0x00000028,0x00000003,0x00000027,0x0004003b,0x00000028,0x00000029,0x00000003,0x00040020, + 0x0000002a,0x00000001,0x00000027,0x0004003b,0x0000002a,0x0000002b,0x00000001,0x0004003b, + 0x00000025,0x0000002d,0x00000003,0x0007002c,0x00000007,0x0000002e,0x0000001f,0x0000001f, + 0x0000001f,0x0000001f,0x00040020,0x0000002f,0x00000003,0x00000006,0x0004003b,0x0000002f, + 0x00000030,0x00000003,0x0004002b,0x00000006,0x00000031,0x00000000,0x00050036,0x00000002, + 0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,0x00050041,0x00000011,0x00000012, + 0x00000010,0x0000000c,0x0004003d,0x0000000d,0x00000013,0x00000012,0x00050041,0x00000017, + 0x00000018,0x00000016,0x0000000c,0x0004003d,0x0000000d,0x00000019,0x00000018,0x00050092, + 0x0000000d,0x0000001a,0x00000013,0x00000019,0x0004003d,0x0000001b,0x0000001e,0x0000001d, + 0x00050051,0x00000006,0x00000020,0x0000001e,0x00000000,0x00050051,0x00000006,0x00000021, + 0x0000001e,0x00000001,0x00050051,0x00000006,0x00000022,0x0000001e,0x00000002,0x00070050, + 0x00000007,0x00000023,0x00000020,0x00000021,0x00000022,0x0000001f,0x00050091,0x00000007, + 0x00000024,0x0000001a,0x00000023,0x00050041,0x00000025,0x00000026,0x0000000a,0x0000000c, + 0x0003003e,0x00000026,0x00000024,0x0004003d,0x00000027,0x0000002c,0x0000002b,0x0003003e, + 0x00000029,0x0000002c,0x0003003e,0x0000002d,0x0000002e,0x0003003e,0x00000030,0x00000031, + 0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/sprite_vert.c b/src/vk/spirv/sprite_vert.c new file mode 100644 index 0000000..090a158 --- /dev/null +++ b/src/vk/spirv/sprite_vert.c @@ -0,0 +1,49 @@ + // 8.13.3559 + #pragma once +const uint32_t sprite_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x0000002f,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x000b000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000a,0x00000016,0x00000022, + 0x00000024,0x00000026,0x0000002d,0x00030003,0x00000002,0x000001c2,0x00090004,0x415f4c47, + 0x735f4252,0x72617065,0x5f657461,0x64616873,0x6f5f7265,0x63656a62,0x00007374,0x00040005, + 0x00000004,0x6e69616d,0x00000000,0x00060005,0x00000008,0x505f6c67,0x65567265,0x78657472, + 0x00000000,0x00060006,0x00000008,0x00000000,0x505f6c67,0x7469736f,0x006e6f69,0x00030005, + 0x0000000a,0x00000000,0x00060005,0x0000000e,0x68737550,0x736e6f43,0x746e6174,0x00000000, + 0x00060006,0x0000000e,0x00000000,0x4d70766d,0x69727461,0x00000078,0x00050006,0x0000000e, + 0x00000001,0x68706c61,0x00000061,0x00030005,0x00000010,0x00006370,0x00050005,0x00000016, + 0x65566e69,0x78657472,0x00000000,0x00050005,0x00000022,0x43786574,0x64726f6f,0x00000000, + 0x00050005,0x00000024,0x65546e69,0x6f6f4378,0x00006472,0x00040005,0x00000026,0x6f6c6f63, + 0x00000072,0x00050005,0x0000002d,0x65725461,0x6c6f6873,0x00000064,0x00050048,0x00000008, + 0x00000000,0x0000000b,0x00000000,0x00030047,0x00000008,0x00000002,0x00040048,0x0000000e, + 0x00000000,0x00000005,0x00050048,0x0000000e,0x00000000,0x00000023,0x00000000,0x00050048, + 0x0000000e,0x00000000,0x00000007,0x00000010,0x00050048,0x0000000e,0x00000001,0x00000023, + 0x00000040,0x00030047,0x0000000e,0x00000002,0x00040047,0x00000016,0x0000001e,0x00000000, + 0x00040047,0x00000022,0x0000001e,0x00000000,0x00040047,0x00000024,0x0000001e,0x00000001, + 0x00040047,0x00000026,0x0000001e,0x00000001,0x00040047,0x0000002d,0x0000001e,0x00000002, + 0x00020013,0x00000002,0x00030021,0x00000003,0x00000002,0x00030016,0x00000006,0x00000020, + 0x00040017,0x00000007,0x00000006,0x00000004,0x0003001e,0x00000008,0x00000007,0x00040020, + 0x00000009,0x00000003,0x00000008,0x0004003b,0x00000009,0x0000000a,0x00000003,0x00040015, + 0x0000000b,0x00000020,0x00000001,0x0004002b,0x0000000b,0x0000000c,0x00000000,0x00040018, + 0x0000000d,0x00000007,0x00000004,0x0004001e,0x0000000e,0x0000000d,0x00000006,0x00040020, + 0x0000000f,0x00000009,0x0000000e,0x0004003b,0x0000000f,0x00000010,0x00000009,0x00040020, + 0x00000011,0x00000009,0x0000000d,0x00040017,0x00000014,0x00000006,0x00000003,0x00040020, + 0x00000015,0x00000001,0x00000014,0x0004003b,0x00000015,0x00000016,0x00000001,0x0004002b, + 0x00000006,0x00000018,0x3f800000,0x00040020,0x0000001e,0x00000003,0x00000007,0x00040017, + 0x00000020,0x00000006,0x00000002,0x00040020,0x00000021,0x00000003,0x00000020,0x0004003b, + 0x00000021,0x00000022,0x00000003,0x00040020,0x00000023,0x00000001,0x00000020,0x0004003b, + 0x00000023,0x00000024,0x00000001,0x0004003b,0x0000001e,0x00000026,0x00000003,0x0004002b, + 0x0000000b,0x00000027,0x00000001,0x00040020,0x00000028,0x00000009,0x00000006,0x00040020, + 0x0000002c,0x00000003,0x00000006,0x0004003b,0x0000002c,0x0000002d,0x00000003,0x0004002b, + 0x00000006,0x0000002e,0x3d886595,0x00050036,0x00000002,0x00000004,0x00000000,0x00000003, + 0x000200f8,0x00000005,0x00050041,0x00000011,0x00000012,0x00000010,0x0000000c,0x0004003d, + 0x0000000d,0x00000013,0x00000012,0x0004003d,0x00000014,0x00000017,0x00000016,0x00050051, + 0x00000006,0x00000019,0x00000017,0x00000000,0x00050051,0x00000006,0x0000001a,0x00000017, + 0x00000001,0x00050051,0x00000006,0x0000001b,0x00000017,0x00000002,0x00070050,0x00000007, + 0x0000001c,0x00000019,0x0000001a,0x0000001b,0x00000018,0x00050091,0x00000007,0x0000001d, + 0x00000013,0x0000001c,0x00050041,0x0000001e,0x0000001f,0x0000000a,0x0000000c,0x0003003e, + 0x0000001f,0x0000001d,0x0004003d,0x00000020,0x00000025,0x00000024,0x0003003e,0x00000022, + 0x00000025,0x00050041,0x00000028,0x00000029,0x00000010,0x00000027,0x0004003d,0x00000006, + 0x0000002a,0x00000029,0x00070050,0x00000007,0x0000002b,0x00000018,0x00000018,0x00000018, + 0x0000002a,0x0003003e,0x00000026,0x0000002b,0x0003003e,0x0000002d,0x0000002e,0x000100fd, + 0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/world_warp_frag.c b/src/vk/spirv/world_warp_frag.c new file mode 100644 index 0000000..0f6ba66 --- /dev/null +++ b/src/vk/spirv/world_warp_frag.c @@ -0,0 +1,154 @@ + // 1011.0.0 + #pragma once +const uint32_t world_warp_frag_spv[] = { + 0x07230203,0x00010000,0x0008000a,0x000000c1,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0007000f,0x00000004,0x00000004,0x6e69616d,0x00000000,0x00000019,0x000000b9,0x00030010, + 0x00000004,0x00000007,0x00030003,0x00000002,0x000001c2,0x00090004,0x415f4c47,0x735f4252, + 0x72617065,0x5f657461,0x64616873,0x6f5f7265,0x63656a62,0x00007374,0x00040005,0x00000004, + 0x6e69616d,0x00000000,0x00040005,0x00000009,0x53726373,0x00657a69,0x00060005,0x0000000a, + 0x68737550,0x736e6f43,0x746e6174,0x00000000,0x00050006,0x0000000a,0x00000000,0x656d6974, + 0x00000000,0x00050006,0x0000000a,0x00000001,0x6c616373,0x00000065,0x00060006,0x0000000a, + 0x00000002,0x57726373,0x68746469,0x00000000,0x00060006,0x0000000a,0x00000003,0x48726373, + 0x68676965,0x00000074,0x00050006,0x0000000a,0x00000004,0x7366666f,0x00587465,0x00050006, + 0x0000000a,0x00000005,0x7366666f,0x00597465,0x00060006,0x0000000a,0x00000006,0x65786970, + 0x7a69536c,0x00000065,0x00050006,0x0000000a,0x00000007,0x64666572,0x00586665,0x00050006, + 0x0000000a,0x00000008,0x64666572,0x00596665,0x00060006,0x0000000a,0x00000009,0x64666572, + 0x69576665,0x00687464,0x00070006,0x0000000a,0x0000000a,0x64666572,0x65486665,0x74686769, + 0x00000000,0x00030005,0x0000000c,0x00006370,0x00050005,0x00000016,0x67617266,0x726f6f43, + 0x00000064,0x00060005,0x00000019,0x465f6c67,0x43676172,0x64726f6f,0x00000000,0x00030005, + 0x00000024,0x00007675,0x00040005,0x00000029,0x6e694d78,0x00000000,0x00040005,0x0000002d, + 0x78614d78,0x00000000,0x00040005,0x00000034,0x6e694d79,0x00000000,0x00040005,0x00000038, + 0x78614d79,0x00000000,0x00030005,0x00000066,0x00007873,0x00030005,0x00000077,0x00007973, + 0x00040005,0x00000086,0x69685378,0x00007466,0x00040005,0x00000091,0x69685379,0x00007466, + 0x00050005,0x0000009a,0x74736964,0x6974726f,0x00006e6f,0x00060005,0x000000b9,0x67617266, + 0x746e656d,0x6f6c6f43,0x00000072,0x00050005,0x000000bd,0x78655473,0x65727574,0x00000000, + 0x00050048,0x0000000a,0x00000000,0x00000023,0x00000044,0x00050048,0x0000000a,0x00000001, + 0x00000023,0x00000048,0x00050048,0x0000000a,0x00000002,0x00000023,0x0000004c,0x00050048, + 0x0000000a,0x00000003,0x00000023,0x00000050,0x00050048,0x0000000a,0x00000004,0x00000023, + 0x00000054,0x00050048,0x0000000a,0x00000005,0x00000023,0x00000058,0x00050048,0x0000000a, + 0x00000006,0x00000023,0x0000005c,0x00050048,0x0000000a,0x00000007,0x00000023,0x00000060, + 0x00050048,0x0000000a,0x00000008,0x00000023,0x00000064,0x00050048,0x0000000a,0x00000009, + 0x00000023,0x00000068,0x00050048,0x0000000a,0x0000000a,0x00000023,0x0000006c,0x00030047, + 0x0000000a,0x00000002,0x00040047,0x00000019,0x0000000b,0x0000000f,0x00040047,0x000000b9, + 0x0000001e,0x00000000,0x00040047,0x000000bd,0x00000022,0x00000000,0x00040047,0x000000bd, + 0x00000021,0x00000000,0x00020013,0x00000002,0x00030021,0x00000003,0x00000002,0x00030016, + 0x00000006,0x00000020,0x00040017,0x00000007,0x00000006,0x00000002,0x00040020,0x00000008, + 0x00000007,0x00000007,0x000d001e,0x0000000a,0x00000006,0x00000006,0x00000006,0x00000006, + 0x00000006,0x00000006,0x00000006,0x00000006,0x00000006,0x00000006,0x00000006,0x00040020, + 0x0000000b,0x00000009,0x0000000a,0x0004003b,0x0000000b,0x0000000c,0x00000009,0x00040015, + 0x0000000d,0x00000020,0x00000001,0x0004002b,0x0000000d,0x0000000e,0x00000002,0x00040020, + 0x0000000f,0x00000009,0x00000006,0x0004002b,0x0000000d,0x00000012,0x00000003,0x00040017, + 0x00000017,0x00000006,0x00000004,0x00040020,0x00000018,0x00000001,0x00000017,0x0004003b, + 0x00000018,0x00000019,0x00000001,0x0004002b,0x0000000d,0x0000001c,0x00000004,0x0004002b, + 0x0000000d,0x0000001f,0x00000005,0x00040020,0x00000028,0x00000007,0x00000006,0x0004002b, + 0x0000000d,0x0000002a,0x00000007,0x0004002b,0x0000000d,0x00000030,0x00000009,0x0004002b, + 0x0000000d,0x00000035,0x00000008,0x0004002b,0x0000000d,0x0000003b,0x0000000a,0x00020014, + 0x0000003f,0x0004002b,0x0000000d,0x00000040,0x00000000,0x0004002b,0x00000006,0x00000043, + 0x00000000,0x00040015,0x00000047,0x00000020,0x00000000,0x0004002b,0x00000047,0x00000048, + 0x00000000,0x0004002b,0x00000047,0x00000057,0x00000001,0x0004002b,0x0000000d,0x00000067, + 0x00000001,0x0004002b,0x00000006,0x0000006c,0x40000000,0x0004002b,0x00000006,0x0000008c, + 0x40490e56,0x0004002b,0x00000006,0x0000008e,0x41200000,0x0004002b,0x00000006,0x000000a4, + 0x3bda3c21,0x0004002b,0x0000000d,0x000000a9,0x00000006,0x0005002c,0x00000007,0x000000b2, + 0x00000043,0x00000043,0x0004002b,0x00000006,0x000000b4,0x3f000000,0x0005002c,0x00000007, + 0x000000b5,0x000000b4,0x000000b4,0x00040020,0x000000b8,0x00000003,0x00000017,0x0004003b, + 0x000000b8,0x000000b9,0x00000003,0x00090019,0x000000ba,0x00000006,0x00000001,0x00000000, + 0x00000000,0x00000000,0x00000001,0x00000000,0x0003001b,0x000000bb,0x000000ba,0x00040020, + 0x000000bc,0x00000000,0x000000bb,0x0004003b,0x000000bc,0x000000bd,0x00000000,0x00050036, + 0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,0x0004003b,0x00000008, + 0x00000009,0x00000007,0x0004003b,0x00000008,0x00000016,0x00000007,0x0004003b,0x00000008, + 0x00000024,0x00000007,0x0004003b,0x00000028,0x00000029,0x00000007,0x0004003b,0x00000028, + 0x0000002d,0x00000007,0x0004003b,0x00000028,0x00000034,0x00000007,0x0004003b,0x00000028, + 0x00000038,0x00000007,0x0004003b,0x00000028,0x00000066,0x00000007,0x0004003b,0x00000028, + 0x00000077,0x00000007,0x0004003b,0x00000028,0x00000086,0x00000007,0x0004003b,0x00000028, + 0x00000091,0x00000007,0x0004003b,0x00000008,0x0000009a,0x00000007,0x00050041,0x0000000f, + 0x00000010,0x0000000c,0x0000000e,0x0004003d,0x00000006,0x00000011,0x00000010,0x00050041, + 0x0000000f,0x00000013,0x0000000c,0x00000012,0x0004003d,0x00000006,0x00000014,0x00000013, + 0x00050050,0x00000007,0x00000015,0x00000011,0x00000014,0x0003003e,0x00000009,0x00000015, + 0x0004003d,0x00000017,0x0000001a,0x00000019,0x0007004f,0x00000007,0x0000001b,0x0000001a, + 0x0000001a,0x00000000,0x00000001,0x00050041,0x0000000f,0x0000001d,0x0000000c,0x0000001c, + 0x0004003d,0x00000006,0x0000001e,0x0000001d,0x00050041,0x0000000f,0x00000020,0x0000000c, + 0x0000001f,0x0004003d,0x00000006,0x00000021,0x00000020,0x00050050,0x00000007,0x00000022, + 0x0000001e,0x00000021,0x00050083,0x00000007,0x00000023,0x0000001b,0x00000022,0x0003003e, + 0x00000016,0x00000023,0x0004003d,0x00000007,0x00000025,0x00000016,0x0004003d,0x00000007, + 0x00000026,0x00000009,0x00050088,0x00000007,0x00000027,0x00000025,0x00000026,0x0003003e, + 0x00000024,0x00000027,0x00050041,0x0000000f,0x0000002b,0x0000000c,0x0000002a,0x0004003d, + 0x00000006,0x0000002c,0x0000002b,0x0003003e,0x00000029,0x0000002c,0x00050041,0x0000000f, + 0x0000002e,0x0000000c,0x0000002a,0x0004003d,0x00000006,0x0000002f,0x0000002e,0x00050041, + 0x0000000f,0x00000031,0x0000000c,0x00000030,0x0004003d,0x00000006,0x00000032,0x00000031, + 0x00050081,0x00000006,0x00000033,0x0000002f,0x00000032,0x0003003e,0x0000002d,0x00000033, + 0x00050041,0x0000000f,0x00000036,0x0000000c,0x00000035,0x0004003d,0x00000006,0x00000037, + 0x00000036,0x0003003e,0x00000034,0x00000037,0x00050041,0x0000000f,0x00000039,0x0000000c, + 0x00000035,0x0004003d,0x00000006,0x0000003a,0x00000039,0x00050041,0x0000000f,0x0000003c, + 0x0000000c,0x0000003b,0x0004003d,0x00000006,0x0000003d,0x0000003c,0x00050081,0x00000006, + 0x0000003e,0x0000003a,0x0000003d,0x0003003e,0x00000038,0x0000003e,0x00050041,0x0000000f, + 0x00000041,0x0000000c,0x00000040,0x0004003d,0x00000006,0x00000042,0x00000041,0x000500ba, + 0x0000003f,0x00000044,0x00000042,0x00000043,0x000300f7,0x00000046,0x00000000,0x000400fa, + 0x00000044,0x00000045,0x00000046,0x000200f8,0x00000045,0x00050041,0x00000028,0x00000049, + 0x00000016,0x00000048,0x0004003d,0x00000006,0x0000004a,0x00000049,0x0004003d,0x00000006, + 0x0000004b,0x00000029,0x000500ba,0x0000003f,0x0000004c,0x0000004a,0x0000004b,0x000200f9, + 0x00000046,0x000200f8,0x00000046,0x000700f5,0x0000003f,0x0000004d,0x00000044,0x00000005, + 0x0000004c,0x00000045,0x000300f7,0x0000004f,0x00000000,0x000400fa,0x0000004d,0x0000004e, + 0x0000004f,0x000200f8,0x0000004e,0x00050041,0x00000028,0x00000050,0x00000016,0x00000048, + 0x0004003d,0x00000006,0x00000051,0x00000050,0x0004003d,0x00000006,0x00000052,0x0000002d, + 0x000500b8,0x0000003f,0x00000053,0x00000051,0x00000052,0x000200f9,0x0000004f,0x000200f8, + 0x0000004f,0x000700f5,0x0000003f,0x00000054,0x0000004d,0x00000046,0x00000053,0x0000004e, + 0x000300f7,0x00000056,0x00000000,0x000400fa,0x00000054,0x00000055,0x00000056,0x000200f8, + 0x00000055,0x00050041,0x00000028,0x00000058,0x00000016,0x00000057,0x0004003d,0x00000006, + 0x00000059,0x00000058,0x0004003d,0x00000006,0x0000005a,0x00000034,0x000500ba,0x0000003f, + 0x0000005b,0x00000059,0x0000005a,0x000200f9,0x00000056,0x000200f8,0x00000056,0x000700f5, + 0x0000003f,0x0000005c,0x00000054,0x0000004f,0x0000005b,0x00000055,0x000300f7,0x0000005e, + 0x00000000,0x000400fa,0x0000005c,0x0000005d,0x0000005e,0x000200f8,0x0000005d,0x00050041, + 0x00000028,0x0000005f,0x00000016,0x00000057,0x0004003d,0x00000006,0x00000060,0x0000005f, + 0x0004003d,0x00000006,0x00000061,0x00000038,0x000500b8,0x0000003f,0x00000062,0x00000060, + 0x00000061,0x000200f9,0x0000005e,0x000200f8,0x0000005e,0x000700f5,0x0000003f,0x00000063, + 0x0000005c,0x00000056,0x00000062,0x0000005d,0x000300f7,0x00000065,0x00000000,0x000400fa, + 0x00000063,0x00000064,0x00000065,0x000200f8,0x00000064,0x00050041,0x0000000f,0x00000068, + 0x0000000c,0x00000067,0x0004003d,0x00000006,0x00000069,0x00000068,0x00050041,0x0000000f, + 0x0000006a,0x0000000c,0x0000000e,0x0004003d,0x00000006,0x0000006b,0x0000006a,0x00050088, + 0x00000006,0x0000006d,0x0000006b,0x0000006c,0x00050041,0x00000028,0x0000006e,0x00000016, + 0x00000048,0x0004003d,0x00000006,0x0000006f,0x0000006e,0x00050083,0x00000006,0x00000070, + 0x0000006d,0x0000006f,0x0006000c,0x00000006,0x00000071,0x00000001,0x00000004,0x00000070, + 0x00050085,0x00000006,0x00000072,0x00000071,0x0000006c,0x00050041,0x0000000f,0x00000073, + 0x0000000c,0x0000000e,0x0004003d,0x00000006,0x00000074,0x00000073,0x00050088,0x00000006, + 0x00000075,0x00000072,0x00000074,0x00050083,0x00000006,0x00000076,0x00000069,0x00000075, + 0x0003003e,0x00000066,0x00000076,0x00050041,0x0000000f,0x00000078,0x0000000c,0x00000067, + 0x0004003d,0x00000006,0x00000079,0x00000078,0x00050041,0x0000000f,0x0000007a,0x0000000c, + 0x00000012,0x0004003d,0x00000006,0x0000007b,0x0000007a,0x00050088,0x00000006,0x0000007c, + 0x0000007b,0x0000006c,0x00050041,0x00000028,0x0000007d,0x00000016,0x00000057,0x0004003d, + 0x00000006,0x0000007e,0x0000007d,0x00050083,0x00000006,0x0000007f,0x0000007c,0x0000007e, + 0x0006000c,0x00000006,0x00000080,0x00000001,0x00000004,0x0000007f,0x00050085,0x00000006, + 0x00000081,0x00000080,0x0000006c,0x00050041,0x0000000f,0x00000082,0x0000000c,0x00000012, + 0x0004003d,0x00000006,0x00000083,0x00000082,0x00050088,0x00000006,0x00000084,0x00000081, + 0x00000083,0x00050083,0x00000006,0x00000085,0x00000079,0x00000084,0x0003003e,0x00000077, + 0x00000085,0x00050041,0x0000000f,0x00000087,0x0000000c,0x00000040,0x0004003d,0x00000006, + 0x00000088,0x00000087,0x00050085,0x00000006,0x00000089,0x0000006c,0x00000088,0x00050041, + 0x00000028,0x0000008a,0x00000024,0x00000057,0x0004003d,0x00000006,0x0000008b,0x0000008a, + 0x00050085,0x00000006,0x0000008d,0x0000008b,0x0000008c,0x00050085,0x00000006,0x0000008f, + 0x0000008d,0x0000008e,0x00050081,0x00000006,0x00000090,0x00000089,0x0000008f,0x0003003e, + 0x00000086,0x00000090,0x00050041,0x0000000f,0x00000092,0x0000000c,0x00000040,0x0004003d, + 0x00000006,0x00000093,0x00000092,0x00050085,0x00000006,0x00000094,0x0000006c,0x00000093, + 0x00050041,0x00000028,0x00000095,0x00000024,0x00000048,0x0004003d,0x00000006,0x00000096, + 0x00000095,0x00050085,0x00000006,0x00000097,0x00000096,0x0000008c,0x00050085,0x00000006, + 0x00000098,0x00000097,0x0000008e,0x00050081,0x00000006,0x00000099,0x00000094,0x00000098, + 0x0003003e,0x00000091,0x00000099,0x0004003d,0x00000006,0x0000009b,0x00000086,0x0006000c, + 0x00000006,0x0000009c,0x00000001,0x0000000d,0x0000009b,0x0004003d,0x00000006,0x0000009d, + 0x00000066,0x00050085,0x00000006,0x0000009e,0x0000009c,0x0000009d,0x0004003d,0x00000006, + 0x0000009f,0x00000091,0x0006000c,0x00000006,0x000000a0,0x00000001,0x0000000d,0x0000009f, + 0x0004003d,0x00000006,0x000000a1,0x00000077,0x00050085,0x00000006,0x000000a2,0x000000a0, + 0x000000a1,0x00050050,0x00000007,0x000000a3,0x0000009e,0x000000a2,0x0005008e,0x00000007, + 0x000000a5,0x000000a3,0x000000a4,0x0003003e,0x0000009a,0x000000a5,0x0004003d,0x00000007, + 0x000000a6,0x0000009a,0x0004003d,0x00000007,0x000000a7,0x00000024,0x00050081,0x00000007, + 0x000000a8,0x000000a7,0x000000a6,0x0003003e,0x00000024,0x000000a8,0x000200f9,0x00000065, + 0x000200f8,0x00000065,0x00050041,0x0000000f,0x000000aa,0x0000000c,0x000000a9,0x0004003d, + 0x00000006,0x000000ab,0x000000aa,0x0004003d,0x00000007,0x000000ac,0x00000024,0x00050050, + 0x00000007,0x000000ad,0x000000ab,0x000000ab,0x00050088,0x00000007,0x000000ae,0x000000ac, + 0x000000ad,0x0003003e,0x00000024,0x000000ae,0x0004003d,0x00000007,0x000000af,0x00000024, + 0x0004003d,0x00000007,0x000000b0,0x00000009,0x00050085,0x00000007,0x000000b1,0x000000af, + 0x000000b0,0x0004003d,0x00000007,0x000000b3,0x00000009,0x00050083,0x00000007,0x000000b6, + 0x000000b3,0x000000b5,0x0008000c,0x00000007,0x000000b7,0x00000001,0x0000002b,0x000000b1, + 0x000000b2,0x000000b6,0x0003003e,0x00000024,0x000000b7,0x0004003d,0x000000bb,0x000000be, + 0x000000bd,0x0004003d,0x00000007,0x000000bf,0x00000024,0x00070058,0x00000017,0x000000c0, + 0x000000be,0x000000bf,0x00000002,0x00000043,0x0003003e,0x000000b9,0x000000c0,0x000100fd, + 0x00010038 +}; \ No newline at end of file diff --git a/src/vk/spirv/world_warp_vert.c b/src/vk/spirv/world_warp_vert.c new file mode 100644 index 0000000..8614c78 --- /dev/null +++ b/src/vk/spirv/world_warp_vert.c @@ -0,0 +1,33 @@ + // 8.13.3559 + #pragma once +const uint32_t world_warp_vert_spv[] = { + 0x07230203,0x00010000,0x00080008,0x00000024,0x00000000,0x00020011,0x00000001,0x0006000b, + 0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001, + 0x0007000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x00000017,0x0000001b,0x00030003, + 0x00000002,0x000001c2,0x00090004,0x415f4c47,0x735f4252,0x72617065,0x5f657461,0x64616873, + 0x6f5f7265,0x63656a62,0x00007374,0x00040005,0x00000004,0x6e69616d,0x00000000,0x00050005, + 0x0000000c,0x69736f70,0x6e6f6974,0x00000073,0x00060005,0x00000015,0x505f6c67,0x65567265, + 0x78657472,0x00000000,0x00060006,0x00000015,0x00000000,0x505f6c67,0x7469736f,0x006e6f69, + 0x00030005,0x00000017,0x00000000,0x00060005,0x0000001b,0x565f6c67,0x65747265,0x646e4978, + 0x00007865,0x00050048,0x00000015,0x00000000,0x0000000b,0x00000000,0x00030047,0x00000015, + 0x00000002,0x00040047,0x0000001b,0x0000000b,0x0000002a,0x00020013,0x00000002,0x00030021, + 0x00000003,0x00000002,0x00030016,0x00000006,0x00000020,0x00040017,0x00000007,0x00000006, + 0x00000004,0x00040015,0x00000008,0x00000020,0x00000000,0x0004002b,0x00000008,0x00000009, + 0x00000003,0x0004001c,0x0000000a,0x00000007,0x00000009,0x00040020,0x0000000b,0x00000007, + 0x0000000a,0x0004002b,0x00000006,0x0000000d,0xbf800000,0x0004002b,0x00000006,0x0000000e, + 0x00000000,0x0004002b,0x00000006,0x0000000f,0x3f800000,0x0007002c,0x00000007,0x00000010, + 0x0000000d,0x0000000d,0x0000000e,0x0000000f,0x0004002b,0x00000006,0x00000011,0x40400000, + 0x0007002c,0x00000007,0x00000012,0x00000011,0x0000000d,0x0000000e,0x0000000f,0x0007002c, + 0x00000007,0x00000013,0x0000000d,0x00000011,0x0000000e,0x0000000f,0x0006002c,0x0000000a, + 0x00000014,0x00000010,0x00000012,0x00000013,0x0003001e,0x00000015,0x00000007,0x00040020, + 0x00000016,0x00000003,0x00000015,0x0004003b,0x00000016,0x00000017,0x00000003,0x00040015, + 0x00000018,0x00000020,0x00000001,0x0004002b,0x00000018,0x00000019,0x00000000,0x00040020, + 0x0000001a,0x00000001,0x00000018,0x0004003b,0x0000001a,0x0000001b,0x00000001,0x0004002b, + 0x00000018,0x0000001d,0x00000003,0x00040020,0x0000001f,0x00000007,0x00000007,0x00040020, + 0x00000022,0x00000003,0x00000007,0x00050036,0x00000002,0x00000004,0x00000000,0x00000003, + 0x000200f8,0x00000005,0x0004003b,0x0000000b,0x0000000c,0x00000007,0x0003003e,0x0000000c, + 0x00000014,0x0004003d,0x00000018,0x0000001c,0x0000001b,0x0005008b,0x00000018,0x0000001e, + 0x0000001c,0x0000001d,0x00050041,0x0000001f,0x00000020,0x0000000c,0x0000001e,0x0004003d, + 0x00000007,0x00000021,0x00000020,0x00050041,0x00000022,0x00000023,0x00000017,0x00000019, + 0x0003003e,0x00000023,0x00000021,0x000100fd,0x00010038 +}; \ No newline at end of file diff --git a/src/vk/vk_buffer.c b/src/vk/vk_buffer.c new file mode 100644 index 0000000..b5f9587 --- /dev/null +++ b/src/vk/vk_buffer.c @@ -0,0 +1,205 @@ +/* +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. +*/ + +#include "header/local.h" + +// internal helper +static void +copyBuffer(const VkBuffer * src, VkBuffer * dst, VkDeviceSize size) +{ + VkCommandBuffer commandBuffer = QVk_CreateCommandBuffer(&vk_transferCommandPool, + VK_COMMAND_BUFFER_LEVEL_PRIMARY); + QVk_BeginCommand(&commandBuffer); + + VkBufferCopy copyRegion = { + .srcOffset = 0, + .dstOffset = 0, + .size = size + }; + vkCmdCopyBuffer(commandBuffer, *src, *dst, 1, ©Region); + + QVk_SubmitCommand(&commandBuffer, &vk_device.transferQueue); + vkFreeCommandBuffers(vk_device.logical, vk_transferCommandPool, 1, + &commandBuffer); +} + +// internal helper +static void +createStagedBuffer(const void *data, VkDeviceSize size, qvkbuffer_t * dstBuffer, + qvkbufferopts_t bufferOpts) +{ + qvkstagingbuffer_t *stgBuffer; + stgBuffer = (qvkstagingbuffer_t *) malloc(sizeof(qvkstagingbuffer_t)); + VK_VERIFY(QVk_CreateStagingBuffer(size, stgBuffer, + VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT | + VK_MEMORY_PROPERTY_HOST_COHERENT_BIT, + VK_MEMORY_PROPERTY_HOST_CACHED_BIT)); + + if (data) + { + void *dst; + // staging buffers in vkQuake2 are required to be host coherent, + // so no flushing/invalidation is involved + dst = buffer_map(&stgBuffer->resource); + memcpy(dst, data, (size_t) size); + buffer_unmap(&stgBuffer->resource); + } + + VK_VERIFY(QVk_CreateBuffer(size, dstBuffer, bufferOpts)); + copyBuffer(&stgBuffer->resource.buffer, &dstBuffer->resource.buffer, size); + + QVk_FreeStagingBuffer(stgBuffer); + free(stgBuffer); +} + +VkResult +QVk_CreateBuffer(VkDeviceSize size, qvkbuffer_t *dstBuffer, + const qvkbufferopts_t options) +{ + VkBufferCreateInfo bcInfo = { + .sType = VK_STRUCTURE_TYPE_BUFFER_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .size = size, + .usage = options.usage, + .sharingMode = VK_SHARING_MODE_EXCLUSIVE, + .queueFamilyIndexCount = 0, + .pQueueFamilyIndices = NULL, + }; + + // separate transfer queue makes sense only if the buffer is targetted + // for being transfered to GPU, so ignore it if it's CPU-only + uint32_t queueFamilies[] = { + (uint32_t)vk_device.gfxFamilyIndex, + (uint32_t)vk_device.transferFamilyIndex + }; + + if (vk_device.gfxFamilyIndex != vk_device.transferFamilyIndex) + { + bcInfo.sharingMode = VK_SHARING_MODE_CONCURRENT; + bcInfo.queueFamilyIndexCount = 2; + bcInfo.pQueueFamilyIndices = queueFamilies; + } + + dstBuffer->currentOffset = 0; + return buffer_create(&dstBuffer->resource, bcInfo, + options.reqMemFlags, options.prefMemFlags); +} + +void +QVk_FreeBuffer(qvkbuffer_t *buffer) +{ + buffer_destroy(&buffer->resource); + buffer->currentOffset = 0; +} + +void +QVk_FreeStagingBuffer(qvkstagingbuffer_t *buffer) +{ + buffer_destroy(&buffer->resource); + buffer->currentOffset = 0; +} + +VkResult +QVk_CreateStagingBuffer(VkDeviceSize size, qvkstagingbuffer_t *dstBuffer, + VkMemoryPropertyFlags reqMemFlags, + VkMemoryPropertyFlags prefMemFlags) +{ + VkBufferCreateInfo bcInfo = { + .sType = VK_STRUCTURE_TYPE_BUFFER_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .size = ROUNDUP(size, 1024), + .usage = VK_BUFFER_USAGE_TRANSFER_SRC_BIT, + .sharingMode = VK_SHARING_MODE_EXCLUSIVE, + .queueFamilyIndexCount = 0, + .pQueueFamilyIndices = NULL, + }; + + reqMemFlags |= VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT | VK_MEMORY_PROPERTY_HOST_COHERENT_BIT; + + dstBuffer->currentOffset = 0; + return buffer_create(&dstBuffer->resource, bcInfo, reqMemFlags, + prefMemFlags); +} + +VkResult +QVk_CreateUniformBuffer(VkDeviceSize size, qvkbuffer_t *dstBuffer, + VkMemoryPropertyFlags reqMemFlags, + VkMemoryPropertyFlags prefMemFlags) +{ + qvkbufferopts_t dstOpts = { + .usage = VK_BUFFER_USAGE_UNIFORM_BUFFER_BIT, + .reqMemFlags = reqMemFlags, + .prefMemFlags = prefMemFlags, + }; + + dstOpts.reqMemFlags |= VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT; + + if((vk_device.properties.deviceType != VK_PHYSICAL_DEVICE_TYPE_INTEGRATED_GPU) || + (dstOpts.prefMemFlags & VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT) == 0) + { + dstOpts.prefMemFlags |= VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT; + } + + return QVk_CreateBuffer(size, dstBuffer, dstOpts); +} + +void +QVk_CreateVertexBuffer(const void *data, VkDeviceSize size, + qvkbuffer_t *dstBuffer, + VkMemoryPropertyFlags reqMemFlags, + VkMemoryPropertyFlags prefMemFlags) +{ + qvkbufferopts_t dstOpts = { + .usage = VK_BUFFER_USAGE_TRANSFER_DST_BIT | + VK_BUFFER_USAGE_VERTEX_BUFFER_BIT, + .reqMemFlags = reqMemFlags, + .prefMemFlags = prefMemFlags, + }; + + if((vk_device.properties.deviceType != VK_PHYSICAL_DEVICE_TYPE_INTEGRATED_GPU) || + (dstOpts.prefMemFlags & VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT) == 0) + { + dstOpts.prefMemFlags |= VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT; + } + + createStagedBuffer(data, size, dstBuffer, dstOpts); +} + +void +QVk_CreateIndexBuffer(const void *data, VkDeviceSize size, + qvkbuffer_t *dstBuffer, + VkMemoryPropertyFlags reqMemFlags, + VkMemoryPropertyFlags prefMemFlags) +{ + qvkbufferopts_t dstOpts = { + .usage = VK_BUFFER_USAGE_TRANSFER_DST_BIT | VK_BUFFER_USAGE_INDEX_BUFFER_BIT, + .reqMemFlags = reqMemFlags, + .prefMemFlags = prefMemFlags, + }; + + if((vk_device.properties.deviceType != VK_PHYSICAL_DEVICE_TYPE_INTEGRATED_GPU) || + (dstOpts.prefMemFlags & VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT) == 0) + { + dstOpts.prefMemFlags |= VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT; + } + + createStagedBuffer(data, size, dstBuffer, dstOpts); +} diff --git a/src/vk/vk_cmd.c b/src/vk/vk_cmd.c new file mode 100644 index 0000000..af55936 --- /dev/null +++ b/src/vk/vk_cmd.c @@ -0,0 +1,91 @@ +/* +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +*/ + +#include "header/local.h" + +VkResult QVk_BeginCommand(const VkCommandBuffer *commandBuffer) +{ + VkCommandBufferBeginInfo cmdInfo = { + .sType = VK_STRUCTURE_TYPE_COMMAND_BUFFER_BEGIN_INFO, + .pNext = NULL, + .flags = VK_COMMAND_BUFFER_USAGE_ONE_TIME_SUBMIT_BIT, + .pInheritanceInfo = NULL + }; + + return vkBeginCommandBuffer(*commandBuffer, &cmdInfo); +} + +void QVk_SubmitCommand(const VkCommandBuffer *commandBuffer, const VkQueue *queue) +{ + VK_VERIFY(vkEndCommandBuffer(*commandBuffer)); + + VkSubmitInfo submitInfo = { + .sType = VK_STRUCTURE_TYPE_SUBMIT_INFO, + .pNext = NULL, + .waitSemaphoreCount = 0, + .pWaitSemaphores = NULL, + .pWaitDstStageMask = NULL, + .commandBufferCount = 1, + .pCommandBuffers = commandBuffer, + .signalSemaphoreCount = 0, + .pSignalSemaphores = NULL + }; + + VkFenceCreateInfo fCreateInfo = { + .sType = VK_STRUCTURE_TYPE_FENCE_CREATE_INFO, + .pNext = NULL, + .flags = 0 + }; + + VkFence queueFence; + VK_VERIFY(vkCreateFence(vk_device.logical, &fCreateInfo, NULL, &queueFence)); + VK_VERIFY(vkQueueSubmit(*queue, 1, &submitInfo, queueFence)); + VK_VERIFY(vkWaitForFences(vk_device.logical, 1, &queueFence, VK_TRUE, UINT64_MAX)); + + vkDestroyFence(vk_device.logical, queueFence, NULL); +} + +VkResult QVk_CreateCommandPool(VkCommandPool *commandPool, uint32_t queueFamilyIndex) +{ + VkCommandPoolCreateInfo cpCreateInfo = { + .sType = VK_STRUCTURE_TYPE_COMMAND_POOL_CREATE_INFO, + .pNext = NULL, + // allow the command pool to be explicitly reset without reallocating it manually during recording each frame + .flags = VK_COMMAND_POOL_CREATE_TRANSIENT_BIT, + .queueFamilyIndex = queueFamilyIndex + }; + + return vkCreateCommandPool(vk_device.logical, &cpCreateInfo, NULL, commandPool); +} + +VkCommandBuffer QVk_CreateCommandBuffer(const VkCommandPool *commandPool, VkCommandBufferLevel level) +{ + VkCommandBuffer commandBuffer = VK_NULL_HANDLE; + VkCommandBufferAllocateInfo allocInfo = { + .sType = VK_STRUCTURE_TYPE_COMMAND_BUFFER_ALLOCATE_INFO, + .pNext = NULL, + .commandPool = *commandPool, + .level = level, + .commandBufferCount = 1 + }; + + VK_VERIFY(vkAllocateCommandBuffers(vk_device.logical, &allocInfo, &commandBuffer)); + return commandBuffer; +} diff --git a/src/vk/vk_common.c b/src/vk/vk_common.c new file mode 100644 index 0000000..ecb6876 --- /dev/null +++ b/src/vk/vk_common.c @@ -0,0 +1,2644 @@ +/* +Copyright (C) 1997-2001 Id Software, Inc. +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. +*/ + +/* +** VK_COMMON.C +** +** This file implements the operating system binding of Vk to QVk function +** pointers. When doing a port of Quake2 you must implement the following +** two functions: +** +** QVk_Init() - loads libraries, assigns function pointers, etc. +** QVk_Shutdown() - unloads libraries, NULLs function pointers +*/ +#include +#include "header/local.h" + +static SDL_Window *vk_window; + +// Vulkan instance, surface and memory allocator +VkInstance vk_instance = VK_NULL_HANDLE; +VkSurfaceKHR vk_surface = VK_NULL_HANDLE; + +// Vulkan device +qvkdevice_t vk_device = { + .physical = VK_NULL_HANDLE, + .logical = VK_NULL_HANDLE, + .gfxQueue = VK_NULL_HANDLE, + .presentQueue = VK_NULL_HANDLE, + .transferQueue = VK_NULL_HANDLE, + .gfxFamilyIndex = -1, + .presentFamilyIndex = -1, + .transferFamilyIndex = -1, + .screenshotSupported = false +}; + +// Vulkan swapchain +qvkswapchain_t vk_swapchain = { + .sc = VK_NULL_HANDLE, + .format = VK_FORMAT_UNDEFINED, + .presentMode = VK_PRESENT_MODE_MAILBOX_KHR, + .extent = { 0, 0 }, + .images = NULL, + .imageCount = 0 +}; + +// Vulkan renderpasses +qvkrenderpass_t vk_renderpasses[RP_COUNT] = { + // RP_WORLD + { + .rp = VK_NULL_HANDLE, + .colorLoadOp = VK_ATTACHMENT_LOAD_OP_LOAD, + .sampleCount = VK_SAMPLE_COUNT_1_BIT + }, + // RP_UI + { + .rp = VK_NULL_HANDLE, + .colorLoadOp = VK_ATTACHMENT_LOAD_OP_LOAD, + .sampleCount = VK_SAMPLE_COUNT_1_BIT + }, + // RP_WORLD_WARP + { + .rp = VK_NULL_HANDLE, + .colorLoadOp = VK_ATTACHMENT_LOAD_OP_LOAD, + .sampleCount = VK_SAMPLE_COUNT_1_BIT + } +}; + +// Vulkan pools +VkCommandPool vk_commandPool[NUM_CMDBUFFERS] = { VK_NULL_HANDLE, VK_NULL_HANDLE }; +VkCommandPool vk_transferCommandPool = VK_NULL_HANDLE; +VkDescriptorPool vk_descriptorPool = VK_NULL_HANDLE; +static VkCommandPool vk_stagingCommandPool[NUM_DYNBUFFERS] = { VK_NULL_HANDLE, VK_NULL_HANDLE }; +// Vulkan image views +static VkImageView *vk_imageviews = NULL; +// Vulkan framebuffers +static VkFramebuffer *vk_framebuffers[RP_COUNT]; +// color buffer containing main game/world view +qvktexture_t vk_colorbuffer = QVVKTEXTURE_INIT; +// color buffer with postprocessed game view +qvktexture_t vk_colorbufferWarp = QVVKTEXTURE_INIT; +// depth buffer +qvktexture_t vk_depthbuffer = QVVKTEXTURE_INIT; +// depth buffer for UI renderpass +static qvktexture_t vk_ui_depthbuffer = QVVKTEXTURE_INIT; +// render target for MSAA resolve +static qvktexture_t vk_msaaColorbuffer = QVVKTEXTURE_INIT; +// viewport and scissor +VkViewport vk_viewport = { .0f, .0f, .0f, .0f, .0f, .0f }; +VkRect2D vk_scissor = { { 0, 0 }, { 0, 0 } }; + +// Vulkan command buffers +static VkCommandBuffer *vk_commandbuffers = NULL; +// command buffer double buffering fences +static VkFence vk_fences[NUM_CMDBUFFERS]; +// semaphore: signal when next image is available for rendering +static VkSemaphore vk_imageAvailableSemaphores[NUM_CMDBUFFERS]; +// semaphore: signal when rendering to current command buffer is complete +static VkSemaphore vk_renderFinishedSemaphores[NUM_CMDBUFFERS]; +// tracker variables +VkCommandBuffer vk_activeCmdbuffer = VK_NULL_HANDLE; +// index of active command buffer +int vk_activeBufferIdx = 0; +// index of currently acquired image +static uint32_t vk_imageIndex = 0; +// index of currently used staging buffer +static int vk_activeStagingBuffer = 0; +// started rendering frame? +qboolean vk_frameStarted = false; +// the renderer needs to be restarted. +qboolean vk_restartNeeded = false; +// is QVk initialized? +qboolean vk_initialized = false; + +// render pipelines +qvkpipeline_t vk_drawTexQuadPipeline[RP_COUNT] = { + QVKPIPELINE_INIT, QVKPIPELINE_INIT, QVKPIPELINE_INIT }; +qvkpipeline_t vk_drawColorQuadPipeline[RP_COUNT] = { + QVKPIPELINE_INIT, QVKPIPELINE_INIT, QVKPIPELINE_INIT }; +qvkpipeline_t vk_drawModelPipelineFan[RP_COUNT] = { + QVKPIPELINE_INIT, QVKPIPELINE_INIT, QVKPIPELINE_INIT }; +qvkpipeline_t vk_drawNoDepthModelPipelineFan = QVKPIPELINE_INIT; +qvkpipeline_t vk_drawLefthandModelPipelineFan = QVKPIPELINE_INIT; +qvkpipeline_t vk_drawNullModelPipeline = QVKPIPELINE_INIT; +qvkpipeline_t vk_drawParticlesPipeline = QVKPIPELINE_INIT; +qvkpipeline_t vk_drawPointParticlesPipeline = QVKPIPELINE_INIT; +qvkpipeline_t vk_drawSpritePipeline = QVKPIPELINE_INIT; +qvkpipeline_t vk_drawPolyPipeline = QVKPIPELINE_INIT; +qvkpipeline_t vk_drawPolyLmapPipeline = QVKPIPELINE_INIT; +qvkpipeline_t vk_drawPolyWarpPipeline = QVKPIPELINE_INIT; +qvkpipeline_t vk_drawPolySolidWarpPipeline = QVKPIPELINE_INIT; +qvkpipeline_t vk_drawBeamPipeline = QVKPIPELINE_INIT; +qvkpipeline_t vk_drawSkyboxPipeline = QVKPIPELINE_INIT; +qvkpipeline_t vk_drawDLightPipeline = QVKPIPELINE_INIT; +qvkpipeline_t vk_showTrisPipeline = QVKPIPELINE_INIT; +qvkpipeline_t vk_shadowsPipelineStrip = QVKPIPELINE_INIT; +qvkpipeline_t vk_shadowsPipelineFan = QVKPIPELINE_INIT; +qvkpipeline_t vk_worldWarpPipeline = QVKPIPELINE_INIT; +qvkpipeline_t vk_postprocessPipeline = QVKPIPELINE_INIT; + +// samplers +static VkSampler vk_samplers[NUM_SAMPLERS]; + +// Vulkan function pointers +PFN_vkCreateDebugUtilsMessengerEXT qvkCreateDebugUtilsMessengerEXT; +PFN_vkDestroyDebugUtilsMessengerEXT qvkDestroyDebugUtilsMessengerEXT; +PFN_vkSetDebugUtilsObjectNameEXT qvkSetDebugUtilsObjectNameEXT; +PFN_vkSetDebugUtilsObjectTagEXT qvkSetDebugUtilsObjectTagEXT; +PFN_vkCmdBeginDebugUtilsLabelEXT qvkCmdBeginDebugUtilsLabelEXT; +PFN_vkCmdEndDebugUtilsLabelEXT qvkCmdEndDebugUtilsLabelEXT; +PFN_vkCmdInsertDebugUtilsLabelEXT qvkInsertDebugUtilsLabelEXT; + +#define VK_INPUTBIND_DESC(s) { \ + .binding = 0, \ + .stride = s, \ + .inputRate = VK_VERTEX_INPUT_RATE_VERTEX \ +}; + +#define VK_INPUTATTR_DESC(l, f, o) { \ + .binding = 0, \ + .location = l, \ + .format = f, \ + .offset = o \ +} + +#define VK_VERTEXINPUT_CINF(b, a) { \ + .sType = VK_STRUCTURE_TYPE_PIPELINE_VERTEX_INPUT_STATE_CREATE_INFO, \ + .pNext = NULL, \ + .flags = 0, \ + .vertexBindingDescriptionCount = 1, \ + .pVertexBindingDescriptions = &b, \ + .vertexAttributeDescriptionCount = sizeof(a) / sizeof(a[0]), \ + .pVertexAttributeDescriptions = a \ +} + +#define VK_VERTINFO(name, bindSize, ...) \ + VkVertexInputAttributeDescription attrDesc##name[] = { __VA_ARGS__ }; \ + VkVertexInputBindingDescription name##bindingDesc = VK_INPUTBIND_DESC(bindSize); \ + VkPipelineVertexInputStateCreateInfo vertInfo##name = VK_VERTEXINPUT_CINF(name##bindingDesc, attrDesc##name); + +#define VK_NULL_VERTEXINPUT_CINF { \ + .sType = VK_STRUCTURE_TYPE_PIPELINE_VERTEX_INPUT_STATE_CREATE_INFO, \ + .pNext = NULL, \ + .flags = 0, \ + .vertexBindingDescriptionCount = 0, \ + .pVertexBindingDescriptions = NULL, \ + .vertexAttributeDescriptionCount = 0, \ + .pVertexAttributeDescriptions = NULL \ +} + +enum { + SHADER_VERT_INDEX = 0, + SHADER_FRAG_INDEX = 1, + SHADER_INDEX_SIZE = 2 +}; + +#define VK_LOAD_VERTFRAG_SHADERS(shaders, namevert, namefrag) \ + DestroyShaderModule(shaders); \ + shaders[SHADER_VERT_INDEX] = QVk_CreateShader(namevert##_vert_spv, namevert##_vert_size, VK_SHADER_STAGE_VERTEX_BIT); \ + shaders[SHADER_FRAG_INDEX] = QVk_CreateShader(namefrag##_frag_spv, namefrag##_frag_size, VK_SHADER_STAGE_FRAGMENT_BIT); \ + QVk_DebugSetObjectName((uint64_t)shaders[SHADER_VERT_INDEX].module, VK_OBJECT_TYPE_SHADER_MODULE, "Shader Module: "#namevert".vert"); \ + QVk_DebugSetObjectName((uint64_t)shaders[SHADER_FRAG_INDEX].module, VK_OBJECT_TYPE_SHADER_MODULE, "Shader Module: "#namefrag".frag"); + +// global static buffers (reused, never changing) +static qvkbuffer_t vk_texRectVbo; +static qvkbuffer_t vk_colorRectVbo; +static qvkbuffer_t vk_rectIbo; + +// global dynamic buffers (double buffered) +static qvkbuffer_t vk_dynVertexBuffers[NUM_DYNBUFFERS]; +static qvkbuffer_t vk_dynIndexBuffers[NUM_DYNBUFFERS]; +static qvkbuffer_t vk_dynUniformBuffers[NUM_DYNBUFFERS]; +static VkDescriptorSet vk_uboDescriptorSets[NUM_DYNBUFFERS]; +static qvkstagingbuffer_t vk_stagingBuffers[NUM_DYNBUFFERS]; +static int vk_activeDynBufferIdx = 0; +static int vk_activeSwapBufferIdx = 0; + +// index buffer for triangle fan/strip emulation - all because Metal/MoltenVK don't support them +static VkBuffer *vk_triangleFanIbo = NULL; +static VkBuffer *vk_triangleStripIbo = NULL; +static uint32_t vk_triangleFanIboUsage = 0; + +// swap buffers used if primary dynamic buffers get full +#define NUM_SWAPBUFFER_SLOTS 4 +static int vk_swapBuffersCnt[NUM_SWAPBUFFER_SLOTS]; +static int vk_swapDescSetsCnt[NUM_SWAPBUFFER_SLOTS]; +static qvkbuffer_t *vk_swapBuffers[NUM_SWAPBUFFER_SLOTS]; +static VkDescriptorSet *vk_swapDescriptorSets[NUM_SWAPBUFFER_SLOTS]; + +// by how much will the dynamic buffers be resized if we run out of space? +#define BUFFER_RESIZE_FACTOR 2.f +// size in bytes used for uniform descriptor update +#define UNIFORM_ALLOC_SIZE 1024 +// start values for dynamic buffer sizes - bound to change if the application runs out of space (sizes in bytes) +#define VERTEX_BUFFER_SIZE (1024 * 1024) +#define INDEX_BUFFER_SIZE (2 * 1024) +#define UNIFORM_BUFFER_SIZE (2048 * 1024) +// staging buffer is constant in size but has a max limit beyond which it will be submitted +#define STAGING_BUFFER_MAXSIZE (8192 * 1024) +// initial index count in triangle fan buffer - assuming 200 indices (200*3 = 600 triangles) per object +#define TRIANGLE_INDEX_CNT 200 + +// Vulkan common descriptor sets for UBO, primary texture sampler and optional lightmap texture +static VkDescriptorSetLayout vk_uboDescSetLayout; +VkDescriptorSetLayout vk_samplerDescSetLayout; +static VkDescriptorSetLayout vk_samplerLightmapDescSetLayout; + +static const char *renderpassObjectNames[] = { + "RP_WORLD", + "RP_UI", + "RP_WORLD_WARP" +}; + +VkFormat QVk_FindDepthFormat() +{ + VkFormat depthFormats[] = { + VK_FORMAT_D32_SFLOAT_S8_UINT, + VK_FORMAT_D32_SFLOAT, + VK_FORMAT_D24_UNORM_S8_UINT, + VK_FORMAT_D16_UNORM_S8_UINT, + VK_FORMAT_D16_UNORM + }; + + for (int i = 0; i < 5; ++i) + { + VkFormatProperties formatProps; + vkGetPhysicalDeviceFormatProperties(vk_device.physical, depthFormats[i], &formatProps); + + if (formatProps.optimalTilingFeatures & VK_FORMAT_FEATURE_DEPTH_STENCIL_ATTACHMENT_BIT) + return depthFormats[i]; + } + + return VK_FORMAT_D16_UNORM; +} + +// internal helper +static VkSampleCountFlagBits GetSampleCount(int msaa, VkSampleCountFlagBits supportedMsaa) +{ + int step = 0, value = 64; + + static VkSampleCountFlagBits msaaModes[] = { + VK_SAMPLE_COUNT_64_BIT, + VK_SAMPLE_COUNT_32_BIT, + VK_SAMPLE_COUNT_16_BIT, + VK_SAMPLE_COUNT_8_BIT, + VK_SAMPLE_COUNT_4_BIT, + VK_SAMPLE_COUNT_2_BIT, + VK_SAMPLE_COUNT_1_BIT + }; + + while ((msaa < value && value > 1) || + ((supportedMsaa & msaaModes[step]) != msaaModes[step])) + { + value >>= 1; + step ++; + } + + R_Printf(PRINT_ALL, "MSAAx%d is used...\n", value); + + return msaaModes[step]; +} + +// internal helper +static void DestroyImageViews() +{ + if(!vk_imageviews) + return; + + for (int i = 0; i < vk_swapchain.imageCount; i++) + { + vkDestroyImageView(vk_device.logical, vk_imageviews[i], NULL); + } + free(vk_imageviews); + vk_imageviews = NULL; +} + +// internal helper +static VkResult CreateImageViews() +{ + VkResult res = VK_SUCCESS; + vk_imageviews = (VkImageView *)malloc(vk_swapchain.imageCount * sizeof(VkImageView)); + + for (size_t i = 0; i < vk_swapchain.imageCount; ++i) + { + res = QVk_CreateImageView(&vk_swapchain.images[i], + VK_IMAGE_ASPECT_COLOR_BIT, &vk_imageviews[i], vk_swapchain.format, 1); + QVk_DebugSetObjectName((uint64_t)vk_swapchain.images[i], + VK_OBJECT_TYPE_IMAGE, va("Swap Chain Image #" YQ2_COM_PRIdS, i)); + QVk_DebugSetObjectName((uint64_t)vk_imageviews[i], + VK_OBJECT_TYPE_IMAGE_VIEW, va("Swap Chain Image View #" YQ2_COM_PRIdS, i)); + + if (res != VK_SUCCESS) + { + DestroyImageViews(); + return res; + } + } + + return res; +} + +// internal helper +static void DestroyFramebuffers() +{ + for (int f = 0; f < RP_COUNT; f++) + { + if (vk_framebuffers[f]) + { + for (int i = 0; i < vk_swapchain.imageCount; ++i) + { + vkDestroyFramebuffer(vk_device.logical, vk_framebuffers[f][i], NULL); + } + + free(vk_framebuffers[f]); + vk_framebuffers[f] = NULL; + } + } +} + +// internal helper +static VkResult CreateFramebuffers() +{ + for(int i = 0; i < RP_COUNT; ++i) + vk_framebuffers[i] = (VkFramebuffer *)malloc(vk_swapchain.imageCount * sizeof(VkFramebuffer)); + + VkFramebufferCreateInfo fbCreateInfos[] = { + // RP_WORLD: main world view framebuffer + { + .sType = VK_STRUCTURE_TYPE_FRAMEBUFFER_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .renderPass = vk_renderpasses[RP_WORLD].rp, + .attachmentCount = (vk_renderpasses[RP_WORLD].sampleCount != VK_SAMPLE_COUNT_1_BIT) ? 3 : 2, + .width = vk_swapchain.extent.width, + .height = vk_swapchain.extent.height, + .layers = 1 + }, + // RP_UI: UI framebuffer + { + .sType = VK_STRUCTURE_TYPE_FRAMEBUFFER_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .renderPass = vk_renderpasses[RP_UI].rp, + .attachmentCount = 3, + .width = vk_swapchain.extent.width, + .height = vk_swapchain.extent.height, + .layers = 1 + }, + // RP_WORLD_WARP: warped main world view (postprocessing) framebuffer + { + .sType = VK_STRUCTURE_TYPE_FRAMEBUFFER_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .renderPass = vk_renderpasses[RP_WORLD_WARP].rp, + .attachmentCount = 2, + .width = vk_swapchain.extent.width, + .height = vk_swapchain.extent.height, + .layers = 1 + } + }; + + VkImageView worldAttachments[] = { vk_colorbuffer.imageView, vk_depthbuffer.imageView, vk_msaaColorbuffer.imageView }; + VkImageView warpAttachments[] = { vk_colorbuffer.imageView, vk_colorbufferWarp.imageView }; + + fbCreateInfos[RP_WORLD].pAttachments = worldAttachments; + fbCreateInfos[RP_WORLD_WARP].pAttachments = warpAttachments; + + for (size_t i = 0; i < vk_swapchain.imageCount; ++i) + { + VkImageView uiAttachments[] = { vk_colorbufferWarp.imageView, vk_ui_depthbuffer.imageView, vk_imageviews[i] }; + fbCreateInfos[RP_UI].pAttachments = uiAttachments; + + for (int j = 0; j < RP_COUNT; ++j) + { + VkResult res = vkCreateFramebuffer(vk_device.logical, &fbCreateInfos[j], NULL, &vk_framebuffers[j][i]); + QVk_DebugSetObjectName((uint64_t)vk_framebuffers[j][i], + VK_OBJECT_TYPE_FRAMEBUFFER, va("Framebuffer #" YQ2_COM_PRIdS "for Render Pass %s", + i, renderpassObjectNames[j])); + + if (res != VK_SUCCESS) + { + R_Printf(PRINT_ALL, "%s(): framebuffer #%d create error: %s\n", __func__, j, QVk_GetError(res)); + DestroyFramebuffers(); + return res; + } + } + } + + return VK_SUCCESS; +} + +// internal helper +static VkResult CreateRenderpasses() +{ + qboolean msaaEnabled = vk_renderpasses[RP_WORLD].sampleCount != VK_SAMPLE_COUNT_1_BIT; + + /* + * world view setup + */ + // The color attachment is loaded from the previous frame and stored + // after the frame is drawn to mask geometry errors in the skybox + // that may leave some pixels without coverage. + VkAttachmentDescription worldAttachments[] = { + // Single-sample color attachment. + { + .flags = 0, + .format = vk_swapchain.format, + .samples = VK_SAMPLE_COUNT_1_BIT, + .loadOp = (msaaEnabled ? VK_ATTACHMENT_LOAD_OP_DONT_CARE : VK_ATTACHMENT_LOAD_OP_LOAD), + .storeOp = VK_ATTACHMENT_STORE_OP_STORE, + .stencilLoadOp = VK_ATTACHMENT_LOAD_OP_DONT_CARE, + .stencilStoreOp = VK_ATTACHMENT_STORE_OP_DONT_CARE, + .initialLayout = VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL, + .finalLayout = VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL + }, + // depth attachment + { + .flags = 0, + .format = QVk_FindDepthFormat(), + .samples = vk_renderpasses[RP_WORLD].sampleCount, + .loadOp = VK_ATTACHMENT_LOAD_OP_CLEAR, + .storeOp = VK_ATTACHMENT_STORE_OP_DONT_CARE, + .stencilLoadOp = VK_ATTACHMENT_LOAD_OP_DONT_CARE, + .stencilStoreOp = VK_ATTACHMENT_STORE_OP_DONT_CARE, + .initialLayout = VK_IMAGE_LAYOUT_UNDEFINED, + .finalLayout = VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL + }, + // MSAA attachment + { + .flags = 0, + .format = vk_swapchain.format, + .samples = vk_renderpasses[RP_WORLD].sampleCount, + .loadOp = (msaaEnabled ? VK_ATTACHMENT_LOAD_OP_DONT_CARE : VK_ATTACHMENT_LOAD_OP_LOAD), + .storeOp = (msaaEnabled ? VK_ATTACHMENT_STORE_OP_DONT_CARE : VK_ATTACHMENT_STORE_OP_STORE), + .stencilLoadOp = VK_ATTACHMENT_LOAD_OP_DONT_CARE, + .stencilStoreOp = VK_ATTACHMENT_STORE_OP_DONT_CARE, + .initialLayout = VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL, + .finalLayout = VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL + } + }; + + VkAttachmentReference worldAttachmentRefs[] = { + // color + { + .attachment = msaaEnabled ? 2 : 0, + .layout = VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL, + }, + // depth + { + .attachment = 1, + .layout = VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL + }, + // MSAA resolve + { + .attachment = 0, + .layout = VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL + } + }; + + // primary renderpass writes to color, depth and optional MSAA resolve + VkSubpassDescription worldSubpassDesc = { + .flags = 0, + .pipelineBindPoint = VK_PIPELINE_BIND_POINT_GRAPHICS, + .inputAttachmentCount = 0, + .pInputAttachments = NULL, + .colorAttachmentCount = 1, + .pColorAttachments = &worldAttachmentRefs[0], + .pResolveAttachments = msaaEnabled ? &worldAttachmentRefs[2] : NULL, + .pDepthStencilAttachment = &worldAttachmentRefs[1], + .preserveAttachmentCount = 0, + .pPreserveAttachments = NULL + }; + + /* + * world warp setup + */ + VkAttachmentDescription warpAttachments[] = { + // color attachment - input from RP_WORLD renderpass + { + .flags = 0, + .format = vk_swapchain.format, + .samples = VK_SAMPLE_COUNT_1_BIT, + .loadOp = VK_ATTACHMENT_LOAD_OP_LOAD, + .storeOp = VK_ATTACHMENT_STORE_OP_STORE, + .stencilLoadOp = VK_ATTACHMENT_LOAD_OP_DONT_CARE, + .stencilStoreOp = VK_ATTACHMENT_STORE_OP_DONT_CARE, + .initialLayout = VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL, + .finalLayout = VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL + }, + // color attachment output - warped/postprocessed image that ends up in RP_UI + { + .flags = 0, + .format = vk_swapchain.format, + .samples = VK_SAMPLE_COUNT_1_BIT, + .loadOp = VK_ATTACHMENT_LOAD_OP_CLEAR, + .storeOp = VK_ATTACHMENT_STORE_OP_STORE, + .stencilLoadOp = VK_ATTACHMENT_LOAD_OP_DONT_CARE, + .stencilStoreOp = VK_ATTACHMENT_STORE_OP_DONT_CARE, + .initialLayout = VK_IMAGE_LAYOUT_UNDEFINED, + .finalLayout = VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL + } + }; + + VkAttachmentReference warpAttachmentRef = { + // output color + .attachment = 1, + .layout = VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL + }; + + // world view postprocess writes to a separate color buffer + VkSubpassDescription warpSubpassDesc = { + .flags = 0, + .pipelineBindPoint = VK_PIPELINE_BIND_POINT_GRAPHICS, + .inputAttachmentCount = 0, + .pInputAttachments = NULL, + .colorAttachmentCount = 1, + .pColorAttachments = &warpAttachmentRef, + .pResolveAttachments = NULL, + .pDepthStencilAttachment = NULL, + .preserveAttachmentCount = 0, + .pPreserveAttachments = NULL + }; + + /* + * UI setup + */ + VkAttachmentDescription uiAttachments[] = { + // color attachment + { + .flags = 0, + .format = vk_swapchain.format, + .samples = VK_SAMPLE_COUNT_1_BIT, + .loadOp = VK_ATTACHMENT_LOAD_OP_LOAD, + .storeOp = VK_ATTACHMENT_STORE_OP_DONT_CARE, + .stencilLoadOp = VK_ATTACHMENT_LOAD_OP_DONT_CARE, + .stencilStoreOp = VK_ATTACHMENT_STORE_OP_DONT_CARE, + .initialLayout = VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL, + .finalLayout = VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL + }, + // depth attachment - because of player model preview in settings screen + { + .flags = 0, + .format = QVk_FindDepthFormat(), + .samples = VK_SAMPLE_COUNT_1_BIT, + .loadOp = VK_ATTACHMENT_LOAD_OP_CLEAR, + .storeOp = VK_ATTACHMENT_STORE_OP_DONT_CARE, + .stencilLoadOp = VK_ATTACHMENT_LOAD_OP_DONT_CARE, + .stencilStoreOp = VK_ATTACHMENT_STORE_OP_DONT_CARE, + .initialLayout = VK_IMAGE_LAYOUT_UNDEFINED, + .finalLayout = VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL + }, + // swapchain presentation + { + .flags = 0, + .format = vk_swapchain.format, + .samples = VK_SAMPLE_COUNT_1_BIT, + .loadOp = VK_ATTACHMENT_LOAD_OP_CLEAR, + .storeOp = VK_ATTACHMENT_STORE_OP_STORE, + .stencilLoadOp = VK_ATTACHMENT_LOAD_OP_DONT_CARE, + .stencilStoreOp = VK_ATTACHMENT_STORE_OP_DONT_CARE, + .initialLayout = VK_IMAGE_LAYOUT_UNDEFINED, + .finalLayout = VK_IMAGE_LAYOUT_PRESENT_SRC_KHR + } + }; + + // UI renderpass writes to depth (for player model in setup screen) and outputs to swapchain + VkAttachmentReference uiAttachmentRefs[] = { + // depth + { + .attachment = 1, + .layout = VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL + }, + // swapchain output + { + .attachment = 2, + .layout = VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL + } + }; + + VkSubpassDescription uiSubpassDesc = { + .flags = 0, + .pipelineBindPoint = VK_PIPELINE_BIND_POINT_GRAPHICS, + .inputAttachmentCount = 0, + .pInputAttachments = NULL, + .colorAttachmentCount = 1, + .pColorAttachments = &uiAttachmentRefs[1], + .pResolveAttachments = NULL, + .pDepthStencilAttachment = &uiAttachmentRefs[0], + .preserveAttachmentCount = 0, + .pPreserveAttachments = NULL + }; + + /* + * create the render passes + */ + // we're using 3 render passes which depend on each other (main color -> warp/postprocessing -> ui) + VkSubpassDependency subpassDeps[2] = { + { + .srcSubpass = VK_SUBPASS_EXTERNAL, + .dstSubpass = 0, + .srcStageMask = VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT, + .dstStageMask = VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT, + .srcAccessMask = VK_ACCESS_SHADER_READ_BIT, + .dstAccessMask = (VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT | VK_ACCESS_COLOR_ATTACHMENT_READ_BIT), + .dependencyFlags = VK_DEPENDENCY_BY_REGION_BIT + }, + { + .srcSubpass = 0, + .dstSubpass = VK_SUBPASS_EXTERNAL, + .srcStageMask = VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT, + .dstStageMask = VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT, + .srcAccessMask = VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT, + .dstAccessMask = VK_ACCESS_SHADER_READ_BIT, + .dependencyFlags = VK_DEPENDENCY_BY_REGION_BIT + } + }; + + VkRenderPassCreateInfo rpCreateInfos[] = { + // offscreen world rendering to color buffer (RP_WORLD) + { + .sType = VK_STRUCTURE_TYPE_RENDER_PASS_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .attachmentCount = msaaEnabled ? 3 : 2, + .pAttachments = worldAttachments, + .subpassCount = 1, + .pSubpasses = &worldSubpassDesc, + .dependencyCount = 2, + .pDependencies = subpassDeps + }, + // UI rendering (RP_UI) + { + .sType = VK_STRUCTURE_TYPE_RENDER_PASS_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .attachmentCount = 3, + .pAttachments = uiAttachments, + .subpassCount = 1, + .pSubpasses = &uiSubpassDesc, + .dependencyCount = 2, + .pDependencies = subpassDeps + }, + // world warp/postprocessing render pass (RP_WORLD_WARP) + { + .sType = VK_STRUCTURE_TYPE_RENDER_PASS_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .attachmentCount = 2, + .pAttachments = warpAttachments, + .subpassCount = 1, + .pSubpasses = &warpSubpassDesc, + .dependencyCount = 2, + .pDependencies = subpassDeps + } + }; + + for (int i = 0; i < RP_COUNT; ++i) + { + VkResult res = vkCreateRenderPass(vk_device.logical, &rpCreateInfos[i], NULL, &vk_renderpasses[i].rp); + if (res != VK_SUCCESS) + { + R_Printf(PRINT_ALL, "%s(): renderpass #%d create error: %s\n", __func__, i, QVk_GetError(res)); + return res; + } + QVk_DebugSetObjectName((uint64_t)vk_renderpasses[i].rp, VK_OBJECT_TYPE_RENDER_PASS, + va("Render Pass: %s", renderpassObjectNames[i])); + } + + return VK_SUCCESS; +} + +// internal helper +static void CreateDrawBuffers() +{ + QVk_CreateDepthBuffer(vk_renderpasses[RP_WORLD].sampleCount, + &vk_depthbuffer); + R_Printf(PRINT_ALL, "...created world depth buffer\n"); + QVk_CreateDepthBuffer(VK_SAMPLE_COUNT_1_BIT, &vk_ui_depthbuffer); + R_Printf(PRINT_ALL, "...created UI depth buffer\n"); + QVk_CreateColorBuffer(VK_SAMPLE_COUNT_1_BIT, &vk_colorbuffer, + VK_IMAGE_USAGE_SAMPLED_BIT); + R_Printf(PRINT_ALL, "...created world color buffer\n"); + QVk_CreateColorBuffer(VK_SAMPLE_COUNT_1_BIT, &vk_colorbufferWarp, + VK_IMAGE_USAGE_SAMPLED_BIT); + R_Printf(PRINT_ALL, "...created world postpocess color buffer\n"); + + if (vk_renderpasses[RP_WORLD].sampleCount > 1) + { + QVk_CreateColorBuffer(vk_renderpasses[RP_WORLD].sampleCount, &vk_msaaColorbuffer, + VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT); + R_Printf(PRINT_ALL, "...created MSAAx%d color buffer\n", + vk_renderpasses[RP_WORLD].sampleCount); + } + + QVk_DebugSetObjectName((uint64_t)vk_depthbuffer.resource.image, + VK_OBJECT_TYPE_IMAGE, "Depth Buffer: World"); + QVk_DebugSetObjectName((uint64_t)vk_depthbuffer.imageView, + VK_OBJECT_TYPE_IMAGE_VIEW, "Image View: World Depth Buffer"); + QVk_DebugSetObjectName((uint64_t)vk_depthbuffer.resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, "Memory: World Depth Buffer"); + QVk_DebugSetObjectName((uint64_t)vk_ui_depthbuffer.resource.image, + VK_OBJECT_TYPE_IMAGE, "Depth Buffer: UI"); + QVk_DebugSetObjectName((uint64_t)vk_ui_depthbuffer.imageView, + VK_OBJECT_TYPE_IMAGE_VIEW, "Image View: UI Depth Buffer"); + QVk_DebugSetObjectName((uint64_t)vk_ui_depthbuffer.resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, "Memory: UI Depth Buffer"); + QVk_DebugSetObjectName((uint64_t)vk_colorbuffer.resource.image, + VK_OBJECT_TYPE_IMAGE, "Color Buffer: World"); + QVk_DebugSetObjectName((uint64_t)vk_colorbuffer.imageView, + VK_OBJECT_TYPE_IMAGE_VIEW, "Image View: World Color Buffer"); + QVk_DebugSetObjectName((uint64_t)vk_colorbuffer.resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, "Memory: World Color Buffer"); + QVk_DebugSetObjectName((uint64_t)vk_colorbufferWarp.resource.image, + VK_OBJECT_TYPE_IMAGE, "Color Buffer: Warp Postprocess"); + QVk_DebugSetObjectName((uint64_t)vk_colorbufferWarp.imageView, + VK_OBJECT_TYPE_IMAGE_VIEW, "Image View: Warp Postprocess Color Buffer"); + QVk_DebugSetObjectName((uint64_t)vk_colorbufferWarp.resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, "Memory: Warp Postprocess Color Buffer"); + + if (vk_renderpasses[RP_WORLD].sampleCount > 1) + { + QVk_DebugSetObjectName((uint64_t)vk_msaaColorbuffer.resource.image, + VK_OBJECT_TYPE_IMAGE, "Color Buffer: MSAA"); + QVk_DebugSetObjectName((uint64_t)vk_msaaColorbuffer.imageView, + VK_OBJECT_TYPE_IMAGE_VIEW, "Image View: MSAA Color Buffer"); + QVk_DebugSetObjectName((uint64_t)vk_msaaColorbuffer.resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, "Memory: MSAA Color Buffer"); + } +} + +// internal helper +static void DestroyDrawBuffer(qvktexture_t *drawBuffer) +{ + if (drawBuffer->imageView != VK_NULL_HANDLE) + { + vkDestroyImageView(vk_device.logical, drawBuffer->imageView, NULL); + drawBuffer->imageView = VK_NULL_HANDLE; + } + + if (drawBuffer->resource.image != VK_NULL_HANDLE) + { + image_destroy(&drawBuffer->resource); + } +} + +// internal helper +static void DestroyDrawBuffers() +{ + DestroyDrawBuffer(&vk_depthbuffer); + DestroyDrawBuffer(&vk_ui_depthbuffer); + DestroyDrawBuffer(&vk_colorbuffer); + DestroyDrawBuffer(&vk_colorbufferWarp); + DestroyDrawBuffer(&vk_msaaColorbuffer); +} + +// internal helper +static void CreateDescriptorSetLayouts() +{ + VkDescriptorSetLayoutBinding layoutBinding = { + .binding = 0, + .descriptorType = VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC, + .descriptorCount = 1, + .stageFlags = VK_SHADER_STAGE_VERTEX_BIT, + .pImmutableSamplers = NULL + }; + + VkDescriptorSetLayoutCreateInfo layoutInfo = { + .sType = VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .bindingCount = 1, + .pBindings = &layoutBinding + }; + + // uniform buffer object layout + VK_VERIFY(vkCreateDescriptorSetLayout(vk_device.logical, &layoutInfo, NULL, &vk_uboDescSetLayout)); + // sampler layout + layoutBinding.descriptorType = VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER; + layoutBinding.stageFlags = VK_SHADER_STAGE_FRAGMENT_BIT; + VK_VERIFY(vkCreateDescriptorSetLayout(vk_device.logical, &layoutInfo, NULL, &vk_samplerDescSetLayout)); + // secondary sampler: lightmaps + VK_VERIFY(vkCreateDescriptorSetLayout(vk_device.logical, &layoutInfo, NULL, &vk_samplerLightmapDescSetLayout)); + + QVk_DebugSetObjectName((uint64_t)vk_uboDescSetLayout, VK_OBJECT_TYPE_DESCRIPTOR_SET_LAYOUT, "Descriptor Set Layout: UBO"); + QVk_DebugSetObjectName((uint64_t)vk_samplerDescSetLayout, VK_OBJECT_TYPE_DESCRIPTOR_SET_LAYOUT, "Descriptor Set Layout: Sampler"); + QVk_DebugSetObjectName((uint64_t)vk_samplerLightmapDescSetLayout, VK_OBJECT_TYPE_DESCRIPTOR_SET_LAYOUT, "Descriptor Set Layout: Sampler + Lightmap"); +} + +// internal helper +static void CreateSamplersHelper(VkSampler *samplers, VkSamplerAddressMode addressMode) +{ + VkSamplerCreateInfo samplerInfo = { + .sType = VK_STRUCTURE_TYPE_SAMPLER_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .magFilter = VK_FILTER_NEAREST, + .minFilter = VK_FILTER_NEAREST, + .mipmapMode = VK_SAMPLER_MIPMAP_MODE_NEAREST, + .addressModeU = addressMode, + .addressModeV = addressMode, + .addressModeW = addressMode, + .mipLodBias = 0.f, + .anisotropyEnable = VK_FALSE, + .maxAnisotropy = 1.f, + .compareEnable = VK_FALSE, + .compareOp = VK_COMPARE_OP_ALWAYS, + .minLod = 0.f, + .maxLod = 1.f, + .borderColor = VK_BORDER_COLOR_INT_OPAQUE_BLACK, + .unnormalizedCoordinates = VK_FALSE + }; + + assert((vk_device.properties.limits.maxSamplerAnisotropy > 1.f) && "maxSamplerAnisotropy is 1"); + if (vk_device.features.samplerAnisotropy && vk_aniso->value > 0.f) + { + const float maxAniso = min(max(vk_aniso->value, 1.f), vk_device.properties.limits.maxSamplerAnisotropy); + samplerInfo.anisotropyEnable = VK_TRUE; + samplerInfo.maxAnisotropy = maxAniso; + } + + VK_VERIFY(vkCreateSampler(vk_device.logical, &samplerInfo, NULL, &samplers[S_NEAREST])); + QVk_DebugSetObjectName((uint64_t)samplers[S_NEAREST], VK_OBJECT_TYPE_SAMPLER, "Sampler: S_NEAREST"); + + { + VkSamplerCreateInfo nuSamplerInfo = samplerInfo; + + // unnormalizedCoordinates set to VK_TRUE forces other parameters to have restricted values. + nuSamplerInfo.minLod = 0.f; + nuSamplerInfo.maxLod = 0.f; + nuSamplerInfo.addressModeU = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE; + nuSamplerInfo.addressModeV = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE; + nuSamplerInfo.addressModeW = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE; + nuSamplerInfo.unnormalizedCoordinates = VK_TRUE; + nuSamplerInfo.anisotropyEnable = VK_FALSE; + nuSamplerInfo.maxAnisotropy = 1.f; + + VK_VERIFY(vkCreateSampler(vk_device.logical, &nuSamplerInfo, NULL, &samplers[S_NEAREST_UNNORMALIZED])); + QVk_DebugSetObjectName((uint64_t)samplers[S_NEAREST_UNNORMALIZED], VK_OBJECT_TYPE_SAMPLER, "Sampler: S_NEAREST_UNNORMALIZED"); + } + + samplerInfo.maxLod = FLT_MAX; + VK_VERIFY(vkCreateSampler(vk_device.logical, &samplerInfo, NULL, &samplers[S_MIPMAP_NEAREST])); + QVk_DebugSetObjectName((uint64_t)samplers[S_MIPMAP_NEAREST], VK_OBJECT_TYPE_SAMPLER, "Sampler: S_MIPMAP_NEAREST"); + + samplerInfo.magFilter = VK_FILTER_LINEAR; + samplerInfo.minFilter = VK_FILTER_LINEAR; + samplerInfo.mipmapMode = VK_SAMPLER_MIPMAP_MODE_LINEAR; + VK_VERIFY(vkCreateSampler(vk_device.logical, &samplerInfo, NULL, &samplers[S_MIPMAP_LINEAR])); + QVk_DebugSetObjectName((uint64_t)samplers[S_MIPMAP_LINEAR], VK_OBJECT_TYPE_SAMPLER, "Sampler: S_MIPMAP_LINEAR"); + + samplerInfo.maxLod = 1.f; + VK_VERIFY(vkCreateSampler(vk_device.logical, &samplerInfo, NULL, &samplers[S_LINEAR])); + QVk_DebugSetObjectName((uint64_t)samplers[S_LINEAR], VK_OBJECT_TYPE_SAMPLER, "Sampler: S_LINEAR"); +} + +// internal helper +static void CreateSamplers() +{ + CreateSamplersHelper(vk_samplers, VK_SAMPLER_ADDRESS_MODE_REPEAT); + CreateSamplersHelper(vk_samplers + S_SAMPLER_CNT, VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE); +} + +// internal helper +static void DestroySamplers() +{ + int i; + for (i = 0; i < NUM_SAMPLERS; ++i) + { + if (vk_samplers[i] != VK_NULL_HANDLE) + vkDestroySampler(vk_device.logical, vk_samplers[i], NULL); + + vk_samplers[i] = VK_NULL_HANDLE; + } +} + +// internal helper +static void CreateDescriptorPool() +{ + VkDescriptorPoolSize poolSizes[] = { + // UBO + { + .type = VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC, + .descriptorCount = 16 + }, + // sampler + { + .type = VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, + .descriptorCount = MAX_VKTEXTURES + 1 + } + }; + + VkDescriptorPoolCreateInfo poolInfo = { + .sType = VK_STRUCTURE_TYPE_DESCRIPTOR_POOL_CREATE_INFO, + .pNext = NULL, + .flags = VK_DESCRIPTOR_POOL_CREATE_FREE_DESCRIPTOR_SET_BIT, + .maxSets = MAX_VKTEXTURES + 32, + .poolSizeCount = sizeof(poolSizes) / sizeof(poolSizes[0]), + .pPoolSizes = poolSizes, + }; + + VK_VERIFY(vkCreateDescriptorPool(vk_device.logical, &poolInfo, NULL, &vk_descriptorPool)); + QVk_DebugSetObjectName((uint64_t)vk_descriptorPool, VK_OBJECT_TYPE_DESCRIPTOR_POOL, "Descriptor Pool: Sampler + UBO"); +} + +// internal helper +static void CreateUboDescriptorSet(VkDescriptorSet *descSet, VkBuffer buffer) +{ + VkDescriptorSetAllocateInfo dsAllocInfo = { + .sType = VK_STRUCTURE_TYPE_DESCRIPTOR_SET_ALLOCATE_INFO, + .pNext = NULL, + .descriptorPool = vk_descriptorPool, + .descriptorSetCount = 1, + .pSetLayouts = &vk_uboDescSetLayout + }; + + VK_VERIFY(vkAllocateDescriptorSets(vk_device.logical, &dsAllocInfo, descSet)); + + VkDescriptorBufferInfo bufferInfo = { + .buffer = buffer, + .offset = 0, + .range = UNIFORM_ALLOC_SIZE + }; + + VkWriteDescriptorSet descriptorWrite = { + .sType = VK_STRUCTURE_TYPE_WRITE_DESCRIPTOR_SET, + .pNext = NULL, + .dstSet = *descSet, + .dstBinding = 0, + .dstArrayElement = 0, + .descriptorCount = 1, + .descriptorType = VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC, + .pImageInfo = NULL, + .pBufferInfo = &bufferInfo, + .pTexelBufferView = NULL, + }; + + vkUpdateDescriptorSets(vk_device.logical, 1, &descriptorWrite, 0, NULL); +} + +// internal helper +static void CreateDynamicBuffers() +{ + for (int i = 0; i < NUM_DYNBUFFERS; ++i) + { + QVk_CreateVertexBuffer(NULL, vk_config.vertex_buffer_size, &vk_dynVertexBuffers[i], + VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT, VK_MEMORY_PROPERTY_HOST_CACHED_BIT); + QVk_CreateIndexBuffer(NULL, vk_config.index_buffer_size, &vk_dynIndexBuffers[i], + VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT, VK_MEMORY_PROPERTY_HOST_CACHED_BIT); + VK_VERIFY(QVk_CreateUniformBuffer(vk_config.uniform_buffer_size, + &vk_dynUniformBuffers[i], VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT, + VK_MEMORY_PROPERTY_HOST_CACHED_BIT)); + // keep dynamic buffers persistently mapped + vk_dynVertexBuffers[i].pMappedData = buffer_map(&vk_dynVertexBuffers[i].resource); + vk_dynIndexBuffers[i].pMappedData = buffer_map(&vk_dynIndexBuffers[i].resource); + vk_dynUniformBuffers[i].pMappedData = buffer_map(&vk_dynUniformBuffers[i].resource); + // create descriptor set for the uniform buffer + CreateUboDescriptorSet(&vk_uboDescriptorSets[i], vk_dynUniformBuffers[i].resource.buffer); + + QVk_DebugSetObjectName((uint64_t)vk_uboDescriptorSets[i], + VK_OBJECT_TYPE_DESCRIPTOR_SET, va("Dynamic UBO Descriptor Set #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_dynVertexBuffers[i].resource.buffer, + VK_OBJECT_TYPE_BUFFER, va("Dynamic Vertex Buffer #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_dynVertexBuffers[i].resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, va("Memory: Dynamic Vertex Buffer #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_dynIndexBuffers[i].resource.buffer, + VK_OBJECT_TYPE_BUFFER, va("Dynamic Index Buffer #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_dynIndexBuffers[i].resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, va("Memory: Dynamic Index Buffer #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_dynUniformBuffers[i].resource.buffer, + VK_OBJECT_TYPE_BUFFER, va("Dynamic Uniform Buffer #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_dynUniformBuffers[i].resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, va("Memory: Dynamic Uniform Buffer #%d", i)); + } +} + +// internal helper +static void ReleaseSwapBuffers() +{ + vk_activeSwapBufferIdx = (vk_activeSwapBufferIdx + 1) % NUM_SWAPBUFFER_SLOTS; + int releaseBufferIdx = (vk_activeSwapBufferIdx + 1) % NUM_SWAPBUFFER_SLOTS; + + if (vk_swapBuffersCnt[releaseBufferIdx] > 0) + { + for (int i = 0; i < vk_swapBuffersCnt[releaseBufferIdx]; i++) + QVk_FreeBuffer(&vk_swapBuffers[releaseBufferIdx][i]); + + free(vk_swapBuffers[releaseBufferIdx]); + vk_swapBuffers[releaseBufferIdx] = NULL; + vk_swapBuffersCnt[releaseBufferIdx] = 0; + } + + if (vk_swapDescSetsCnt[releaseBufferIdx] > 0) + { + vkFreeDescriptorSets(vk_device.logical, vk_descriptorPool, vk_swapDescSetsCnt[releaseBufferIdx], vk_swapDescriptorSets[releaseBufferIdx]); + + free(vk_swapDescriptorSets[releaseBufferIdx]); + vk_swapDescriptorSets[releaseBufferIdx] = NULL; + vk_swapDescSetsCnt[releaseBufferIdx] = 0; + } +} + +// internal helper +static int NextPow2(int v) +{ + v--; + v |= v >> 1; + v |= v >> 2; + v |= v >> 4; + v |= v >> 8; + v |= v >> 16; + v++; + return v; +} + +// internal helper +static uint8_t *QVk_GetIndexBuffer(VkDeviceSize size, VkDeviceSize *dstOffset, int currentBufferIdx); +static void RebuildTriangleIndexBuffer() +{ + int idx = 0; + VkDeviceSize dstOffset = 0; + VkDeviceSize bufferSize = 3 * vk_config.triangle_index_count * sizeof(uint16_t); + uint16_t *iboData = NULL; + uint16_t *fanData = malloc(bufferSize); + uint16_t *stripData = malloc(bufferSize); + + // fill the index buffer so that we can emulate triangle fans via triangle lists + for (int i = 0; i < vk_config.triangle_index_count; ++i) + { + fanData[idx++] = 0; + fanData[idx++] = i + 1; + fanData[idx++] = i + 2; + } + + // fill the index buffer so that we can emulate triangle strips via triangle lists + idx = 0; + for (int i = 2; i < (vk_config.triangle_index_count + 2); ++i) + { + if ((i%2) == 0) + { + stripData[idx++] = i - 2; + stripData[idx++] = i - 1; + stripData[idx++] = i; + } + else + { + stripData[idx++] = i; + stripData[idx++] = i - 1; + stripData[idx++] = i - 2; + } + } + + for (int i = 0; i < NUM_DYNBUFFERS; ++i) + { + VK_VERIFY(buffer_invalidate(&vk_dynIndexBuffers[i].resource)); + + iboData = (uint16_t *)QVk_GetIndexBuffer(bufferSize, &dstOffset, i); + if ((i%2) == 0) + { + memcpy(iboData, fanData, bufferSize); + } + else + { + memcpy(iboData, stripData, bufferSize); + } + + VK_VERIFY(buffer_flush(&vk_dynIndexBuffers[i].resource)); + } + + vk_triangleFanIbo = &vk_dynIndexBuffers[0].resource.buffer; + vk_triangleStripIbo = &vk_dynIndexBuffers[1].resource.buffer; + vk_triangleFanIboUsage = ((bufferSize % 4) == 0) ? bufferSize : (bufferSize + 4 - (bufferSize % 4)); + + free(fanData); + free(stripData); +} + +static void CreateStagingBuffer(VkDeviceSize size, qvkstagingbuffer_t *dstBuffer, int i) +{ + VkFenceCreateInfo fCreateInfo = { + .sType = VK_STRUCTURE_TYPE_FENCE_CREATE_INFO, + .flags = 0 + }; + + VK_VERIFY(QVk_CreateStagingBuffer(size, + dstBuffer, + VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT | VK_MEMORY_PROPERTY_HOST_COHERENT_BIT, + VK_MEMORY_PROPERTY_HOST_CACHED_BIT)); + dstBuffer->pMappedData = buffer_map(&dstBuffer->resource); + dstBuffer->submitted = false; + + VK_VERIFY(vkCreateFence(vk_device.logical, &fCreateInfo, + NULL, &dstBuffer->fence)); + + dstBuffer->cmdBuffer = QVk_CreateCommandBuffer(&vk_stagingCommandPool[i], + VK_COMMAND_BUFFER_LEVEL_PRIMARY); + VK_VERIFY(QVk_BeginCommand(&dstBuffer->cmdBuffer)); + + QVk_DebugSetObjectName((uint64_t)dstBuffer->fence, + VK_OBJECT_TYPE_FENCE, va("Fence: Staging Buffer #%d", i)); + QVk_DebugSetObjectName((uint64_t)dstBuffer->resource.buffer, + VK_OBJECT_TYPE_BUFFER, va("Staging Buffer #%d", i)); + QVk_DebugSetObjectName((uint64_t)dstBuffer->resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, va("Memory: Staging Buffer #%d", i)); + QVk_DebugSetObjectName((uintptr_t)dstBuffer->cmdBuffer, + VK_OBJECT_TYPE_COMMAND_BUFFER, va("Command Buffer: Staging Buffer #%d", i)); +} + +// internal helper +static void CreateStagingBuffers() +{ + for (int i = 0; i < NUM_DYNBUFFERS; ++i) + { + VK_VERIFY(QVk_CreateCommandPool(&vk_stagingCommandPool[i], + vk_device.gfxFamilyIndex)); + QVk_DebugSetObjectName((uint64_t)vk_stagingCommandPool[i], + VK_OBJECT_TYPE_COMMAND_POOL, va("Command Pool #%d: Staging", i)); + + CreateStagingBuffer(STAGING_BUFFER_MAXSIZE, &vk_stagingBuffers[i], i); + } +} + +// Records a memory barrier in the given command buffer. +void Qvk_MemoryBarrier(VkCommandBuffer cmdBuffer, VkPipelineStageFlags srcStageMask, VkPipelineStageFlags dstStageMask, VkAccessFlags srcAccessMask, VkAccessFlags dstAccessMask) +{ + const VkMemoryBarrier memBarrier = { + .sType = VK_STRUCTURE_TYPE_MEMORY_BARRIER, + .pNext = NULL, + .srcAccessMask = srcAccessMask, + .dstAccessMask = dstAccessMask, + }; + vkCmdPipelineBarrier(cmdBuffer, srcStageMask, dstStageMask, 0u, 1u, &memBarrier, 0u, NULL, 0u, NULL); +} + +// internal helper +static void SubmitStagingBuffer(int index) +{ + if (vk_stagingBuffers[index].submitted) + { + // buffer is alredy submitted + return; + } + + Qvk_MemoryBarrier(vk_stagingBuffers[index].cmdBuffer, + VK_PIPELINE_STAGE_TRANSFER_BIT, + VK_PIPELINE_STAGE_VERTEX_INPUT_BIT, + VK_ACCESS_TRANSFER_WRITE_BIT, + (VK_ACCESS_INDEX_READ_BIT | VK_ACCESS_VERTEX_ATTRIBUTE_READ_BIT)); + + VK_VERIFY(vkEndCommandBuffer(vk_stagingBuffers[index].cmdBuffer)); + + VkSubmitInfo submitInfo = { + .sType = VK_STRUCTURE_TYPE_SUBMIT_INFO, + .waitSemaphoreCount = 0, + .pWaitSemaphores = NULL, + .signalSemaphoreCount = 0, + .pSignalSemaphores = NULL, + .pWaitDstStageMask = NULL, + .commandBufferCount = 1, + .pCommandBuffers = &vk_stagingBuffers[index].cmdBuffer + }; + + VK_VERIFY(vkQueueSubmit(vk_device.gfxQueue, 1, &submitInfo, vk_stagingBuffers[index].fence)); + + vk_stagingBuffers[index].submitted = true; + vk_activeStagingBuffer = (vk_activeStagingBuffer + 1) % NUM_DYNBUFFERS; +} + +// internal helper +static void CreateStaticBuffers() +{ + const float texVerts[] = { -1., -1., 0., 0., + 1., 1., 1., 1., + -1., 1., 0., 1., + 1., -1., 1., 0. }; + + const float colorVerts[] = { -1., -1., + 1., 1., + -1., 1., + 1., -1. }; + + const uint32_t indices[] = { 0, 1, 2, 0, 3, 1 }; + + QVk_CreateVertexBuffer(texVerts, sizeof(texVerts), + &vk_texRectVbo, VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT, 0); + QVk_CreateVertexBuffer(colorVerts, sizeof(colorVerts), + &vk_colorRectVbo, VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT, 0); + QVk_CreateIndexBuffer(indices, sizeof(indices), + &vk_rectIbo, VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT, 0); + + QVk_DebugSetObjectName((uint64_t)vk_texRectVbo.resource.buffer, + VK_OBJECT_TYPE_BUFFER, "Static Buffer: Textured Rectangle VBO"); + QVk_DebugSetObjectName((uint64_t)vk_texRectVbo.resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, "Memory: Textured Rectangle VBO"); + QVk_DebugSetObjectName((uint64_t)vk_colorRectVbo.resource.buffer, + VK_OBJECT_TYPE_BUFFER, "Static Buffer: Colored Rectangle VBO"); + QVk_DebugSetObjectName((uint64_t)vk_colorRectVbo.resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, "Memory: Colored Rectangle VBO"); + QVk_DebugSetObjectName((uint64_t)vk_rectIbo.resource.buffer, + VK_OBJECT_TYPE_BUFFER, "Static Buffer: Rectangle IBO"); + QVk_DebugSetObjectName((uint64_t)vk_rectIbo.resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, "Memory: Rectangle IBO"); +} + +static void +DestroyShaderModule(qvkshader_t *shaders) +{ + // final shader cleanup + for (int i = 0; i < SHADER_INDEX_SIZE; ++i) + { + if (shaders[i].module) + { + vkDestroyShaderModule(vk_device.logical, shaders[i].module, NULL); + memset(&shaders[i], 0, sizeof(qvkshader_t)); + } + } +} + +// internal helper +static void CreatePipelines() +{ + // shared pipeline vertex input state create infos + VK_VERTINFO(RG, sizeof(float) * 2, VK_INPUTATTR_DESC(0, VK_FORMAT_R32G32_SFLOAT, 0)); + + VK_VERTINFO(RGB, sizeof(float) * 3, VK_INPUTATTR_DESC(0, VK_FORMAT_R32G32B32_SFLOAT, 0)); + + VK_VERTINFO(RG_RG, sizeof(float) * 4, VK_INPUTATTR_DESC(0, VK_FORMAT_R32G32_SFLOAT, 0), + VK_INPUTATTR_DESC(1, VK_FORMAT_R32G32_SFLOAT, sizeof(float) * 2)); + + VK_VERTINFO(RGB_RG, sizeof(float) * 5, VK_INPUTATTR_DESC(0, VK_FORMAT_R32G32B32_SFLOAT, 0), + VK_INPUTATTR_DESC(1, VK_FORMAT_R32G32_SFLOAT, sizeof(float) * 3)); + + VK_VERTINFO(RGB_RGB, sizeof(float) * 6, VK_INPUTATTR_DESC(0, VK_FORMAT_R32G32B32_SFLOAT, 0), + VK_INPUTATTR_DESC(1, VK_FORMAT_R32G32B32_SFLOAT, sizeof(float) * 3)); + + VK_VERTINFO(RGB_RGBA, sizeof(float) * 7, VK_INPUTATTR_DESC(0, VK_FORMAT_R32G32B32_SFLOAT, 0), + VK_INPUTATTR_DESC(1, VK_FORMAT_R32G32B32A32_SFLOAT, sizeof(float) * 3)); + + VK_VERTINFO(RGB_RG_RG, sizeof(float) * 7, VK_INPUTATTR_DESC(0, VK_FORMAT_R32G32B32_SFLOAT, 0), + VK_INPUTATTR_DESC(1, VK_FORMAT_R32G32_SFLOAT, sizeof(float) * 3), + VK_INPUTATTR_DESC(2, VK_FORMAT_R32G32_SFLOAT, sizeof(float) * 5)); + + VK_VERTINFO(RGB_RGBA_RG, sizeof(float) * 9, VK_INPUTATTR_DESC(0, VK_FORMAT_R32G32B32_SFLOAT, 0), + VK_INPUTATTR_DESC(1, VK_FORMAT_R32G32B32A32_SFLOAT, sizeof(float) * 3), + VK_INPUTATTR_DESC(2, VK_FORMAT_R32G32_SFLOAT, sizeof(float) * 7)); + // no vertices passed to the pipeline (postprocessing) + VkPipelineVertexInputStateCreateInfo vertInfoNull = VK_NULL_VERTEXINPUT_CINF; + + // shared descriptor set layouts + VkDescriptorSetLayout samplerUboDsLayouts[] = { vk_samplerDescSetLayout, vk_uboDescSetLayout }; + VkDescriptorSetLayout samplerUboLmapDsLayouts[] = { vk_samplerDescSetLayout, vk_uboDescSetLayout, vk_samplerLightmapDescSetLayout }; + + // shader array (vertex and fragment, no compute... yet) + qvkshader_t shaders[SHADER_INDEX_SIZE] = {0}; + + // textured quad pipeline + VK_LOAD_VERTFRAG_SHADERS(shaders, basic, basic); + for (int i = 0; i < RP_COUNT; ++i) + { + vk_drawTexQuadPipeline[i].depthTestEnable = VK_FALSE; + QVk_CreatePipeline(samplerUboDsLayouts, 2, &vertInfoRG_RG, &vk_drawTexQuadPipeline[i], &vk_renderpasses[i], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawTexQuadPipeline[i].layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, + va("Pipeline Layout: textured quad (%s)", renderpassObjectNames[i])); + QVk_DebugSetObjectName((uint64_t)vk_drawTexQuadPipeline[i].pl, VK_OBJECT_TYPE_PIPELINE, + va("Pipeline: textured quad (%s)", renderpassObjectNames[i])); + } + + // draw particles pipeline (using a texture) + VK_LOAD_VERTFRAG_SHADERS(shaders, particle, basic); + vk_drawParticlesPipeline.depthWriteEnable = VK_TRUE; + vk_drawParticlesPipeline.blendOpts.blendEnable = VK_TRUE; + QVk_CreatePipeline(&vk_samplerDescSetLayout, 1, &vertInfoRGB_RGBA_RG, &vk_drawParticlesPipeline, &vk_renderpasses[RP_WORLD], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawParticlesPipeline.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: textured particles"); + QVk_DebugSetObjectName((uint64_t)vk_drawParticlesPipeline.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: textured particles"); + + // draw particles pipeline (using point list) + VK_LOAD_VERTFRAG_SHADERS(shaders, point_particle, point_particle); + vk_drawPointParticlesPipeline.topology = VK_PRIMITIVE_TOPOLOGY_POINT_LIST; + vk_drawPointParticlesPipeline.depthWriteEnable = VK_TRUE; + vk_drawPointParticlesPipeline.blendOpts.blendEnable = VK_TRUE; + QVk_CreatePipeline(&vk_uboDescSetLayout, 1, &vertInfoRGB_RGBA, &vk_drawPointParticlesPipeline, &vk_renderpasses[RP_WORLD], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawPointParticlesPipeline.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: point particles"); + QVk_DebugSetObjectName((uint64_t)vk_drawPointParticlesPipeline.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: point particles"); + + // colored quad pipeline + VK_LOAD_VERTFRAG_SHADERS(shaders, basic_color_quad, basic_color_quad); + for (int i = 0; i < RP_COUNT; ++i) + { + vk_drawColorQuadPipeline[i].depthTestEnable = VK_FALSE; + vk_drawColorQuadPipeline[i].blendOpts.blendEnable = VK_TRUE; + QVk_CreatePipeline(&vk_uboDescSetLayout, 1, &vertInfoRG, &vk_drawColorQuadPipeline[i], &vk_renderpasses[i], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawColorQuadPipeline[i].layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, + va("Pipeline Layout: colored quad (%s)", renderpassObjectNames[i])); + QVk_DebugSetObjectName((uint64_t)vk_drawColorQuadPipeline[i].pl, VK_OBJECT_TYPE_PIPELINE, + va("Pipeline: colored quad (%s)", renderpassObjectNames[i])); + } + + // untextured null model + VK_LOAD_VERTFRAG_SHADERS(shaders, nullmodel, basic_color_quad); + vk_drawNullModelPipeline.cullMode = VK_CULL_MODE_NONE; + vk_drawNullModelPipeline.topology = VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST; + QVk_CreatePipeline(&vk_uboDescSetLayout, 1, &vertInfoRGB_RGB, &vk_drawNullModelPipeline, &vk_renderpasses[RP_WORLD], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawNullModelPipeline.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: null model"); + QVk_DebugSetObjectName((uint64_t)vk_drawNullModelPipeline.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: null model"); + + // textured model + VK_LOAD_VERTFRAG_SHADERS(shaders, model, model); + for (int i = 0; i < RP_COUNT; ++i) + { + vk_drawModelPipelineFan[i].topology = VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST; + vk_drawModelPipelineFan[i].blendOpts.blendEnable = VK_TRUE; + QVk_CreatePipeline(samplerUboDsLayouts, 2, &vertInfoRGB_RGBA_RG, &vk_drawModelPipelineFan[i], &vk_renderpasses[i], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawModelPipelineFan[i].layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, + va("Pipeline Layout: draw model: fan (%s)", renderpassObjectNames[i])); + QVk_DebugSetObjectName((uint64_t)vk_drawModelPipelineFan[i].pl, VK_OBJECT_TYPE_PIPELINE, + va("Pipeline: draw model: fan (%s)", renderpassObjectNames[i])); + } + + // dedicated model pipelines for translucent objects with depth write disabled + vk_drawNoDepthModelPipelineFan.topology = VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST; + vk_drawNoDepthModelPipelineFan.depthWriteEnable = VK_FALSE; + vk_drawNoDepthModelPipelineFan.blendOpts.blendEnable = VK_TRUE; + QVk_CreatePipeline(samplerUboDsLayouts, 2, &vertInfoRGB_RGBA_RG, &vk_drawNoDepthModelPipelineFan, &vk_renderpasses[RP_WORLD], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawNoDepthModelPipelineFan.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: translucent model: fan"); + QVk_DebugSetObjectName((uint64_t)vk_drawNoDepthModelPipelineFan.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: translucent model: fan"); + + // dedicated model pipelines for when left-handed weapon model is drawn + vk_drawLefthandModelPipelineFan.topology = VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST; + vk_drawLefthandModelPipelineFan.cullMode = VK_CULL_MODE_FRONT_BIT; + QVk_CreatePipeline(samplerUboDsLayouts, 2, &vertInfoRGB_RGBA_RG, &vk_drawLefthandModelPipelineFan, &vk_renderpasses[RP_WORLD], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawLefthandModelPipelineFan.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: left-handed model: fan"); + QVk_DebugSetObjectName((uint64_t)vk_drawLefthandModelPipelineFan.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: left-handed model: fan"); + + // draw sprite pipeline + VK_LOAD_VERTFRAG_SHADERS(shaders, sprite, basic); + vk_drawSpritePipeline.blendOpts.blendEnable = VK_TRUE; + QVk_CreatePipeline(&vk_samplerDescSetLayout, 1, &vertInfoRGB_RG, &vk_drawSpritePipeline, &vk_renderpasses[RP_WORLD], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawSpritePipeline.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: sprite"); + QVk_DebugSetObjectName((uint64_t)vk_drawSpritePipeline.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: sprite"); + + // draw polygon pipeline + VK_LOAD_VERTFRAG_SHADERS(shaders, polygon, basic); + vk_drawPolyPipeline.topology = VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST; + vk_drawPolyPipeline.blendOpts.blendEnable = VK_TRUE; + QVk_CreatePipeline(samplerUboDsLayouts, 2, &vertInfoRGB_RG, &vk_drawPolyPipeline, &vk_renderpasses[RP_WORLD], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawPolyPipeline.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: polygon"); + QVk_DebugSetObjectName((uint64_t)vk_drawPolyPipeline.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: polygon"); + + // draw lightmapped polygon + VK_LOAD_VERTFRAG_SHADERS(shaders, polygon_lmap, polygon_lmap); + vk_drawPolyLmapPipeline.topology = VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST; + QVk_CreatePipeline(samplerUboLmapDsLayouts, 3, &vertInfoRGB_RG_RG, &vk_drawPolyLmapPipeline, &vk_renderpasses[RP_WORLD], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawPolyLmapPipeline.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: lightmapped polygon"); + QVk_DebugSetObjectName((uint64_t)vk_drawPolyLmapPipeline.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: lightmapped polygon"); + + // draw polygon with warp effect (liquid) pipeline + VK_LOAD_VERTFRAG_SHADERS(shaders, polygon_warp, basic); + vk_drawPolyWarpPipeline.topology = VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST; + vk_drawPolyWarpPipeline.blendOpts.blendEnable = VK_TRUE; + QVk_CreatePipeline(samplerUboLmapDsLayouts, 2, &vertInfoRGB_RG, &vk_drawPolyWarpPipeline, &vk_renderpasses[RP_WORLD], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawPolyWarpPipeline.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: warped polygon (liquids)"); + QVk_DebugSetObjectName((uint64_t)vk_drawPolyWarpPipeline.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: warped polygon (liquids)"); + + // draw solid polygon with warp effect (liquid) pipeline + VK_LOAD_VERTFRAG_SHADERS(shaders, polygon_warp, basic); + vk_drawPolySolidWarpPipeline.topology = VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST; + QVk_CreatePipeline(samplerUboLmapDsLayouts, 2, &vertInfoRGB_RG, + &vk_drawPolySolidWarpPipeline, &vk_renderpasses[RP_WORLD], + shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawPolySolidWarpPipeline.layout, + VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: warped solid polygon (liquids)"); + QVk_DebugSetObjectName((uint64_t)vk_drawPolySolidWarpPipeline.pl, + VK_OBJECT_TYPE_PIPELINE, "Pipeline: warped solid polygon (liquids)"); + + // draw beam pipeline + VK_LOAD_VERTFRAG_SHADERS(shaders, beam, basic_color_quad); + vk_drawBeamPipeline.topology = VK_PRIMITIVE_TOPOLOGY_TRIANGLE_STRIP; + vk_drawBeamPipeline.depthWriteEnable = VK_FALSE; + vk_drawBeamPipeline.blendOpts.blendEnable = VK_TRUE; + QVk_CreatePipeline(&vk_uboDescSetLayout, 1, &vertInfoRGB, &vk_drawBeamPipeline, &vk_renderpasses[RP_WORLD], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawBeamPipeline.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: beam"); + QVk_DebugSetObjectName((uint64_t)vk_drawBeamPipeline.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: beam"); + + // draw skybox pipeline + VK_LOAD_VERTFRAG_SHADERS(shaders, skybox, basic); + QVk_CreatePipeline(samplerUboDsLayouts, 2, &vertInfoRGB_RG, &vk_drawSkyboxPipeline, &vk_renderpasses[RP_WORLD], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawSkyboxPipeline.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: skybox"); + QVk_DebugSetObjectName((uint64_t)vk_drawSkyboxPipeline.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: skybox"); + + // draw dynamic light pipeline + VK_LOAD_VERTFRAG_SHADERS(shaders, d_light, basic_color_quad); + vk_drawDLightPipeline.topology = VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST; + vk_drawDLightPipeline.depthWriteEnable = VK_FALSE; + vk_drawDLightPipeline.cullMode = VK_CULL_MODE_FRONT_BIT; + vk_drawDLightPipeline.blendOpts.blendEnable = VK_TRUE; + vk_drawDLightPipeline.blendOpts.srcColorBlendFactor = VK_BLEND_FACTOR_ONE; + vk_drawDLightPipeline.blendOpts.dstColorBlendFactor = VK_BLEND_FACTOR_ONE; + vk_drawDLightPipeline.blendOpts.srcAlphaBlendFactor = VK_BLEND_FACTOR_ONE; + vk_drawDLightPipeline.blendOpts.dstAlphaBlendFactor = VK_BLEND_FACTOR_ONE; + QVk_CreatePipeline(&vk_uboDescSetLayout, 1, &vertInfoRGB_RGB, &vk_drawDLightPipeline, &vk_renderpasses[RP_WORLD], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_drawDLightPipeline.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: dynamic light"); + QVk_DebugSetObjectName((uint64_t)vk_drawDLightPipeline.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: dynamic light"); + + // vk_showtris render pipeline + VK_LOAD_VERTFRAG_SHADERS(shaders, d_light, basic_color_quad); + vk_showTrisPipeline.cullMode = VK_CULL_MODE_NONE; + vk_showTrisPipeline.depthTestEnable = VK_FALSE; + vk_showTrisPipeline.depthWriteEnable = VK_FALSE; + vk_showTrisPipeline.topology = VK_PRIMITIVE_TOPOLOGY_LINE_STRIP; + QVk_CreatePipeline(&vk_uboDescSetLayout, 1, &vertInfoRGB_RGB, &vk_showTrisPipeline, &vk_renderpasses[RP_WORLD], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_showTrisPipeline.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: show triangles"); + QVk_DebugSetObjectName((uint64_t)vk_showTrisPipeline.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: show triangles"); + + //vk_shadows render pipeline + VK_LOAD_VERTFRAG_SHADERS(shaders, shadows, basic_color_quad); + vk_shadowsPipelineFan.blendOpts.blendEnable = VK_TRUE; + vk_shadowsPipelineFan.topology = VK_PRIMITIVE_TOPOLOGY_TRIANGLE_LIST; + QVk_CreatePipeline(&vk_uboDescSetLayout, 1, &vertInfoRGB, &vk_shadowsPipelineFan, &vk_renderpasses[RP_WORLD], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_shadowsPipelineFan.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: draw shadows: fan"); + QVk_DebugSetObjectName((uint64_t)vk_shadowsPipelineFan.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: draw shadows: fan"); + + // underwater world warp pipeline (postprocess) + VK_LOAD_VERTFRAG_SHADERS(shaders, world_warp, world_warp); + vk_worldWarpPipeline.depthTestEnable = VK_FALSE; + vk_worldWarpPipeline.depthWriteEnable = VK_FALSE; + vk_worldWarpPipeline.cullMode = VK_CULL_MODE_NONE; + QVk_CreatePipeline(&vk_samplerDescSetLayout, 1, &vertInfoNull, &vk_worldWarpPipeline, &vk_renderpasses[RP_WORLD_WARP], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_worldWarpPipeline.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: underwater view warp"); + QVk_DebugSetObjectName((uint64_t)vk_worldWarpPipeline.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: underwater view warp"); + + // postprocessing pipeline + VK_LOAD_VERTFRAG_SHADERS(shaders, postprocess, postprocess); + vk_postprocessPipeline.depthTestEnable = VK_FALSE; + vk_postprocessPipeline.depthWriteEnable = VK_FALSE; + vk_postprocessPipeline.cullMode = VK_CULL_MODE_NONE; + QVk_CreatePipeline(&vk_samplerDescSetLayout, 1, &vertInfoNull, &vk_postprocessPipeline, &vk_renderpasses[RP_UI], shaders, 2); + QVk_DebugSetObjectName((uint64_t)vk_postprocessPipeline.layout, VK_OBJECT_TYPE_PIPELINE_LAYOUT, "Pipeline Layout: world postprocess"); + QVk_DebugSetObjectName((uint64_t)vk_postprocessPipeline.pl, VK_OBJECT_TYPE_PIPELINE, "Pipeline: world postprocess"); + + DestroyShaderModule(shaders); +} + +static void DestroyStagingBuffer(qvkstagingbuffer_t *dstBuffer) +{ + if (dstBuffer->resource.buffer != VK_NULL_HANDLE) + { + // wait only if something is submitted + if (dstBuffer->submitted) + { + VK_VERIFY(vkWaitForFences(vk_device.logical, 1, &dstBuffer->fence, + VK_TRUE, UINT64_MAX)); + } + + buffer_unmap(&dstBuffer->resource); + QVk_FreeStagingBuffer(dstBuffer); + vkDestroyFence(vk_device.logical, dstBuffer->fence, NULL); + } +} + +/* +** QVk_Shutdown +** +** Destroy all Vulkan related resources. +*/ +void QVk_Shutdown( void ) +{ + if (!vk_initialized) + { + return; + } + + if (vk_instance != VK_NULL_HANDLE) + { + R_Printf(PRINT_ALL, "Shutting down Vulkan\n"); + + for (int i = 0; i < RP_COUNT; ++i) + { + QVk_DestroyPipeline(&vk_drawColorQuadPipeline[i]); + QVk_DestroyPipeline(&vk_drawModelPipelineFan[i]); + QVk_DestroyPipeline(&vk_drawTexQuadPipeline[i]); + } + QVk_DestroyPipeline(&vk_drawNullModelPipeline); + QVk_DestroyPipeline(&vk_drawNoDepthModelPipelineFan); + QVk_DestroyPipeline(&vk_drawLefthandModelPipelineFan); + QVk_DestroyPipeline(&vk_drawParticlesPipeline); + QVk_DestroyPipeline(&vk_drawPointParticlesPipeline); + QVk_DestroyPipeline(&vk_drawSpritePipeline); + QVk_DestroyPipeline(&vk_drawPolyPipeline); + QVk_DestroyPipeline(&vk_drawPolyLmapPipeline); + QVk_DestroyPipeline(&vk_drawPolyWarpPipeline); + QVk_DestroyPipeline(&vk_drawPolySolidWarpPipeline); + QVk_DestroyPipeline(&vk_drawBeamPipeline); + QVk_DestroyPipeline(&vk_drawSkyboxPipeline); + QVk_DestroyPipeline(&vk_drawDLightPipeline); + QVk_DestroyPipeline(&vk_showTrisPipeline); + QVk_DestroyPipeline(&vk_shadowsPipelineFan); + QVk_DestroyPipeline(&vk_worldWarpPipeline); + QVk_DestroyPipeline(&vk_postprocessPipeline); + QVk_FreeBuffer(&vk_texRectVbo); + QVk_FreeBuffer(&vk_colorRectVbo); + QVk_FreeBuffer(&vk_rectIbo); + for (int i = 0; i < NUM_DYNBUFFERS; ++i) + { + if (vk_dynUniformBuffers[i].resource.buffer != VK_NULL_HANDLE) + { + buffer_unmap(&vk_dynUniformBuffers[i].resource); + QVk_FreeBuffer(&vk_dynUniformBuffers[i]); + } + if (vk_dynIndexBuffers[i].resource.buffer != VK_NULL_HANDLE) + { + buffer_unmap(&vk_dynIndexBuffers[i].resource); + QVk_FreeBuffer(&vk_dynIndexBuffers[i]); + } + if (vk_dynVertexBuffers[i].resource.buffer != VK_NULL_HANDLE) + { + buffer_unmap(&vk_dynVertexBuffers[i].resource); + QVk_FreeBuffer(&vk_dynVertexBuffers[i]); + } + DestroyStagingBuffer(&vk_stagingBuffers[i]); + if (vk_stagingCommandPool[i] != VK_NULL_HANDLE) + { + vkDestroyCommandPool(vk_device.logical, vk_stagingCommandPool[i], NULL); + vk_stagingCommandPool[i] = VK_NULL_HANDLE; + } + } + if (vk_descriptorPool != VK_NULL_HANDLE) + vkDestroyDescriptorPool(vk_device.logical, vk_descriptorPool, NULL); + if (vk_uboDescSetLayout != VK_NULL_HANDLE) + vkDestroyDescriptorSetLayout(vk_device.logical, vk_uboDescSetLayout, NULL); + if (vk_samplerDescSetLayout != VK_NULL_HANDLE) + vkDestroyDescriptorSetLayout(vk_device.logical, vk_samplerDescSetLayout, NULL); + if (vk_samplerLightmapDescSetLayout != VK_NULL_HANDLE) + vkDestroyDescriptorSetLayout(vk_device.logical, vk_samplerLightmapDescSetLayout, NULL); + for (int i = 0; i < RP_COUNT; i++) + { + if (vk_renderpasses[i].rp != VK_NULL_HANDLE) + vkDestroyRenderPass(vk_device.logical, vk_renderpasses[i].rp, NULL); + vk_renderpasses[i].rp = VK_NULL_HANDLE; + } + + for (int i = 0; i < NUM_CMDBUFFERS; i++) + { + if (vk_commandPool[i] != VK_NULL_HANDLE) + { + vkFreeCommandBuffers(vk_device.logical, vk_commandPool[i], 1, &vk_commandbuffers[i]); + vkDestroyCommandPool(vk_device.logical, vk_commandPool[i], NULL); + vk_commandPool[i] = VK_NULL_HANDLE; + } + } + if (vk_commandbuffers != NULL) + { + free(vk_commandbuffers); + vk_commandbuffers = NULL; + } + if (vk_transferCommandPool != VK_NULL_HANDLE) + vkDestroyCommandPool(vk_device.logical, vk_transferCommandPool, NULL); + DestroySamplers(); + DestroyFramebuffers(); + DestroyImageViews(); + DestroyDrawBuffers(); + if (vk_swapchain.sc != VK_NULL_HANDLE) + { + vkDestroySwapchainKHR(vk_device.logical, vk_swapchain.sc, NULL); + free(vk_swapchain.images); + vk_swapchain.sc = VK_NULL_HANDLE; + vk_swapchain.images = NULL; + vk_swapchain.imageCount = 0; + } + if (vk_device.logical != VK_NULL_HANDLE) + { + for (int i = 0; i < NUM_CMDBUFFERS; ++i) + { + vkDestroySemaphore(vk_device.logical, vk_imageAvailableSemaphores[i], NULL); + vkDestroySemaphore(vk_device.logical, vk_renderFinishedSemaphores[i], NULL); + vkDestroyFence(vk_device.logical, vk_fences[i], NULL); + } + } + // free all memory + vulkan_memory_delete(); + + if (vk_device.logical != VK_NULL_HANDLE) + vkDestroyDevice(vk_device.logical, NULL); + if(vk_surface != VK_NULL_HANDLE) + vkDestroySurfaceKHR(vk_instance, vk_surface, NULL); + QVk_DestroyValidationLayers(); + + vkDestroyInstance(vk_instance, NULL); + vk_instance = VK_NULL_HANDLE; + vk_activeCmdbuffer = VK_NULL_HANDLE; + vk_descriptorPool = VK_NULL_HANDLE; + vk_uboDescSetLayout = VK_NULL_HANDLE; + vk_samplerDescSetLayout = VK_NULL_HANDLE; + vk_samplerLightmapDescSetLayout = VK_NULL_HANDLE; + vk_transferCommandPool = VK_NULL_HANDLE; + vk_activeBufferIdx = 0; + vk_imageIndex = 0; + } +} + +void QVk_SetWindow(SDL_Window *window) +{ + vk_window = window; +} + +void QVk_WaitAndShutdownAll (void) +{ + if (!vk_initialized) + { + return; + } + + if (vk_device.logical != VK_NULL_HANDLE) + { + vkDeviceWaitIdle(vk_device.logical); + } + + Mod_FreeAll(); + Mod_FreeModelsKnown(); + Vk_ShutdownImages(); + Mesh_Free(); + QVk_Shutdown(); + + vk_frameStarted = false; + vk_initialized = false; +} + +void QVk_Restart(void) +{ + QVk_WaitAndShutdownAll(); + if (!QVk_Init()) + ri.Sys_Error(ERR_FATAL, "Unable to restart Vulkan renderer"); + QVk_PostInit(); + ri.Vid_RequestRestart(RESTART_PARTIAL); +} + +void QVk_PostInit(void) +{ + Mesh_Init(); + Vk_InitImages(); + Mod_Init(); + RE_InitParticleTexture(); + Draw_InitLocal(); +} + +/* +** QVk_Init +** +** This is responsible for initializing Vulkan. +** +*/ +qboolean QVk_Init(void) +{ + PFN_vkEnumerateInstanceVersion vkEnumerateInstanceVersion = (PFN_vkEnumerateInstanceVersion)vkGetInstanceProcAddr(NULL, "vkEnumerateInstanceVersion"); + uint32_t instanceVersion = VK_API_VERSION_1_0; + + if (vkEnumerateInstanceVersion) + { + VK_VERIFY(vkEnumerateInstanceVersion(&instanceVersion)); + } + + VkApplicationInfo appInfo = { + .sType = VK_STRUCTURE_TYPE_APPLICATION_INFO, + .pNext = NULL, + .pApplicationName = "Quake 2", + .applicationVersion = VK_MAKE_VERSION(3, 21, 0), + .pEngineName = "id Tech 2", + .engineVersion = VK_MAKE_VERSION(2, 0, 0), + .apiVersion = instanceVersion + }; + + uint32_t extCount; + const char **wantedExtensions; + memset(vk_config.supported_present_modes, 0, sizeof(vk_config.supported_present_modes)); + memset(vk_config.extensions, 0, sizeof(vk_config.extensions)); + memset(vk_config.layers, 0, sizeof(vk_config.layers)); + vk_config.vk_version = instanceVersion; + vk_config.vertex_buffer_usage = 0; + vk_config.vertex_buffer_max_usage = 0; + vk_config.vertex_buffer_size = VERTEX_BUFFER_SIZE; + vk_config.index_buffer_usage = 0; + vk_config.index_buffer_max_usage = 0; + vk_config.index_buffer_size = INDEX_BUFFER_SIZE; + vk_config.uniform_buffer_usage = 0; + vk_config.uniform_buffer_max_usage = 0; + vk_config.uniform_buffer_size = UNIFORM_BUFFER_SIZE; + vk_config.triangle_index_usage = 0; + vk_config.triangle_index_max_usage = 0; + vk_config.triangle_index_count = TRIANGLE_INDEX_CNT; + + if (!SDL_Vulkan_GetInstanceExtensions(vk_window, &extCount, NULL)) + { + R_Printf(PRINT_ALL, "%s() SDL_Vulkan_GetInstanceExtensions failed: %s", + __func__, SDL_GetError()); + return false; + } + + // add space for validation layer + if (vk_validation->value) + extCount++; + + wantedExtensions = malloc(extCount * sizeof(const char *)); + if (!SDL_Vulkan_GetInstanceExtensions(vk_window, &extCount, wantedExtensions)) + { + R_Printf(PRINT_ALL, "%s() SDL_Vulkan_GetInstanceExtensions failed: %s", + __func__, SDL_GetError()); + return false; + } + + // restore extensions count + if (vk_validation->value) + { + extCount++; + wantedExtensions[extCount - 1] = VK_EXT_DEBUG_UTILS_EXTENSION_NAME; + } + + R_Printf(PRINT_ALL, "Enabled extensions: "); + for (int i = 0; i < extCount; i++) + { + R_Printf(PRINT_ALL, "%s ", wantedExtensions[i]); + vk_config.extensions[i] = wantedExtensions[i]; + } + R_Printf(PRINT_ALL, "\n"); + + VkInstanceCreateInfo createInfo = { + .sType = VK_STRUCTURE_TYPE_INSTANCE_CREATE_INFO, + .pNext = NULL, + .pApplicationInfo = &appInfo, + .enabledLayerCount = 0, + .ppEnabledLayerNames = NULL, + .enabledExtensionCount = extCount, + .ppEnabledExtensionNames = (const char* const*)wantedExtensions + }; + +// introduced in SDK 1.1.121 +#if VK_HEADER_VERSION > 114 + VkValidationFeatureEnableEXT validationFeaturesEnable[] = { VK_VALIDATION_FEATURE_ENABLE_BEST_PRACTICES_EXT }; + VkValidationFeaturesEXT validationFeatures = { + .sType = VK_STRUCTURE_TYPE_VALIDATION_FEATURES_EXT, + .pNext = NULL, + .enabledValidationFeatureCount = sizeof(validationFeaturesEnable) / sizeof(validationFeaturesEnable[0]), + .pEnabledValidationFeatures = validationFeaturesEnable, + .disabledValidationFeatureCount = 0, + .pDisabledValidationFeatures = NULL + }; + + if (vk_validation->value > 1) + { + createInfo.pNext = &validationFeatures; + } +#endif + +// introduced in SDK 1.1.106 +#if VK_HEADER_VERSION > 101 + const char *validationLayers[] = { "VK_LAYER_KHRONOS_validation" }; +#else + const char *validationLayers[] = { "VK_LAYER_LUNARG_standard_validation" }; +#endif + + if (vk_validation->value) + { + createInfo.enabledLayerCount = sizeof(validationLayers) / sizeof(validationLayers[0]); + createInfo.ppEnabledLayerNames = validationLayers; + for (int i = 0; i < createInfo.enabledLayerCount; i++) + { + vk_config.layers[i] = validationLayers[i]; + } + } + + VkResult res = vkCreateInstance(&createInfo, NULL, &vk_instance); + free(wantedExtensions); + + volkLoadInstance(vk_instance); + + if (res != VK_SUCCESS) + { + R_Printf(PRINT_ALL, "%s(): Could not create Vulkan instance: %s\n", __func__, QVk_GetError(res)); + return false; + } + R_Printf(PRINT_ALL, "...created Vulkan instance\n"); + + if (vk_validation->value) + { + // initialize function pointers + qvkCreateDebugUtilsMessengerEXT = (PFN_vkCreateDebugUtilsMessengerEXT)vkGetInstanceProcAddr(vk_instance, "vkCreateDebugUtilsMessengerEXT"); + qvkDestroyDebugUtilsMessengerEXT = (PFN_vkDestroyDebugUtilsMessengerEXT)vkGetInstanceProcAddr(vk_instance, "vkDestroyDebugUtilsMessengerEXT"); + qvkSetDebugUtilsObjectNameEXT = (PFN_vkSetDebugUtilsObjectNameEXT)vkGetInstanceProcAddr(vk_instance, "vkSetDebugUtilsObjectNameEXT"); + qvkSetDebugUtilsObjectTagEXT = (PFN_vkSetDebugUtilsObjectTagEXT)vkGetInstanceProcAddr(vk_instance, "vkSetDebugUtilsObjectTagEXT"); + qvkCmdBeginDebugUtilsLabelEXT = (PFN_vkCmdBeginDebugUtilsLabelEXT)vkGetInstanceProcAddr(vk_instance, "vkCmdBeginDebugUtilsLabelEXT"); + qvkCmdEndDebugUtilsLabelEXT = (PFN_vkCmdEndDebugUtilsLabelEXT)vkGetInstanceProcAddr(vk_instance, "vkCmdEndDebugUtilsLabelEXT"); + qvkInsertDebugUtilsLabelEXT = (PFN_vkCmdInsertDebugUtilsLabelEXT)vkGetInstanceProcAddr(vk_instance, "vkCmdInsertDebugUtilsLabelEXT"); + } + else + { + qvkCreateDebugUtilsMessengerEXT = NULL; + qvkDestroyDebugUtilsMessengerEXT = NULL; + qvkSetDebugUtilsObjectNameEXT = NULL; + qvkSetDebugUtilsObjectTagEXT = NULL; + qvkCmdBeginDebugUtilsLabelEXT = NULL; + qvkCmdEndDebugUtilsLabelEXT = NULL; + qvkInsertDebugUtilsLabelEXT = NULL; + } + + if (vk_validation->value) + QVk_CreateValidationLayers(); + + if (!Vkimp_CreateSurface(vk_window)) + { + return false; + } + R_Printf(PRINT_ALL, "...created Vulkan surface\n"); + + // create Vulkan device - see if the user prefers any specific device if there's more than one GPU in the system + if (!QVk_CreateDevice((int)vk_device_idx->value)) + { + return false; + } + QVk_DebugSetObjectName((uintptr_t)vk_device.physical, VK_OBJECT_TYPE_PHYSICAL_DEVICE, va("Physical Device: %s", vk_config.vendor_name)); + + if (vk_validation->value) + vulkan_memory_types_show(); + + // setup swapchain + res = QVk_CreateSwapchain(); + if (res != VK_SUCCESS) + { + R_Printf(PRINT_ALL, "%s(): Could not create Vulkan swapchain: %s\n", __func__, QVk_GetError(res)); + return false; + } + R_Printf(PRINT_ALL, "...created Vulkan swapchain\n"); + + // set viewport and scissor + if (vid_fullscreen->value == 2) + { + // Center viewport in "keep resolution mode". + vk_viewport.x = max(0.f, (float)(vk_swapchain.extent.width - (uint32_t)(vid.width)) / 2.0f); + vk_viewport.y = max(0.f, (float)(vk_swapchain.extent.height - (uint32_t)(vid.height)) / 2.0f); + } + else + { + vk_viewport.x = 0.f; + vk_viewport.y = 0.f; + } + vk_viewport.minDepth = 0.f; + vk_viewport.maxDepth = 1.f; + vk_viewport.width = min((float)vid.width, (float)(vk_swapchain.extent.width) - vk_viewport.x); + vk_viewport.height = min((float)vid.height, (float)(vk_swapchain.extent.height) - vk_viewport.y); + vk_scissor.offset.x = 0; + vk_scissor.offset.y = 0; + vk_scissor.extent = vk_swapchain.extent; + + // setup fences and semaphores + VkFenceCreateInfo fCreateInfo = { + .sType = VK_STRUCTURE_TYPE_FENCE_CREATE_INFO, + .pNext = NULL, + .flags = VK_FENCE_CREATE_SIGNALED_BIT + }; + VkSemaphoreCreateInfo sCreateInfo = { + .sType = VK_STRUCTURE_TYPE_SEMAPHORE_CREATE_INFO, + .pNext = NULL, + .flags = 0 + }; + for (int i = 0; i < NUM_CMDBUFFERS; ++i) + { + VK_VERIFY(vkCreateFence(vk_device.logical, &fCreateInfo, NULL, &vk_fences[i])); + VK_VERIFY(vkCreateSemaphore(vk_device.logical, &sCreateInfo, NULL, &vk_imageAvailableSemaphores[i])); + VK_VERIFY(vkCreateSemaphore(vk_device.logical, &sCreateInfo, NULL, &vk_renderFinishedSemaphores[i])); + + QVk_DebugSetObjectName((uint64_t)vk_fences[i], + VK_OBJECT_TYPE_FENCE, va("Fence #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_imageAvailableSemaphores[i], + VK_OBJECT_TYPE_SEMAPHORE, va("Semaphore: image available #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_renderFinishedSemaphores[i], + VK_OBJECT_TYPE_SEMAPHORE, va("Semaphore: render finished #%d", i)); + } + R_Printf(PRINT_ALL, "...created synchronization objects\n"); + + // setup render passes + for (int i = 0; i < RP_COUNT; ++i) + { + vk_renderpasses[i].colorLoadOp = r_clear->value ? VK_ATTACHMENT_LOAD_OP_CLEAR : VK_ATTACHMENT_LOAD_OP_DONT_CARE; + } + + VkSampleCountFlagBits msaaMode = GetSampleCount((int)vk_msaa->value, + vk_device.properties.limits.framebufferColorSampleCounts); + + // MSAA setting will be only relevant for the primary world render pass + vk_renderpasses[RP_WORLD].sampleCount = msaaMode; + + res = CreateRenderpasses(); + if (res != VK_SUCCESS) + { + R_Printf(PRINT_ALL, "%s(): Could not create Vulkan render passes: %s\n", __func__, QVk_GetError(res)); + return false; + } + R_Printf(PRINT_ALL, "...created %d Vulkan render passes\n", RP_COUNT); + + // setup command pools + for (int i = 0; i < NUM_CMDBUFFERS; i++) + { + res = QVk_CreateCommandPool(&vk_commandPool[i], vk_device.gfxFamilyIndex); + if (res != VK_SUCCESS) + { + R_Printf(PRINT_ALL, "%s(): Could not create Vulkan command pool #%d for graphics: %s\n", __func__, i, QVk_GetError(res)); + return false; + } + QVk_DebugSetObjectName((uint64_t)vk_commandPool[i], + VK_OBJECT_TYPE_COMMAND_POOL, va("Command Pool #%d: Graphics", i)); + } + + res = QVk_CreateCommandPool(&vk_transferCommandPool, vk_device.transferFamilyIndex); + if (res != VK_SUCCESS) + { + R_Printf(PRINT_ALL, "%s(): Could not create Vulkan command pool for transfer: %s\n", __func__, QVk_GetError(res)); + return false; + } + + QVk_DebugSetObjectName((uint64_t)vk_transferCommandPool, + VK_OBJECT_TYPE_COMMAND_POOL, "Command Pool: Transfer"); + R_Printf(PRINT_ALL, "...created Vulkan command pools\n"); + + // setup draw buffers + CreateDrawBuffers(); + + // setup image views + res = CreateImageViews(); + if (res != VK_SUCCESS) + { + R_Printf(PRINT_ALL, "%s(): Could not create Vulkan image views: %s\n", __func__, QVk_GetError(res)); + return false; + } + R_Printf(PRINT_ALL, "...created %d Vulkan image view(s)\n", vk_swapchain.imageCount); + + // setup framebuffers + res = CreateFramebuffers(); + if (res != VK_SUCCESS) + { + R_Printf(PRINT_ALL, "%s(): Could not create Vulkan framebuffers: %s\n", __func__, QVk_GetError(res)); + return false; + } + R_Printf(PRINT_ALL, "...created %d Vulkan framebuffers\n", vk_swapchain.imageCount); + + // setup command buffers (double buffering) + vk_commandbuffers = (VkCommandBuffer *)malloc(NUM_CMDBUFFERS * sizeof(VkCommandBuffer)); + + for (int i = 0; i < NUM_CMDBUFFERS; i++) + { + VkCommandBufferAllocateInfo cbInfo = { + .sType = VK_STRUCTURE_TYPE_COMMAND_BUFFER_ALLOCATE_INFO, + .pNext = NULL, + .commandPool = vk_commandPool[i], + .level = VK_COMMAND_BUFFER_LEVEL_PRIMARY, + .commandBufferCount = 1 + }; + + res = vkAllocateCommandBuffers(vk_device.logical, &cbInfo, &vk_commandbuffers[i]); + if (res != VK_SUCCESS) + { + R_Printf(PRINT_ALL, "%s(): Could not create Vulkan commandbuffers: %s\n", __func__, QVk_GetError(res)); + free(vk_commandbuffers); + vk_commandbuffers = NULL; + return false; + } + } + R_Printf(PRINT_ALL, "...created %d Vulkan commandbuffers\n", NUM_CMDBUFFERS); + + // initialize tracker variables + vk_activeCmdbuffer = vk_commandbuffers[vk_activeBufferIdx]; + + CreateDescriptorSetLayouts(); + CreateDescriptorPool(); + // create static vertex/index buffers reused in the games + CreateStaticBuffers(); + // create vertex, index and uniform buffer pools + CreateDynamicBuffers(); + // create staging buffers + CreateStagingBuffers(); + // assign a dynamic index buffer for triangle fan emulation + RebuildTriangleIndexBuffer(); + CreatePipelines(); + CreateSamplers(); + + // main and world warp color buffers will be sampled for postprocessing effects, so they need descriptors and samplers + VkDescriptorSetAllocateInfo dsAllocInfo = { + .sType = VK_STRUCTURE_TYPE_DESCRIPTOR_SET_ALLOCATE_INFO, + .pNext = NULL, + .descriptorPool = vk_descriptorPool, + .descriptorSetCount = 1, + .pSetLayouts = &vk_samplerDescSetLayout + }; + + VK_VERIFY(vkAllocateDescriptorSets(vk_device.logical, &dsAllocInfo, &vk_colorbuffer.descriptorSet)); + QVk_UpdateTextureSampler(&vk_colorbuffer, S_NEAREST_UNNORMALIZED, false); + VK_VERIFY(vkAllocateDescriptorSets(vk_device.logical, &dsAllocInfo, &vk_colorbufferWarp.descriptorSet)); + QVk_UpdateTextureSampler(&vk_colorbufferWarp, S_NEAREST_UNNORMALIZED, false); + + QVk_DebugSetObjectName((uint64_t)vk_colorbuffer.descriptorSet, + VK_OBJECT_TYPE_DESCRIPTOR_SET, "Descriptor Set: World Color Buffer"); + QVk_DebugSetObjectName((uint64_t)vk_colorbufferWarp.descriptorSet, + VK_OBJECT_TYPE_DESCRIPTOR_SET, "Descriptor Set: Warp Postprocess Color Buffer"); + + vk_frameStarted = false; + vk_initialized = true; + return true; +} + +VkResult QVk_BeginFrame(const VkViewport* viewport, const VkRect2D* scissor) +{ + // reset tracking variables + vk_state.current_pipeline = VK_NULL_HANDLE; + vk_state.current_renderpass = RP_COUNT; + vk_config.vertex_buffer_usage = 0; + // triangle fan index buffer data will not be cleared between frames unless the buffer itself is too small + vk_config.index_buffer_usage = vk_triangleFanIboUsage; + vk_config.uniform_buffer_usage = 0; + vk_config.triangle_index_usage = 0; + + ReleaseSwapBuffers(); + + static int restartcount; + VkResult result = vkAcquireNextImageKHR(vk_device.logical, vk_swapchain.sc, 500000000, vk_imageAvailableSemaphores[vk_activeBufferIdx], VK_NULL_HANDLE, &vk_imageIndex); + if (result == VK_ERROR_OUT_OF_DATE_KHR || result == VK_SUBOPTIMAL_KHR || result == VK_ERROR_SURFACE_LOST_KHR || result == VK_TIMEOUT) + { + if (restartcount > 2) + { + Sys_Error("%s(): tried to restart 3 times after vkAcquireNextImageKHR: %s", __func__, QVk_GetError(result)); + } + else + { + restartcount++; + } + + // for VK_OUT_OF_DATE_KHR and VK_SUBOPTIMAL_KHR it'd be fine to just rebuild the swapchain but let's take the easy way out and restart Vulkan. + R_Printf(PRINT_ALL, "%s(): received %s after vkAcquireNextImageKHR - restarting video!\n", __func__, QVk_GetError(result)); + return result; + } + else if (result != VK_SUCCESS) + { + Sys_Error("%s(): unexpected error after vkAcquireNextImageKHR: %s", __func__, QVk_GetError(result)); + } + + restartcount = 0; + vk_activeCmdbuffer = vk_commandbuffers[vk_activeBufferIdx]; + + // swap dynamic buffers + vk_activeDynBufferIdx = (vk_activeDynBufferIdx + 1) % NUM_DYNBUFFERS; + vk_dynUniformBuffers[vk_activeDynBufferIdx].currentOffset = 0; + vk_dynVertexBuffers[vk_activeDynBufferIdx].currentOffset = 0; + // triangle fan index data is placed in the beginning of the buffer + vk_dynIndexBuffers[vk_activeDynBufferIdx].currentOffset = vk_triangleFanIboUsage; + VK_VERIFY(buffer_invalidate(&vk_dynUniformBuffers[vk_activeDynBufferIdx].resource)); + VK_VERIFY(buffer_invalidate(&vk_dynVertexBuffers[vk_activeDynBufferIdx].resource)); + VK_VERIFY(buffer_invalidate(&vk_dynIndexBuffers[vk_activeDynBufferIdx].resource)); + + VK_VERIFY(vkWaitForFences(vk_device.logical, 1, &vk_fences[vk_activeBufferIdx], VK_TRUE, UINT32_MAX)); + VK_VERIFY(vkResetFences(vk_device.logical, 1, &vk_fences[vk_activeBufferIdx])); + + // setup command buffers and render pass for drawing + VkCommandBufferBeginInfo beginInfo = { + .sType = VK_STRUCTURE_TYPE_COMMAND_BUFFER_BEGIN_INFO, + .pNext = NULL, + .flags = VK_COMMAND_BUFFER_USAGE_ONE_TIME_SUBMIT_BIT, + .pInheritanceInfo = NULL + }; + + // Command buffers are implicitly reset by vkBeginCommandBuffer if VK_COMMAND_POOL_CREATE_RESET_COMMAND_BUFFER_BIT is set - this is expensive. + // It's more efficient to reset entire pool instead and it also fixes the VK_VALIDATION_FEATURE_ENABLE_BEST_PRACTICES_EXT performance warning. + // see also: https://github.com/KhronosGroup/Vulkan-Samples/blob/master/samples/performance/command_buffer_usage/command_buffer_usage_tutorial.md + vkResetCommandPool(vk_device.logical, vk_commandPool[vk_activeBufferIdx], 0); + VK_VERIFY(vkBeginCommandBuffer(vk_commandbuffers[vk_activeBufferIdx], &beginInfo)); + + vkCmdSetViewport(vk_commandbuffers[vk_activeBufferIdx], 0, 1, viewport); + vkCmdSetScissor(vk_commandbuffers[vk_activeBufferIdx], 0, 1, scissor); + + vk_frameStarted = true; + return VK_SUCCESS; +} + +VkResult QVk_EndFrame(qboolean force) +{ + // continue only if QVk_BeginFrame() had been previously issued + if (!vk_frameStarted) + return VK_SUCCESS; + + // this may happen if Sys_Error is issued mid-frame, so we need to properly advance the draw pipeline + if (force) + { + if(!RE_EndWorldRenderpass()) + // buffers is not initialized + return VK_NOT_READY; + } + + // submit + QVk_SubmitStagingBuffers(); + VK_VERIFY(buffer_flush(&vk_dynUniformBuffers[vk_activeDynBufferIdx].resource)); + VK_VERIFY(buffer_flush(&vk_dynVertexBuffers[vk_activeDynBufferIdx].resource)); + VK_VERIFY(buffer_flush(&vk_dynIndexBuffers[vk_activeDynBufferIdx].resource)); + + vkCmdEndRenderPass(vk_commandbuffers[vk_activeBufferIdx]); + QVk_DebugLabelEnd(&vk_commandbuffers[vk_activeBufferIdx]); + VK_VERIFY(vkEndCommandBuffer(vk_commandbuffers[vk_activeBufferIdx])); + + VkPipelineStageFlags waitStages = VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT; + VkSubmitInfo submitInfo = { + .sType = VK_STRUCTURE_TYPE_SUBMIT_INFO, + .waitSemaphoreCount = 1, + .pWaitSemaphores = &vk_imageAvailableSemaphores[vk_activeBufferIdx], + .signalSemaphoreCount = 1, + .pSignalSemaphores = &vk_renderFinishedSemaphores[vk_activeBufferIdx], + .pWaitDstStageMask = &waitStages, + .commandBufferCount = 1, + .pCommandBuffers = &vk_commandbuffers[vk_activeBufferIdx] + }; + + VK_VERIFY(vkQueueSubmit(vk_device.gfxQueue, 1, &submitInfo, vk_fences[vk_activeBufferIdx])); + + // present + VkPresentInfoKHR presentInfo = { + .sType = VK_STRUCTURE_TYPE_PRESENT_INFO_KHR, + .waitSemaphoreCount = 1, + .pWaitSemaphores = &vk_renderFinishedSemaphores[vk_activeBufferIdx], + .swapchainCount = 1, + .pSwapchains = &vk_swapchain.sc, + .pImageIndices = &vk_imageIndex, + .pResults = NULL + }; + + VkResult renderResult = vkQueuePresentKHR(vk_device.presentQueue, &presentInfo); + + // for VK_OUT_OF_DATE_KHR and VK_SUBOPTIMAL_KHR it'd be fine to just rebuild the swapchain but let's take the easy way out and restart video system + if (renderResult == VK_ERROR_OUT_OF_DATE_KHR || renderResult == VK_SUBOPTIMAL_KHR || renderResult == VK_ERROR_SURFACE_LOST_KHR) + { + R_Printf(PRINT_ALL, "%s(): received %s after vkQueuePresentKHR - will restart video!\n", __func__, QVk_GetError(renderResult)); + vk_restartNeeded = true; + } + else if (renderResult != VK_SUCCESS) + { + Sys_Error("%s(): unexpected error after vkQueuePresentKHR: %s", __func__, QVk_GetError(renderResult)); + } + + vk_activeBufferIdx = (vk_activeBufferIdx + 1) % NUM_CMDBUFFERS; + + vk_frameStarted = false; + return renderResult; +} + +void QVk_BeginRenderpass(qvkrenderpasstype_t rpType) +{ + const VkClearValue clearColors[3] = { + {.color = {.float32 = { 1.f, .0f, .5f, 1.f } } }, + {.depthStencil = { 1.f, 0 } }, + {.color = {.float32 = { 1.f, .0f, .5f, 1.f } } }, + }; + + const VkClearValue warpClearColors[2] = { + clearColors[0], + {.color = {.float32 = { 0.f, .0f, .0f, 1.f } } }, + }; + + const VkClearValue uiClearColors[3] = { + clearColors[0], + clearColors[1], + {.color = {.float32 = { 0.f, .0f, .0f, 1.f } } }, + }; + + VkRenderPassBeginInfo renderBeginInfo[] = { + // RP_WORLD + { + .sType = VK_STRUCTURE_TYPE_RENDER_PASS_BEGIN_INFO, + .renderPass = vk_renderpasses[RP_WORLD].rp, + .framebuffer = vk_framebuffers[RP_WORLD][vk_imageIndex], + .renderArea.offset = { 0, 0 }, + .renderArea.extent = vk_swapchain.extent, + .clearValueCount = vk_renderpasses[RP_WORLD].sampleCount != VK_SAMPLE_COUNT_1_BIT ? 3 : 2, + .pClearValues = clearColors + }, + // RP_UI + { + .sType = VK_STRUCTURE_TYPE_RENDER_PASS_BEGIN_INFO, + .renderPass = vk_renderpasses[RP_UI].rp, + .framebuffer = vk_framebuffers[RP_UI][vk_imageIndex], + .renderArea.offset = { 0, 0 }, + .renderArea.extent = vk_swapchain.extent, + .clearValueCount = 3, + .pClearValues = uiClearColors + }, + // RP_WORLD_WARP + { + .sType = VK_STRUCTURE_TYPE_RENDER_PASS_BEGIN_INFO, + .renderPass = vk_renderpasses[RP_WORLD_WARP].rp, + .framebuffer = vk_framebuffers[RP_WORLD_WARP][vk_imageIndex], + .renderArea.offset = { 0, 0 }, + .renderArea.extent = vk_swapchain.extent, + .clearValueCount = 2, + .pClearValues = warpClearColors + } + }; + + if (rpType == RP_WORLD) { + QVk_DebugLabelBegin(&vk_commandbuffers[vk_activeBufferIdx], "Draw World", 0.f, 1.f, 0.f); + } + if (rpType == RP_UI) { + QVk_DebugLabelEnd(&vk_commandbuffers[vk_activeBufferIdx]); + QVk_DebugLabelBegin(&vk_commandbuffers[vk_activeBufferIdx], "Draw UI", 1.f, 1.f, 0.f); + } + if (rpType == RP_WORLD_WARP) { + QVk_DebugLabelEnd(&vk_commandbuffers[vk_activeBufferIdx]); + QVk_DebugLabelBegin(&vk_commandbuffers[vk_activeBufferIdx], "Draw View Warp", 1.f, 0.f, .5f); + } + + vkCmdBeginRenderPass(vk_commandbuffers[vk_activeBufferIdx], &renderBeginInfo[rpType], VK_SUBPASS_CONTENTS_INLINE); + vk_state.current_renderpass = rpType; +} + +#if 0 +void QVk_RecreateSwapchain() +{ + vkDeviceWaitIdle( vk_device.logical ); + DestroyFramebuffers(); + DestroyImageViews(); + VK_VERIFY(QVk_CreateSwapchain()); + vk_viewport.width = (float)vid.width; + vk_viewport.height = (float)vid.height; + vk_scissor.extent = vk_swapchain.extent; + DestroyDrawBuffers(); + CreateDrawBuffers(); + VK_VERIFY(CreateImageViews()); + VK_VERIFY(CreateFramebuffers()); +} +#endif + +uint8_t *QVk_GetVertexBuffer(VkDeviceSize size, VkBuffer *dstBuffer, VkDeviceSize *dstOffset) +{ + if (vk_dynVertexBuffers[vk_activeDynBufferIdx].currentOffset + size > vk_config.vertex_buffer_size) + { + vk_config.vertex_buffer_size = max(vk_config.vertex_buffer_size * BUFFER_RESIZE_FACTOR, NextPow2(size)); + + R_Printf(PRINT_ALL, "Resizing dynamic vertex buffer to %ukB\n", vk_config.vertex_buffer_size / 1024); + int swapBufferOffset = vk_swapBuffersCnt[vk_activeSwapBufferIdx]; + vk_swapBuffersCnt[vk_activeSwapBufferIdx] += NUM_DYNBUFFERS; + + if (vk_swapBuffers[vk_activeSwapBufferIdx] == NULL) + vk_swapBuffers[vk_activeSwapBufferIdx] = malloc(sizeof(qvkbuffer_t) * vk_swapBuffersCnt[vk_activeSwapBufferIdx]); + else + vk_swapBuffers[vk_activeSwapBufferIdx] = realloc(vk_swapBuffers[vk_activeSwapBufferIdx], sizeof(qvkbuffer_t) * vk_swapBuffersCnt[vk_activeSwapBufferIdx]); + + for (int i = 0; i < NUM_DYNBUFFERS; ++i) + { + // need unmap before copy to swapBuffers + buffer_unmap(&vk_dynVertexBuffers[i].resource); + vk_swapBuffers[vk_activeSwapBufferIdx][swapBufferOffset + i] = vk_dynVertexBuffers[i]; + + QVk_CreateVertexBuffer(NULL, vk_config.vertex_buffer_size, + &vk_dynVertexBuffers[i], VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT, VK_MEMORY_PROPERTY_HOST_CACHED_BIT); + vk_dynVertexBuffers[i].pMappedData = buffer_map(&vk_dynVertexBuffers[i].resource); + + QVk_DebugSetObjectName((uint64_t)vk_dynVertexBuffers[i].resource.buffer, + VK_OBJECT_TYPE_BUFFER, va("Dynamic Vertex Buffer #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_dynVertexBuffers[i].resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, va("Memory: Dynamic Vertex Buffer #%d", i)); + } + } + + *dstOffset = vk_dynVertexBuffers[vk_activeDynBufferIdx].currentOffset; + *dstBuffer = vk_dynVertexBuffers[vk_activeDynBufferIdx].resource.buffer; + vk_dynVertexBuffers[vk_activeDynBufferIdx].currentOffset += size; + + vk_config.vertex_buffer_usage = vk_dynVertexBuffers[vk_activeDynBufferIdx].currentOffset; + if (vk_config.vertex_buffer_max_usage < vk_config.vertex_buffer_usage) + vk_config.vertex_buffer_max_usage = vk_config.vertex_buffer_usage; + + return (uint8_t *)vk_dynVertexBuffers[vk_activeDynBufferIdx].pMappedData + (*dstOffset); +} + +static uint8_t *QVk_GetIndexBuffer(VkDeviceSize size, VkDeviceSize *dstOffset, int currentBufferIdx) +{ + // align to 4 bytes, so that we can reuse the buffer for both VK_INDEX_TYPE_UINT16 and VK_INDEX_TYPE_UINT32 + const uint32_t aligned_size = ROUNDUP(size, 4); + + if (vk_dynIndexBuffers[currentBufferIdx].currentOffset + aligned_size > vk_config.index_buffer_size) + { + vk_config.index_buffer_size = max(vk_config.index_buffer_size * BUFFER_RESIZE_FACTOR, NextPow2(size)); + + R_Printf(PRINT_ALL, "Resizing dynamic index buffer to %ukB\n", vk_config.index_buffer_size / 1024); + int swapBufferOffset = vk_swapBuffersCnt[vk_activeSwapBufferIdx]; + vk_swapBuffersCnt[vk_activeSwapBufferIdx] += NUM_DYNBUFFERS; + + if (vk_swapBuffers[vk_activeSwapBufferIdx] == NULL) + vk_swapBuffers[vk_activeSwapBufferIdx] = malloc(sizeof(qvkbuffer_t) * vk_swapBuffersCnt[vk_activeSwapBufferIdx]); + else + vk_swapBuffers[vk_activeSwapBufferIdx] = realloc(vk_swapBuffers[vk_activeSwapBufferIdx], sizeof(qvkbuffer_t) * vk_swapBuffersCnt[vk_activeSwapBufferIdx]); + + for (int i = 0; i < NUM_DYNBUFFERS; ++i) + { + // need unmap before copy to swapBuffers + buffer_unmap(&vk_dynIndexBuffers[i].resource); + vk_swapBuffers[vk_activeSwapBufferIdx][swapBufferOffset + i] = vk_dynIndexBuffers[i]; + + QVk_CreateIndexBuffer(NULL, vk_config.index_buffer_size, + &vk_dynIndexBuffers[i], VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT, VK_MEMORY_PROPERTY_HOST_CACHED_BIT); + vk_dynIndexBuffers[i].pMappedData = buffer_map(&vk_dynIndexBuffers[i].resource); + + QVk_DebugSetObjectName((uint64_t)vk_dynIndexBuffers[i].resource.buffer, + VK_OBJECT_TYPE_BUFFER, va("Dynamic Index Buffer #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_dynIndexBuffers[i].resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, va("Memory: Dynamic Index Buffer #%d", i)); + } + } + + *dstOffset = vk_dynIndexBuffers[currentBufferIdx].currentOffset; + vk_dynIndexBuffers[currentBufferIdx].currentOffset += aligned_size; + + vk_config.index_buffer_usage = vk_dynIndexBuffers[currentBufferIdx].currentOffset; + if (vk_config.index_buffer_max_usage < vk_config.index_buffer_usage) + vk_config.index_buffer_max_usage = vk_config.index_buffer_usage; + + return (uint8_t *)vk_dynIndexBuffers[currentBufferIdx].pMappedData + (*dstOffset); +} + +uint8_t *QVk_GetUniformBuffer(VkDeviceSize size, uint32_t *dstOffset, VkDescriptorSet *dstUboDescriptorSet) +{ + // 0x100 alignment is required by Vulkan spec + const uint32_t aligned_size = ROUNDUP(size, 0x100); + + if (vk_dynUniformBuffers[vk_activeDynBufferIdx].currentOffset + UNIFORM_ALLOC_SIZE > vk_config.uniform_buffer_size) + { + vk_config.uniform_buffer_size = max(vk_config.uniform_buffer_size * BUFFER_RESIZE_FACTOR, NextPow2(size)); + + R_Printf(PRINT_ALL, "Resizing dynamic uniform buffer to %ukB\n", vk_config.uniform_buffer_size / 1024); + int swapBufferOffset = vk_swapBuffersCnt[vk_activeSwapBufferIdx]; + int swapDescSetsOffset = vk_swapDescSetsCnt[vk_activeSwapBufferIdx]; + vk_swapBuffersCnt[vk_activeSwapBufferIdx] += NUM_DYNBUFFERS; + vk_swapDescSetsCnt[vk_activeSwapBufferIdx] += NUM_DYNBUFFERS; + + if (vk_swapBuffers[vk_activeSwapBufferIdx] == NULL) + vk_swapBuffers[vk_activeSwapBufferIdx] = malloc(sizeof(qvkbuffer_t) * vk_swapBuffersCnt[vk_activeSwapBufferIdx]); + else + vk_swapBuffers[vk_activeSwapBufferIdx] = realloc(vk_swapBuffers[vk_activeSwapBufferIdx], sizeof(qvkbuffer_t) * vk_swapBuffersCnt[vk_activeSwapBufferIdx]); + + if (vk_swapDescriptorSets[vk_activeSwapBufferIdx] == NULL) + vk_swapDescriptorSets[vk_activeSwapBufferIdx] = malloc(sizeof(VkDescriptorSet) * vk_swapDescSetsCnt[vk_activeSwapBufferIdx]); + else + vk_swapDescriptorSets[vk_activeSwapBufferIdx] = realloc(vk_swapDescriptorSets[vk_activeSwapBufferIdx], sizeof(VkDescriptorSet) * vk_swapDescSetsCnt[vk_activeSwapBufferIdx]); + + for (int i = 0; i < NUM_DYNBUFFERS; ++i) + { + // need unmap before copy to swapBuffers + buffer_unmap(&vk_dynUniformBuffers[i].resource); + vk_swapBuffers[vk_activeSwapBufferIdx][swapBufferOffset + i] = vk_dynUniformBuffers[i]; + vk_swapDescriptorSets[vk_activeSwapBufferIdx][swapDescSetsOffset + i] = vk_uboDescriptorSets[i]; + + VK_VERIFY(QVk_CreateUniformBuffer(vk_config.uniform_buffer_size, + &vk_dynUniformBuffers[i], VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT, VK_MEMORY_PROPERTY_HOST_CACHED_BIT)); + vk_dynUniformBuffers[i].pMappedData = buffer_map(&vk_dynUniformBuffers[i].resource); + CreateUboDescriptorSet(&vk_uboDescriptorSets[i], + vk_dynUniformBuffers[i].resource.buffer); + + QVk_DebugSetObjectName((uint64_t)vk_uboDescriptorSets[i], + VK_OBJECT_TYPE_DESCRIPTOR_SET, va("Dynamic UBO Descriptor Set #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_dynUniformBuffers[i].resource.buffer, + VK_OBJECT_TYPE_BUFFER, va("Dynamic Uniform Buffer #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_dynUniformBuffers[i].resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, va("Memory: Dynamic Uniform Buffer #%d", i)); + } + } + + *dstOffset = vk_dynUniformBuffers[vk_activeDynBufferIdx].currentOffset; + *dstUboDescriptorSet = vk_uboDescriptorSets[vk_activeDynBufferIdx]; + vk_dynUniformBuffers[vk_activeDynBufferIdx].currentOffset += aligned_size; + + vk_config.uniform_buffer_usage = vk_dynUniformBuffers[vk_activeDynBufferIdx].currentOffset; + if (vk_config.uniform_buffer_max_usage < vk_config.uniform_buffer_usage) + vk_config.uniform_buffer_max_usage = vk_config.uniform_buffer_usage; + + return (uint8_t *)vk_dynUniformBuffers[vk_activeDynBufferIdx].pMappedData + (*dstOffset); +} + +uint8_t *QVk_GetStagingBuffer(VkDeviceSize size, int alignment, VkCommandBuffer *cmdBuffer, VkBuffer *buffer, uint32_t *dstOffset) +{ + qvkstagingbuffer_t * stagingBuffer = &vk_stagingBuffers[vk_activeStagingBuffer]; + stagingBuffer->currentOffset = ROUNDUP(stagingBuffer->currentOffset, alignment); + + if (((stagingBuffer->currentOffset + size) >= stagingBuffer->resource.size) && !stagingBuffer->submitted) + SubmitStagingBuffer(vk_activeStagingBuffer); + + stagingBuffer = &vk_stagingBuffers[vk_activeStagingBuffer]; + if (size > stagingBuffer->resource.size) + { + R_Printf(PRINT_ALL, "%s: %d: Resize stanging buffer" YQ2_COM_PRId64 "->" YQ2_COM_PRId64 "\n", + __func__, vk_activeStagingBuffer, stagingBuffer->resource.size, size); + + DestroyStagingBuffer(stagingBuffer); + CreateStagingBuffer(size, stagingBuffer, vk_activeStagingBuffer); + } + else + { + if (stagingBuffer->submitted) + { + VK_VERIFY(vkWaitForFences(vk_device.logical, 1, &stagingBuffer->fence, VK_TRUE, UINT64_MAX)); + VK_VERIFY(vkResetFences(vk_device.logical, 1, &stagingBuffer->fence)); + + stagingBuffer->currentOffset = 0; + stagingBuffer->submitted = false; + + VkCommandBufferBeginInfo beginInfo = { + .sType = VK_STRUCTURE_TYPE_COMMAND_BUFFER_BEGIN_INFO, + .pNext = NULL, + .flags = VK_COMMAND_BUFFER_USAGE_ONE_TIME_SUBMIT_BIT, + .pInheritanceInfo = NULL + }; + + // Command buffers are implicitly reset by vkBeginCommandBuffer if VK_COMMAND_POOL_CREATE_RESET_COMMAND_BUFFER_BIT is set - this is expensive. + // It's more efficient to reset entire pool instead and it also fixes the VK_VALIDATION_FEATURE_ENABLE_BEST_PRACTICES_EXT performance warning. + // see also: https://github.com/KhronosGroup/Vulkan-Samples/blob/master/samples/performance/command_buffer_usage/command_buffer_usage_tutorial.md + vkResetCommandPool(vk_device.logical, vk_stagingCommandPool[vk_activeStagingBuffer], 0); + VK_VERIFY(vkBeginCommandBuffer(stagingBuffer->cmdBuffer, &beginInfo)); + } + } + + if (cmdBuffer) + *cmdBuffer = stagingBuffer->cmdBuffer; + if (buffer) + *buffer = stagingBuffer->resource.buffer; + if (dstOffset) + *dstOffset = stagingBuffer->currentOffset; + + unsigned char *data = (uint8_t *)stagingBuffer->pMappedData + stagingBuffer->currentOffset; + stagingBuffer->currentOffset += size; + + return data; +} + +static void +QVk_CheckTriangleIbo(VkDeviceSize indexCount) +{ + if (indexCount > vk_config.triangle_index_usage) + vk_config.triangle_index_usage = indexCount; + + if (vk_config.triangle_index_usage > vk_config.triangle_index_max_usage) + vk_config.triangle_index_max_usage = vk_config.triangle_index_usage; + + if (indexCount > vk_config.triangle_index_count) + { + vk_config.triangle_index_count *= BUFFER_RESIZE_FACTOR; + R_Printf(PRINT_ALL, "Resizing triangle index buffer to %u indices.\n", vk_config.triangle_index_count); + RebuildTriangleIndexBuffer(); + } +} + +VkBuffer QVk_GetTriangleFanIbo(VkDeviceSize indexCount) +{ + QVk_CheckTriangleIbo(indexCount); + + return *vk_triangleFanIbo; +} + +VkBuffer QVk_GetTriangleStripIbo(VkDeviceSize indexCount) +{ + QVk_CheckTriangleIbo(indexCount); + + return *vk_triangleStripIbo; +} + +void QVk_SubmitStagingBuffers() +{ + for (int i = 0; i < NUM_DYNBUFFERS; ++i) + { + if (!vk_stagingBuffers[i].submitted && vk_stagingBuffers[i].currentOffset > 0) + SubmitStagingBuffer(i); + } +} + +VkSampler QVk_UpdateTextureSampler(qvktexture_t *texture, qvksampler_t samplerType, qboolean clampToEdge) +{ + const int samplerIndex = samplerType + (clampToEdge ? S_SAMPLER_CNT : 0); + + assert((vk_samplers[samplerIndex] != VK_NULL_HANDLE) && "Sampler is VK_NULL_HANDLE!"); + + texture->clampToEdge = clampToEdge; + + VkDescriptorImageInfo dImgInfo = { + .sampler = vk_samplers[samplerIndex], + .imageView = texture->imageView, + .imageLayout = VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL + }; + + VkWriteDescriptorSet writeSet = { + .sType = VK_STRUCTURE_TYPE_WRITE_DESCRIPTOR_SET, + .pNext = NULL, + .dstSet = texture->descriptorSet, + .dstBinding = 0, + .dstArrayElement = 0, + .descriptorCount = 1, + .descriptorType = VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER, + .pImageInfo = &dImgInfo, + .pBufferInfo = NULL, + .pTexelBufferView = NULL + }; + + vkUpdateDescriptorSets(vk_device.logical, 1, &writeSet, 0, NULL); + + return vk_samplers[samplerIndex]; +} + +void QVk_DrawColorRect(float *ubo, VkDeviceSize uboSize, qvkrenderpasstype_t rpType) +{ + uint32_t uboOffset; + VkDescriptorSet uboDescriptorSet; + uint8_t *vertData = QVk_GetUniformBuffer(uboSize, + &uboOffset, &uboDescriptorSet); + memcpy(vertData, ubo, uboSize); + + QVk_BindPipeline(&vk_drawColorQuadPipeline[rpType]); + VkDeviceSize offsets = 0; + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, + vk_drawColorQuadPipeline[rpType].layout, 0, 1, &uboDescriptorSet, 1, &uboOffset); + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, + &vk_colorRectVbo.resource.buffer, &offsets); + vkCmdBindIndexBuffer(vk_activeCmdbuffer, vk_rectIbo.resource.buffer, + 0, VK_INDEX_TYPE_UINT32); + vkCmdDrawIndexed(vk_activeCmdbuffer, 6, 1, 0, 0, 0); +} + +void QVk_DrawTexRect(const float *ubo, VkDeviceSize uboSize, qvktexture_t *texture) +{ + uint32_t uboOffset; + VkDescriptorSet uboDescriptorSet; + uint8_t *uboData = QVk_GetUniformBuffer(uboSize, &uboOffset, &uboDescriptorSet); + memcpy(uboData, ubo, uboSize); + + QVk_BindPipeline(&vk_drawTexQuadPipeline[vk_state.current_renderpass]); + VkDeviceSize offsets = 0; + VkDescriptorSet descriptorSets[] = { texture->descriptorSet, uboDescriptorSet }; + + float gamma = 2.1F - vid_gamma->value; + + vkCmdPushConstants(vk_activeCmdbuffer, vk_drawTexQuadPipeline[vk_state.current_renderpass].layout, + VK_SHADER_STAGE_FRAGMENT_BIT, 17 * sizeof(float), sizeof(gamma), &gamma); + + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_drawTexQuadPipeline[vk_state.current_renderpass].layout, 0, 2, descriptorSets, 1, &uboOffset); + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, + &vk_texRectVbo.resource.buffer, &offsets); + vkCmdBindIndexBuffer(vk_activeCmdbuffer, + vk_rectIbo.resource.buffer, 0, VK_INDEX_TYPE_UINT32); + vkCmdDrawIndexed(vk_activeCmdbuffer, 6, 1, 0, 0, 0); +} + +void QVk_BindPipeline(qvkpipeline_t *pipeline) +{ + if (vk_state.current_pipeline != pipeline->pl) + { + vkCmdBindPipeline(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, pipeline->pl); + vk_state.current_pipeline = pipeline->pl; + } +} + +const char *QVk_GetError(VkResult errorCode) +{ +#define ERRSTR(r) case VK_ ##r: return "VK_"#r + switch (errorCode) + { + ERRSTR(SUCCESS); + ERRSTR(NOT_READY); + ERRSTR(TIMEOUT); + ERRSTR(EVENT_SET); + ERRSTR(EVENT_RESET); + ERRSTR(INCOMPLETE); + ERRSTR(ERROR_OUT_OF_HOST_MEMORY); + ERRSTR(ERROR_OUT_OF_DEVICE_MEMORY); + ERRSTR(ERROR_INITIALIZATION_FAILED); + ERRSTR(ERROR_DEVICE_LOST); + ERRSTR(ERROR_MEMORY_MAP_FAILED); + ERRSTR(ERROR_LAYER_NOT_PRESENT); + ERRSTR(ERROR_EXTENSION_NOT_PRESENT); + ERRSTR(ERROR_FEATURE_NOT_PRESENT); + ERRSTR(ERROR_INCOMPATIBLE_DRIVER); + ERRSTR(ERROR_TOO_MANY_OBJECTS); + ERRSTR(ERROR_FORMAT_NOT_SUPPORTED); + ERRSTR(ERROR_SURFACE_LOST_KHR); + ERRSTR(ERROR_NATIVE_WINDOW_IN_USE_KHR); + ERRSTR(SUBOPTIMAL_KHR); + ERRSTR(ERROR_OUT_OF_DATE_KHR); + ERRSTR(ERROR_INCOMPATIBLE_DISPLAY_KHR); + ERRSTR(ERROR_VALIDATION_FAILED_EXT); + ERRSTR(ERROR_INVALID_SHADER_NV); + default: return ""; + } +#undef ERRSTR + return "UNKNOWN ERROR"; +} diff --git a/src/vk/vk_device.c b/src/vk/vk_device.c new file mode 100644 index 0000000..cbd75a1 --- /dev/null +++ b/src/vk/vk_device.c @@ -0,0 +1,365 @@ +/* +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. +*/ + +#include "header/local.h" + +// internal helper +static qboolean deviceExtensionsSupported(const VkPhysicalDevice *physicalDevice, const char* extensionName) +{ + uint32_t availableExtCount = 0; + qboolean vk_extension_available = false; + VK_VERIFY(vkEnumerateDeviceExtensionProperties(*physicalDevice, NULL, &availableExtCount, NULL)); + + if (availableExtCount > 0) + { + VkExtensionProperties *extensions = (VkExtensionProperties *)malloc(availableExtCount * sizeof(VkExtensionProperties)); + VK_VERIFY(vkEnumerateDeviceExtensionProperties(*physicalDevice, NULL, &availableExtCount, extensions)); + + for (uint32_t i = 0; i < availableExtCount; ++i) + { + vk_extension_available |= strcmp(extensions[i].extensionName, extensionName) == 0; + } + + free(extensions); + } + + // lack of extension disqualifies the device + return vk_extension_available; +} + +// internal helper +static void getBestPhysicalDevice(const VkPhysicalDevice *devices, int preferredIdx, int count) +{ + VkPhysicalDeviceProperties deviceProperties; + VkPhysicalDeviceFeatures deviceFeatures; + uint32_t queueFamilyCount = 0; + + for (int i = 0; i < count; ++i) + { + vkGetPhysicalDeviceProperties(devices[i], &deviceProperties); + vkGetPhysicalDeviceFeatures(devices[i], &deviceFeatures); + vkGetPhysicalDeviceQueueFamilyProperties(devices[i], &queueFamilyCount, NULL); + + if (queueFamilyCount == 0) + continue; + + // prefer discrete GPU but if it's the only one available then don't be picky + // also - if the user specifies a preferred device, select it + qboolean bestProperties = deviceProperties.deviceType == VK_PHYSICAL_DEVICE_TYPE_DISCRETE_GPU; + if (preferredIdx == i || (bestProperties && preferredIdx < 0) || count == 1) + { + uint32_t formatCount = 0; + uint32_t presentModesCount = 0; + + // check if requested device extensions are present + qboolean extSupported = deviceExtensionsSupported(&devices[i], VK_KHR_SWAPCHAIN_EXTENSION_NAME); + + // no required extensions? try next device + if (!extSupported) + continue; + + // if extensions are fine, query surface formats and present modes to see if the device can be used + VK_VERIFY(vkGetPhysicalDeviceSurfaceFormatsKHR(devices[i], vk_surface, &formatCount, NULL)); + VK_VERIFY(vkGetPhysicalDeviceSurfacePresentModesKHR(devices[i], vk_surface, &presentModesCount, NULL)); + + if (formatCount == 0 || presentModesCount == 0) + continue; + + VkQueueFamilyProperties *queueFamilies = (VkQueueFamilyProperties *)malloc(queueFamilyCount * sizeof(VkQueueFamilyProperties)); + vkGetPhysicalDeviceQueueFamilyProperties(devices[i], &queueFamilyCount, queueFamilies); + + // secondary check - device is OK if there's at least on queue with VK_QUEUE_GRAPHICS_BIT set + for (uint32_t j = 0; j < queueFamilyCount; ++j) + { + // check if this queue family has support for presentation + VkBool32 presentSupported; + VK_VERIFY(vkGetPhysicalDeviceSurfaceSupportKHR(devices[i], j, vk_surface, &presentSupported)); + + // good optimization would be to find a queue where presentIdx == gfxQueueIdx for less overhead + if (vk_device.presentFamilyIndex < 0 && queueFamilies[j].queueCount > 0 && presentSupported) + { + vk_device.presentFamilyIndex = j; + } + + if (vk_device.gfxFamilyIndex < 0 && queueFamilies[j].queueCount > 0 && (queueFamilies[j].queueFlags & VK_QUEUE_GRAPHICS_BIT)) + { + vk_device.gfxFamilyIndex = j; + } + + if (vk_device.transferFamilyIndex < 0 && queueFamilies[j].queueCount > 0 && !(queueFamilies[j].queueFlags & VK_QUEUE_GRAPHICS_BIT) && (queueFamilies[j].queueFlags & VK_QUEUE_TRANSFER_BIT)) + { + vk_device.transferFamilyIndex = j; + } + } + + free(queueFamilies); + + // accept only device that has support for presentation and drawing + if (vk_device.presentFamilyIndex >= 0 && vk_device.gfxFamilyIndex >= 0) + { + if (vk_device.transferFamilyIndex < 0) + { + vk_device.transferFamilyIndex = vk_device.gfxFamilyIndex; + } + + vk_device.physical = devices[i]; + vk_device.properties = deviceProperties; + vk_device.features = deviceFeatures; + return; + } + } + } +} + +// internal helper +static qboolean selectPhysicalDevice(int preferredDeviceIdx) +{ + uint32_t physicalDeviceCount = 0; + VK_VERIFY(vkEnumeratePhysicalDevices(vk_instance, &physicalDeviceCount, NULL)); + + if (physicalDeviceCount == 0) + { + R_Printf(PRINT_ALL, "No Vulkan-capable devices found!\n"); + return false; + } + + R_Printf(PRINT_ALL, "...found %d Vulkan-capable device(s)\n", physicalDeviceCount); + + VkPhysicalDevice *physicalDevices = (VkPhysicalDevice *)malloc(physicalDeviceCount * sizeof(VkPhysicalDevice)); + VK_VERIFY(vkEnumeratePhysicalDevices(vk_instance, &physicalDeviceCount, physicalDevices)); + + getBestPhysicalDevice(physicalDevices, preferredDeviceIdx < physicalDeviceCount ? preferredDeviceIdx : -1, physicalDeviceCount); + free(physicalDevices); + + if (vk_device.physical == VK_NULL_HANDLE) + { + R_Printf(PRINT_ALL, "Could not find a suitable physical device!\n"); + return false; + } + + if (!vk_device.features.samplerAnisotropy) + { + R_Printf(PRINT_ALL, "...anisotropy filtering is unsupported.\n"); + } + + return true; +} + +// internal helper +static VkResult createLogicalDevice() +{ + // at least one queue (graphics and present combined) has to be present + uint32_t numQueues = 1; + float queuePriority = 1.f; + VkDeviceQueueCreateInfo queueCreateInfo[3]; + queueCreateInfo[0].sType = VK_STRUCTURE_TYPE_DEVICE_QUEUE_CREATE_INFO; + queueCreateInfo[0].pNext = NULL; + queueCreateInfo[0].flags = 0; + queueCreateInfo[0].queueFamilyIndex = vk_device.gfxFamilyIndex; + queueCreateInfo[0].queueCount = 1; + queueCreateInfo[0].pQueuePriorities = &queuePriority; + queueCreateInfo[1].sType = VK_STRUCTURE_TYPE_DEVICE_QUEUE_CREATE_INFO; + queueCreateInfo[1].pNext = NULL; + queueCreateInfo[1].flags = 0; + queueCreateInfo[1].queueFamilyIndex = 0; + queueCreateInfo[1].queueCount = 1; + queueCreateInfo[1].pQueuePriorities = &queuePriority; + queueCreateInfo[2].sType = VK_STRUCTURE_TYPE_DEVICE_QUEUE_CREATE_INFO; + queueCreateInfo[2].pNext = NULL; + queueCreateInfo[2].flags = 0; + queueCreateInfo[2].queueFamilyIndex = 0; + queueCreateInfo[2].queueCount = 1; + queueCreateInfo[2].pQueuePriorities = &queuePriority; + + VkPhysicalDeviceFeatures wantedDeviceFeatures = { + .samplerAnisotropy = vk_device.features.samplerAnisotropy, + .fillModeNonSolid = vk_device.features.fillModeNonSolid, // for wireframe rendering + .sampleRateShading = vk_device.features.sampleRateShading, // for sample shading + }; + + // a graphics and present queue are different - two queues have to be created + if (vk_device.gfxFamilyIndex != vk_device.presentFamilyIndex) + { + queueCreateInfo[numQueues++].queueFamilyIndex = vk_device.presentFamilyIndex; + } + + // a separate transfer queue exists that's different from present and graphics queue? + if (vk_device.transferFamilyIndex != vk_device.gfxFamilyIndex && vk_device.transferFamilyIndex != vk_device.presentFamilyIndex) + { + queueCreateInfo[numQueues++].queueFamilyIndex = vk_device.transferFamilyIndex; + } + + const char *deviceExtensions[] = { VK_KHR_SWAPCHAIN_EXTENSION_NAME }; + + VkDeviceCreateInfo deviceCreateInfo = { + .sType = VK_STRUCTURE_TYPE_DEVICE_CREATE_INFO, + .pEnabledFeatures = &wantedDeviceFeatures, + .ppEnabledExtensionNames = deviceExtensions, + .enabledExtensionCount = 1, + .enabledLayerCount = 0, + .ppEnabledLayerNames = NULL, + .queueCreateInfoCount = numQueues, + .pQueueCreateInfos = queueCreateInfo + }; + +#if VK_HEADER_VERSION > 101 + const char *validationLayers[] = { "VK_LAYER_KHRONOS_validation" }; +#else + const char *validationLayers[] = { "VK_LAYER_LUNARG_standard_validation" }; +#endif + + if (vk_validation->value) + { + deviceCreateInfo.enabledLayerCount = sizeof(validationLayers)/sizeof(validationLayers[0]); + deviceCreateInfo.ppEnabledLayerNames = validationLayers; + } + + return vkCreateDevice(vk_device.physical, &deviceCreateInfo, NULL, &vk_device.logical); +} + +// internal helper +static const char *deviceTypeString(VkPhysicalDeviceType dType) +{ +#define DEVTYPESTR(r) case VK_ ##r: return "VK_"#r + switch (dType) + { + DEVTYPESTR(PHYSICAL_DEVICE_TYPE_OTHER); + DEVTYPESTR(PHYSICAL_DEVICE_TYPE_INTEGRATED_GPU); + DEVTYPESTR(PHYSICAL_DEVICE_TYPE_DISCRETE_GPU); + DEVTYPESTR(PHYSICAL_DEVICE_TYPE_VIRTUAL_GPU); + DEVTYPESTR(PHYSICAL_DEVICE_TYPE_CPU); + default: return ""; + } +#undef DEVTYPESTR + return "UNKNOWN DEVICE"; +} + +// internal helper +static const char *vendorNameString(uint32_t vendorId) +{ + switch (vendorId) + { + case 0x1002: return "AMD"; + case 0x1010: return "ImgTec"; + case 0x10DE: return "NVIDIA"; + case 0x13B5: return "ARM"; + case 0x5143: return "Qualcomm"; + case 0x8086: return "Intel"; + default: return "unknown"; + } +} + +qboolean QVk_CreateDevice(int preferredDeviceIdx) +{ + if (!selectPhysicalDevice(preferredDeviceIdx)) + return false; + + vk_config.vendor_name = vendorNameString(vk_device.properties.vendorID); + vk_config.device_type = deviceTypeString(vk_device.properties.deviceType); + + VkResult res = createLogicalDevice(); + if (res != VK_SUCCESS) + { + R_Printf(PRINT_ALL, "Could not create Vulkan logical device: %s\n", QVk_GetError(res)); + return false; + } + + vkGetDeviceQueue(vk_device.logical, vk_device.gfxFamilyIndex, 0, &vk_device.gfxQueue); + vkGetDeviceQueue(vk_device.logical, vk_device.presentFamilyIndex, 0, &vk_device.presentQueue); + vkGetDeviceQueue(vk_device.logical, vk_device.transferFamilyIndex, 0, &vk_device.transferQueue); + vkGetPhysicalDeviceMemoryProperties(vk_device.physical, &vk_device.mem_properties); + // init our memory management + vulkan_memory_init(); + + return true; +} + +// debug label related functions +void QVk_DebugSetObjectName(uint64_t obj, VkObjectType objType, const char *objName) +{ + if (qvkSetDebugUtilsObjectNameEXT) + { + VkDebugUtilsObjectNameInfoEXT oNameInf = { + .sType = VK_STRUCTURE_TYPE_DEBUG_UTILS_OBJECT_NAME_INFO_EXT, + .pNext = NULL, + .objectType = objType, + .objectHandle = obj, + .pObjectName = objName + }; + + qvkSetDebugUtilsObjectNameEXT(vk_device.logical, &oNameInf); + } +} + +void QVk_DebugSetObjectTag(uint64_t obj, VkObjectType objType, uint64_t tagName, + size_t tagSize, const void *tagData) +{ + if (qvkSetDebugUtilsObjectTagEXT) + { + VkDebugUtilsObjectTagInfoEXT oTagInf = { + .sType = VK_STRUCTURE_TYPE_DEBUG_UTILS_OBJECT_TAG_INFO_EXT, + .pNext = NULL, + .objectType = objType, + .objectHandle = obj, + .tagName = tagName, + .tagSize = tagSize, + .pTag = tagData + }; + + qvkSetDebugUtilsObjectTagEXT(vk_device.logical, &oTagInf); + } +} + +void QVk_DebugLabelBegin(const VkCommandBuffer *cmdBuffer, const char *labelName, const float r, const float g, const float b) +{ + if (qvkCmdBeginDebugUtilsLabelEXT) + { + VkDebugUtilsLabelEXT labelInfo = { + .sType = VK_STRUCTURE_TYPE_DEBUG_UTILS_LABEL_EXT, + .pNext = NULL, + .pLabelName = labelName, + .color = { r, g, b, 1.f } + }; + + qvkCmdBeginDebugUtilsLabelEXT(*cmdBuffer, &labelInfo); + } +} + +void QVk_DebugLabelEnd(const VkCommandBuffer *cmdBuffer) +{ + if (qvkCmdEndDebugUtilsLabelEXT) + { + qvkCmdEndDebugUtilsLabelEXT(*cmdBuffer); + } +} + +void QVk_DebugLabelInsert(const VkCommandBuffer *cmdBuffer, const char *labelName, const float r, const float g, const float b) +{ + if (qvkInsertDebugUtilsLabelEXT) + { + VkDebugUtilsLabelEXT labelInfo = { + .sType = VK_STRUCTURE_TYPE_DEBUG_UTILS_LABEL_EXT, + .pNext = NULL, + .pLabelName = labelName, + .color = { r, g, b, 1.f } + }; + + qvkInsertDebugUtilsLabelEXT(*cmdBuffer, &labelInfo); + } +} diff --git a/src/vk/vk_draw.c b/src/vk/vk_draw.c new file mode 100644 index 0000000..6d12ff6 --- /dev/null +++ b/src/vk/vk_draw.c @@ -0,0 +1,371 @@ +/* +Copyright (C) 1997-2001 Id Software, Inc. +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. +*/ + +// vk_draw.c + +#include "header/local.h" + +static image_t *draw_chars; + +/* +=============== +Draw_InitLocal +=============== +*/ +void Draw_InitLocal (void) +{ + draw_chars = Vk_FindImage("pics/conchars.pcx", it_pic); + if (!draw_chars) + { + ri.Sys_Error(ERR_FATAL, "%s: Couldn't load pics/conchars.pcx", __func__); + } +} + + + +/* +================ +RE_Draw_CharScaled + +Draws one 8*8 graphics character with 0 being transparent. +It can be clipped to the top of the screen to allow the console to be +smoothly scrolled off. +================ +*/ +void RE_Draw_CharScaled (int x, int y, int num, float scale) +{ + int row, col; + float frow, fcol, size; + + if (!vk_frameStarted) + return; + + num &= 255; + + if ((num & 127) == 32) + return; // space + + if (y <= -8) + return; // totally off screen + + row = num >> 4; + col = num & 15; + + frow = row * 0.0625; + fcol = col * 0.0625; + size = 0.0625; + + float imgTransform[] = { (float)x / vid.width, (float)y / vid.height, + 8.f * scale / vid.width, 8.f * scale / vid.height, + fcol, frow, size, size }; + QVk_DrawTexRect(imgTransform, sizeof(imgTransform), &draw_chars->vk_texture); +} + +/* +============= +RE_Draw_FindPic +============= +*/ +image_t *RE_Draw_FindPic (char *name) +{ + image_t *vk; + + if (name[0] != '/' && name[0] != '\\') + { + char fullname[MAX_QPATH]; + + Com_sprintf(fullname, sizeof(fullname), "pics/%s.pcx", name); + vk = Vk_FindImage(fullname, it_pic); + } + else + vk = Vk_FindImage(name + 1, it_pic); + + return vk; +} + +/* +============= +RE_Draw_GetPicSize +============= +*/ +void RE_Draw_GetPicSize (int *w, int *h, char *name) +{ + image_t *vk; + + vk = RE_Draw_FindPic(name); + if (!vk) + { + *w = *h = -1; + return; + } + + *w = vk->width; + *h = vk->height; +} + +/* +============= +RE_Draw_StretchPic +============= +*/ +void RE_Draw_StretchPic (int x, int y, int w, int h, char *name) +{ + image_t *vk; + + if (!vk_frameStarted) + return; + + vk = RE_Draw_FindPic(name); + if (!vk) + { + R_Printf(PRINT_ALL, "%s(): Can't find pic: %s\n", __func__, name); + return; + } + + float imgTransform[] = { (float)x / vid.width, (float)y / vid.height, + (float)w / vid.width, (float)h / vid.height, + 0, 0, 1, 1 }; + QVk_DrawTexRect(imgTransform, sizeof(imgTransform), &vk->vk_texture); +} + + +/* +============= +RE_Draw_PicScaled +============= +*/ +void RE_Draw_PicScaled (int x, int y, char *name, float scale) +{ + image_t *vk; + + vk = RE_Draw_FindPic(name); + if (!vk) + { + R_Printf(PRINT_ALL, "%s(): Can't find pic: %s\n", __func__, name); + return; + } + + RE_Draw_StretchPic(x, y, vk->width*scale, vk->height*scale, name); +} + +/* +============= +RE_Draw_TileClear + +This repeats a 64*64 tile graphic to fill the screen around a sized down +refresh window. +============= +*/ +void RE_Draw_TileClear (int x, int y, int w, int h, char *name) +{ + image_t *image; + + if (!vk_frameStarted) + return; + + image = RE_Draw_FindPic(name); + if (!image) + { + R_Printf(PRINT_ALL, "%s(): Can't find pic: %s\n", __func__, name); + return; + } + + // Change viewport and scissor to draw in the top left corner as the world view. + VkViewport tileViewport = vk_viewport; + VkRect2D tileScissor = vk_scissor; + + tileViewport.x = 0.f; + tileViewport.y = 0.f; + tileScissor.offset.x = 0; + tileScissor.offset.y = 0; + + vkCmdSetViewport(vk_activeCmdbuffer, 0u, 1u, &tileViewport); + vkCmdSetScissor(vk_activeCmdbuffer, 0u, 1u, &tileScissor); + + const float divisor = (vk_pixel_size->value < 1.0f ? 1.0f : vk_pixel_size->value); + float imgTransform[] = { (float)x / (vid.width * divisor), (float)y / (vid.height * divisor), + (float)w / (vid.width * divisor), (float)h / (vid.height * divisor), + (float)x / (64.0 * divisor), (float)y / (64.0 * divisor), + (float)w / (64.0 * divisor), (float)h / (64.0 * divisor) }; + QVk_DrawTexRect(imgTransform, sizeof(imgTransform), &image->vk_texture); + + // Restore viewport and scissor. + vkCmdSetViewport(vk_activeCmdbuffer, 0u, 1u, &vk_viewport); + vkCmdSetScissor(vk_activeCmdbuffer, 0u, 1u, &vk_scissor); +} + + +/* +============= +RE_Draw_Fill + +Fills a box of pixels with a single color +============= +*/ +void RE_Draw_Fill (int x, int y, int w, int h, int c) +{ + union + { + unsigned c; + byte v[4]; + } color; + + if (!vk_frameStarted) + return; + + if ((unsigned)c > 255) + ri.Sys_Error(ERR_FATAL, "%s: bad color", __func__); + + color.c = d_8to24table[c]; + + float imgTransform[] = { (float)x / vid.width, (float)y / vid.height, + (float)w / vid.width, (float)h / vid.height, + color.v[0] / 255.f, color.v[1] / 255.f, color.v[2] / 255.f, 1.f }; + QVk_DrawColorRect(imgTransform, sizeof(imgTransform), RP_UI); +} + +//============================================================================= + +/* +================ +RE_Draw_FadeScreen + +================ +*/ +void RE_Draw_FadeScreen (void) +{ + float imgTransform[] = { 0.f, 0.f, vid.width, vid.height, 0.f, 0.f, 0.f, .8f }; + + if (!vk_frameStarted) + return; + + QVk_DrawColorRect(imgTransform, sizeof(imgTransform), RP_UI); +} + + +//==================================================================== + + +/* +============= +RE_Draw_StretchRaw +============= +*/ +extern unsigned r_rawpalette[256]; +extern qvktexture_t vk_rawTexture; +static int scaled_size = 512; + +void RE_Draw_StretchRaw (int x, int y, int w, int h, int cols, int rows, byte *data) +{ + + int i, j; + int hscale, vscale; + unsigned *dest; + byte *source; + byte *image_scaled; + unsigned *raw_image32; + + if (!vk_frameStarted) + return; + + if (vk_retexturing->value) + { + // triple scaling + if (cols < (vid.width / 3) || rows < (vid.height / 3)) + { + image_scaled = malloc(cols * rows * 9); + + scale3x(data, image_scaled, cols, rows); + + cols = cols * 3; + rows = rows * 3; + } + else + // double scaling + { + image_scaled = malloc(cols * rows * 4); + + scale2x(data, image_scaled, cols, rows); + + cols = cols * 2; + rows = rows * 2; + } + } + else + { + image_scaled = data; + } + + if (vk_rawTexture.resource.image == VK_NULL_HANDLE) + { + // get power of two image size, + // could be updated only size only after recreate texture + for (scaled_size = 512; scaled_size < cols && scaled_size < rows; scaled_size <<= 1) + ; + } + + raw_image32 = malloc(scaled_size * scaled_size * sizeof(unsigned)); + + hscale = cols * 0x10000 / scaled_size; + vscale = rows * 0x10000 / scaled_size; + + source = image_scaled; + dest = raw_image32; + for (i = 0; i < scaled_size; i++) + { + for (j = 0; j < scaled_size; j++) + { + *dest = r_rawpalette[*(source + ((j * hscale) >> 16))]; + dest ++; + } + source = image_scaled + (((i * vscale) >> 16) * cols); + } + + if (vk_retexturing->value) + { + free(image_scaled); + SmoothColorImage(raw_image32, scaled_size * scaled_size, scaled_size >> 7); + } + + if (vk_rawTexture.resource.image != VK_NULL_HANDLE) + { + QVk_UpdateTextureData(&vk_rawTexture, (unsigned char*)raw_image32, 0, 0, scaled_size, scaled_size); + } + else + { + QVVKTEXTURE_CLEAR(vk_rawTexture); + QVk_CreateTexture(&vk_rawTexture, (unsigned char*)raw_image32, scaled_size, scaled_size, + vk_current_sampler, false); + QVk_DebugSetObjectName((uint64_t)vk_rawTexture.resource.image, + VK_OBJECT_TYPE_IMAGE, "Image: raw texture"); + QVk_DebugSetObjectName((uint64_t)vk_rawTexture.imageView, + VK_OBJECT_TYPE_IMAGE_VIEW, "Image View: raw texture"); + QVk_DebugSetObjectName((uint64_t)vk_rawTexture.descriptorSet, + VK_OBJECT_TYPE_DESCRIPTOR_SET, "Descriptor Set: raw texture"); + QVk_DebugSetObjectName((uint64_t)vk_rawTexture.resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, "Memory: raw texture"); + } + free(raw_image32); + + float imgTransform[] = { (float)x / vid.width, (float)y / vid.height, + (float)w / vid.width, (float)h / vid.height, + 0.f, 0.f, 1.f, 1.0f }; + QVk_DrawTexRect(imgTransform, sizeof(imgTransform), &vk_rawTexture); +} diff --git a/src/vk/vk_image.c b/src/vk/vk_image.c new file mode 100644 index 0000000..4dc555a --- /dev/null +++ b/src/vk/vk_image.c @@ -0,0 +1,1584 @@ +/* +Copyright (C) 1997-2001 Id Software, Inc. +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. +*/ + +#include "header/local.h" + +image_t vktextures[MAX_VKTEXTURES]; +int numvktextures = 0; +static int img_loaded = 0; +static int image_max = 0; + +// texture for storing raw image data (cinematics, endscreens, etc.) +qvktexture_t vk_rawTexture = QVVKTEXTURE_INIT; + +static byte intensitytable[256]; +static unsigned char overbrightable[256]; + +static cvar_t *intensity; +extern cvar_t *vk_mip_nearfilter; + +unsigned d_8to24table[256]; + +// default global texture and lightmap samplers +qvksampler_t vk_current_sampler = S_MIPMAP_LINEAR; +qvksampler_t vk_current_lmap_sampler = S_MIPMAP_LINEAR; + +// internal helper +static VkImageAspectFlags getDepthStencilAspect(VkFormat depthFormat) +{ + switch (depthFormat) + { + case VK_FORMAT_D32_SFLOAT_S8_UINT: + case VK_FORMAT_D24_UNORM_S8_UINT: + case VK_FORMAT_D16_UNORM_S8_UINT: + return VK_IMAGE_ASPECT_DEPTH_BIT | VK_IMAGE_ASPECT_STENCIL_BIT; + default: + return VK_IMAGE_ASPECT_DEPTH_BIT; + } +} + +// internal helper +static void transitionImageLayout(const VkCommandBuffer *cmdBuffer, const VkQueue *queue, const qvktexture_t *texture, const VkImageLayout oldLayout, const VkImageLayout newLayout) +{ + VkPipelineStageFlags srcStage = 0; + VkPipelineStageFlags dstStage = 0; + + VkImageMemoryBarrier imgBarrier = { + .sType = VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER, + .pNext = NULL, + .oldLayout = oldLayout, + .newLayout = newLayout, + .srcQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED, + .dstQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED, + .image = texture->resource.image, + .subresourceRange.baseMipLevel = 0, // no mip mapping levels + .subresourceRange.baseArrayLayer = 0, + .subresourceRange.layerCount = 1, + .subresourceRange.levelCount = texture->mipLevels + }; + + if (newLayout == VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL) + { + imgBarrier.subresourceRange.aspectMask = getDepthStencilAspect(texture->format); + } + else + { + imgBarrier.subresourceRange.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT; + } + + if (oldLayout == VK_IMAGE_LAYOUT_UNDEFINED && newLayout == VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL) + { + imgBarrier.srcAccessMask = 0; + imgBarrier.dstAccessMask = VK_ACCESS_TRANSFER_WRITE_BIT; + srcStage = VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT; + dstStage = VK_PIPELINE_STAGE_TRANSFER_BIT; + } + // transiton that may occur when updating existing image + else if (oldLayout == VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL && newLayout == VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL) + { + imgBarrier.srcAccessMask = 0; + imgBarrier.dstAccessMask = VK_ACCESS_TRANSFER_WRITE_BIT; + srcStage = VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT; + dstStage = VK_PIPELINE_STAGE_TRANSFER_BIT; + } + else if (oldLayout == VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL && newLayout == VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL) + { + if (vk_device.transferQueue == vk_device.gfxQueue) + { + imgBarrier.srcAccessMask = VK_ACCESS_TRANSFER_WRITE_BIT; + imgBarrier.dstAccessMask = VK_ACCESS_SHADER_READ_BIT; + srcStage = VK_PIPELINE_STAGE_TRANSFER_BIT; + dstStage = VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT; + } + else + { + if (vk_device.transferQueue == *queue) + { + // if the image is exclusively shared, start queue ownership transfer process (release) - only for VK_SHARING_MODE_EXCLUSIVE + imgBarrier.srcAccessMask = VK_ACCESS_TRANSFER_WRITE_BIT; + imgBarrier.dstAccessMask = 0; + imgBarrier.srcQueueFamilyIndex = vk_device.transferFamilyIndex; + imgBarrier.dstQueueFamilyIndex = vk_device.gfxFamilyIndex; + srcStage = VK_PIPELINE_STAGE_TRANSFER_BIT; + dstStage = VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT; + } + else + { + // continuing queue transfer (acquisition) - this will only happen for VK_SHARING_MODE_EXCLUSIVE images + if (texture->sharingMode == VK_SHARING_MODE_EXCLUSIVE) + { + imgBarrier.srcAccessMask = 0; + imgBarrier.dstAccessMask = VK_ACCESS_SHADER_READ_BIT; + imgBarrier.srcQueueFamilyIndex = vk_device.transferFamilyIndex; + imgBarrier.dstQueueFamilyIndex = vk_device.gfxFamilyIndex; + srcStage = VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT; + dstStage = VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT; + } + else + { + imgBarrier.srcAccessMask = VK_ACCESS_TRANSFER_WRITE_BIT; + imgBarrier.dstAccessMask = VK_ACCESS_SHADER_READ_BIT; + srcStage = VK_PIPELINE_STAGE_TRANSFER_BIT; + dstStage = VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT; + } + } + } + } + else if (oldLayout == VK_IMAGE_LAYOUT_UNDEFINED && newLayout == VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL) + { + imgBarrier.srcAccessMask = 0; + imgBarrier.dstAccessMask = VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_READ_BIT | VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_WRITE_BIT; + srcStage = VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT; + dstStage = VK_PIPELINE_STAGE_EARLY_FRAGMENT_TESTS_BIT; + } + else if (oldLayout == VK_IMAGE_LAYOUT_UNDEFINED && newLayout == VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL) + { + imgBarrier.srcAccessMask = 0; + imgBarrier.dstAccessMask = VK_ACCESS_COLOR_ATTACHMENT_READ_BIT | VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT; + srcStage = VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT; + dstStage = VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT; + } + else + { + assert(0 && !"Invalid image stage!"); + } + + vkCmdPipelineBarrier(*cmdBuffer, srcStage, dstStage, 0, 0, NULL, 0, NULL, 1, &imgBarrier); +} + +// internal helper +static void generateMipmaps(const VkCommandBuffer *cmdBuffer, const qvktexture_t *texture, uint32_t width, uint32_t height) +{ + int32_t mipWidth = width; + int32_t mipHeight = height; + VkFilter mipFilter = vk_mip_nearfilter->value > 0 ? VK_FILTER_NEAREST : VK_FILTER_LINEAR; + + VkImageMemoryBarrier imgBarrier = { + .sType = VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER, + .pNext = NULL, + .srcQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED, + .dstQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED, + .image = texture->resource.image, + .subresourceRange.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT, + .subresourceRange.levelCount = 1, + .subresourceRange.baseArrayLayer = 0, + .subresourceRange.layerCount = 1 + }; + + // copy rescaled mip data between consecutive levels (each higher level is half the size of the previous level) + for (uint32_t i = 1; i < texture->mipLevels; ++i) + { + imgBarrier.srcAccessMask = VK_ACCESS_TRANSFER_WRITE_BIT; + imgBarrier.dstAccessMask = VK_ACCESS_TRANSFER_READ_BIT; + imgBarrier.oldLayout = VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL; + imgBarrier.newLayout = VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL; + imgBarrier.subresourceRange.baseMipLevel = i - 1; + + vkCmdPipelineBarrier(*cmdBuffer, VK_PIPELINE_STAGE_TRANSFER_BIT, VK_PIPELINE_STAGE_TRANSFER_BIT, 0, 0, NULL, 0, NULL, 1, &imgBarrier); + + VkImageBlit blit = { + .srcSubresource.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT, + .srcSubresource.mipLevel = i - 1, + .srcSubresource.baseArrayLayer = 0, + .srcSubresource.layerCount = 1, + .srcOffsets[0] = { 0, 0, 0 }, + .srcOffsets[1] = { mipWidth, mipHeight, 1 }, + .dstSubresource.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT, + .dstSubresource.mipLevel = i, + .dstSubresource.baseArrayLayer = 0, + .dstSubresource.layerCount = 1, + .dstOffsets[0] = { 0, 0, 0 }, + .dstOffsets[1] = { mipWidth > 1 ? mipWidth >> 1 : 1, + mipHeight > 1 ? mipHeight >> 1 : 1, 1 } // each mip level is half the size of the previous level + }; + + // src image == dst image, because we're blitting between different mip levels of the same image + vkCmdBlitImage(*cmdBuffer, texture->resource.image, VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL, + texture->resource.image, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL, 1, &blit, mipFilter); + + imgBarrier.srcAccessMask = VK_ACCESS_TRANSFER_READ_BIT; + imgBarrier.dstAccessMask = VK_ACCESS_SHADER_READ_BIT; + imgBarrier.oldLayout = VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL; + imgBarrier.newLayout = VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL; + + vkCmdPipelineBarrier(*cmdBuffer, VK_PIPELINE_STAGE_TRANSFER_BIT, VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT, 0, 0, NULL, 0, NULL, 1, &imgBarrier); + + // avoid zero-sized mip levels + if (mipWidth > 1) mipWidth >>= 1; + if (mipHeight > 1) mipHeight >>= 1; + } + + imgBarrier.srcAccessMask = VK_ACCESS_TRANSFER_WRITE_BIT; + imgBarrier.dstAccessMask = VK_ACCESS_SHADER_READ_BIT; + imgBarrier.oldLayout = VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL; + imgBarrier.newLayout = VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL; + imgBarrier.subresourceRange.baseMipLevel = texture->mipLevels - 1; + + vkCmdPipelineBarrier(*cmdBuffer, VK_PIPELINE_STAGE_TRANSFER_BIT, VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT, 0, 0, NULL, 0, NULL, 1, &imgBarrier); +} + +// internal helper +static void createTextureImage(qvktexture_t *dstTex, const unsigned char *data, uint32_t width, uint32_t height) +{ + int unifiedTransferAndGfx = vk_device.transferQueue == vk_device.gfxQueue ? 1 : 0; + // assuming 32bit images + uint32_t imageSize = width * height * 4; + + VkBuffer staging_buffer; + VkCommandBuffer command_buffer; + uint32_t staging_offset; + void *imgData = QVk_GetStagingBuffer(imageSize, 4, &command_buffer, &staging_buffer, &staging_offset); + if (!imgData) + Sys_Error("%s: Staging buffers is smaller than image: %d.\n", __func__, imageSize); + + memcpy(imgData, data, (size_t)imageSize); + + VkImageUsageFlags imageUsage = VK_IMAGE_USAGE_TRANSFER_DST_BIT | VK_IMAGE_USAGE_SAMPLED_BIT; + // set extra image usage flag if we're dealing with mipmapped image - will need it for copying data between mip levels + if (dstTex->mipLevels > 1) + imageUsage |= VK_IMAGE_USAGE_TRANSFER_SRC_BIT; + + VK_VERIFY(QVk_CreateImage(width, height, dstTex->format, VK_IMAGE_TILING_OPTIMAL, imageUsage, dstTex)); + + transitionImageLayout(&command_buffer, &vk_device.transferQueue, dstTex, VK_IMAGE_LAYOUT_UNDEFINED, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL); + // copy buffer to image + VkBufferImageCopy region = { + .bufferOffset = staging_offset, + .bufferRowLength = 0, + .bufferImageHeight = 0, + .imageSubresource.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT, + .imageSubresource.mipLevel = 0, + .imageSubresource.baseArrayLayer = 0, + .imageSubresource.layerCount = 1, + .imageOffset = { 0, 0, 0 }, + .imageExtent = { width, height, 1 } + }; + + vkCmdCopyBufferToImage(command_buffer, staging_buffer, dstTex->resource.image, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL, 1, ®ion); + + if (dstTex->mipLevels > 1) + { + // vkCmdBlitImage requires a queue with GRAPHICS_BIT present + generateMipmaps(&command_buffer, dstTex, width, height); + } + else + { + // for non-unified transfer and graphics, this step begins queue ownership transfer to graphics queue (for exclusive sharing only) + if (unifiedTransferAndGfx || dstTex->sharingMode == VK_SHARING_MODE_EXCLUSIVE) + transitionImageLayout(&command_buffer, &vk_device.transferQueue, dstTex, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL, VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL); + + if (!unifiedTransferAndGfx) + { + transitionImageLayout(&command_buffer, &vk_device.gfxQueue, dstTex, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL, VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL); + } + } +} + +VkResult QVk_CreateImageView(const VkImage *image, VkImageAspectFlags aspectFlags, VkImageView *imageView, VkFormat format, uint32_t mipLevels) +{ + VkImageViewCreateInfo ivCreateInfo = { + .sType = VK_STRUCTURE_TYPE_IMAGE_VIEW_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .image = *image, + .viewType = VK_IMAGE_VIEW_TYPE_2D, + .format = format, + .components.r = VK_COMPONENT_SWIZZLE_IDENTITY, + .components.g = VK_COMPONENT_SWIZZLE_IDENTITY, + .components.b = VK_COMPONENT_SWIZZLE_IDENTITY, + .components.a = VK_COMPONENT_SWIZZLE_IDENTITY, + .subresourceRange.aspectMask = aspectFlags, + .subresourceRange.baseArrayLayer = 0, + .subresourceRange.baseMipLevel = 0, + .subresourceRange.layerCount = 1, + .subresourceRange.levelCount = mipLevels + }; + + return vkCreateImageView(vk_device.logical, &ivCreateInfo, NULL, imageView); +} + +VkResult QVk_CreateImage(uint32_t width, uint32_t height, VkFormat format, VkImageTiling tiling, VkImageUsageFlags usage, qvktexture_t *texture) +{ + VkImageCreateInfo imageInfo = { + .sType = VK_STRUCTURE_TYPE_IMAGE_CREATE_INFO, + .imageType = VK_IMAGE_TYPE_2D, + .extent.width = width, + .extent.height = height, + .extent.depth = 1, + .mipLevels = texture->mipLevels, + .arrayLayers = 1, + .format = format, + .tiling = tiling, + .initialLayout = VK_IMAGE_LAYOUT_UNDEFINED, + .usage = usage, + .sharingMode = VK_SHARING_MODE_EXCLUSIVE, + .samples = texture->sampleCount, + .flags = 0 + }; + + uint32_t queueFamilies[] = { (uint32_t)vk_device.gfxFamilyIndex, (uint32_t)vk_device.transferFamilyIndex }; + if (vk_device.gfxFamilyIndex != vk_device.transferFamilyIndex) + { + imageInfo.sharingMode = VK_SHARING_MODE_CONCURRENT; + imageInfo.queueFamilyIndexCount = 2; + imageInfo.pQueueFamilyIndices = queueFamilies; + } + + texture->sharingMode = imageInfo.sharingMode; + return image_create( + &texture->resource, imageInfo, + /*mem_properties*/ 0, + /*mem_preferences*/ VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT); +} + +void QVk_CreateDepthBuffer(VkSampleCountFlagBits sampleCount, qvktexture_t *depthBuffer) +{ + depthBuffer->format = QVk_FindDepthFormat(); + depthBuffer->sampleCount = sampleCount; + + VK_VERIFY(QVk_CreateImage(vk_swapchain.extent.width, vk_swapchain.extent.height, depthBuffer->format, VK_IMAGE_TILING_OPTIMAL, VK_IMAGE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT, depthBuffer)); + VK_VERIFY(QVk_CreateImageView(&depthBuffer->resource.image, getDepthStencilAspect(depthBuffer->format), &depthBuffer->imageView, depthBuffer->format, depthBuffer->mipLevels)); +} + +static void ChangeColorBufferLayout(VkImage image, VkImageLayout fromLayout, VkImageLayout toLayout) +{ + VkCommandBuffer commandBuffer = QVk_CreateCommandBuffer(&vk_transferCommandPool, VK_COMMAND_BUFFER_LEVEL_PRIMARY); + QVk_BeginCommand(&commandBuffer); + + const VkImageSubresourceRange subresourceRange = { + .aspectMask = VK_IMAGE_ASPECT_COLOR_BIT, + .baseMipLevel = 0u, + .levelCount = 1u, + .baseArrayLayer = 0u, + .layerCount = 1u, + }; + + const VkImageMemoryBarrier imageBarrier = { + .sType = VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER, + .pNext = NULL, + .srcAccessMask = 0u, + .dstAccessMask = 0u, + .oldLayout = fromLayout, + .newLayout = toLayout, + .srcQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED, + .dstQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED, + .image = image, + .subresourceRange = subresourceRange, + }; + + vkCmdPipelineBarrier(commandBuffer, VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT, VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT, 0u, 0u, NULL, 0u, NULL, 1u, &imageBarrier); + + QVk_SubmitCommand(&commandBuffer, &vk_device.transferQueue); + vkFreeCommandBuffers(vk_device.logical, vk_transferCommandPool, 1, &commandBuffer); +} + +void QVk_CreateColorBuffer(VkSampleCountFlagBits sampleCount, qvktexture_t *colorBuffer, int extraFlags) +{ + colorBuffer->format = vk_swapchain.format; + colorBuffer->sampleCount = sampleCount; + + VK_VERIFY(QVk_CreateImage(vk_swapchain.extent.width, vk_swapchain.extent.height, colorBuffer->format, VK_IMAGE_TILING_OPTIMAL, extraFlags | VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT, colorBuffer)); + VK_VERIFY(QVk_CreateImageView(&colorBuffer->resource.image, VK_IMAGE_ASPECT_COLOR_BIT, &colorBuffer->imageView, colorBuffer->format, colorBuffer->mipLevels)); + + // The single-sample color attachment ends every frame as + // SHADER_READ_ONLY_OPTIMAL as used in the post-processing stage, so its + // initial image layout is specified as such. + // + // The multi-sample color attachment is always COLOR_ATTACHMENT_OPTIMAL. + // + const VkImageLayout newLayout = ((sampleCount == VK_SAMPLE_COUNT_1_BIT) ? VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL : VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL); + ChangeColorBufferLayout(colorBuffer->resource.image, VK_IMAGE_LAYOUT_UNDEFINED, newLayout); +} + +void QVk_CreateTexture(qvktexture_t *texture, const unsigned char *data, uint32_t width, uint32_t height, qvksampler_t samplerType, qboolean clampToEdge) +{ + createTextureImage(texture, data, width, height); + VK_VERIFY(QVk_CreateImageView(&texture->resource.image, VK_IMAGE_ASPECT_COLOR_BIT, &texture->imageView, texture->format, texture->mipLevels)); + + // create descriptor set for the texture + VkDescriptorSetAllocateInfo dsAllocInfo = { + .sType = VK_STRUCTURE_TYPE_DESCRIPTOR_SET_ALLOCATE_INFO, + .pNext = NULL, + .descriptorPool = vk_descriptorPool, + .descriptorSetCount = 1, + .pSetLayouts = &vk_samplerDescSetLayout + }; + + VK_VERIFY(vkAllocateDescriptorSets(vk_device.logical, &dsAllocInfo, &texture->descriptorSet)); + + // attach sampler + QVk_UpdateTextureSampler(texture, samplerType, clampToEdge); +} + +void QVk_UpdateTextureData(qvktexture_t *texture, const unsigned char *data, uint32_t offset_x, uint32_t offset_y, uint32_t width, uint32_t height) +{ + int unifiedTransferAndGfx = vk_device.transferQueue == vk_device.gfxQueue ? 1 : 0; + // assuming 32bit images + uint32_t imageSize = width * height * 4; + + VkBuffer staging_buffer; + VkCommandBuffer command_buffer; + uint32_t staging_offset; + void *imgData = QVk_GetStagingBuffer(imageSize, 4, &command_buffer, &staging_buffer, &staging_offset); + if (!imgData) + Sys_Error("%s: Staging buffers is smaller than image: %d.\n", __func__, imageSize); + + memcpy(imgData, data, (size_t)imageSize); + + transitionImageLayout(&command_buffer, &vk_device.transferQueue, texture, VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL); + // copy buffer to image + VkBufferImageCopy region = { + .bufferOffset = staging_offset, + .bufferRowLength = 0, + .bufferImageHeight = 0, + .imageSubresource.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT, + .imageSubresource.mipLevel = 0, + .imageSubresource.baseArrayLayer = 0, + .imageSubresource.layerCount = 1, + .imageOffset = { offset_x, offset_y, 0 }, + .imageExtent = { width, height, 1 } + }; + + vkCmdCopyBufferToImage(command_buffer, staging_buffer, texture->resource.image, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL, 1, ®ion); + + if (texture->mipLevels > 1) + { + // vkCmdBlitImage requires a queue with GRAPHICS_BIT present + generateMipmaps(&command_buffer, texture, width, height); + } + else + { + // for non-unified transfer and graphics, this step begins queue ownership transfer to graphics queue (for exclusive sharing only) + if (unifiedTransferAndGfx || texture->sharingMode == VK_SHARING_MODE_EXCLUSIVE) + transitionImageLayout(&command_buffer, &vk_device.transferQueue, texture, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL, VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL); + + if (!unifiedTransferAndGfx) + { + transitionImageLayout(&command_buffer, &vk_device.gfxQueue, texture, VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL, VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL); + } + } +} + +static void QVk_ReleaseTexture(qvktexture_t *texture) +{ + QVk_SubmitStagingBuffers(); + if (vk_device.logical != VK_NULL_HANDLE) + { + vkDeviceWaitIdle(vk_device.logical); + } + + if (texture->imageView != VK_NULL_HANDLE) + { + vkDestroyImageView(vk_device.logical, texture->imageView, NULL); + texture->imageView = VK_NULL_HANDLE; + } + + if (texture->resource.image != VK_NULL_HANDLE) + image_destroy(&texture->resource); + + if (texture->descriptorSet != VK_NULL_HANDLE) + vkFreeDescriptorSets(vk_device.logical, vk_descriptorPool, 1, &texture->descriptorSet); + + texture->descriptorSet = VK_NULL_HANDLE; +} + +void QVk_ReadPixels(uint8_t *dstBuffer, const VkOffset2D *offset, const VkExtent2D *extent) +{ + BufferResource_t buff; + uint8_t *pMappedData; + VkCommandBuffer cmdBuffer; + extern int vk_activeBufferIdx; + + VkBufferCreateInfo bcInfo = { + .sType = VK_STRUCTURE_TYPE_BUFFER_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .size = extent->width * extent->height * 4, + .usage = VK_BUFFER_USAGE_TRANSFER_DST_BIT, + .sharingMode = VK_SHARING_MODE_EXCLUSIVE, + .queueFamilyIndexCount = 0, + .pQueueFamilyIndices = NULL, + }; + + VK_VERIFY(buffer_create(&buff, bcInfo, VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT | VK_MEMORY_PROPERTY_HOST_COHERENT_BIT, VK_MEMORY_PROPERTY_HOST_CACHED_BIT)); + + cmdBuffer = QVk_CreateCommandBuffer(&vk_commandPool[vk_activeBufferIdx], VK_COMMAND_BUFFER_LEVEL_PRIMARY); + VK_VERIFY(QVk_BeginCommand(&cmdBuffer)); + + // transition the current swapchain image to be a source of data transfer to our buffer + VkImageMemoryBarrier imgBarrier = { + .sType = VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER, + .pNext = NULL, + .srcAccessMask = VK_ACCESS_COLOR_ATTACHMENT_READ_BIT | VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT, + .dstAccessMask = VK_ACCESS_TRANSFER_READ_BIT, + .oldLayout = VK_IMAGE_LAYOUT_PRESENT_SRC_KHR, + .newLayout = VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL, + .srcQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED, + .dstQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED, + .image = vk_swapchain.images[vk_activeBufferIdx], + .subresourceRange.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT, + .subresourceRange.baseMipLevel = 0, + .subresourceRange.baseArrayLayer = 0, + .subresourceRange.layerCount = 1, + .subresourceRange.levelCount = 1 + }; + + vkCmdPipelineBarrier(cmdBuffer, VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT, VK_PIPELINE_STAGE_TRANSFER_BIT, 0, 0, NULL, 0, NULL, 1, &imgBarrier); + + VkBufferImageCopy region = { + .bufferOffset = 0, + .bufferRowLength = extent->width, + .bufferImageHeight = extent->height, + .imageSubresource.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT, + .imageSubresource.mipLevel = 0, + .imageSubresource.baseArrayLayer = 0, + .imageSubresource.layerCount = 1, + .imageOffset = { offset->x, offset->y, 0 }, + .imageExtent = { extent->width, extent->height, 1 } + }; + + // copy the swapchain image + vkCmdCopyImageToBuffer(cmdBuffer, vk_swapchain.images[vk_activeBufferIdx], VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL, buff.buffer, 1, ®ion); + VK_VERIFY(vkDeviceWaitIdle(vk_device.logical)); + QVk_SubmitCommand(&cmdBuffer, &vk_device.gfxQueue); + + // store image in destination buffer + pMappedData = buffer_map(&buff); + memcpy(dstBuffer, pMappedData, extent->width * extent->height * 4); + buffer_unmap(&buff); + + buffer_destroy(&buff); +} + +/* +=============== +Vk_ImageList_f +=============== +*/ +void Vk_ImageList_f (void) +{ + int i, used, texels; + image_t *image; + qboolean freeup; + + R_Printf(PRINT_ALL, "------------------\n"); + texels = 0; + used = 0; + + for (i = 0, image = vktextures; i < numvktextures; i++, image++) + { + char *in_use = ""; + + if (image->vk_texture.resource.image == VK_NULL_HANDLE) + continue; + + if (image->registration_sequence == registration_sequence) + { + in_use = "*"; + used++; + } + + texels += image->upload_width*image->upload_height; + switch (image->type) + { + case it_skin: + R_Printf(PRINT_ALL, "M"); + break; + case it_sprite: + R_Printf(PRINT_ALL, "S"); + break; + case it_wall: + R_Printf(PRINT_ALL, "W"); + break; + case it_pic: + R_Printf(PRINT_ALL, "P"); + break; + default: + R_Printf(PRINT_ALL, " "); + break; + } + + R_Printf(PRINT_ALL, " %4i %4i RGB: %s (%dx%d) %s\n", + image->upload_width, image->upload_height, image->name, + image->width, image->height, in_use); + } + R_Printf(PRINT_ALL, "Total texel count (not counting mipmaps): %i in %d images\n", texels, img_loaded); + freeup = Vk_ImageHasFreeSpace(); + R_Printf(PRINT_ALL, "Used %d of %d images%s.\n", used, image_max, freeup ? ", has free space" : ""); +} + +typedef struct +{ + char *name; + qvksampler_t samplerType; +} vkmode_t; + +static vkmode_t modes[] = { + {"VK_NEAREST", S_NEAREST }, + {"VK_LINEAR", S_LINEAR }, + {"VK_MIPMAP_NEAREST", S_MIPMAP_NEAREST }, + {"VK_MIPMAP_LINEAR", S_MIPMAP_LINEAR } +}; + +#define NUM_VK_MODES (sizeof(modes) / sizeof (vkmode_t)) + +/* + =============== + Vk_TextureMode + =============== + */ +void Vk_TextureMode( char *string ) +{ + int i,j; + image_t *image; + static char prev_mode[32] = { "VK_MIPMAP_LINEAR" }; + + for (i = 0; i < NUM_VK_MODES; i++) + { + if (!Q_stricmp(modes[i].name, string)) + break; + } + + if (i == NUM_VK_MODES) + { + R_Printf(PRINT_ALL, "bad filter name (valid values: VK_NEAREST, VK_LINEAR, VK_MIPMAP_NEAREST, VK_MIPMAP_LINEAR)\n"); + ri.Cvar_Set("vk_texturemode", prev_mode); + return; + } + + memcpy(prev_mode, string, strlen(string)); + prev_mode[strlen(string)] = '\0'; + + vk_current_sampler = i; + + vkDeviceWaitIdle(vk_device.logical); + for (j = 0, image = vktextures; j < numvktextures; j++, image++) + { + // skip console characters - we want them unfiltered at all times + if (image->vk_texture.resource.image != VK_NULL_HANDLE && Q_stricmp(image->name, "pics/conchars.pcx")) + QVk_UpdateTextureSampler(&image->vk_texture, i, image->vk_texture.clampToEdge); + } + + if (vk_rawTexture.resource.image != VK_NULL_HANDLE) + QVk_UpdateTextureSampler(&vk_rawTexture, i, vk_rawTexture.clampToEdge); +} + +/* + =============== + Vk_LmapTextureMode + =============== + */ +void Vk_LmapTextureMode( char *string ) +{ + int i,j; + static char prev_mode[32] = { "VK_MIPMAP_LINEAR" }; + + for (i = 0; i < NUM_VK_MODES; i++) + { + if (!Q_stricmp(modes[i].name, string)) + break; + } + + if (i == NUM_VK_MODES) + { + R_Printf(PRINT_ALL, "bad filter name (valid values: VK_NEAREST, VK_LINEAR, VK_MIPMAP_NEAREST, VK_MIPMAP_LINEAR)\n"); + ri.Cvar_Set("vk_lmaptexturemode", prev_mode); + return; + } + + memcpy(prev_mode, string, strlen(string)); + prev_mode[strlen(string)] = '\0'; + + vk_current_lmap_sampler = i; + + vkDeviceWaitIdle(vk_device.logical); + for (j = 0; j < MAX_LIGHTMAPS*2; j++) + { + if (vk_state.lightmap_textures[j].resource.image != VK_NULL_HANDLE) + QVk_UpdateTextureSampler(&vk_state.lightmap_textures[j], i, vk_state.lightmap_textures[j].clampToEdge); + } +} + +/* +==================================================================== + +IMAGE FLOOD FILLING + +==================================================================== +*/ + + +/* +================= +Mod_FloodFillSkin + +Fill background pixels so mipmapping doesn't have haloes +================= +*/ + +typedef struct +{ + short x, y; +} floodfill_t; + +/* must be a power of 2 */ +#define FLOODFILL_FIFO_SIZE 0x1000 +#define FLOODFILL_FIFO_MASK (FLOODFILL_FIFO_SIZE - 1) + +#define FLOODFILL_STEP(off, dx, dy) \ + { \ + if (pos[off] == fillcolor) \ + { \ + pos[off] = 255; \ + fifo[inpt].x = x + (dx), fifo[inpt].y = y + (dy); \ + inpt = (inpt + 1) & FLOODFILL_FIFO_MASK; \ + } \ + else if (pos[off] != 255) \ + { \ + fdc = pos[off]; \ + } \ + } + +/* + * Fill background pixels so mipmapping doesn't have haloes + */ +static void +FloodFillSkin(byte *skin, int skinwidth, int skinheight) +{ + byte fillcolor = *skin; /* assume this is the pixel to fill */ + floodfill_t fifo[FLOODFILL_FIFO_SIZE]; + int inpt = 0, outpt = 0; + int filledcolor = 0; + int i; + + /* attempt to find opaque black */ + for (i = 0; i < 256; ++i) + { + if (LittleLong(d_8to24table[i]) == (255 << 0)) /* alpha 1.0 */ + { + filledcolor = i; + break; + } + } + + /* can't fill to filled color or to transparent color (used as visited marker) */ + if ((fillcolor == filledcolor) || (fillcolor == 255)) + { + return; + } + + fifo[inpt].x = 0, fifo[inpt].y = 0; + inpt = (inpt + 1) & FLOODFILL_FIFO_MASK; + + while (outpt != inpt) + { + int x = fifo[outpt].x, y = fifo[outpt].y; + int fdc = filledcolor; + byte *pos = &skin[x + skinwidth * y]; + + outpt = (outpt + 1) & FLOODFILL_FIFO_MASK; + + if (x > 0) + { + FLOODFILL_STEP(-1, -1, 0); + } + + if (x < skinwidth - 1) + { + FLOODFILL_STEP(1, 1, 0); + } + + if (y > 0) + { + FLOODFILL_STEP(-skinwidth, 0, -1); + } + + if (y < skinheight - 1) + { + FLOODFILL_STEP(skinwidth, 0, 1); + } + + skin[x + skinwidth * y] = fdc; + } +} + + +/* +================ +Vk_LightScaleTexture + +Scale up the pixel values in a texture to increase the +lighting range +================ +*/ +static void Vk_LightScaleTexture (byte *in, int inwidth, int inheight) +{ + int i, c; + byte *p; + + p = (byte *)in; + + c = inwidth*inheight; + for (i=0 ; i>= (int)vk_picmip->value; + scaled_height >>= (int)vk_picmip->value; + } + + if (scaled_width < 1) + scaled_width = 1; + if (scaled_height < 1) + scaled_height = 1; + + *texBuffer = malloc(scaled_width * scaled_height * 4); + if (!*texBuffer) + ri.Sys_Error(ERR_DROP, "%s: too big", __func__); + + *upload_width = scaled_width; + *upload_height = scaled_height; + + if (scaled_width == width && scaled_height == height) + { + memcpy(*texBuffer, data, scaled_width * scaled_height * 4); + } + else + { + ResizeSTB(data, width, height, + *texBuffer, scaled_width, scaled_height); + } + + // world textures + if (type != it_pic && type != it_sky) + { + Vk_LightScaleTexture(*texBuffer, scaled_width, scaled_height); + } + + while (scaled_width > 1 || scaled_height > 1) + { + scaled_width >>= 1; + scaled_height >>= 1; + if (scaled_width < 1) + scaled_width = 1; + if (scaled_height < 1) + scaled_height = 1; + miplevel++; + } + + return miplevel; +} + +/* +=============== +Vk_Upload8 + +Returns number of mip levels +=============== +*/ + +static uint32_t Vk_Upload8 (byte *data, int width, int height, imagetype_t type, + byte **texBuffer, int *upload_width, int *upload_height) +{ + unsigned *trans; + int i, s; + int miplevel; + + s = width * height; + + trans = malloc(s * sizeof(*trans)); + if (!trans) + ri.Sys_Error(ERR_DROP, "%s: too large", __func__); + + for (i = 0; i < s; i++) + { + int p; + + p = data[i]; + trans[i] = d_8to24table[p]; + } + + if (type != it_sky && type != it_wall) + { + for (i = 0; i < s; i++) + { + int p; + + p = data[i]; + + if (p == 255) + { // transparent, so scan around for another color + // to avoid alpha fringes + // FIXME: do a full flood fill so mips work... + if (i > width && data[i - width] != 255) + p = data[i - width]; + else if (i < s - width && data[i + width] != 255) + p = data[i + width]; + else if (i > 0 && data[i - 1] != 255) + p = data[i - 1]; + else if (i < s - 1 && data[i + 1] != 255) + p = data[i + 1]; + else + p = 0; + // copy rgb components + ((byte *)&trans[i])[0] = ((byte *)&d_8to24table[p])[0]; + ((byte *)&trans[i])[1] = ((byte *)&d_8to24table[p])[1]; + ((byte *)&trans[i])[2] = ((byte *)&d_8to24table[p])[2]; + } + } + } + + // optimize 8bit images only when we forced such logic + if (vk_retexturing->value >= 2) + { + SmoothColorImage(trans, s, s >> 7); + } + + miplevel = Vk_Upload32((byte *)trans, width, height, type, texBuffer, upload_width, upload_height); + free(trans); + return miplevel; +} + + +/* +================ +Vk_LoadPic + +This is also used as an entry point for the generated r_notexture +================ +*/ +image_t * +Vk_LoadPic(char *name, byte *pic, int width, int realwidth, + int height, int realheight, imagetype_t type, + int bits) +{ + image_t *image; + byte *texBuffer; + int upload_width, upload_height; + + qboolean nolerp = false; + + if (vk_nolerp_list != NULL && vk_nolerp_list->string != NULL) + { + nolerp = strstr(vk_nolerp_list->string, name) != NULL; + } + + { + int i; + // find a free image_t + for (i = 0, image = vktextures; ivk_texture.resource.image == VK_NULL_HANDLE) + break; + } + if (i == numvktextures) + { + if (numvktextures == MAX_VKTEXTURES) + ri.Sys_Error(ERR_DROP, "%s: MAX_VKTEXTURES", __func__); + numvktextures++; + } + image = &vktextures[i]; + } + + if (strlen(name) >= sizeof(image->name)) + ri.Sys_Error(ERR_DROP, "%s: \"%s\" is too long", __func__, name); + strcpy(image->name, name); + image->registration_sequence = registration_sequence; + // zero-clear Vulkan texture handle + QVVKTEXTURE_CLEAR(image->vk_texture); + image->width = realwidth; + image->height = realheight; + image->type = type; + // update count of loaded images + img_loaded ++; + if (vk_validation->value) + { + R_Printf(PRINT_ALL, "%s: Load %s[%d]\n", __func__, image->name, img_loaded); + } + + if (type == it_skin && bits == 8) + FloodFillSkin(pic, width, height); + + if (bits == 8) + { + // resize 8bit images only when we forced such logic + if (vk_retexturing->value >= 2) + { + byte *image_converted = malloc(width * height * 4); + scale2x(pic, image_converted, width, height); + image->vk_texture.mipLevels = Vk_Upload8(image_converted, width * 2, height * 2, image->type, &texBuffer, &upload_width, &upload_height); + free(image_converted); + } + else + { + image->vk_texture.mipLevels = Vk_Upload8(pic, width, height, image->type, &texBuffer, &upload_width, &upload_height); + } + } + else + image->vk_texture.mipLevels = Vk_Upload32(pic, width, height, image->type, &texBuffer, &upload_width, &upload_height); + + image->upload_width = upload_width; // after power of 2 and scales + image->upload_height = upload_height; + + QVk_CreateTexture(&image->vk_texture, (unsigned char*)texBuffer, + image->upload_width, image->upload_height, + nolerp ? S_NEAREST : vk_current_sampler, (type == it_sky)); + QVk_DebugSetObjectName((uint64_t)image->vk_texture.resource.image, + VK_OBJECT_TYPE_IMAGE, va("Image: %s", name)); + QVk_DebugSetObjectName((uint64_t)image->vk_texture.imageView, + VK_OBJECT_TYPE_IMAGE_VIEW, va("Image View: %s", name)); + QVk_DebugSetObjectName((uint64_t)image->vk_texture.descriptorSet, + VK_OBJECT_TYPE_DESCRIPTOR_SET, va("Descriptor Set: %s", name)); + QVk_DebugSetObjectName((uint64_t)image->vk_texture.resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, "Memory: game textures"); + + if (texBuffer) + { + free(texBuffer); + } + return image; +} + + +/* +================ +Vk_LoadWal +================ +*/ +static image_t *Vk_LoadWal (char *name, imagetype_t type) +{ + miptex_t *mt; + int width, height, ofs; + image_t *image; + + ri.FS_LoadFile (name, (void **)&mt); + if (!mt) + { + R_Printf(PRINT_ALL, "%s: can't load %s\n", __func__, name); + return r_notexture; + } + + width = LittleLong (mt->width); + height = LittleLong (mt->height); + ofs = LittleLong (mt->offsets[0]); + + image = Vk_LoadPic(name, (byte *)mt + ofs, + width, width, + height, height, + type, 8); + + ri.FS_FreeFile ((void *)mt); + + return image; +} + +static image_t * +Vk_LoadM8(char *origname, imagetype_t type) +{ + m8tex_t *mt; + int width, height, ofs, size; + image_t *image; + char name[256]; + unsigned char *image_buffer = NULL; + + Q_strlcpy(name, origname, sizeof(name)); + + /* Add the extension */ + if (strcmp(COM_FileExtension(name), "m8")) + { + Q_strlcat(name, ".m8", sizeof(name)); + } + + size = ri.FS_LoadFile(name, (void **)&mt); + + if (!mt) + { + R_Printf(PRINT_ALL, "%s: can't load %s\n", __func__, name); + return r_notexture; + } + + if (size < sizeof(m8tex_t)) + { + R_Printf(PRINT_ALL, "%s: can't load %s, small header\n", __func__, name); + ri.FS_FreeFile((void *)mt); + return r_notexture; + } + + if (LittleLong (mt->version) != M8_VERSION) + { + R_Printf(PRINT_ALL, "LoadWal: can't load %s, wrong magic value.\n", name); + ri.FS_FreeFile ((void *)mt); + return r_notexture; + } + + width = LittleLong(mt->width[0]); + height = LittleLong(mt->height[0]); + ofs = LittleLong(mt->offsets[0]); + + if ((ofs <= 0) || (width <= 0) || (height <= 0) || + (((size - ofs) / height) < width)) + { + R_Printf(PRINT_ALL, "%s: can't load %s, small body\n", __func__, name); + ri.FS_FreeFile((void *)mt); + return r_notexture; + } + + image_buffer = malloc (width * height * 4); + for(int i=0; ipalette[value].r; + image_buffer[i * 4 + 1] = mt->palette[value].g; + image_buffer[i * 4 + 2] = mt->palette[value].b; + image_buffer[i * 4 + 3] = value == 255 ? 0 : 255; + } + + image = Vk_LoadPic(name, image_buffer, width, width, height, height, type, 32); + free(image_buffer); + + ri.FS_FreeFile((void *)mt); + + return image; +} + +static image_t* +Vk_LoadHiColorImage(char *name, const char* namewe, const char *ext, imagetype_t type) +{ + image_t *image = NULL; + byte *pic = NULL; + int realwidth = 0, realheight = 0; + int width = 0, height = 0; + + if (strcmp(ext, "pcx") == 0) + { + /* Get size of the original texture */ + GetPCXInfo(name, &realwidth, &realheight); + } + else if (strcmp(ext, "wal") == 0) + { + /* Get size of the original texture */ + GetWalInfo(name, &realwidth, &realheight); + } + else if (strcmp(ext, "m8") == 0) + { + /* Get size of the original texture */ + GetM8Info(name, &realwidth, &realheight); + } + + /* try to load a tga, png or jpg (in that order/priority) */ + if ( LoadSTB(namewe, "tga", &pic, &width, &height) + || LoadSTB(namewe, "png", &pic, &width, &height) + || LoadSTB(namewe, "jpg", &pic, &width, &height) ) + { + if (width >= realwidth && height >= realheight) + { + if (realheight == 0 || realwidth == 0) + { + realheight = height; + realwidth = width; + } + + image = Vk_LoadPic(name, pic, + width, realwidth, + height, realheight, + type, 32); + } + } + + if (pic) + { + free(pic); + } + + return image; +} + +static image_t* +Vk_LoadImage(char *name, const char* namewe, const char *ext, imagetype_t type) +{ + image_t *image = NULL; + + // with retexturing + if (vk_retexturing->value) + { + image = Vk_LoadHiColorImage(name, namewe, ext, type); + } + + if (!image) + { + byte *pic; + int width, height; + + // + // load the pic from disk + // + pic = NULL; + if (!strcmp(ext, "pcx")) + { + byte *palette; + palette = NULL; + + LoadPCX (name, &pic, &palette, &width, &height); + if (!pic) + return NULL; + image = Vk_LoadPic(name, pic, + width, width, + height, height, + type, 8); + + if (palette) + free(palette); + } + else if (!strcmp(ext, "wal")) + { + image = Vk_LoadWal (name, type); + } + else if (!strcmp(ext, "m8")) + { + image = Vk_LoadM8 (name, type); + } + else if (!strcmp(ext, "tga")) + { + if (!LoadSTB (namewe, "tga", &pic, &width, &height)) + return NULL; + if (!pic) + return NULL; + image = Vk_LoadPic(name, pic, + width, width, + height, height, + type, 32); + } + + if (pic) + free(pic); + } + + return image; +} + +/* +=============== +Vk_FindImage + +Finds or loads the given image +=============== +*/ +image_t *Vk_FindImage (char *name, imagetype_t type) +{ + image_t *image; + int i, len; + char *ptr; + char namewe[256]; + const char* ext; + + if (!name) + { + return NULL; + } + + ext = COM_FileExtension(name); + if(!ext[0]) + { + /* file has no extension */ + return NULL; + } + + len = strlen(name); + + /* Remove the extension */ + memset(namewe, 0, 256); + memcpy(namewe, name, len - (strlen(ext) + 1)); + + if (len < 5) + { + return NULL; + } + + /* fix backslashes */ + while ((ptr = strchr(name, '\\'))) + { + *ptr = '/'; + } + + // look for it + for (i=0, image=vktextures ; iname)) + { + image->registration_sequence = registration_sequence; + return image; + } + } + + // + // load the pic from disk + // + return Vk_LoadImage(name, namewe, ext, type); +} + + +/* +=============== +RE_RegisterSkin +=============== +*/ +struct image_s *RE_RegisterSkin (char *name) +{ + return Vk_FindImage (name, it_skin); +} + +qboolean Vk_ImageHasFreeSpace(void) +{ + int i, used; + image_t *image; + + used = 0; + + for (i = 0, image = vktextures; i < numvktextures; i++, image++) + { + if (!image->name[0]) + continue; + if (image->registration_sequence == registration_sequence) + { + used ++; + } + } + + if (image_max < used) + { + image_max = used; + } + + // should same size of free slots as currently used + return (img_loaded + used) < MAX_VKTEXTURES; +} + +/* +================ +Vk_FreeUnusedImages + +Any image that was not touched on this registration sequence +will be freed. +================ +*/ +void Vk_FreeUnusedImages (void) +{ + int i; + image_t *image; + + if (Vk_ImageHasFreeSpace()) + { + // should be enough space for load next images + return; + } + + // never free r_notexture or particle texture + r_notexture->registration_sequence = registration_sequence; + r_particletexture->registration_sequence = registration_sequence; + r_squaretexture->registration_sequence = registration_sequence; + + for (i = 0, image = vktextures; i < numvktextures; i++, image++) + { + if (image->registration_sequence == registration_sequence) + continue; // used this sequence + if (!image->registration_sequence) + continue; // free image_t slot + if (image->type == it_pic) + continue; // don't free pics + + if (vk_validation->value) + { + R_Printf(PRINT_ALL, "%s: Unload %s[%d]\n", __func__, image->name, img_loaded); + } + + // free it + QVk_ReleaseTexture(&image->vk_texture); + memset(image, 0, sizeof(*image)); + + img_loaded --; + if (img_loaded < 0) + { + ri.Sys_Error (ERR_DROP, "%s: Broken unload", __func__); + } + } + + // free all unused blocks + vulkan_memory_free_unused(); +} + + +/* +=============== +Draw_GetPalette +=============== +*/ +static int Draw_GetPalette (void) +{ + int i; + byte *pic, *pal; + int width, height; + + // get the palette + + LoadPCX ("pics/colormap.pcx", &pic, &pal, &width, &height); + if (!pal) + ri.Sys_Error (ERR_FATAL, "Couldn't load pics/colormap.pcx"); + + for (i=0 ; i<256 ; i++) + { + unsigned v; + int r, g, b; + + r = pal[i*3+0]; + g = pal[i*3+1]; + b = pal[i*3+2]; + + v = (255<<24) + (r<<0) + (g<<8) + (b<<16); + d_8to24table[i] = LittleLong(v); + } + + d_8to24table[255] &= LittleLong(0xffffff); // 255 is transparent + + free (pic); + free (pal); + + return 0; +} + + +/* +=============== +Vk_InitImages +=============== +*/ +void Vk_InitImages (void) +{ + int i; + float overbright; + + numvktextures = 0; + img_loaded = 0; + image_max = 0; + registration_sequence = 1; + + // init intensity conversions + intensity = ri.Cvar_Get("vk_intensity", "2", 0); + + if (intensity->value <= 1) + ri.Cvar_Set("vk_intensity", "1"); + + vk_state.inverse_intensity = 1 / intensity->value; + + for (i = 0; i<256; i++) + { + int j; + + j = i * intensity->value; + if (j > 255) + j = 255; + intensitytable[i] = j; + } + + Draw_GetPalette(); + + overbright = vk_overbrightbits->value; + + if(overbright < 0.5) + overbright = 0.5; + + if(overbright > 4.0) + overbright = 4.0; + + for (i=0 ; i<256 ; i++) { + int inf; + + inf = i * overbright; + + if (inf < 0) + inf = 0; + if (inf > 255) + inf = 255; + + overbrightable[i] = inf; + } +} + +/* +=============== +Vk_ShutdownImages +=============== +*/ +void Vk_ShutdownImages (void) +{ + int i; + image_t *image; + + for (i = 0, image = vktextures; iregistration_sequence) + continue; // free image_t slot + + if (vk_validation->value) + { + R_Printf(PRINT_ALL, "%s: Unload %s[%d]\n", __func__, image->name, img_loaded); + } + + QVk_ReleaseTexture(&image->vk_texture); + memset(image, 0, sizeof(*image)); + + img_loaded --; + if (img_loaded < 0) + { + ri.Sys_Error (ERR_DROP, "%s: Broken unload", __func__); + } + } + + QVk_ReleaseTexture(&vk_rawTexture); + + for(i = 0; i < MAX_LIGHTMAPS*2; i++) + QVk_ReleaseTexture(&vk_state.lightmap_textures[i]); +} + diff --git a/src/vk/vk_light.c b/src/vk/vk_light.c new file mode 100644 index 0000000..1a17929 --- /dev/null +++ b/src/vk/vk_light.c @@ -0,0 +1,626 @@ +/* +Copyright (C) 1997-2001 Id Software, Inc. +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +*/ +// vk_light.c + +#include "header/local.h" + +static int r_dlightframecount; + +#define DLIGHT_CUTOFF 64 + +/* +============================================================================= + +DYNAMIC LIGHTS BLEND RENDERING + +============================================================================= +*/ + +void R_RenderDlight (dlight_t *light) +{ + int i, j; + float rad; + + rad = light->intensity * 0.35; + + struct { + vec3_t verts; + float color[3]; + } lightVerts[18]; + + for (i = 0; i < 3; i++) + lightVerts[0].verts[i] = light->origin[i] - vpn[i] * rad; + + lightVerts[0].color[0] = light->color[0] * 0.2; + lightVerts[0].color[1] = light->color[1] * 0.2; + lightVerts[0].color[2] = light->color[2] * 0.2; + + for (i = 16; i >= 0; i--) + { + float a; + + a = i / 16.0 * M_PI * 2; + for (j = 0; j < 3; j++) + { + lightVerts[i+1].verts[j] = light->origin[j] + vright[j] * cos(a)*rad + + vup[j] * sin(a)*rad; + lightVerts[i+1].color[j] = 0.f; + } + } + + QVk_BindPipeline(&vk_drawDLightPipeline); + + VkBuffer vbo; + VkDeviceSize vboOffset; + uint32_t uboOffset; + VkDescriptorSet uboDescriptorSet; + uint8_t *vertData = QVk_GetVertexBuffer(sizeof(lightVerts), &vbo, &vboOffset); + uint8_t *uboData = QVk_GetUniformBuffer(sizeof(r_viewproj_matrix), &uboOffset, &uboDescriptorSet); + memcpy(vertData, lightVerts, sizeof(lightVerts)); + memcpy(uboData, r_viewproj_matrix, sizeof(r_viewproj_matrix)); + + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_drawDLightPipeline.layout, 0, 1, &uboDescriptorSet, 1, &uboOffset); + vkCmdBindIndexBuffer(vk_activeCmdbuffer, QVk_GetTriangleFanIbo(48), 0, VK_INDEX_TYPE_UINT16); + vkCmdDrawIndexed(vk_activeCmdbuffer, 48, 1, 0, 0, 0); +} + +/* +============= +R_RenderDlights +============= +*/ +void R_RenderDlights (void) +{ + int i; + dlight_t *l; + + if (!vk_flashblend->value) + return; + + r_dlightframecount = r_framecount + 1; // because the count hasn't + // advanced yet for this frame + l = r_newrefdef.dlights; + for (i=0 ; icontents != -1) + return; + + splitplane = node->plane; + dist = DotProduct (light->origin, splitplane->normal) - splitplane->dist; + + if (dist > light->intensity-DLIGHT_CUTOFF) + { + R_MarkLights (light, bit, node->children[0]); + return; + } + if (dist < -light->intensity+DLIGHT_CUTOFF) + { + R_MarkLights (light, bit, node->children[1]); + return; + } + + // mark the polygons + surf = r_worldmodel->surfaces + node->firstsurface; + for (i=0 ; inumsurfaces ; i++, surf++) + { + if (surf->dlightframe != r_dlightframecount) + { + surf->dlightbits = 0; + surf->dlightframe = r_dlightframecount; + } + surf->dlightbits |= bit; + } + + R_MarkLights (light, bit, node->children[0]); + R_MarkLights (light, bit, node->children[1]); +} + + +/* +============= +R_PushDlights +============= +*/ +void R_PushDlights (void) +{ + int i; + dlight_t *l; + + if (vk_flashblend->value) + return; + + r_dlightframecount = r_framecount + 1; // because the count hasn't + // advanced yet for this frame + l = r_newrefdef.dlights; + for (i=0 ; inodes ); +} + + +/* +============================================================================= + +LIGHT SAMPLING + +============================================================================= +*/ + +vec3_t pointcolor; +cplane_t *lightplane; // used as shadow plane +vec3_t lightspot; + +static int RecursiveLightPoint (mnode_t *node, vec3_t start, vec3_t end) +{ + float front, back, frac; + int side; + cplane_t *plane; + vec3_t mid; + msurface_t *surf; + int s, t, ds, dt; + int i; + mtexinfo_t *tex; + byte *lightmap; + int maps; + int r; + + if (node->contents != -1) + return -1; // didn't hit anything + +// calculate mid point + +// FIXME: optimize for axial + plane = node->plane; + front = DotProduct (start, plane->normal) - plane->dist; + back = DotProduct (end, plane->normal) - plane->dist; + side = front < 0; + + if ( (back < 0) == side) + return RecursiveLightPoint (node->children[side], start, end); + + frac = front / (front-back); + mid[0] = start[0] + (end[0] - start[0])*frac; + mid[1] = start[1] + (end[1] - start[1])*frac; + mid[2] = start[2] + (end[2] - start[2])*frac; + + // go down front side + r = RecursiveLightPoint (node->children[side], start, mid); + if (r >= 0) + return r; // hit something + + // check for impact on this node + VectorCopy (mid, lightspot); + lightplane = plane; + + surf = r_worldmodel->surfaces + node->firstsurface; + for (i=0 ; inumsurfaces ; i++, surf++) + { + vec3_t scale; + + if (surf->flags&(SURF_DRAWTURB|SURF_DRAWSKY)) + continue; // no lightmaps + + tex = surf->texinfo; + + s = DotProduct (mid, tex->vecs[0]) + tex->vecs[0][3]; + t = DotProduct (mid, tex->vecs[1]) + tex->vecs[1][3];; + + if (s < surf->texturemins[0] || + t < surf->texturemins[1]) + continue; + + ds = s - surf->texturemins[0]; + dt = t - surf->texturemins[1]; + + if ( ds > surf->extents[0] || dt > surf->extents[1] ) + continue; + + if (!surf->samples) + return 0; + + ds >>= 4; + dt >>= 4; + + lightmap = surf->samples; + VectorCopy (vec3_origin, pointcolor); + + lightmap += 3*(dt * ((surf->extents[0]>>4)+1) + ds); + + for (maps = 0 ; maps < MAXLIGHTMAPS && surf->styles[maps] != 255 ; + maps++) + { + for (int j=0 ; j<3 ; j++) + scale[j] = r_modulate->value*r_newrefdef.lightstyles[surf->styles[maps]].rgb[j]; + + pointcolor[0] += lightmap[0] * scale[0] * (1.0/255); + pointcolor[1] += lightmap[1] * scale[1] * (1.0/255); + pointcolor[2] += lightmap[2] * scale[2] * (1.0/255); + lightmap += 3*((surf->extents[0]>>4)+1) * + ((surf->extents[1]>>4)+1); + } + + return 1; + } + +// go down back side + return RecursiveLightPoint (node->children[!side], mid, end); +} + +/* +=============== +R_LightPoint +=============== +*/ +void R_LightPoint (vec3_t p, vec3_t color, entity_t *currententity) +{ + vec3_t end; + float r; + int lnum; + dlight_t *dl; + vec3_t dist; + + if (!r_worldmodel->lightdata || !currententity) + { + color[0] = color[1] = color[2] = 1.0; + return; + } + + end[0] = p[0]; + end[1] = p[1]; + end[2] = p[2] - 2048; + + r = RecursiveLightPoint (r_worldmodel->nodes, p, end); + + if (r == -1) + { + VectorCopy (vec3_origin, color); + } + else + { + VectorCopy (pointcolor, color); + } + + // + // add dynamic lights + // + dl = r_newrefdef.dlights; + for (lnum=0 ; lnumorigin, + dl->origin, + dist); + add = dl->intensity - VectorLength(dist); + add *= (1.0/256); + if (add > 0) + { + VectorMA (color, add, dl->color, color); + } + } + + VectorScale (color, r_modulate->value, color); +} + + +//=================================================================== + +static float s_blocklights[34*34*3]; +/* +=============== +R_AddDynamicLights +=============== +*/ +static void R_AddDynamicLights (msurface_t *surf) +{ + int lnum; + int sd, td; + float fdist, frad, fminlight; + vec3_t impact, local; + int s, t; + int i; + int smax, tmax; + mtexinfo_t *tex; + dlight_t *dl; + float *pfBL; + float fsacc, ftacc; + + smax = (surf->extents[0]>>4)+1; + tmax = (surf->extents[1]>>4)+1; + tex = surf->texinfo; + + for (lnum=0 ; lnumdlightbits & (1<intensity; + fdist = DotProduct (dl->origin, surf->plane->normal) - + surf->plane->dist; + frad -= fabs(fdist); + // rad is now the highest intensity on the plane + + fminlight = DLIGHT_CUTOFF; // FIXME: make configurable? + if (frad < fminlight) + continue; + fminlight = frad - fminlight; + + for (i=0 ; i<3 ; i++) + { + impact[i] = dl->origin[i] - + surf->plane->normal[i]*fdist; + } + + local[0] = DotProduct (impact, tex->vecs[0]) + tex->vecs[0][3] - surf->texturemins[0]; + local[1] = DotProduct (impact, tex->vecs[1]) + tex->vecs[1][3] - surf->texturemins[1]; + + pfBL = s_blocklights; + for (t = 0, ftacc = 0 ; t td) + fdist = sd + (td>>1); + else + fdist = td + (sd>>1); + + if ( fdist < fminlight ) + { + pfBL[0] += ( frad - fdist ) * dl->color[0]; + pfBL[1] += ( frad - fdist ) * dl->color[1]; + pfBL[2] += ( frad - fdist ) * dl->color[2]; + } + } + } + } +} + + +/* +** R_SetCacheState +*/ +void R_SetCacheState( msurface_t *surf ) +{ + int maps; + + for (maps = 0 ; maps < MAXLIGHTMAPS && surf->styles[maps] != 255 ; + maps++) + { + surf->cached_light[maps] = r_newrefdef.lightstyles[surf->styles[maps]].white; + } +} + +/* +=============== +R_BuildLightMap + +Combine and scale multiple lightmaps into the floating format in blocklights +=============== +*/ +void R_BuildLightMap (msurface_t *surf, byte *dest, int stride) +{ + int smax, tmax; + int r, g, b, a, max; + int i, j, size; + byte *lightmap; + float scale[4]; + int nummaps; + float *bl; + + if ( surf->texinfo->flags & (SURF_SKY|SURF_TRANS33|SURF_TRANS66|SURF_WARP) ) + ri.Sys_Error (ERR_DROP, "R_BuildLightMap called for non-lit surface"); + + smax = (surf->extents[0]>>4)+1; + tmax = (surf->extents[1]>>4)+1; + size = smax*tmax; + if (size > (sizeof(s_blocklights)>>4) ) + ri.Sys_Error (ERR_DROP, "Bad s_blocklights size"); + + // set to full bright if no light data + if (!surf->samples) + { + for (i=0 ; istyles[nummaps] != 255 ; + nummaps++) + ; + + lightmap = surf->samples; + + // add all the lightmaps + if ( nummaps == 1 ) + { + int maps; + + for (maps = 0 ; maps < MAXLIGHTMAPS && surf->styles[maps] != 255 ; + maps++) + { + bl = s_blocklights; + + for (i=0 ; i<3 ; i++) + scale[i] = r_modulate->value*r_newrefdef.lightstyles[surf->styles[maps]].rgb[i]; + + if ( scale[0] == 1.0F && + scale[1] == 1.0F && + scale[2] == 1.0F ) + { + for (i=0 ; istyles[maps] != 255 ; + maps++) + { + bl = s_blocklights; + + for (i=0 ; i<3 ; i++) + scale[i] = r_modulate->value*r_newrefdef.lightstyles[surf->styles[maps]].rgb[i]; + + if ( scale[0] == 1.0F && + scale[1] == 1.0F && + scale[2] == 1.0F ) + { + for (i=0 ; idlightframe == r_framecount) + R_AddDynamicLights (surf); + +store: + // put into texture format + stride -= (smax<<2); + bl = s_blocklights; + + for (i = 0; i < tmax; i++, dest += stride) + { + for (j = 0; j < smax; j++) + { + + r = Q_ftol(bl[0]); + g = Q_ftol(bl[1]); + b = Q_ftol(bl[2]); + + // catch negative lights + if (r < 0) + r = 0; + if (g < 0) + g = 0; + if (b < 0) + b = 0; + + /* + ** determine the brightest of the three color components + */ + if (r > g) + max = r; + else + max = g; + if (b > max) + max = b; + + /* + ** alpha is ONLY used for the mono lightmap case. For this reason + ** we set it to the brightest of the color components so that + ** things don't get too dim. + */ + a = max; + + /* + ** rescale all the color components if the intensity of the greatest + ** channel exceeds 1.0 + */ + if (max > 255) + { + float t = 255.0F / max; + + r = r * t; + g = g * t; + b = b * t; + a = a * t; + } + + dest[0] = r; + dest[1] = g; + dest[2] = b; + dest[3] = a; + + bl += 3; + dest += 4; + } + } +} + diff --git a/src/vk/vk_mesh.c b/src/vk/vk_mesh.c new file mode 100644 index 0000000..d390db2 --- /dev/null +++ b/src/vk/vk_mesh.c @@ -0,0 +1,999 @@ +/* +Copyright (C) 1997-2001 Id Software, Inc. +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +*/ +// vk_mesh.c: triangle model functions + +#include "header/local.h" + +/* +============================================================= + + ALIAS MODELS + +============================================================= +*/ + +#define NUMVERTEXNORMALS 162 + +static float r_avertexnormals[NUMVERTEXNORMALS][3] = { +#include "../constants/anorms.h" +}; + +typedef float vec4_t[4]; + +enum { + TRIANGLE_STRIP = 0, + TRIANGLE_FAN = 1 +} pipelineIdx; + +typedef struct { + vec3_t vertex; + float color[4]; + float texCoord[2]; +} modelvert; + +typedef struct { + int vertexCount; + int firstVertex; +} drawinfo_t; + +polyvert_t *verts_buffer = NULL; +lmappolyvert_t *lmappolyverts_buffer = NULL; +static drawinfo_t *drawInfo[2] = {NULL, NULL}; +static modelvert *vertList[2] = {NULL, NULL}; +static vec4_t *s_lerped = NULL; +static vec3_t *shadowverts = NULL; +static int verts_count = 0; + +vec3_t shadevector; +float shadelight[3]; + +// precalculated dot products for quantized angles +#define SHADEDOT_QUANT 16 +static float r_avertexnormal_dots[SHADEDOT_QUANT][256] = { +#include "../constants/anormtab.h" +}; + +float *shadedots = r_avertexnormal_dots[0]; + +extern float r_view_matrix[16]; +extern float r_projection_matrix[16]; +extern float r_viewproj_matrix[16]; + +// correction matrix with "hacked depth" for models with RF_DEPTHHACK flag set +static float r_vulkan_correction_dh[16] = { + 1.f, 0.f, 0.f, 0.f, + 0.f, -1.f, 0.f, 0.f, + 0.f, 0.f, .3f, 0.f, + 0.f, 0.f, .3f, 1.f +}; + +int +Mesh_VertsRealloc(int count) +{ + void *ptr; + + if (verts_count > count) + { + return 0; + } + + verts_count = ROUNDUP(count * 2, 256); + + ptr = realloc(s_lerped, verts_count * sizeof(vec4_t)); + if (!ptr) + { + return -1; + } + s_lerped = ptr; + + ptr = realloc(shadowverts, verts_count * sizeof(vec3_t)); + if (!ptr) + { + return -1; + } + shadowverts = ptr; + + ptr = realloc(verts_buffer, verts_count * sizeof(polyvert_t)); + if (!ptr) + { + return -1; + } + verts_buffer = ptr; + + ptr = realloc(lmappolyverts_buffer, verts_count * sizeof(polyvert_t)); + if (!ptr) + { + return -1; + } + lmappolyverts_buffer = ptr; + + ptr = realloc(vertList[0], verts_count * sizeof(modelvert)); + if (!ptr) + { + return -1; + } + vertList[0] = ptr; + + ptr = realloc(vertList[1], verts_count * sizeof(modelvert)); + if (!ptr) + { + return -1; + } + vertList[1] = ptr; + + ptr = realloc(drawInfo[0], verts_count * sizeof(drawinfo_t)); + if (!ptr) + { + return -1; + } + drawInfo[0] = ptr; + + ptr = realloc(drawInfo[1], verts_count * sizeof(drawinfo_t)); + if (!ptr) + { + return -1; + } + drawInfo[1] = ptr; + + return 0; +} + +/* +=============== +Mesh_Init +=============== +*/ +void Mesh_Init (void) +{ + s_lerped = NULL; + shadowverts = NULL; + verts_buffer = NULL; + lmappolyverts_buffer = NULL; + vertList[0] = NULL; + vertList[1] = NULL; + drawInfo[0] = NULL; + drawInfo[1] = NULL; + + verts_count = 0; + + if (Mesh_VertsRealloc(MAX_VERTS)) + { + ri.Sys_Error(ERR_FATAL, "%s: can't allocate memory", __func__); + } +} + +/* +================ +Mesh_Free +================ +*/ +void Mesh_Free (void) +{ + if (vk_validation->value > 1) + { + R_Printf(PRINT_ALL, "%s: Deallocated %d mesh verts\n", + __func__, verts_count); + } + verts_count = 0; + + if (shadowverts) + { + free(shadowverts); + } + shadowverts = NULL; + + if (s_lerped) + { + free(s_lerped); + } + s_lerped = NULL; + + if (verts_buffer) + { + free(verts_buffer); + } + verts_buffer = NULL; + + if (lmappolyverts_buffer) + { + free(lmappolyverts_buffer); + } + lmappolyverts_buffer = NULL; + + if (vertList[0]) + { + free(vertList[0]); + } + if (vertList[1]) + { + free(vertList[1]); + } + vertList[0] = NULL; + vertList[1] = NULL; + + if (drawInfo[0]) + { + free(drawInfo[0]); + } + if (drawInfo[1]) + { + free(drawInfo[1]); + } + drawInfo[0] = NULL; + drawInfo[1] = NULL; +} + + +static void Vk_LerpVerts( int nverts, dtrivertx_t *v, dtrivertx_t *ov, dtrivertx_t *verts, float *lerp, float move[3], float frontv[3], float backv[3], entity_t *currententity ) +{ + int i; + + //PMM -- added RF_SHELL_DOUBLE, RF_SHELL_HALF_DAM + if ( currententity->flags & ( RF_SHELL_RED | RF_SHELL_GREEN | RF_SHELL_BLUE | RF_SHELL_DOUBLE | RF_SHELL_HALF_DAM) ) + { + for (i=0 ; i < nverts; i++, v++, ov++, lerp+=4 ) + { + float *normal = r_avertexnormals[verts[i].lightnormalindex]; + + lerp[0] = move[0] + ov->v[0]*backv[0] + v->v[0]*frontv[0] + normal[0] * POWERSUIT_SCALE; + lerp[1] = move[1] + ov->v[1]*backv[1] + v->v[1]*frontv[1] + normal[1] * POWERSUIT_SCALE; + lerp[2] = move[2] + ov->v[2]*backv[2] + v->v[2]*frontv[2] + normal[2] * POWERSUIT_SCALE; + } + } + else + { + for (i=0 ; i < nverts; i++, v++, ov++, lerp+=4) + { + lerp[0] = move[0] + ov->v[0]*backv[0] + v->v[0]*frontv[0]; + lerp[1] = move[1] + ov->v[1]*backv[1] + v->v[1]*frontv[1]; + lerp[2] = move[2] + ov->v[2]*backv[2] + v->v[2]*frontv[2]; + } + } + +} + +/* +============= +Vk_DrawAliasFrameLerp + +interpolates between two frames and origins +FIXME: batch lerp all vertexes +============= +*/ +static void Vk_DrawAliasFrameLerp (dmdl_t *paliashdr, float backlerp, image_t *skin, float *modelMatrix, int leftHandOffset, int translucentIdx, entity_t *currententity) +{ + daliasframe_t *frame, *oldframe; + dtrivertx_t *v, *ov, *verts; + int *order; + float frontlerp; + float alpha; + vec3_t move, delta, vectors[3]; + vec3_t frontv, backv; + int i; + float *lerp; + + frame = (daliasframe_t *)((byte *)paliashdr + paliashdr->ofs_frames + + currententity->frame * paliashdr->framesize); + verts = v = frame->verts; + + oldframe = (daliasframe_t *)((byte *)paliashdr + paliashdr->ofs_frames + + currententity->oldframe * paliashdr->framesize); + ov = oldframe->verts; + + order = (int *)((byte *)paliashdr + paliashdr->ofs_glcmds); + + if (currententity->flags & RF_TRANSLUCENT) + alpha = currententity->alpha; + else + alpha = 1.0; + + frontlerp = 1.0 - backlerp; + + // move should be the delta back to the previous frame * backlerp + VectorSubtract (currententity->oldorigin, currententity->origin, delta); + AngleVectors (currententity->angles, vectors[0], vectors[1], vectors[2]); + + move[0] = DotProduct (delta, vectors[0]); // forward + move[1] = -DotProduct (delta, vectors[1]); // left + move[2] = DotProduct (delta, vectors[2]); // up + + VectorAdd (move, oldframe->translate, move); + + for (i=0 ; i<3 ; i++) + { + move[i] = backlerp*move[i] + frontlerp*frame->translate[i]; + } + + for (i=0 ; i<3 ; i++) + { + frontv[i] = frontlerp*frame->scale[i]; + backv[i] = backlerp*oldframe->scale[i]; + } + + if (Mesh_VertsRealloc(paliashdr->num_xyz)) + { + ri.Sys_Error(ERR_FATAL, "%s: can't allocate memory", __func__); + } + + lerp = s_lerped[0]; + + Vk_LerpVerts( paliashdr->num_xyz, v, ov, verts, lerp, move, frontv, backv, currententity ); + + int vertCounts[2] = { 0, 0 }; + int pipeCounters[2] = { 0, 0 }; + VkDeviceSize maxTriangleFanIdxCnt = 0; + + if (Mesh_VertsRealloc(1)) + { + ri.Sys_Error(ERR_FATAL, "%s: can't allocate memory", __func__); + } + + drawInfo[0][0].firstVertex = 0; + drawInfo[1][0].firstVertex = 0; + + struct { + float model[16]; + int textured; + } meshUbo; + + while (1) + { + int count; + + // get the vertex count and primitive type + count = *order++; + if (!count) + break; // done + if (count < 0) + { + count = -count; + pipelineIdx = TRIANGLE_FAN; + } + else + { + pipelineIdx = TRIANGLE_STRIP; + } + + if (Mesh_VertsRealloc(pipeCounters[pipelineIdx])) + { + ri.Sys_Error(ERR_FATAL, "%s: can't allocate memory", __func__); + } + + drawInfo[pipelineIdx][pipeCounters[pipelineIdx]].vertexCount = count; + maxTriangleFanIdxCnt = max(maxTriangleFanIdxCnt, ((count - 2) * 3)); + + if (currententity->flags & (RF_SHELL_RED | RF_SHELL_GREEN | RF_SHELL_BLUE)) + { + meshUbo.textured = 0; + do + { + int vertIdx = vertCounts[pipelineIdx]; + int index_xyz = order[2]; + + if (Mesh_VertsRealloc(vertIdx)) + { + ri.Sys_Error(ERR_FATAL, "%s: can't allocate memory", __func__); + } + + // unused in this case, since texturing is disabled + vertList[pipelineIdx][vertIdx].texCoord[0] = 0.f; + vertList[pipelineIdx][vertIdx].texCoord[1] = 0.f; + + vertList[pipelineIdx][vertIdx].color[0] = shadelight[0]; + vertList[pipelineIdx][vertIdx].color[1] = shadelight[1]; + vertList[pipelineIdx][vertIdx].color[2] = shadelight[2]; + vertList[pipelineIdx][vertIdx].color[3] = alpha; + + if (verts_count < index_xyz) + { + R_Printf(PRINT_ALL, "%s: Model has issues with lerped index\n", __func__); + return; + } + + vertList[pipelineIdx][vertIdx].vertex[0] = s_lerped[index_xyz][0]; + vertList[pipelineIdx][vertIdx].vertex[1] = s_lerped[index_xyz][1]; + vertList[pipelineIdx][vertIdx].vertex[2] = s_lerped[index_xyz][2]; + vertCounts[pipelineIdx]++; + order += 3; + } while (--count); + } + else + { + meshUbo.textured = 1; + do + { + int vertIdx = vertCounts[pipelineIdx]; + int index_xyz = order[2]; + float l; + + if (Mesh_VertsRealloc(vertIdx)) + { + ri.Sys_Error(ERR_FATAL, "%s: can't allocate memory", __func__); + } + + // texture coordinates come from the draw list + vertList[pipelineIdx][vertIdx].texCoord[0] = ((float *)order)[0]; + vertList[pipelineIdx][vertIdx].texCoord[1] = ((float *)order)[1]; + + // normals and vertexes come from the frame list + l = shadedots[verts[index_xyz].lightnormalindex]; + + vertList[pipelineIdx][vertIdx].color[0] = l * shadelight[0]; + vertList[pipelineIdx][vertIdx].color[1] = l * shadelight[1]; + vertList[pipelineIdx][vertIdx].color[2] = l * shadelight[2]; + vertList[pipelineIdx][vertIdx].color[3] = alpha; + + if (verts_count < index_xyz) + { + R_Printf(PRINT_ALL, "%s: Model has issues with lerped index\n", __func__); + return; + } + + vertList[pipelineIdx][vertIdx].vertex[0] = s_lerped[index_xyz][0]; + vertList[pipelineIdx][vertIdx].vertex[1] = s_lerped[index_xyz][1]; + vertList[pipelineIdx][vertIdx].vertex[2] = s_lerped[index_xyz][2]; + vertCounts[pipelineIdx]++; + order += 3; + } while (--count); + } + + if (Mesh_VertsRealloc(pipeCounters[pipelineIdx] + 1)) + { + ri.Sys_Error(ERR_FATAL, "%s: can't allocate memory", __func__); + } + + pipeCounters[pipelineIdx]++; + drawInfo[pipelineIdx][pipeCounters[pipelineIdx]].firstVertex = vertCounts[pipelineIdx]; + } + + uint32_t uboOffset; + VkDescriptorSet uboDescriptorSet; + uint8_t *uboData = QVk_GetUniformBuffer(sizeof(meshUbo), &uboOffset, &uboDescriptorSet); + memcpy(meshUbo.model, modelMatrix, sizeof(float) * 16); + memcpy(uboData, &meshUbo, sizeof(meshUbo)); + + // player configuration screen model is using the UI renderpass + int pidx = (r_newrefdef.rdflags & RDF_NOWORLDMODEL) ? RP_UI : RP_WORLD; + // non-depth write alias models don't occur with RF_WEAPONMODEL set, so no need for additional left-handed pipelines + qvkpipeline_t pipelines[2][2] = { + { vk_drawModelPipelineFan[pidx], vk_drawLefthandModelPipelineFan }, + { vk_drawNoDepthModelPipelineFan, vk_drawLefthandModelPipelineFan } }; + for (int p = 0; p < 2; p++) + { + VkDeviceSize vaoSize = sizeof(modelvert) * vertCounts[p]; + VkBuffer vbo; + VkDeviceSize vboOffset; + uint8_t *vertData = QVk_GetVertexBuffer(vaoSize, &vbo, &vboOffset); + memcpy(vertData, vertList[p], vaoSize); + + QVk_BindPipeline(&pipelines[translucentIdx][leftHandOffset]); + VkDescriptorSet descriptorSets[] = { skin->vk_texture.descriptorSet, uboDescriptorSet }; + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, pipelines[translucentIdx][leftHandOffset].layout, 0, 2, descriptorSets, 1, &uboOffset); + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + + if (p == TRIANGLE_STRIP) + { + vkCmdBindIndexBuffer(vk_activeCmdbuffer, QVk_GetTriangleStripIbo(maxTriangleFanIdxCnt), 0, VK_INDEX_TYPE_UINT16); + + for (i = 0; i < pipeCounters[p]; i++) + { + vkCmdDrawIndexed(vk_activeCmdbuffer, (drawInfo[p][i].vertexCount - 2) * 3, 1, 0, drawInfo[p][i].firstVertex, 0); + } + } + else + { + vkCmdBindIndexBuffer(vk_activeCmdbuffer, QVk_GetTriangleFanIbo(maxTriangleFanIdxCnt), 0, VK_INDEX_TYPE_UINT16); + + for (i = 0; i < pipeCounters[p]; i++) + { + vkCmdDrawIndexed(vk_activeCmdbuffer, (drawInfo[p][i].vertexCount - 2) * 3, 1, 0, drawInfo[p][i].firstVertex, 0); + } + } + } +} + + +/* +============= +Vk_DrawAliasShadow +============= +*/ +extern vec3_t lightspot; + +static void Vk_DrawAliasShadow (dmdl_t *paliashdr, int posenum, float *modelMatrix, entity_t *currententity) +{ + int *order; + vec3_t point; + float height, lheight; + + lheight = currententity->origin[2] - lightspot[2]; + + height = 0; + + order = (int *)((byte *)paliashdr + paliashdr->ofs_glcmds); + + height = -lheight + 1.0; + + uint32_t uboOffset; + VkDescriptorSet uboDescriptorSet; + uint8_t *uboData = QVk_GetUniformBuffer(sizeof(float) * 16, &uboOffset, &uboDescriptorSet); + memcpy(uboData, modelMatrix, sizeof(float) * 16); + + while (1) + { + int i; + int count; + + i = 0; + // get the vertex count and primitive type + count = *order++; + if (!count) + break; // done + if (count < 0) + { + count = -count; + pipelineIdx = TRIANGLE_FAN; + } + else + { + pipelineIdx = TRIANGLE_STRIP; + } + + if (Mesh_VertsRealloc(count)) + { + ri.Sys_Error(ERR_FATAL, "%s: can't allocate memory", __func__); + } + + do + { + if (Mesh_VertsRealloc(order[2])) + { + ri.Sys_Error(ERR_FATAL, "%s: can't allocate memory", __func__); + } + + // normals and vertexes come from the frame list + memcpy( point, s_lerped[order[2]], sizeof( point ) ); + + point[0] -= shadevector[0]*(point[2]+lheight); + point[1] -= shadevector[1]*(point[2]+lheight); + point[2] = height; + + shadowverts[i][0] = point[0]; + shadowverts[i][1] = point[1]; + shadowverts[i][2] = point[2]; + + order += 3; + i++; + } while (--count); + + if (i > 0) + { + VkDeviceSize vaoSize = sizeof(vec3_t) * i; + VkBuffer vbo; + VkDeviceSize vboOffset; + uint8_t *vertData = QVk_GetVertexBuffer(vaoSize, &vbo, &vboOffset); + memcpy(vertData, shadowverts, vaoSize); + + QVk_BindPipeline(&vk_shadowsPipelineFan); + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_shadowsPipelineFan.layout, 0, 1, &uboDescriptorSet, 1, &uboOffset); + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + + if (pipelineIdx == TRIANGLE_STRIP) + { + vkCmdBindIndexBuffer(vk_activeCmdbuffer, QVk_GetTriangleStripIbo((i - 2) * 3), 0, VK_INDEX_TYPE_UINT16); + vkCmdDrawIndexed(vk_activeCmdbuffer, (i - 2) * 3, 1, 0, 0, 0); + } + else + { + vkCmdBindIndexBuffer(vk_activeCmdbuffer, QVk_GetTriangleFanIbo((i - 2) * 3), 0, VK_INDEX_TYPE_UINT16); + vkCmdDrawIndexed(vk_activeCmdbuffer, (i - 2) * 3, 1, 0, 0, 0); + } + } + } +} + + +/* +** R_CullAliasModel +*/ +static qboolean R_CullAliasModel( vec3_t bbox[8], entity_t *e, model_t *currentmodel ) +{ + int i; + vec3_t mins, maxs; + dmdl_t *paliashdr; + vec3_t vectors[3]; + vec3_t thismins, oldmins, thismaxs, oldmaxs; + daliasframe_t *pframe, *poldframe; + vec3_t angles; + + paliashdr = (dmdl_t *)currentmodel->extradata; + + if ( ( e->frame >= paliashdr->num_frames ) || ( e->frame < 0 ) ) + { + R_Printf(PRINT_ALL, "%s %s: no such frame %d\n", + __func__, currentmodel->name, e->frame); + e->frame = 0; + } + if ( ( e->oldframe >= paliashdr->num_frames ) || ( e->oldframe < 0 ) ) + { + R_Printf(PRINT_ALL, "%s %s: no such oldframe %d\n", + __func__, currentmodel->name, e->oldframe); + e->oldframe = 0; + } + + pframe = ( daliasframe_t * ) ( ( byte * ) paliashdr + + paliashdr->ofs_frames + + e->frame * paliashdr->framesize); + + poldframe = ( daliasframe_t * ) ( ( byte * ) paliashdr + + paliashdr->ofs_frames + + e->oldframe * paliashdr->framesize); + + /* + ** compute axially aligned mins and maxs + */ + if ( pframe == poldframe ) + { + for ( i = 0; i < 3; i++ ) + { + mins[i] = pframe->translate[i]; + maxs[i] = mins[i] + pframe->scale[i]*255; + } + } + else + { + for ( i = 0; i < 3; i++ ) + { + thismins[i] = pframe->translate[i]; + thismaxs[i] = thismins[i] + pframe->scale[i]*255; + + oldmins[i] = poldframe->translate[i]; + oldmaxs[i] = oldmins[i] + poldframe->scale[i]*255; + + if ( thismins[i] < oldmins[i] ) + mins[i] = thismins[i]; + else + mins[i] = oldmins[i]; + + if ( thismaxs[i] > oldmaxs[i] ) + maxs[i] = thismaxs[i]; + else + maxs[i] = oldmaxs[i]; + } + } + + /* + ** compute a full bounding box + */ + for ( i = 0; i < 8; i++ ) + { + vec3_t tmp; + + if ( i & 1 ) + tmp[0] = mins[0]; + else + tmp[0] = maxs[0]; + + if ( i & 2 ) + tmp[1] = mins[1]; + else + tmp[1] = maxs[1]; + + if ( i & 4 ) + tmp[2] = mins[2]; + else + tmp[2] = maxs[2]; + + VectorCopy( tmp, bbox[i] ); + } + + /* + ** rotate the bounding box + */ + VectorCopy( e->angles, angles ); + angles[YAW] = -angles[YAW]; + AngleVectors( angles, vectors[0], vectors[1], vectors[2] ); + + for ( i = 0; i < 8; i++ ) + { + vec3_t tmp; + + VectorCopy( bbox[i], tmp ); + + bbox[i][0] = DotProduct( vectors[0], tmp ); + bbox[i][1] = -DotProduct( vectors[1], tmp ); + bbox[i][2] = DotProduct( vectors[2], tmp ); + + VectorAdd( e->origin, bbox[i], bbox[i] ); + } + + { + int p, f, aggregatemask = ~0; + + for ( p = 0; p < 8; p++ ) + { + int mask = 0; + + for ( f = 0; f < 4; f++ ) + { + float dp = DotProduct( frustum[f].normal, bbox[p] ); + + if ( ( dp - frustum[f].dist ) < 0 ) + { + mask |= ( 1 << f ); + } + } + + aggregatemask &= mask; + } + + if ( aggregatemask ) + { + return true; + } + + return false; + } +} + +/* +================= +R_DrawAliasModel + +================= +*/ +void R_DrawAliasModel (entity_t *currententity, model_t *currentmodel) +{ + int i; + int leftHandOffset = 0; + dmdl_t *paliashdr; + float an; + vec3_t bbox[8]; + float prev_viewproj[16]; + + if ( !( currententity->flags & RF_WEAPONMODEL ) ) + { + if ( R_CullAliasModel( bbox, currententity, currentmodel ) ) + return; + } + + if ( currententity->flags & RF_WEAPONMODEL ) + { + if ( r_lefthand->value == 2 ) + return; + } + + paliashdr = (dmdl_t *)currentmodel->extradata; + + // + // get lighting information + // + // PMM - rewrote, reordered to handle new shells & mixing + // PMM - 3.20 code .. replaced with original way of doing it to keep mod authors happy + // + if ( currententity->flags & ( RF_SHELL_HALF_DAM | RF_SHELL_GREEN | RF_SHELL_RED | RF_SHELL_BLUE | RF_SHELL_DOUBLE ) ) + { + VectorClear (shadelight); + if (currententity->flags & RF_SHELL_HALF_DAM) + { + shadelight[0] = 0.56; + shadelight[1] = 0.59; + shadelight[2] = 0.45; + } + if ( currententity->flags & RF_SHELL_DOUBLE ) + { + shadelight[0] = 0.9; + shadelight[1] = 0.7; + } + if ( currententity->flags & RF_SHELL_RED ) + shadelight[0] = 1.0; + if ( currententity->flags & RF_SHELL_GREEN ) + shadelight[1] = 1.0; + if ( currententity->flags & RF_SHELL_BLUE ) + shadelight[2] = 1.0; + } + else if ( currententity->flags & RF_FULLBRIGHT ) + { + for (i=0 ; i<3 ; i++) + shadelight[i] = 1.0; + } + else + { + R_LightPoint (currententity->origin, shadelight, currententity); + + // player lighting hack for communication back to server + // big hack! + if ( currententity->flags & RF_WEAPONMODEL ) + { + // pick the greatest component, which should be the same + // as the mono value returned by software + if (shadelight[0] > shadelight[1]) + { + if (shadelight[0] > shadelight[2]) + r_lightlevel->value = 150*shadelight[0]; + else + r_lightlevel->value = 150*shadelight[2]; + } + else + { + if (shadelight[1] > shadelight[2]) + r_lightlevel->value = 150*shadelight[1]; + else + r_lightlevel->value = 150*shadelight[2]; + } + } + } + + if ( currententity->flags & RF_MINLIGHT ) + { + for (i=0 ; i<3 ; i++) + if (shadelight[i] > 0.1) + break; + if (i == 3) + { + shadelight[0] = 0.1; + shadelight[1] = 0.1; + shadelight[2] = 0.1; + } + } + + if ( currententity->flags & RF_GLOW ) + { // bonus items will pulse with time + float scale; + + scale = 0.1 * sin(r_newrefdef.time*7); + for (i=0 ; i<3 ; i++) + { + float min; + + min = shadelight[i] * 0.8; + shadelight[i] += scale; + if (shadelight[i] < min) + shadelight[i] = min; + } + } + + // ================= + // PGM ir goggles color override + if ( r_newrefdef.rdflags & RDF_IRGOGGLES && currententity->flags & RF_IR_VISIBLE) + { + shadelight[0] = 1.0; + shadelight[1] = 0.0; + shadelight[2] = 0.0; + } + // PGM + // ================= + + shadedots = r_avertexnormal_dots[((int)(currententity->angles[1] * (SHADEDOT_QUANT / 360.0))) & (SHADEDOT_QUANT - 1)]; + + an = currententity->angles[1]/180*M_PI; + shadevector[0] = cos(-an); + shadevector[1] = sin(-an); + shadevector[2] = 1; + VectorNormalize (shadevector); + + // + // locate the proper data + // + + c_alias_polys += paliashdr->num_tris; + + // + // draw all the triangles + // + if (currententity->flags & RF_DEPTHHACK || r_newrefdef.rdflags & RDF_NOWORLDMODEL) { // hack the depth range to prevent view model from poking into walls + float r_proj_aspect = (float)r_newrefdef.width / r_newrefdef.height; + float r_proj_fovy = r_newrefdef.fov_y; + float dist = (r_farsee->value == 0) ? 4096.0f : 8192.0f; + // use different range for player setup screen so it doesn't collide with the viewmodel + r_vulkan_correction_dh[10] = 0.3f - (r_newrefdef.rdflags & RDF_NOWORLDMODEL) * 0.1f; + r_vulkan_correction_dh[14] = 0.3f - (r_newrefdef.rdflags & RDF_NOWORLDMODEL) * 0.1f; + + memcpy(prev_viewproj, r_viewproj_matrix, sizeof(r_viewproj_matrix)); + if (currententity->flags & RF_WEAPONMODEL && r_gunfov->value < 0) + { + Mat_Perspective(r_projection_matrix, r_vulkan_correction_dh, r_proj_fovy, r_proj_aspect, 4, dist); + } + else + { + Mat_Perspective(r_projection_matrix, r_vulkan_correction_dh, r_gunfov->value, r_proj_aspect, 4, dist); + } + Mat_Mul(r_view_matrix, r_projection_matrix, r_viewproj_matrix); + vkCmdPushConstants(vk_activeCmdbuffer, vk_drawTexQuadPipeline[vk_state.current_renderpass].layout, VK_SHADER_STAGE_VERTEX_BIT, 0, sizeof(r_viewproj_matrix), r_viewproj_matrix); + } + + if ( ( currententity->flags & RF_WEAPONMODEL ) && ( r_lefthand->value == 1.0F ) ) + { + Mat_Scale(r_viewproj_matrix, -1.f, 1.f, 1.f); + vkCmdPushConstants(vk_activeCmdbuffer, vk_drawTexQuadPipeline[vk_state.current_renderpass].layout, VK_SHADER_STAGE_VERTEX_BIT, 0, sizeof(r_viewproj_matrix), r_viewproj_matrix); + leftHandOffset = 1; + } + + currententity->angles[PITCH] = -currententity->angles[PITCH]; // sigh. + { + float model[16]; + image_t *skin; + Mat_Identity(model); + R_RotateForEntity (currententity, model); + + currententity->angles[PITCH] = -currententity->angles[PITCH]; // sigh. + + // select skin + if (currententity->skin) + skin = currententity->skin; // custom player skin + else + { + if (currententity->skinnum >= MAX_MD2SKINS) + skin = currentmodel->skins[0]; + else + { + skin = currentmodel->skins[currententity->skinnum]; + if (!skin) + skin = currentmodel->skins[0]; + } + } + if (!skin) + skin = r_notexture; // fallback... + + // draw it + if ( (currententity->frame >= paliashdr->num_frames) + || (currententity->frame < 0) ) + { + R_Printf(PRINT_ALL, "%s %s: no such frame %d\n", + __func__, currentmodel->name, currententity->frame); + currententity->frame = 0; + currententity->oldframe = 0; + } + + if ( (currententity->oldframe >= paliashdr->num_frames) + || (currententity->oldframe < 0)) + { + R_Printf(PRINT_ALL, "%s %s: no such oldframe %d\n", + __func__, currentmodel->name, currententity->oldframe); + currententity->frame = 0; + currententity->oldframe = 0; + } + + if ( !r_lerpmodels->value ) + currententity->backlerp = 0; + Vk_DrawAliasFrameLerp (paliashdr, currententity->backlerp, skin, model, leftHandOffset, (currententity->flags & RF_TRANSLUCENT) ? 1 : 0, currententity); + } + + if ( ( currententity->flags & RF_WEAPONMODEL ) && ( r_lefthand->value == 1.0F ) ) + { + Mat_Scale(r_viewproj_matrix, -1.f, 1.f, 1.f); + vkCmdPushConstants(vk_activeCmdbuffer, vk_drawTexQuadPipeline[vk_state.current_renderpass].layout, VK_SHADER_STAGE_VERTEX_BIT, 0, sizeof(r_viewproj_matrix), r_viewproj_matrix); + } + + if (currententity->flags & RF_DEPTHHACK || r_newrefdef.rdflags & RDF_NOWORLDMODEL) + { + memcpy(r_viewproj_matrix, prev_viewproj, sizeof(prev_viewproj)); + vkCmdPushConstants(vk_activeCmdbuffer, vk_drawTexQuadPipeline[vk_state.current_renderpass].layout, VK_SHADER_STAGE_VERTEX_BIT, 0, sizeof(r_viewproj_matrix), r_viewproj_matrix); + } + + if (vk_shadows->value && !(currententity->flags & (RF_TRANSLUCENT | RF_WEAPONMODEL))) + { + float model[16]; + Mat_Identity(model); + R_RotateForEntity(currententity, model); + Vk_DrawAliasShadow (paliashdr, currententity->frame, model, currententity); + } +} diff --git a/src/vk/vk_model.c b/src/vk/vk_model.c new file mode 100644 index 0000000..01a6577 --- /dev/null +++ b/src/vk/vk_model.c @@ -0,0 +1,1525 @@ +/* +Copyright (C) 1997-2001 Id Software, Inc. +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +*/ +// vk_model.c -- model loading and caching + +#include "header/local.h" + +YQ2_ALIGNAS_TYPE(int) static byte mod_novis[MAX_MAP_LEAFS/8]; + +#define MAX_MOD_KNOWN 512 +static model_t *models_known; +static int mod_numknown = 0; +static int mod_loaded = 0; +static int mod_max = 0; +static int models_known_max = 0; + +int registration_sequence; + +/* +=============== +Mod_PointInLeaf +=============== +*/ +mleaf_t *Mod_PointInLeaf (vec3_t p, model_t *model) +{ + mnode_t *node; + + if (!model || !model->nodes) + ri.Sys_Error (ERR_DROP, "%s: bad model", __func__); + + node = model->nodes; + while (1) + { + cplane_t *plane; + float d; + + if (node->contents != -1) + return (mleaf_t *)node; + plane = node->plane; + d = DotProduct (p,plane->normal) - plane->dist; + if (d > 0) + node = node->children[0]; + else + node = node->children[1]; + } + + return NULL; // never reached +} + + +/* +============== +Mod_ClusterPVS +============== +*/ +byte *Mod_ClusterPVS (int cluster, model_t *model) +{ + if (cluster == -1 || !model->vis) + return mod_novis; + + return Mod_DecompressVis ( (byte *)model->vis + + model->vis->bitofs[cluster][DVIS_PVS], + (model->vis->numclusters+7)>>3); +} + + +//=============================================================================== + +/* +=============== +Mod_Reallocate +=============== +*/ +void Mod_Reallocate (void) +{ + if ((models_known_max >= (mod_max * 4)) && (models_known != NULL)) + { + return; + } + + if (models_known) + { + // Always up if already allocated + models_known_max *= 2; + // free up + Mod_FreeAll(); + Mod_FreeModelsKnown(); + } + + if (models_known_max < (mod_max * 4)) + { + // double is not enough? + models_known_max = ROUNDUP(mod_max * 4, 16); + } + + R_Printf(PRINT_ALL, "Reallocate space for %d models.\n", models_known_max); + + models_known = malloc(models_known_max * sizeof(model_t)); + memset(models_known, 0, models_known_max * sizeof(model_t)); +} + +/* +=============== +Mod_Init +=============== +*/ +void Mod_Init (void) +{ + memset (mod_novis, 0xff, sizeof(mod_novis)); + mod_numknown = 0; + mod_loaded = 0; + mod_max = 0; + models_known_max = MAX_MOD_KNOWN; + models_known = NULL; + + Mod_Reallocate (); +} + +/* +================ +Mod_Free +================ +*/ +static void Mod_Free (model_t *mod) +{ + if (!mod->extradata) + { + // looks as empty model + memset (mod, 0, sizeof(*mod)); + return; + } + + if (vk_validation->value) + { + R_Printf(PRINT_ALL, "%s: Unload %s[%d]\n", __func__, mod->name, mod_loaded); + } + + Hunk_Free (mod->extradata); + memset (mod, 0, sizeof(*mod)); + mod_loaded --; + if (mod_loaded < 0) + { + ri.Sys_Error (ERR_DROP, "%s: Broken unload", __func__); + } +} + +/* +================ +Mod_FreeAll +================ +*/ +void Mod_FreeAll (void) +{ + int i; + + for (i=0 ; ifilelen) + { + loadmodel->lightdata = NULL; + return; + } + loadmodel->lightdata = Hunk_Alloc ( l->filelen); + memcpy (loadmodel->lightdata, mod_base + l->fileofs, l->filelen); +} + + +/* +================= +Mod_LoadVisibility +================= +*/ +static void Mod_LoadVisibility (model_t *loadmodel, byte *mod_base, lump_t *l) +{ + int i; + + if (!l->filelen) + { + loadmodel->vis = NULL; + return; + } + loadmodel->vis = Hunk_Alloc ( l->filelen); + memcpy (loadmodel->vis, mod_base + l->fileofs, l->filelen); + + loadmodel->vis->numclusters = LittleLong (loadmodel->vis->numclusters); + for (i=0 ; ivis->numclusters ; i++) + { + loadmodel->vis->bitofs[i][0] = LittleLong (loadmodel->vis->bitofs[i][0]); + loadmodel->vis->bitofs[i][1] = LittleLong (loadmodel->vis->bitofs[i][1]); + } +} + + +/* +================= +Mod_LoadVertexes +================= +*/ +static void Mod_LoadVertexes (model_t *loadmodel, byte *mod_base, lump_t *l) +{ + dvertex_t *in; + mvertex_t *out; + int i, count; + + in = (void *)(mod_base + l->fileofs); + if (l->filelen % sizeof(*in)) + { + ri.Sys_Error(ERR_DROP, "%s: funny lump size in %s", + __func__, loadmodel->name); + } + + count = l->filelen / sizeof(*in); + out = Hunk_Alloc ( count*sizeof(*out)); + + loadmodel->vertexes = out; + loadmodel->numvertexes = count; + + for ( i=0 ; iposition[0] = LittleFloat (in->point[0]); + out->position[1] = LittleFloat (in->point[1]); + out->position[2] = LittleFloat (in->point[2]); + } +} + +/* +================= +RadiusFromBounds +================= +*/ +static float RadiusFromBounds (vec3_t mins, vec3_t maxs) +{ + int i; + vec3_t corner; + + for (i=0 ; i<3 ; i++) + { + corner[i] = fabs(mins[i]) > fabs(maxs[i]) ? fabs(mins[i]) : fabs(maxs[i]); + } + + return VectorLength (corner); +} + + +/* +================= +Mod_LoadSubmodels +================= +*/ +static void Mod_LoadSubmodels (model_t *loadmodel, byte *mod_base, lump_t *l) +{ + dmodel_t *in; + model_t *out; + int i, j, count; + + in = (void *)(mod_base + l->fileofs); + if (l->filelen % sizeof(*in)) + { + ri.Sys_Error(ERR_DROP, "%s: funny lump size in %s", + __func__, loadmodel->name); + } + + count = l->filelen / sizeof(*in); + out = Hunk_Alloc ( count*sizeof(*out)); + + loadmodel->submodels = out; + loadmodel->numsubmodels = count; + + for ( i=0 ; isubmodels, sizeof(*out)); + } + + Com_sprintf (out->name, sizeof(out->name), "*%d", i); + + for (j=0 ; j<3 ; j++) + { // spread the mins / maxs by a pixel + out->mins[j] = LittleFloat (in->mins[j]) - 1; + out->maxs[j] = LittleFloat (in->maxs[j]) + 1; + out->origin[j] = LittleFloat (in->origin[j]); + } + + out->radius = RadiusFromBounds (out->mins, out->maxs); + out->firstnode = LittleLong (in->headnode); + out->firstmodelsurface = LittleLong (in->firstface); + out->nummodelsurfaces = LittleLong (in->numfaces); + // visleafs + out->numleafs = 0; + // check limits + if (out->firstnode >= loadmodel->numnodes) + { + ri.Sys_Error(ERR_DROP, "%s: Inline model %i has bad firstnode", + __func__, i); + } + } +} + +/* +================= +Mod_LoadEdges +================= +*/ +static void Mod_LoadEdges (model_t *loadmodel, byte *mod_base, lump_t *l) +{ + dedge_t *in; + medge_t *out; + int i, count; + + in = (void *)(mod_base + l->fileofs); + if (l->filelen % sizeof(*in)) + { + ri.Sys_Error(ERR_DROP, "%s: funny lump size in %s", + __func__, loadmodel->name); + } + + count = l->filelen / sizeof(*in); + out = Hunk_Alloc ( (count + 1) * sizeof(*out)); + + loadmodel->edges = out; + loadmodel->numedges = count; + + for ( i=0 ; iv[0] = (unsigned short)LittleShort(in->v[0]); + out->v[1] = (unsigned short)LittleShort(in->v[1]); + } +} + +/* +================= +Mod_LoadTexinfo +================= +*/ +static void Mod_LoadTexinfo (model_t *loadmodel, byte *mod_base, lump_t *l) +{ + texinfo_t *in; + mtexinfo_t *out, *step; + int i, j, count; + char name[MAX_QPATH]; + + in = (void *)(mod_base + l->fileofs); + if (l->filelen % sizeof(*in)) + { + ri.Sys_Error(ERR_DROP, "%s: funny lump size in %s", + __func__, loadmodel->name); + } + + count = l->filelen / sizeof(*in); + out = Hunk_Alloc ( count*sizeof(*out)); + + loadmodel->texinfo = out; + loadmodel->numtexinfo = count; + + for ( i=0 ; ivecs[0][j] = LittleFloat (in->vecs[0][j]); + out->vecs[1][j] = LittleFloat (in->vecs[1][j]); + } + + out->flags = LittleLong (in->flags); + next = LittleLong (in->nexttexinfo); + if (next > 0) + out->next = loadmodel->texinfo + next; + else + out->next = NULL; + Com_sprintf (name, sizeof(name), "textures/%s.wal", in->texture); + + out->image = Vk_FindImage (name, it_wall); + if (!out->image || out->image == r_notexture) + { + Com_sprintf (name, sizeof(name), "textures/%s.m8", in->texture); + out->image = Vk_FindImage (name, it_wall); + } + + if (!out->image) + { + R_Printf(PRINT_ALL, "Couldn't load %s\n", name); + out->image = r_notexture; + } + } + + // count animation frames + for (i=0 ; itexinfo[i]; + out->numframes = 1; + for (step = out->next ; step && step != out ; step=step->next) + out->numframes++; + } +} + +/* +================ +CalcSurfaceExtents + +Fills in s->texturemins[] and s->extents[] +================ +*/ +static void CalcSurfaceExtents (model_t *loadmodel, msurface_t *s) +{ + float mins[2], maxs[2], val; + int i; + mtexinfo_t *tex; + int bmins[2], bmaxs[2]; + + mins[0] = mins[1] = 999999; + maxs[0] = maxs[1] = -99999; + + tex = s->texinfo; + + for (i=0 ; inumedges ; i++) + { + int e, j; + mvertex_t *v; + + e = loadmodel->surfedges[s->firstedge+i]; + if (e >= 0) + v = &loadmodel->vertexes[loadmodel->edges[e].v[0]]; + else + v = &loadmodel->vertexes[loadmodel->edges[-e].v[1]]; + + for (j=0 ; j<2 ; j++) + { + val = v->position[0] * tex->vecs[j][0] + + v->position[1] * tex->vecs[j][1] + + v->position[2] * tex->vecs[j][2] + + tex->vecs[j][3]; + if (val < mins[j]) + mins[j] = val; + if (val > maxs[j]) + maxs[j] = val; + } + } + + for (i=0 ; i<2 ; i++) + { + bmins[i] = floor(mins[i]/16); + bmaxs[i] = ceil(maxs[i]/16); + + s->texturemins[i] = bmins[i] * 16; + s->extents[i] = (bmaxs[i] - bmins[i]) * 16; + } +} + +static int calcTexinfoAndFacesSize(const lump_t *fl, byte *mod_base, const lump_t *tl) +{ + dface_t* face_in = (void *)(mod_base + fl->fileofs); + texinfo_t* texinfo_in = (void *)(mod_base + tl->fileofs); + + if (fl->filelen % sizeof(*face_in) || tl->filelen % sizeof(*texinfo_in)) + { + // will error out when actually loading it + return 0; + } + + int ret = 0; + + int face_count = fl->filelen / sizeof(*face_in); + int texinfo_count = tl->filelen / sizeof(*texinfo_in); + + { + // out = Hunk_Alloc(count * sizeof(*out)); + int baseSize = face_count * sizeof(msurface_t); + baseSize = (baseSize + 31) & ~31; + ret += baseSize; + + int ti_size = texinfo_count * sizeof(mtexinfo_t); + ti_size = (ti_size + 31) & ~31; + ret += ti_size; + } + + int numWarpFaces = 0; + + for (int surfnum = 0; surfnum < face_count; surfnum++, face_in++) + { + int numverts = LittleShort(face_in->numedges); + int ti = LittleShort(face_in->texinfo); + if ((ti < 0) || (ti >= texinfo_count)) + { + return 0; // will error out + } + int texFlags = LittleLong(texinfo_in[ti].flags); + + /* set the drawing flags */ + if (texFlags & SURF_WARP) + { + if (numverts > 60) + return 0; // will error out in R_SubdividePolygon() + + // Vk_SubdivideSurface(out, loadmodel); /* cut up polygon for warps */ + // for each (pot. recursive) call to R_SubdividePolygon(): + // sizeof(vkpoly_t) + ((numverts - 4) + 2) * VERTEXSIZE*sizeof(float) + + // this is tricky, how much is allocated depends on the size of the surface + // which we don't know (we'd need the vertices etc to know, but we can't load + // those without allocating...) + // so we just count warped faces and use a generous estimate below + + ++numWarpFaces; + } + else + { + // Vk_BuildPolygonFromSurface(out); + // => poly = Hunk_Alloc(sizeof(vkpoly_t) + (numverts - 4) * VERTEXSIZE*sizeof(float)); + int polySize = sizeof(vkpoly_t) + (numverts - 4) * VERTEXSIZE*sizeof(float); + polySize = (polySize + 31) & ~31; + ret += polySize; + } + } + + // yeah, this is a bit hacky, but it looks like for each warped face + // 256-55000 bytes are allocated (usually on the lower end), + // so just assume 48k per face to be safe + ret += numWarpFaces * 49152; + ret += 1000000; // and 1MB extra just in case + + return ret; +} + +/* +================= +Mod_LoadFaces +================= +*/ +static void Mod_LoadFaces (model_t *loadmodel, byte *mod_base, lump_t *l) +{ + dface_t *in; + msurface_t *out; + int i, count, surfnum; + + in = (void *)(mod_base + l->fileofs); + if (l->filelen % sizeof(*in)) + { + ri.Sys_Error(ERR_DROP, "%s: funny lump size in %s", + __func__, loadmodel->name); + } + + count = l->filelen / sizeof(*in); + out = Hunk_Alloc(count * sizeof(*out)); + + loadmodel->surfaces = out; + loadmodel->numsurfaces = count; + + Vk_BeginBuildingLightmaps(loadmodel); + + for (surfnum = 0; surfnumfirstedge = LittleLong(in->firstedge); + out->numedges = LittleShort(in->numedges); + if (out->numedges < 3) + { + ri.Sys_Error(ERR_DROP, "%s: Surface with %d edges", + __func__, out->numedges); + } + out->flags = 0; + out->polys = NULL; + + planenum = LittleShort(in->planenum); + side = LittleShort(in->side); + if (side) + out->flags |= SURF_PLANEBACK; + + out->plane = loadmodel->planes + planenum; + + ti = LittleShort(in->texinfo); + if (ti < 0 || ti >= loadmodel->numtexinfo) + { + ri.Sys_Error(ERR_DROP, "%s: bad texinfo number", __func__); + } + out->texinfo = loadmodel->texinfo + ti; + + CalcSurfaceExtents(loadmodel, out); + + // lighting info + + for (i = 0; istyles[i] = in->styles[i]; + i = LittleLong(in->lightofs); + if (i == -1) + out->samples = NULL; + else + out->samples = loadmodel->lightdata + i; + + // set the drawing flags + + if (out->texinfo->flags & SURF_WARP) + { + out->flags |= SURF_DRAWTURB; + for (i = 0; i<2; i++) + { + out->extents[i] = 16384; + out->texturemins[i] = -8192; + } + Vk_SubdivideSurface(out, loadmodel); // cut up polygon for warps + } + + if (r_fixsurfsky->value) + { + if (out->texinfo->flags & SURF_SKY) + { + out->flags |= SURF_DRAWSKY; + } + } + + // create lightmaps and polygons + if (!(out->texinfo->flags & (SURF_SKY | SURF_TRANS33 | SURF_TRANS66 | SURF_WARP))) + Vk_CreateSurfaceLightmap(out); + + if (!(out->texinfo->flags & SURF_WARP)) + Vk_BuildPolygonFromSurface(out, loadmodel); + + } + + Vk_EndBuildingLightmaps(); +} + + +/* +================= +Mod_SetParent +================= +*/ +static void Mod_SetParent (mnode_t *node, mnode_t *parent) +{ + node->parent = parent; + if (node->contents != -1) + return; + Mod_SetParent (node->children[0], node); + Mod_SetParent (node->children[1], node); +} + +/* +================= +Mod_LoadNodes +================= +*/ +static void Mod_LoadNodes (model_t *loadmodel, byte *mod_base, lump_t *l) +{ + int i, j, count; + dnode_t *in; + mnode_t *out; + + in = (void *)(mod_base + l->fileofs); + if (l->filelen % sizeof(*in)) + { + ri.Sys_Error(ERR_DROP, "%s: funny lump size in %s", + __func__, loadmodel->name); + } + + count = l->filelen / sizeof(*in); + out = Hunk_Alloc ( count*sizeof(*out)); + + loadmodel->nodes = out; + loadmodel->numnodes = count; + + for ( i=0 ; iminmaxs[j] = LittleShort (in->mins[j]); + out->minmaxs[3+j] = LittleShort (in->maxs[j]); + } + + p = LittleLong(in->planenum); + out->plane = loadmodel->planes + p; + + out->firstsurface = LittleShort (in->firstface); + out->numsurfaces = LittleShort (in->numfaces); + out->contents = -1; // differentiate from leafs + + for (j=0 ; j<2 ; j++) + { + p = LittleLong (in->children[j]); + if (p >= 0) + out->children[j] = loadmodel->nodes + p; + else + out->children[j] = (mnode_t *)(loadmodel->leafs + (-1 - p)); + } + } + + Mod_SetParent (loadmodel->nodes, NULL); // sets nodes and leafs +} + +/* +================= +Mod_LoadLeafs +================= +*/ +static void Mod_LoadLeafs (model_t *loadmodel, byte *mod_base, lump_t *l) +{ + dleaf_t *in; + mleaf_t *out; + int i, j, count; + + in = (void *)(mod_base + l->fileofs); + if (l->filelen % sizeof(*in)) + { + ri.Sys_Error(ERR_DROP, "%s: funny lump size in %s", + __func__, loadmodel->name); + } + + count = l->filelen / sizeof(*in); + out = Hunk_Alloc ( count*sizeof(*out)); + + loadmodel->leafs = out; + loadmodel->numleafs = count; + + for ( i=0 ; iminmaxs[j] = LittleShort (in->mins[j]); + out->minmaxs[3+j] = LittleShort (in->maxs[j]); + } + + out->contents = LittleLong(in->contents); + out->cluster = LittleShort(in->cluster); + out->area = LittleShort(in->area); + + // make unsigned long from signed short + firstleafface = LittleShort(in->firstleafface) & 0xFFFF; + out->nummarksurfaces = LittleShort(in->numleaffaces) & 0xFFFF; + + out->firstmarksurface = loadmodel->marksurfaces + firstleafface; + if ((firstleafface + out->nummarksurfaces) > loadmodel->nummarksurfaces) + { + ri.Sys_Error(ERR_DROP, "%s: wrong marksurfaces position in %s", + __func__, loadmodel->name); + } + + // gl underwater warp +#if 0 + if (out->contents & (CONTENTS_WATER|CONTENTS_SLIME|CONTENTS_LAVA|CONTENTS_THINWATER) ) + { + for (j=0 ; jnummarksurfaces ; j++) + { + out->firstmarksurface[j]->flags |= SURF_UNDERWATER; + for (poly = out->firstmarksurface[j]->polys ; poly ; poly=poly->next) + poly->flags |= SURF_UNDERWATER; + } + } +#endif + } +} + +/* +================= +Mod_LoadMarksurfaces +================= +*/ +static void Mod_LoadMarksurfaces (model_t *loadmodel, byte *mod_base, lump_t *l) +{ + int i, count; + short *in; + msurface_t **out; + + in = (void *)(mod_base + l->fileofs); + + if (l->filelen % sizeof(*in)) + { + ri.Sys_Error(ERR_DROP, "%s: funny lump size in %s", + __func__, loadmodel->name); + } + + count = l->filelen / sizeof(*in); + out = Hunk_Alloc ( count*sizeof(*out)); + + loadmodel->marksurfaces = out; + loadmodel->nummarksurfaces = count; + + for ( i=0 ; i= loadmodel->numsurfaces) + { + ri.Sys_Error(ERR_DROP, "%s: bad surface number", __func__); + } + out[i] = loadmodel->surfaces + j; + } +} + +/* +================= +Mod_LoadSurfedges +================= +*/ +static void Mod_LoadSurfedges (model_t *loadmodel, byte *mod_base, lump_t *l) +{ + int i, count; + int *in, *out; + + in = (void *)(mod_base + l->fileofs); + if (l->filelen % sizeof(*in)) + { + ri.Sys_Error(ERR_DROP, "%s: funny lump size in %s", + __func__, loadmodel->name); + } + + count = l->filelen / sizeof(*in); + if (count < 1 || count >= MAX_MAP_SURFEDGES) + ri.Sys_Error (ERR_DROP, "%s: bad surfedges count in %s: %i", + __func__, loadmodel->name, count); + + out = Hunk_Alloc ( count*sizeof(*out)); + + loadmodel->surfedges = out; + loadmodel->numsurfedges = count; + + for ( i=0 ; ifileofs); + + if (l->filelen % sizeof(*in)) + { + ri.Sys_Error(ERR_DROP, "%s: funny lump size in %s", + __func__, loadmodel->name); + } + + count = l->filelen / sizeof(*in); + out = Hunk_Alloc ( count*2*sizeof(*out)); + + loadmodel->planes = out; + loadmodel->numplanes = count; + + for ( i=0 ; inormal[j] = LittleFloat (in->normal[j]); + if (out->normal[j] < 0) + bits |= 1<dist = LittleFloat (in->dist); + out->type = LittleLong (in->type); + out->signbits = bits; + } +} + + +// calculate the size that Hunk_Alloc(), called by Mod_Load*() from Mod_LoadBrushModel(), +// will use (=> includes its padding), so we'll know how big the hunk needs to be +static int calcLumpHunkSize(const lump_t *l, int inSize, int outSize) +{ + if (l->filelen % inSize) + { + // Mod_Load*() will error out on this because of "funny size" + // don't error out here because in Mod_Load*() it can print the functionname + // (=> tells us what kind of lump) before shutting down the game + return 0; + } + + int count = l->filelen / inSize; + int size = count * outSize; + + // round to cacheline, like Hunk_Alloc() does + size = (size + 31) & ~31; + return size; +} + +/* +================= +Mod_LoadBrushModel +================= +*/ +static void Mod_LoadBrushModel (model_t *loadmodel, void *buffer, int modfilelen) +{ + int i; + dheader_t *header; + byte *mod_base; + + header = (dheader_t *)buffer; + + i = LittleLong(header->version); + + if (i != BSPVERSION) + { + ri.Sys_Error(ERR_DROP, "%s: %s has wrong version number (%i should be %i)", + __func__, loadmodel->name, i, BSPVERSION); + } + + // swap all the lumps + mod_base = (byte *)header; + + for (i = 0; i < sizeof(dheader_t) / 4; i++) + { + ((int *)header)[i] = LittleLong(((int *)header)[i]); + } + + // calculate the needed hunksize from the lumps + int hunkSize = 0; + hunkSize += calcLumpHunkSize(&header->lumps[LUMP_VERTEXES], sizeof(dvertex_t), sizeof(mvertex_t)); + hunkSize += calcLumpHunkSize(&header->lumps[LUMP_EDGES], sizeof(dedge_t), sizeof(medge_t)); + hunkSize += sizeof(medge_t) + 31; // for count+1 in Mod_LoadEdges() + int surfEdgeCount = (header->lumps[LUMP_SURFEDGES].filelen+sizeof(int)-1)/sizeof(int); + if(surfEdgeCount < MAX_MAP_SURFEDGES) // else it errors out later anyway + hunkSize += calcLumpHunkSize(&header->lumps[LUMP_SURFEDGES], sizeof(int), sizeof(int)); + hunkSize += calcLumpHunkSize(&header->lumps[LUMP_LIGHTING], 1, 1); + hunkSize += calcLumpHunkSize(&header->lumps[LUMP_PLANES], sizeof(dplane_t), sizeof(cplane_t)*2); + hunkSize += calcTexinfoAndFacesSize(&header->lumps[LUMP_FACES], mod_base, &header->lumps[LUMP_TEXINFO]); + hunkSize += calcLumpHunkSize(&header->lumps[LUMP_LEAFFACES], sizeof(short), sizeof(msurface_t *)); // yes, out is indeeed a pointer! + hunkSize += calcLumpHunkSize(&header->lumps[LUMP_VISIBILITY], 1, 1); + hunkSize += calcLumpHunkSize(&header->lumps[LUMP_LEAFS], sizeof(dleaf_t), sizeof(mleaf_t)); + hunkSize += calcLumpHunkSize(&header->lumps[LUMP_NODES], sizeof(dnode_t), sizeof(mnode_t)); + hunkSize += calcLumpHunkSize(&header->lumps[LUMP_MODELS], sizeof(dmodel_t), sizeof(model_t)); + + loadmodel->extradata = Hunk_Begin(hunkSize); + loadmodel->type = mod_brush; + loadmodel->numframes = 2; // regular and alternate animation + + // load into heap + Mod_LoadVertexes (loadmodel, mod_base, &header->lumps[LUMP_VERTEXES]); + Mod_LoadEdges (loadmodel, mod_base, &header->lumps[LUMP_EDGES]); + Mod_LoadSurfedges (loadmodel, mod_base, &header->lumps[LUMP_SURFEDGES]); + Mod_LoadLighting (loadmodel, mod_base, &header->lumps[LUMP_LIGHTING]); + Mod_LoadPlanes (loadmodel, mod_base, &header->lumps[LUMP_PLANES]); + Mod_LoadTexinfo (loadmodel, mod_base, &header->lumps[LUMP_TEXINFO]); + Mod_LoadFaces (loadmodel, mod_base, &header->lumps[LUMP_FACES]); + Mod_LoadMarksurfaces (loadmodel, mod_base, &header->lumps[LUMP_LEAFFACES]); + Mod_LoadVisibility (loadmodel, mod_base, &header->lumps[LUMP_VISIBILITY]); + Mod_LoadLeafs (loadmodel, mod_base, &header->lumps[LUMP_LEAFS]); + Mod_LoadNodes (loadmodel, mod_base, &header->lumps[LUMP_NODES]); + Mod_LoadSubmodels (loadmodel, mod_base, &header->lumps[LUMP_MODELS]); +} + +/* +============================================================================== + +ALIAS MODELS + +============================================================================== +*/ + +/* +================= +Mod_LoadAliasModel +================= +*/ +static void Mod_LoadAliasModel (model_t *mod, void *buffer, int modfilelen) +{ + int i, j; + dmdl_t *pinmodel, *pheader; + dstvert_t *pinst, *poutst; + dtriangle_t *pintri, *pouttri; + int *pincmd, *poutcmd; + int version; + int ofs_end; + + pinmodel = (dmdl_t *)buffer; + + version = LittleLong (pinmodel->version); + if (version != ALIAS_VERSION) + { + ri.Sys_Error(ERR_DROP, "%s: %s has wrong version number (%i should be %i)", + __func__, mod->name, version, ALIAS_VERSION); + } + + ofs_end = LittleLong(pinmodel->ofs_end); + if (ofs_end < 0 || ofs_end > modfilelen) + { + ri.Sys_Error(ERR_DROP, "%s: model %s file size(%d) too small, should be %d", + __func__, mod->name, modfilelen, ofs_end); + } + + mod->extradata = Hunk_Begin(modfilelen); + pheader = Hunk_Alloc(ofs_end); + + // byte swap the header fields and sanity check + for (i=0 ; iskinheight > MAX_LBM_HEIGHT) + { + ri.Sys_Error(ERR_DROP, "%s: model %s has a skin taller than %d", + __func__, mod->name, MAX_LBM_HEIGHT); + } + + if (pheader->num_xyz <= 0) + { + ri.Sys_Error(ERR_DROP, "%s: model %s has no vertices", + __func__, mod->name); + } + + if (pheader->num_xyz > MAX_VERTS) + { + ri.Sys_Error(ERR_DROP, "%s: model %s has too many vertices", + __func__, mod->name); + } + + if (pheader->num_st <= 0) + { + ri.Sys_Error(ERR_DROP, "%s: model %s has no st vertices", + __func__, mod->name); + } + + if (pheader->num_tris <= 0) + { + ri.Sys_Error(ERR_DROP, "%s: model %s has no triangles", + __func__, mod->name); + } + + if (pheader->num_frames <= 0) + { + ri.Sys_Error(ERR_DROP, "%s: model %s has no frames", + __func__, mod->name); + } + + // + // load base s and t vertices (not used in gl version) + // + pinst = (dstvert_t *) ((byte *)pinmodel + pheader->ofs_st); + poutst = (dstvert_t *) ((byte *)pheader + pheader->ofs_st); + + for (i=0 ; inum_st ; i++) + { + poutst[i].s = LittleShort (pinst[i].s); + poutst[i].t = LittleShort (pinst[i].t); + } + + // + // load triangle lists + // + pintri = (dtriangle_t *) ((byte *)pinmodel + pheader->ofs_tris); + pouttri = (dtriangle_t *) ((byte *)pheader + pheader->ofs_tris); + + for (i=0 ; inum_tris ; i++) + { + for (j=0 ; j<3 ; j++) + { + pouttri[i].index_xyz[j] = LittleShort (pintri[i].index_xyz[j]); + pouttri[i].index_st[j] = LittleShort (pintri[i].index_st[j]); + } + } + + // + // load the frames + // + for (i=0 ; inum_frames ; i++) + { + daliasframe_t *pinframe, *poutframe; + + pinframe = (daliasframe_t *) ((byte *)pinmodel + + pheader->ofs_frames + i * pheader->framesize); + poutframe = (daliasframe_t *) ((byte *)pheader + + pheader->ofs_frames + i * pheader->framesize); + + memcpy (poutframe->name, pinframe->name, sizeof(poutframe->name)); + for (j=0 ; j<3 ; j++) + { + poutframe->scale[j] = LittleFloat (pinframe->scale[j]); + poutframe->translate[j] = LittleFloat (pinframe->translate[j]); + } + // verts are all 8 bit, so no swapping needed + memcpy (poutframe->verts, pinframe->verts, + pheader->num_xyz*sizeof(dtrivertx_t)); + + } + + mod->type = mod_alias; + + // + // load the glcmds + // + pincmd = (int *) ((byte *)pinmodel + pheader->ofs_glcmds); + poutcmd = (int *) ((byte *)pheader + pheader->ofs_glcmds); + for (i=0; i < pheader->num_glcmds; i++) + { + poutcmd[i] = LittleLong (pincmd[i]); + } + + if (poutcmd[pheader->num_glcmds-1] != 0) + { + R_Printf(PRINT_ALL, "%s: Entity %s has possible last element issues with %d verts.\n", + __func__, + mod->name, + poutcmd[pheader->num_glcmds-1]); + } + + // register all skins + memcpy ((char *)pheader + pheader->ofs_skins, (char *)pinmodel + pheader->ofs_skins, + pheader->num_skins*MAX_SKINNAME); + for (i=0 ; inum_skins ; i++) + { + mod->skins[i] = Vk_FindImage ((char *)pheader + pheader->ofs_skins + i*MAX_SKINNAME + , it_skin); + } + + mod->mins[0] = -32; + mod->mins[1] = -32; + mod->mins[2] = -32; + mod->maxs[0] = 32; + mod->maxs[1] = 32; + mod->maxs[2] = 32; +} + +/* +============================================================================== + +SPRITE MODELS + +============================================================================== +*/ + +/* +================= +Mod_LoadSpriteModel +================= +*/ +static void Mod_LoadSpriteModel (model_t *mod, void *buffer, int modfilelen) +{ + dsprite_t *sprin, *sprout; + int i; + + sprin = (dsprite_t *)buffer; + mod->extradata = Hunk_Begin(modfilelen); + sprout = Hunk_Alloc(modfilelen); + + sprout->ident = LittleLong (sprin->ident); + sprout->version = LittleLong (sprin->version); + sprout->numframes = LittleLong (sprin->numframes); + + if (sprout->version != SPRITE_VERSION) + { + ri.Sys_Error(ERR_DROP, "%s: %s has wrong version number (%i should be %i)", + __func__, mod->name, sprout->version, SPRITE_VERSION); + } + + if (sprout->numframes > MAX_MD2SKINS) + { + ri.Sys_Error(ERR_DROP, "%s: %s has too many frames (%i > %i)", + __func__, mod->name, sprout->numframes, MAX_MD2SKINS); + } + + // byte swap everything + for (i=0 ; inumframes ; i++) + { + sprout->frames[i].width = LittleLong (sprin->frames[i].width); + sprout->frames[i].height = LittleLong (sprin->frames[i].height); + sprout->frames[i].origin_x = LittleLong (sprin->frames[i].origin_x); + sprout->frames[i].origin_y = LittleLong (sprin->frames[i].origin_y); + memcpy (sprout->frames[i].name, sprin->frames[i].name, MAX_SKINNAME); + mod->skins[i] = Vk_FindImage (sprout->frames[i].name, + it_sprite); + } + + mod->type = mod_sprite; +} + +//============================================================================= + +/* +================== +Mod_ForName + +Loads in a model for the given name +================== +*/ +static model_t *Mod_ForName (char *name, model_t *parent_model, qboolean crash) +{ + model_t *mod; + unsigned *buf; + int i, modfilelen; + + if (!name[0]) + { + ri.Sys_Error(ERR_DROP, "%s: NULL name", __func__); + } + + // + // inline models are grabbed only from worldmodel + // + if (name[0] == '*' && parent_model) + { + i = atoi(name+1); + if (i < 1 || i >= parent_model->numsubmodels) + ri.Sys_Error (ERR_DROP, "bad inline model number"); + return &parent_model->submodels[i]; + } + + // + // search the currently loaded models + // + for (i=0 , mod=models_known ; iname[0]) + continue; + if (!strcmp (mod->name, name) ) + return mod; + } + + // + // find a free model slot spot + // + for (i=0 , mod=models_known ; iname[0]) + break; // free spot + } + if (i == mod_numknown) + { + if (mod_numknown == models_known_max) + ri.Sys_Error(ERR_DROP, "%s: mod_numknown == models_known_max", __func__); + mod_numknown++; + } + strcpy (mod->name, name); + + // + // load the file + // + modfilelen = ri.FS_LoadFile (mod->name, (void **)&buf); + if (!buf) + { + if (crash) + { + ri.Sys_Error(ERR_DROP, "%s: %s not found", + __func__, mod->name); + } + + if (vk_validation->value) + { + R_Printf(PRINT_ALL, "%s: Can't load %s\n", __func__, mod->name); + } + memset (mod->name, 0, sizeof(mod->name)); + return NULL; + } + + // update count of loaded models + mod_loaded ++; + if (vk_validation->value) + { + R_Printf(PRINT_ALL, "%s: Load %s[%d]\n", __func__, mod->name, mod_loaded); + } + + // + // fill it in + // + + // call the apropriate loader + + switch (LittleLong(*(unsigned *)buf)) + { + case IDALIASHEADER: + Mod_LoadAliasModel(mod, buf, modfilelen); + break; + + case IDSPRITEHEADER: + Mod_LoadSpriteModel(mod, buf, modfilelen); + break; + + case IDBSPHEADER: + Mod_LoadBrushModel(mod, buf, modfilelen); + break; + + default: + ri.Sys_Error(ERR_DROP, "%s: unknown fileid for %s", + __func__, mod->name); + break; + } + + mod->extradatasize = Hunk_End (); + + ri.FS_FreeFile(buf); + + return mod; +} + +/* +===================== +RE_BeginRegistration + +Specifies the model that will be used as the world +===================== +*/ +void +RE_BeginRegistration (char *model) +{ + char fullname[MAX_QPATH]; + cvar_t *flushmap; + + Mod_Reallocate(); + + registration_sequence++; + r_oldviewcluster = -1; // force markleafs + + Com_sprintf (fullname, sizeof(fullname), "maps/%s.bsp", model); + + // explicitly free the old map if different + // this guarantees that models_known[0] is the world map + flushmap = ri.Cvar_Get ("flushmap", "0", 0); + if ( strcmp(models_known[0].name, fullname) || flushmap->value) + Mod_Free (&models_known[0]); + r_worldmodel = Mod_ForName(fullname, NULL, true); + if (r_worldmodel != models_known) + ri.Sys_Error(ERR_DROP, "%s: Loaded a brush model after the world", __func__); + + r_viewcluster = -1; +} + +/* +===================== +RE_RegisterModel + +===================== +*/ +struct model_s *RE_RegisterModel (char *name) +{ + model_t *mod; + + mod = Mod_ForName (name, r_worldmodel, false); + if (mod) + { + int i; + + mod->registration_sequence = registration_sequence; + + // register any images used by the models + if (mod->type == mod_sprite) + { + dsprite_t *sprout; + + sprout = (dsprite_t *)mod->extradata; + for (i=0 ; inumframes ; i++) + mod->skins[i] = Vk_FindImage (sprout->frames[i].name, it_sprite); + } + else if (mod->type == mod_alias) + { + dmdl_t *pheader; + + pheader = (dmdl_t *)mod->extradata; + for (i=0 ; inum_skins ; i++) + mod->skins[i] = Vk_FindImage ((char *)pheader + pheader->ofs_skins + i*MAX_SKINNAME, it_skin); +//PGM + mod->numframes = pheader->num_frames; +//PGM + } + else if (mod->type == mod_brush) + { + for (i=0 ; inumtexinfo ; i++) + mod->texinfo[i].image->registration_sequence = registration_sequence; + } + } + return mod; +} + +static qboolean +Mod_HasFreeSpace(void) +{ + int i, used; + model_t *mod; + + used = 0; + + for (i=0, mod=models_known ; iname[0]) + continue; + if (mod->registration_sequence == registration_sequence) + { + used ++; + } + } + + if (mod_max < used) + { + mod_max = used; + } + + // should same size of free slots as currently used + return (mod_loaded + mod_max) < models_known_max; +} + +/* +================ +Mod_Modellist_f +================ +*/ +void Mod_Modellist_f (void) +{ + int i, total, used; + model_t *mod; + qboolean freeup; + + total = 0; + used = 0; + + R_Printf(PRINT_ALL,"Loaded models:\n"); + for (i=0, mod=models_known ; i < mod_numknown ; i++, mod++) + { + char *in_use = ""; + + if (mod->registration_sequence == registration_sequence) + { + in_use = "*"; + used ++; + } + + if (!mod->name[0]) + continue; + R_Printf(PRINT_ALL, "%8i : %s %s\n", + mod->extradatasize, mod->name, in_use); + total += mod->extradatasize; + } + R_Printf(PRINT_ALL, "Total resident: %i in %d models\n", total, mod_loaded); + // update statistics + freeup = Mod_HasFreeSpace(); + R_Printf(PRINT_ALL, "Used %d of %d models%s.\n", used, mod_max, freeup ? ", has free space" : ""); +} + +/* +===================== +RE_EndRegistration + +===================== +*/ +void RE_EndRegistration (void) +{ + int i; + model_t *mod; + + if (Mod_HasFreeSpace() && Vk_ImageHasFreeSpace()) + { + // should be enough space for load next maps + return; + } + + for (i=0, mod=models_known ; iname[0]) + continue; + if (mod->registration_sequence != registration_sequence) + { // don't need this model + Mod_Free (mod); + } + } + + Vk_FreeUnusedImages (); +} diff --git a/src/vk/vk_pipeline.c b/src/vk/vk_pipeline.c new file mode 100644 index 0000000..ee7d3f2 --- /dev/null +++ b/src/vk/vk_pipeline.c @@ -0,0 +1,222 @@ +/* +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +*/ + +#include "header/local.h" + +qvkshader_t QVk_CreateShader(const uint32_t *shaderSrc, size_t shaderCodeSize, VkShaderStageFlagBits shaderStage) +{ + qvkshader_t shader; + VkShaderModuleCreateInfo smCreateInfo = { + .sType = VK_STRUCTURE_TYPE_SHADER_MODULE_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .codeSize = shaderCodeSize, + .pCode = shaderSrc + }; + + VK_VERIFY(vkCreateShaderModule(vk_device.logical, &smCreateInfo, NULL, &shader.module)); + + VkPipelineShaderStageCreateInfo vssCreateInfo = { + .sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .stage = shaderStage, + .module = shader.module, + .pName = "main", + .pSpecializationInfo = NULL + }; + + shader.createInfo = vssCreateInfo; + + return shader; +} + +void QVk_CreatePipeline(const VkDescriptorSetLayout *descriptorLayout, const uint32_t descLayoutCount, const VkPipelineVertexInputStateCreateInfo *vertexInputInfo, + qvkpipeline_t *pipeline, const qvkrenderpass_t *renderpass, const qvkshader_t *shaders, uint32_t shaderCount) +{ + VkPipelineShaderStageCreateInfo *ssCreateInfos = (VkPipelineShaderStageCreateInfo *)malloc(shaderCount * sizeof(VkPipelineShaderStageCreateInfo)); + for (int i = 0; i < shaderCount; i++) + { + ssCreateInfos[i] = shaders[i].createInfo; + } + + VkPipelineInputAssemblyStateCreateInfo iaCreateInfo = { + .sType = VK_STRUCTURE_TYPE_PIPELINE_INPUT_ASSEMBLY_STATE_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .topology = pipeline->topology, + .primitiveRestartEnable = VK_FALSE + }; + + VkViewport viewport = { + .x = 0.f, + .y = 0.f, + .width = (float)vid.width, + .height = (float)vid.height, + .minDepth = 0.f, + .maxDepth = 1.f, + }; + + VkRect2D scissor = { + .offset.x = 0, + .offset.y = 0, + .extent = vk_swapchain.extent + }; + + VkPipelineViewportStateCreateInfo vpCreateInfo = { + .sType = VK_STRUCTURE_TYPE_PIPELINE_VIEWPORT_STATE_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .viewportCount = 1, + .pViewports = &viewport, + .scissorCount = 1, + .pScissors = &scissor + }; + + VkPipelineRasterizationStateCreateInfo rCreateInfo = { + .sType = VK_STRUCTURE_TYPE_PIPELINE_RASTERIZATION_STATE_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .depthClampEnable = VK_FALSE, + .rasterizerDiscardEnable = VK_FALSE, + .polygonMode = VK_POLYGON_MODE_FILL, + .cullMode = pipeline->cullMode, + .frontFace = VK_FRONT_FACE_CLOCKWISE, + .depthBiasEnable = VK_FALSE, + .depthBiasConstantFactor = 0.f, + .depthBiasClamp = 0.f, + .depthBiasSlopeFactor = 0.f, + .lineWidth = 1.f + }; + + VkPipelineMultisampleStateCreateInfo msCreateInfo = { + .sType = VK_STRUCTURE_TYPE_PIPELINE_MULTISAMPLE_STATE_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .rasterizationSamples = renderpass->sampleCount, + .sampleShadingEnable = (vk_sampleshading->value > 0 && vk_device.features.sampleRateShading) ? VK_TRUE : VK_FALSE, + .minSampleShading = (vk_sampleshading->value > 0 && vk_device.features.sampleRateShading) ? 1.f : 0.f, + .pSampleMask = NULL, + .alphaToCoverageEnable = VK_FALSE, + .alphaToOneEnable = VK_FALSE + }; + + VkPipelineDepthStencilStateCreateInfo dCreateInfo = { + .sType = VK_STRUCTURE_TYPE_PIPELINE_DEPTH_STENCIL_STATE_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .depthTestEnable = pipeline->depthTestEnable, + .depthWriteEnable = pipeline->depthTestEnable == VK_TRUE ? pipeline->depthWriteEnable : VK_FALSE, // there should be NO depth writes if depthTestEnable is false but Intel seems to not follow the specs fully... + .depthCompareOp = VK_COMPARE_OP_LESS, + .depthBoundsTestEnable = VK_FALSE, + .stencilTestEnable = VK_FALSE, + .front = { VK_STENCIL_OP_KEEP, VK_STENCIL_OP_KEEP, VK_STENCIL_OP_KEEP, VK_COMPARE_OP_NEVER, 0, 0, 0 }, + .back = { VK_STENCIL_OP_KEEP, VK_STENCIL_OP_KEEP, VK_STENCIL_OP_KEEP, VK_COMPARE_OP_NEVER, 0, 0, 0 }, + .minDepthBounds = 0.f, + .maxDepthBounds = 1.f + }; + + VkPipelineColorBlendStateCreateInfo cbsCreateInfo = { + .sType = VK_STRUCTURE_TYPE_PIPELINE_COLOR_BLEND_STATE_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .logicOpEnable = VK_FALSE, + .logicOp = VK_LOGIC_OP_COPY, + .attachmentCount = 1, + .pAttachments = &pipeline->blendOpts, + .blendConstants[0] = 0.f, + .blendConstants[1] = 0.f, + .blendConstants[2] = 0.f, + .blendConstants[3] = 0.f + }; + + VkDynamicState dynamicStates[] = { VK_DYNAMIC_STATE_VIEWPORT, VK_DYNAMIC_STATE_SCISSOR }; + VkPipelineDynamicStateCreateInfo dsCreateInfo = { + .sType = VK_STRUCTURE_TYPE_PIPELINE_DYNAMIC_STATE_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .dynamicStateCount = 2, + .pDynamicStates = dynamicStates + }; + + // push constant sizes accomodate for maximum number of uploaded elements (should probably be checked against the hardware's maximum supported value) + VkPushConstantRange pushConstantRange[] = { + { + .stageFlags = VK_SHADER_STAGE_VERTEX_BIT, + .offset = 0, + .size = 17 * sizeof(float) + }, + { + .stageFlags = VK_SHADER_STAGE_FRAGMENT_BIT, + .offset = 17 * sizeof(float), + .size = 11 * sizeof(float) + }}; + + VkPipelineLayoutCreateInfo plCreateInfo = { + .sType = VK_STRUCTURE_TYPE_PIPELINE_LAYOUT_CREATE_INFO, + .pNext = NULL, + .flags = 0, + .setLayoutCount = descLayoutCount, + .pSetLayouts = descriptorLayout, + // for simplicity assume only two push constant range is passed, + // so it's not the most flexible approach + .pushConstantRangeCount = 2, + .pPushConstantRanges = pushConstantRange + }; + + VK_VERIFY(vkCreatePipelineLayout(vk_device.logical, &plCreateInfo, NULL, &pipeline->layout)); + + // create THE pipeline + VkGraphicsPipelineCreateInfo pCreateInfo = { + .sType = VK_STRUCTURE_TYPE_GRAPHICS_PIPELINE_CREATE_INFO, + .pNext = NULL, + .flags = pipeline->flags, + .stageCount = shaderCount, + .pStages = ssCreateInfos, + .pVertexInputState = vertexInputInfo, + .pInputAssemblyState = &iaCreateInfo, + .pTessellationState = NULL, + .pViewportState = &vpCreateInfo, + .pRasterizationState = &rCreateInfo, + .pMultisampleState = &msCreateInfo, + .pDepthStencilState = &dCreateInfo, + .pColorBlendState = &cbsCreateInfo, + .pDynamicState = &dsCreateInfo, + .layout = pipeline->layout, + .renderPass = renderpass->rp, + .subpass = 0, + .basePipelineHandle = VK_NULL_HANDLE, + .basePipelineIndex = -1 + }; + + VK_VERIFY(vkCreateGraphicsPipelines(vk_device.logical, VK_NULL_HANDLE, 1, &pCreateInfo, NULL, &pipeline->pl)); + free(ssCreateInfos); +} + +void QVk_DestroyPipeline(qvkpipeline_t *pipeline) +{ + if (pipeline->layout != VK_NULL_HANDLE) + vkDestroyPipelineLayout(vk_device.logical, pipeline->layout, NULL); + if (pipeline->pl != VK_NULL_HANDLE) + vkDestroyPipeline(vk_device.logical, pipeline->pl, NULL); + + pipeline->layout = VK_NULL_HANDLE; + pipeline->pl = VK_NULL_HANDLE; +} diff --git a/src/vk/vk_rmain.c b/src/vk/vk_rmain.c new file mode 100644 index 0000000..046b527 --- /dev/null +++ b/src/vk/vk_rmain.c @@ -0,0 +1,1761 @@ +/* +Copyright (C) 1997-2001 Id Software, Inc. +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +*/ +// vk_rmain.c + +#include "header/local.h" + +#define REF_VERSION "Yamagi Quake II Vulkan Refresher (based on vkQuake2 v1.4.3)" + +// world rendered and ready to render 2d elements +static qboolean world_rendered; + +viddef_t vid; + +refimport_t ri; + +model_t *r_worldmodel; + +vkconfig_t vk_config; +vkstate_t vk_state; + +image_t *r_notexture; // use for bad textures +image_t *r_particletexture; // little dot for particles +image_t *r_squaretexture; // rectangle for particles + +cplane_t frustum[4]; + +int r_visframecount; // bumped when going to a new PVS +int r_framecount; // used for dlight push checking + +int c_brush_polys, c_alias_polys; + +static float v_blend[4]; // final blending color + +void Vk_Strings_f(void); +void Vk_Mem_f(void); + +// +// view origin +// +vec3_t vup; +vec3_t vpn; +vec3_t vright; +vec3_t r_origin; + +float r_projection_matrix[16]; +float r_view_matrix[16]; +float r_viewproj_matrix[16]; +// correction matrix for perspective in Vulkan +static float r_vulkan_correction[16] = { 1.f, 0.f, 0.f, 0.f, + 0.f, -1.f, 0.f, 0.f, + 0.f, 0.f, .5f, 0.f, + 0.f, 0.f, .5f, 1.f + }; +// +// screen size info +// +refdef_t r_newrefdef; + +int r_viewcluster, r_viewcluster2, r_oldviewcluster, r_oldviewcluster2; + +static cvar_t *r_norefresh; +static cvar_t *r_drawentities; +cvar_t *r_drawworld; +static cvar_t *r_speeds; +static cvar_t *r_fullbright; +cvar_t *r_novis; +static cvar_t *r_nocull; +cvar_t *r_lerpmodels; +cvar_t *r_lefthand; +cvar_t *r_vsync; +static cvar_t *r_mode; +cvar_t *r_gunfov; +cvar_t *r_farsee; +static cvar_t *r_customwidth; +static cvar_t *r_customheight; + +cvar_t *r_lightlevel; // FIXME: This is a HACK to get the client's light level + +cvar_t *vk_overbrightbits; +cvar_t *vk_validation; +cvar_t *vk_picmip; +cvar_t *vk_skymip; +cvar_t *vk_flashblend; +cvar_t *vk_finish; +cvar_t *r_clear; +cvar_t *r_lockpvs; +static cvar_t *vk_polyblend; +cvar_t *r_modulate; +cvar_t *vk_shadows; +cvar_t *vk_pixel_size; +static cvar_t *vk_particle_size; +static cvar_t *vk_particle_att_a; +static cvar_t *vk_particle_att_b; +static cvar_t *vk_particle_att_c; +static cvar_t *vk_particle_min_size; +static cvar_t *vk_particle_max_size; +static cvar_t *vk_custom_particles; +cvar_t *vk_postprocess; +cvar_t *vk_dynamic; +cvar_t *vk_msaa; +cvar_t *vk_showtris; +cvar_t *vk_lightmap; +cvar_t *vk_texturemode; +cvar_t *vk_lmaptexturemode; +cvar_t *vk_aniso; +cvar_t *vk_mip_nearfilter; +cvar_t *vk_sampleshading; +cvar_t *vk_device_idx; +cvar_t *vk_retexturing; +static cvar_t *vk_underwater; +cvar_t *vk_nolerp_list; +cvar_t *r_fixsurfsky; + +cvar_t *vid_fullscreen; +cvar_t *vid_gamma; +static cvar_t *viewsize; + +/* +================= +R_CullBox + +Returns true if the box is completely outside the frustom +================= +*/ +qboolean R_CullBox (vec3_t mins, vec3_t maxs) +{ + int i; + + if (r_nocull->value) + return false; + + for (i=0 ; i<4 ; i++) + if ( BOX_ON_PLANE_SIDE(mins, maxs, &frustum[i]) == 2) + return true; + return false; +} + + +void R_RotateForEntity (entity_t *e, float *mvMatrix) +{ + Mat_Rotate(mvMatrix, -e->angles[2], 1.f, 0.f, 0.f); + Mat_Rotate(mvMatrix, -e->angles[0], 0.f, 1.f, 0.f); + Mat_Rotate(mvMatrix, e->angles[1], 0.f, 0.f, 1.f); + Mat_Translate(mvMatrix, e->origin[0], e->origin[1], e->origin[2]); +} + +/* +============================================================= + + SPRITE MODELS + +============================================================= +*/ + + +/* +================= +R_DrawSpriteModel + +================= +*/ +void R_DrawSpriteModel (entity_t *currententity, model_t *currentmodel) +{ + float alpha = 1.0F; + vec3_t point; + dsprframe_t *frame; + float *up, *right; + dsprite_t *psprite; + + // don't even bother culling, because it's just a single + // polygon without a surface cache + + psprite = (dsprite_t *)currentmodel->extradata; + + currententity->frame %= psprite->numframes; + + frame = &psprite->frames[currententity->frame]; + + // normal sprite + up = vup; + right = vright; + + if (currententity->flags & RF_TRANSLUCENT) + alpha = currententity->alpha; + + vec3_t spriteQuad[4]; + + VectorMA(currententity->origin, -frame->origin_y, up, point); + VectorMA(point, -frame->origin_x, right, spriteQuad[0]); + VectorMA(currententity->origin, frame->height - frame->origin_y, up, point); + VectorMA(point, -frame->origin_x, right, spriteQuad[1]); + VectorMA(currententity->origin, frame->height - frame->origin_y, up, point); + VectorMA(point, frame->width - frame->origin_x, right, spriteQuad[2]); + VectorMA(currententity->origin, -frame->origin_y, up, point); + VectorMA(point, frame->width - frame->origin_x, right, spriteQuad[3]); + + float quadVerts[] = { spriteQuad[0][0], spriteQuad[0][1], spriteQuad[0][2], 0.f, 1.f, + spriteQuad[1][0], spriteQuad[1][1], spriteQuad[1][2], 0.f, 0.f, + spriteQuad[2][0], spriteQuad[2][1], spriteQuad[2][2], 1.f, 0.f, + spriteQuad[0][0], spriteQuad[0][1], spriteQuad[0][2], 0.f, 1.f, + spriteQuad[2][0], spriteQuad[2][1], spriteQuad[2][2], 1.f, 0.f, + spriteQuad[3][0], spriteQuad[3][1], spriteQuad[3][2], 1.f, 1.f }; + + vkCmdPushConstants(vk_activeCmdbuffer, vk_drawSpritePipeline.layout, + VK_SHADER_STAGE_VERTEX_BIT, sizeof(r_viewproj_matrix), sizeof(float), &alpha); + QVk_BindPipeline(&vk_drawSpritePipeline); + + VkBuffer vbo; + VkDeviceSize vboOffset; + uint8_t *vertData = QVk_GetVertexBuffer(sizeof(quadVerts), &vbo, &vboOffset); + memcpy(vertData, quadVerts, sizeof(quadVerts)); + + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + + float gamma = 2.1F - vid_gamma->value; + + vkCmdPushConstants(vk_activeCmdbuffer, vk_drawTexQuadPipeline[vk_state.current_renderpass].layout, + VK_SHADER_STAGE_FRAGMENT_BIT, 17 * sizeof(float), sizeof(gamma), &gamma); + + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_drawSpritePipeline.layout, 0, 1, ¤tmodel->skins[currententity->frame]->vk_texture.descriptorSet, 0, NULL); + vkCmdDraw(vk_activeCmdbuffer, 6, 1, 0, 0); +} + +//================================================================================== + +/* +============= +R_DrawNullModel +============= +*/ +void R_DrawNullModel (entity_t *currententity) +{ + vec3_t shadelight; + int i,j; + + if (currententity->flags & RF_FULLBRIGHT) + shadelight[0] = shadelight[1] = shadelight[2] = 1.0F; + else + R_LightPoint(currententity->origin, shadelight, currententity); + + float model[16]; + Mat_Identity(model); + R_RotateForEntity(currententity, model); + + vec3_t verts[24]; + verts[0][0] = 0.f; + verts[0][1] = 0.f; + verts[0][2] = -16.f; + verts[1][0] = shadelight[0]; + verts[1][1] = shadelight[1]; + verts[1][2] = shadelight[2]; + + for (i = 2, j = 0; i < 12; i+=2, j++) + { + verts[i][0] = 16 * cos(j*M_PI / 2); + verts[i][1] = 16 * sin(j*M_PI / 2); + verts[i][2] = 0.f; + verts[i+1][0] = shadelight[0]; + verts[i+1][1] = shadelight[1]; + verts[i+1][2] = shadelight[2]; + } + + verts[12][0] = 0.f; + verts[12][1] = 0.f; + verts[12][2] = 16.f; + verts[13][0] = shadelight[0]; + verts[13][1] = shadelight[1]; + verts[13][2] = shadelight[2]; + + for (i = 23, j = 4; i > 13; i-=2, j--) + { + verts[i-1][0] = 16 * cos(j*M_PI / 2); + verts[i-1][1] = 16 * sin(j*M_PI / 2); + verts[i-1][2] = 0.f; + verts[i][0] = shadelight[0]; + verts[i][1] = shadelight[1]; + verts[i][2] = shadelight[2]; + } + + VkBuffer vbo; + VkDeviceSize vboOffset; + uint32_t uboOffset; + VkDescriptorSet uboDescriptorSet; + uint8_t *vertData = QVk_GetVertexBuffer(sizeof(verts), &vbo, &vboOffset); + uint8_t *uboData = QVk_GetUniformBuffer(sizeof(model), &uboOffset, &uboDescriptorSet); + memcpy(vertData, verts, sizeof(verts)); + memcpy(uboData, model, sizeof(model)); + + QVk_BindPipeline(&vk_drawNullModelPipeline); + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_drawNullModelPipeline.layout, 0, 1, &uboDescriptorSet, 1, &uboOffset); + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + vkCmdBindIndexBuffer(vk_activeCmdbuffer, QVk_GetTriangleFanIbo(12), 0, VK_INDEX_TYPE_UINT16); + vkCmdDrawIndexed(vk_activeCmdbuffer, 12, 1, 0, 0, 0); + vkCmdDrawIndexed(vk_activeCmdbuffer, 12, 1, 0, 6, 0); +} + +/* +============= +R_DrawEntitiesOnList +============= +*/ +void R_DrawEntitiesOnList (void) +{ + int i; + + if (!r_drawentities->value) + return; + + // draw non-transparent first + for (i = 0; iflags & RF_TRANSLUCENT) + continue; // solid + + if (currententity->flags & RF_BEAM) + { + R_DrawBeam(currententity); + } + else + { + model_t *currentmodel = currententity->model; + if (!currentmodel) + { + R_DrawNullModel(currententity); + continue; + } + switch (currentmodel->type) + { + case mod_alias: + R_DrawAliasModel(currententity, currentmodel); + break; + case mod_brush: + R_DrawBrushModel(currententity, currentmodel); + break; + case mod_sprite: + R_DrawSpriteModel(currententity, currentmodel); + break; + default: + R_Printf(PRINT_ALL, "%s: Bad modeltype %d\n", + __func__, currentmodel->type); + return; + } + } + } + + // draw transparent entities + // we could sort these if it ever becomes a problem... + for (i = 0; iflags & RF_TRANSLUCENT)) + continue; // solid + + if (currententity->flags & RF_BEAM) + { + R_DrawBeam(currententity); + } + else + { + model_t *currentmodel = currententity->model; + + if (!currentmodel) + { + R_DrawNullModel(currententity); + continue; + } + switch (currentmodel->type) + { + case mod_alias: + R_DrawAliasModel(currententity, currentmodel); + break; + case mod_brush: + R_DrawBrushModel(currententity, currentmodel); + break; + case mod_sprite: + R_DrawSpriteModel(currententity, currentmodel); + break; + default: + R_Printf(PRINT_ALL, "%s: Bad modeltype %d\n", + __func__, currentmodel->type); + return; + } + } + } +} + +/* +** Vk_DrawParticles +** +*/ +void Vk_DrawParticles(int num_particles, const particle_t particles[], const unsigned colortable[768]) +{ + typedef struct { + float x,y,z,r,g,b,a,u,v; + } pvertex; + + const particle_t *p; + int i; + vec3_t up, right; + byte color[4]; + pvertex* currentvertex; + + if (!num_particles) + return; + + VectorScale(vup, 1.5, up); + VectorScale(vright, 1.5, right); + + static pvertex visibleParticles[MAX_PARTICLES * 3]; + + if (num_particles > MAX_PARTICLES) + { + num_particles = MAX_PARTICLES; + } + + currentvertex = visibleParticles; + for (p = particles, i = 0; i < num_particles; i++, p++) + { + float scale; + + // hack a scale up to keep particles from disapearing + scale = (p->origin[0] - r_origin[0]) * vpn[0] + + (p->origin[1] - r_origin[1]) * vpn[1] + + (p->origin[2] - r_origin[2]) * vpn[2]; + + if (scale < 20) + scale = 1; + else + scale = 1 + scale * 0.004; + + *(int *)color = colortable[p->color]; + + float r = color[0] / 255.f; + float g = color[1] / 255.f; + float b = color[2] / 255.f; + + currentvertex->x = p->origin[0]; + currentvertex->y = p->origin[1]; + currentvertex->z = p->origin[2]; + currentvertex->r = r; + currentvertex->g = g; + currentvertex->b = b; + currentvertex->a = p->alpha; + currentvertex->u = 0.0625; + currentvertex->v = 0.0625; + currentvertex++; + + currentvertex->x = p->origin[0] + up[0] * scale; + currentvertex->y = p->origin[1] + up[1] * scale; + currentvertex->z = p->origin[2] + up[2] * scale; + currentvertex->r = r; + currentvertex->g = g; + currentvertex->b = b; + currentvertex->a = p->alpha; + currentvertex->u = 1.0625; + currentvertex->v = 0.0625; + currentvertex++; + + currentvertex->x = p->origin[0] + right[0] * scale; + currentvertex->y = p->origin[1] + right[1] * scale; + currentvertex->z = p->origin[2] + right[2] * scale; + currentvertex->r = r; + currentvertex->g = g; + currentvertex->b = b; + currentvertex->a = p->alpha; + currentvertex->u = 0.0625; + currentvertex->v = 1.0625; + currentvertex++; + } + + QVk_BindPipeline(&vk_drawParticlesPipeline); + + VkBuffer vbo; + VkDeviceSize vboOffset; + uint8_t *vertData = QVk_GetVertexBuffer((currentvertex - visibleParticles) * sizeof(pvertex), &vbo, &vboOffset); + memcpy(vertData, &visibleParticles, (currentvertex - visibleParticles) * sizeof(pvertex)); + + float gamma = 2.1F - vid_gamma->value; + + vkCmdPushConstants(vk_activeCmdbuffer, vk_drawTexQuadPipeline[vk_state.current_renderpass].layout, + VK_SHADER_STAGE_FRAGMENT_BIT, 17 * sizeof(float), sizeof(gamma), &gamma); + + if (vk_custom_particles->value == 2) + { + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, + vk_drawParticlesPipeline.layout, 0, 1, + &r_squaretexture->vk_texture.descriptorSet, 0, NULL); + } + else + { + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, + vk_drawParticlesPipeline.layout, 0, 1, + &r_particletexture->vk_texture.descriptorSet, 0, NULL); + } + + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + vkCmdDraw(vk_activeCmdbuffer, (currentvertex - visibleParticles), 1, 0, 0); +} + +/* +=============== +R_DrawParticles +=============== +*/ +void R_DrawParticles (void) +{ + if (vk_custom_particles->value == 1) + { + int i; + unsigned char color[4]; + const particle_t *p; + + if (!r_newrefdef.num_particles) + return; + + typedef struct { + float x, y, z, r, g, b, a; + } ppoint; + + struct { + float particleSize; + float particleScale; + float minPointSize; + float maxPointSize; + float att_a; + float att_b; + float att_c; + } particleUbo; + + // Particle size needs to be scaled down proportionally to vk_pixel_size. + const float divisor = (vk_pixel_size->value < 1.0f ? 1.0f : vk_pixel_size->value); + particleUbo.particleSize = vk_particle_size->value / divisor; + particleUbo.particleScale = vid.width * ri.Cvar_Get("viewsize", "100", CVAR_ARCHIVE)->value / 102400; + particleUbo.minPointSize = vk_particle_min_size->value; + particleUbo.maxPointSize = vk_particle_max_size->value; + particleUbo.att_a = vk_particle_att_a->value; + particleUbo.att_b = vk_particle_att_b->value; + particleUbo.att_c = vk_particle_att_c->value; + + static ppoint visibleParticles[MAX_PARTICLES]; + + for (i = 0, p = r_newrefdef.particles; i < r_newrefdef.num_particles; i++, p++) + { + *(int *)color = d_8to24table[p->color]; + + float r = color[0] / 255.f; + float g = color[1] / 255.f; + float b = color[2] / 255.f; + + visibleParticles[i].x = p->origin[0]; + visibleParticles[i].y = p->origin[1]; + visibleParticles[i].z = p->origin[2]; + visibleParticles[i].r = r; + visibleParticles[i].g = g; + visibleParticles[i].b = b; + visibleParticles[i].a = p->alpha; + } + + QVk_BindPipeline(&vk_drawPointParticlesPipeline); + + VkBuffer vbo; + VkDeviceSize vboOffset; + uint32_t uboOffset; + VkDescriptorSet uboDescriptorSet; + uint8_t *vertData = QVk_GetVertexBuffer(sizeof(ppoint) * r_newrefdef.num_particles, &vbo, &vboOffset); + uint8_t *uboData = QVk_GetUniformBuffer(sizeof(particleUbo), &uboOffset, &uboDescriptorSet); + memcpy(vertData, &visibleParticles, sizeof(ppoint) * r_newrefdef.num_particles); + memcpy(uboData, &particleUbo, sizeof(particleUbo)); + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_drawPointParticlesPipeline.layout, 0, 1, &uboDescriptorSet, 1, &uboOffset); + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + vkCmdDraw(vk_activeCmdbuffer, r_newrefdef.num_particles, 1, 0, 0); + } + else + { + Vk_DrawParticles(r_newrefdef.num_particles, r_newrefdef.particles, d_8to24table); + } +} + +/* +============ +R_PolyBlend +============ +*/ +void R_PolyBlend (void) +{ + if (!vk_polyblend->value) + return; + if (!v_blend[3]) + return; + + float polyTransform[] = { 0.f, 0.f, vid.width, vid.height, v_blend[0], v_blend[1], v_blend[2], v_blend[3] }; + QVk_DrawColorRect(polyTransform, sizeof(polyTransform), RP_WORLD); +} + +//======================================================================= + +static int +SignbitsForPlane (cplane_t *out) +{ + int bits, j; + + // for fast box on planeside test + + bits = 0; + for (j = 0; j<3; j++) + { + if (out->normal[j] < 0) + bits |= 1 << j; + } + return bits; +} + + +static void +R_SetFrustum (float fovx, float fovy) +{ + // rotate VPN right by FOV_X/2 degrees + RotatePointAroundVector(frustum[0].normal, vup, vpn, -(90 - fovx / 2)); + // rotate VPN left by FOV_X/2 degrees + RotatePointAroundVector(frustum[1].normal, vup, vpn, 90 - fovx / 2); + // rotate VPN up by FOV_X/2 degrees + RotatePointAroundVector(frustum[2].normal, vright, vpn, 90 - fovy / 2); + // rotate VPN down by FOV_X/2 degrees + RotatePointAroundVector(frustum[3].normal, vright, vpn, -(90 - fovy / 2)); + + for (int i = 0; i < 4; i++) + { + frustum[i].type = PLANE_ANYZ; + frustum[i].dist = DotProduct(r_origin, frustum[i].normal); + frustum[i].signbits = SignbitsForPlane(&frustum[i]); + } +} + +//======================================================================= + +/* +=============== +R_SetupFrame +=============== +*/ +void R_SetupFrame (void) +{ + int i; + mleaf_t *leaf; + + r_framecount++; + + // build the transformation matrix for the given view angles + VectorCopy(r_newrefdef.vieworg, r_origin); + + AngleVectors(r_newrefdef.viewangles, vpn, vright, vup); + + // current viewcluster + if (!(r_newrefdef.rdflags & RDF_NOWORLDMODEL)) + { + r_oldviewcluster = r_viewcluster; + r_oldviewcluster2 = r_viewcluster2; + leaf = Mod_PointInLeaf(r_origin, r_worldmodel); + r_viewcluster = r_viewcluster2 = leaf->cluster; + + // check above and below so crossing solid water doesn't draw wrong + if (!leaf->contents) + { // look down a bit + vec3_t temp; + + VectorCopy(r_origin, temp); + temp[2] -= 16; + leaf = Mod_PointInLeaf(temp, r_worldmodel); + if (!(leaf->contents & CONTENTS_SOLID) && + (leaf->cluster != r_viewcluster2)) + r_viewcluster2 = leaf->cluster; + } + else + { // look up a bit + vec3_t temp; + + VectorCopy(r_origin, temp); + temp[2] += 16; + leaf = Mod_PointInLeaf(temp, r_worldmodel); + if (!(leaf->contents & CONTENTS_SOLID) && + (leaf->cluster != r_viewcluster2)) + r_viewcluster2 = leaf->cluster; + } + } + + for (i = 0; i < 4; i++) + v_blend[i] = r_newrefdef.blend[i]; + + c_brush_polys = 0; + c_alias_polys = 0; + + // clear out the portion of the screen that the NOWORLDMODEL defines + // unlike OpenGL, draw a rectangle in proper location - it's easier to do in Vulkan + if (r_newrefdef.rdflags & RDF_NOWORLDMODEL) + { + float clearArea[] = { (float)r_newrefdef.x / vid.width, (float)r_newrefdef.y / vid.height, + (float)r_newrefdef.width / vid.width, (float)r_newrefdef.height / vid.height, + .3f, .3f, .3f, 1.f }; + QVk_DrawColorRect(clearArea, sizeof(clearArea), RP_UI); + } +} + +void Mat_Identity(float *matrix) +{ + matrix[0] = 1.f; + matrix[1] = 0.f; + matrix[2] = 0.f; + matrix[3] = 0.f; + matrix[4] = 0.f; + matrix[5] = 1.f; + matrix[6] = 0.f; + matrix[7] = 0.f; + matrix[8] = 0.f; + matrix[9] = 0.f; + matrix[10] = 1.f; + matrix[11] = 0.f; + matrix[12] = 0.f; + matrix[13] = 0.f; + matrix[14] = 0.f; + matrix[15] = 1.f; +} + +void Mat_Mul(float *m1, float *m2, float *res) +{ + float mul[16] = { m1[0] * m2[0] + m1[1] * m2[4] + m1[2] * m2[8] + m1[3] * m2[12], + m1[0] * m2[1] + m1[1] * m2[5] + m1[2] * m2[9] + m1[3] * m2[13], + m1[0] * m2[2] + m1[1] * m2[6] + m1[2] * m2[10] + m1[3] * m2[14], + m1[0] * m2[3] + m1[1] * m2[7] + m1[2] * m2[11] + m1[3] * m2[15], + m1[4] * m2[0] + m1[5] * m2[4] + m1[6] * m2[8] + m1[7] * m2[12], + m1[4] * m2[1] + m1[5] * m2[5] + m1[6] * m2[9] + m1[7] * m2[13], + m1[4] * m2[2] + m1[5] * m2[6] + m1[6] * m2[10] + m1[7] * m2[14], + m1[4] * m2[3] + m1[5] * m2[7] + m1[6] * m2[11] + m1[7] * m2[15], + m1[8] * m2[0] + m1[9] * m2[4] + m1[10] * m2[8] + m1[11] * m2[12], + m1[8] * m2[1] + m1[9] * m2[5] + m1[10] * m2[9] + m1[11] * m2[13], + m1[8] * m2[2] + m1[9] * m2[6] + m1[10] * m2[10] + m1[11] * m2[14], + m1[8] * m2[3] + m1[9] * m2[7] + m1[10] * m2[11] + m1[11] * m2[15], + m1[12] * m2[0] + m1[13] * m2[4] + m1[14] * m2[8] + m1[15] * m2[12], + m1[12] * m2[1] + m1[13] * m2[5] + m1[14] * m2[9] + m1[15] * m2[13], + m1[12] * m2[2] + m1[13] * m2[6] + m1[14] * m2[10] + m1[15] * m2[14], + m1[12] * m2[3] + m1[13] * m2[7] + m1[14] * m2[11] + m1[15] * m2[15] + }; + + memcpy(res, mul, sizeof(float) * 16); +} + +void Mat_Translate(float *matrix, float x, float y, float z) +{ + float t[16] = { 1.f, 0.f, 0.f, 0.f, + 0.f, 1.f, 0.f, 0.f, + 0.f, 0.f, 1.f, 0.f, + x, y, z, 1.f }; + + Mat_Mul(matrix, t, matrix); +} + +void Mat_Rotate(float *matrix, float deg, float x, float y, float z) +{ + double c = cos(deg * M_PI / 180.0); + double s = sin(deg * M_PI / 180.0); + double cd = 1.0 - c; + vec3_t r = { x, y, z }; + VectorNormalize(r); + + float rot[16] = { r[0]*r[0]*cd + c, r[1]*r[0]*cd + r[2]*s, r[0]*r[2]*cd - r[1]*s, 0.f, + r[0]*r[1]*cd - r[2]*s, r[1]*r[1]*cd + c, r[1]*r[2]*cd + r[0]*s, 0.f, + r[0]*r[2]*cd + r[1]*s, r[1]*r[2]*cd - r[0]*s, r[2]*r[2]*cd + c, 0.f, + 0.f, 0.f, 0.f, 1.f + }; + + Mat_Mul(matrix, rot, matrix); +} + +void Mat_Scale(float *matrix, float x, float y, float z) +{ + float s[16] = { x, 0.f, 0.f, 0.f, + 0.f, y, 0.f, 0.f, + 0.f, 0.f, z, 0.f, + 0.f, 0.f, 0.f, 1.f + }; + + Mat_Mul(matrix, s, matrix); +} + +void Mat_Perspective(float *matrix, float *correction_matrix, float fovy, float aspect, + float zNear, float zFar) +{ + float xmin, xmax, ymin, ymax; + + ymax = zNear * tan(fovy * M_PI / 360.0); + ymin = -ymax; + + xmin = ymin * aspect; + xmax = ymax * aspect; + + xmin += -(2 * vk_state.camera_separation) / zNear; + xmax += -(2 * vk_state.camera_separation) / zNear; + + float proj[16]; + memset(proj, 0, sizeof(float) * 16); + proj[0] = 2.f * zNear / (xmax - xmin); + proj[2] = (xmax + xmin) / (xmax - xmin); + proj[5] = 2.f * zNear / (ymax - ymin); + proj[6] = (ymax + ymin) / (ymax - ymin); + proj[10] = -(zFar + zNear) / (zFar - zNear); + proj[11] = -1.f; + proj[14] = -2.f * zFar * zNear / (zFar - zNear); + + // Convert projection matrix to Vulkan coordinate system (https://matthewwellings.com/blog/the-new-vulkan-coordinate-system/) + Mat_Mul(proj, correction_matrix, matrix); +} + +void Mat_Ortho(float *matrix, float left, float right, float bottom, float top, + float zNear, float zFar) +{ + float proj[16]; + memset(proj, 0, sizeof(float) * 16); + proj[0] = 2.f / (right - left); + proj[3] = (right + left) / (right - left); + proj[5] = 2.f / (top - bottom); + proj[7] = (top + bottom) / (top - bottom); + proj[10] = -2.f / (zFar - zNear); + proj[11] = -(zFar + zNear) / (zFar - zNear); + proj[15] = 1.f; + + // Convert projection matrix to Vulkan coordinate system (https://matthewwellings.com/blog/the-new-vulkan-coordinate-system/) + Mat_Mul(proj, r_vulkan_correction, matrix); +} + +/* +============= +R_SetupVulkan +============= +*/ +static void +R_SetupVulkan (void) +{ + float r_proj_aspect; + float r_proj_fovx; + float r_proj_fovy; + int x, x2, y2, y, w, h; + float dist = (r_farsee->value == 0) ? 4096.0f : 8192.0f; + + // + // set up viewport + // + x = floor(r_newrefdef.x * vid.width / vid.width); + x2 = ceil((r_newrefdef.x + r_newrefdef.width) * vid.width / vid.width); + y = floor(vid.height - r_newrefdef.y * vid.height / vid.height); + y2 = ceil(vid.height - (r_newrefdef.y + r_newrefdef.height) * vid.height / vid.height); + + w = x2 - x; + h = y - y2; + + VkViewport viewport = { + .x = x, + .y = vid.height - h - y2, + .width = w, + .height = h, + .minDepth = 0.f, + .maxDepth = 1.f, + }; + + // When rendering the world, reduce viewport size proportionally to vk_pixel_size. + if (vk_state.current_renderpass == RP_WORLD) + { + const float divisor = (vk_pixel_size->value < 1.0f ? 1.0f : vk_pixel_size->value); + viewport.x /= divisor; + viewport.y /= divisor; + viewport.width /= divisor; + viewport.height /= divisor; + } + + vkCmdSetViewport(vk_activeCmdbuffer, 0, 1, &viewport); + + // set up projection matrix + r_proj_fovx = r_newrefdef.fov_x; + r_proj_fovy = r_newrefdef.fov_y; + r_proj_aspect = (float)r_newrefdef.width / r_newrefdef.height; + Mat_Perspective(r_projection_matrix, r_vulkan_correction, r_proj_fovy, r_proj_aspect, 4, dist); + + R_SetFrustum(r_proj_fovx, r_proj_fovy); + + // set up view matrix + Mat_Identity(r_view_matrix); + // put Z going up + Mat_Translate(r_view_matrix, -r_newrefdef.vieworg[0], -r_newrefdef.vieworg[1], -r_newrefdef.vieworg[2]); + Mat_Rotate(r_view_matrix, -r_newrefdef.viewangles[1], 0.f, 0.f, 1.f); + Mat_Rotate(r_view_matrix, -r_newrefdef.viewangles[0], 0.f, 1.f, 0.f); + Mat_Rotate(r_view_matrix, -r_newrefdef.viewangles[2], 1.f, 0.f, 0.f); + Mat_Rotate(r_view_matrix, 90.f, 0.f, 0.f, 1.f); + Mat_Rotate(r_view_matrix, -90.f, 1.f, 0.f, 0.f); + + // precalculate view-projection matrix + Mat_Mul(r_view_matrix, r_projection_matrix, r_viewproj_matrix); + // view-projection matrix will always be stored as the first push constant item, so set no offset + vkCmdPushConstants(vk_activeCmdbuffer, vk_drawTexQuadPipeline[vk_state.current_renderpass].layout, VK_SHADER_STAGE_VERTEX_BIT, 0, sizeof(r_viewproj_matrix), r_viewproj_matrix); +} + +static void R_Flash( void ) +{ + R_PolyBlend (); +} + +/* +================ +RE_RenderView + +r_newrefdef must be set before the first call +================ +*/ +static void RE_RenderView (refdef_t *fd) +{ + if (r_norefresh->value) + return; + + r_newrefdef = *fd; + + if (!r_worldmodel && !(r_newrefdef.rdflags & RDF_NOWORLDMODEL)) + ri.Sys_Error(ERR_DROP, "%s: NULL worldmodel", __func__); + + if (r_speeds->value) + { + c_brush_polys = 0; + c_alias_polys = 0; + } + + VkRect2D scissor = { + .offset = { r_newrefdef.x, r_newrefdef.y }, + .extent = { r_newrefdef.width, r_newrefdef.height } + }; + + // When rendering the world, scale down scissor proportionally to vk_pixel_size. + if (vk_state.current_renderpass == RP_WORLD) + { + const float divisor = (vk_pixel_size->value < 1.0f ? 1.0f : vk_pixel_size->value); + scissor.offset.x = (int32_t)floorf(scissor.offset.x / divisor); + scissor.offset.y = (int32_t)floorf(scissor.offset.y / divisor); + scissor.extent.width = (uint32_t)ceilf(scissor.extent.width / divisor); + scissor.extent.height = (uint32_t)ceilf(scissor.extent.height / divisor); + } + + vkCmdSetScissor(vk_activeCmdbuffer, 0, 1, &scissor); + + R_PushDlights(); + + // added for compatibility sake with OpenGL implementation - don't use it! + if (vk_finish->value) + vkDeviceWaitIdle(vk_device.logical); + + R_SetupFrame(); + + R_SetupVulkan(); + + R_MarkLeaves(); // done here so we know if we're in water + + R_DrawWorld(); + + R_DrawEntitiesOnList(); + + R_RenderDlights(); + + R_DrawParticles(); + + R_DrawAlphaSurfaces(); + + R_Flash(); + + if (r_speeds->value) + { + R_Printf(PRINT_ALL, "%4i wpoly %4i epoly %i tex %i lmaps\n", + c_brush_polys, + c_alias_polys, + c_visible_textures, + c_visible_lightmaps); + } +} + +qboolean RE_EndWorldRenderpass(void) +{ + // still some issues? + if (!vk_frameStarted) + { + // we can't start 2d rendering + return false; + } + + // 3d world has alredy rendered and 2d already initialized + if (world_rendered) + { + return true; + } + + world_rendered = true; + + // finish rendering world view to offsceen buffer + vkCmdEndRenderPass(vk_activeCmdbuffer); + + // apply postprocessing effects to offscreen buffer: + // * underwater view warp if the player is submerged in liquid + // * restore world view to the full screen size when vk_pixel_size is >1.0 + QVk_BeginRenderpass(RP_WORLD_WARP); + float underwaterTime; + if (vk_underwater->value) + { + underwaterTime = (r_newrefdef.rdflags & RDF_UNDERWATER) ? r_newrefdef.time : 0.f; + } + else + { + underwaterTime = 0.f; + }; + float pushConsts[] = + { + underwaterTime, + viewsize->value / 100.0f, + vid.width, + vid.height, + vk_viewport.x, + vk_viewport.y, + (vk_pixel_size->value < 1.0f ? 1.0f : vk_pixel_size->value), + r_newrefdef.x, + r_newrefdef.y, + r_newrefdef.width, + r_newrefdef.height, + }; + vkCmdPushConstants(vk_activeCmdbuffer, vk_worldWarpPipeline.layout, + VK_SHADER_STAGE_FRAGMENT_BIT, 17 * sizeof(float), sizeof(pushConsts), pushConsts); + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_worldWarpPipeline.layout, 0, 1, &vk_colorbuffer.descriptorSet, 0, NULL); + QVk_BindPipeline(&vk_worldWarpPipeline); + // Restore full viewport for future steps. + vkCmdSetViewport(vk_activeCmdbuffer, 0u, 1u, &vk_viewport); + vkCmdSetScissor(vk_activeCmdbuffer, 0u, 1u, &vk_scissor); + vkCmdDraw(vk_activeCmdbuffer, 3, 1, 0, 0); + vkCmdEndRenderPass(vk_activeCmdbuffer); + + // start drawing UI + QVk_BeginRenderpass(RP_UI); + + return true; +} + +static void R_SetVulkan2D (const VkViewport* viewport, const VkRect2D* scissor) +{ + // player configuration screen renders a model using the UI renderpass, so skip finishing RP_WORLD twice + if (!(r_newrefdef.rdflags & RDF_NOWORLDMODEL)) + if(!RE_EndWorldRenderpass()) + // buffers is not initialized + return; + + vkCmdSetViewport(vk_activeCmdbuffer, 0, 1, viewport); + vkCmdSetScissor(vk_activeCmdbuffer, 0, 1, scissor); + + // first, blit offscreen color buffer with warped/postprocessed world view + // skip this step if we're in player config screen since it uses RP_UI and draws directly to swapchain + if (!(r_newrefdef.rdflags & RDF_NOWORLDMODEL)) + { + float pushConsts[] = { vk_postprocess->value, (2.1 - vid_gamma->value)}; + vkCmdPushConstants(vk_activeCmdbuffer, vk_postprocessPipeline.layout, + VK_SHADER_STAGE_FRAGMENT_BIT, 17 * sizeof(float), sizeof(pushConsts), pushConsts); + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_postprocessPipeline.layout, 0, 1, &vk_colorbufferWarp.descriptorSet, 0, NULL); + QVk_BindPipeline(&vk_postprocessPipeline); + vkCmdDraw(vk_activeCmdbuffer, 3, 1, 0, 0); + } +} + + +/* +==================== +R_SetLightLevel + +==================== +*/ +static void +R_SetLightLevel (void) +{ + vec3_t shadelight; + + if (r_newrefdef.rdflags & RDF_NOWORLDMODEL) + return; + + // save off light value for server to look at (BIG HACK!) + + R_LightPoint(r_newrefdef.vieworg, shadelight, NULL); + + // pick the greatest component, which should be the same + // as the mono value returned by software + if (shadelight[0] > shadelight[1]) + { + if (shadelight[0] > shadelight[2]) + r_lightlevel->value = 150 * shadelight[0]; + else + r_lightlevel->value = 150 * shadelight[2]; + } + else + { + if (shadelight[1] > shadelight[2]) + r_lightlevel->value = 150 * shadelight[1]; + else + r_lightlevel->value = 150 * shadelight[2]; + } +} + +/* +===================== +RE_RenderFrame + +===================== +*/ +static void +RE_RenderFrame (refdef_t *fd) +{ + if (!vk_frameStarted) + return; + + RE_RenderView( fd ); + R_SetLightLevel (); + R_SetVulkan2D (&vk_viewport, &vk_scissor); +} + + +static void +R_Register( void ) +{ + r_lefthand = ri.Cvar_Get("hand", "0", CVAR_USERINFO | CVAR_ARCHIVE); + r_norefresh = ri.Cvar_Get("r_norefresh", "0", 0); + r_fullbright = ri.Cvar_Get("r_fullbright", "0", 0); + r_drawentities = ri.Cvar_Get("r_drawentities", "1", 0); + r_drawworld = ri.Cvar_Get("r_drawworld", "1", 0); + r_novis = ri.Cvar_Get("r_novis", "0", 0); + r_nocull = ri.Cvar_Get("r_nocull", "0", 0); + r_lerpmodels = ri.Cvar_Get("r_lerpmodels", "1", 0); + r_speeds = ri.Cvar_Get("r_speeds", "0", 0); + r_lightlevel = ri.Cvar_Get("r_lightlevel", "0", 0); + r_mode = ri.Cvar_Get("r_mode", "11", CVAR_ARCHIVE); + r_vsync = ri.Cvar_Get("r_vsync", "0", CVAR_ARCHIVE); + r_gunfov = ri.Cvar_Get("r_gunfov", "80", CVAR_ARCHIVE); + r_farsee = ri.Cvar_Get("r_farsee", "0", CVAR_LATCH | CVAR_ARCHIVE); + r_customwidth = ri.Cvar_Get("r_customwidth", "1024", CVAR_ARCHIVE); + r_customheight = ri.Cvar_Get("r_customheight", "768", CVAR_ARCHIVE); + + vk_overbrightbits = ri.Cvar_Get("vk_overbrightbits", "1.0", CVAR_ARCHIVE); + vk_validation = ri.Cvar_Get("vk_validation", "0", CVAR_ARCHIVE); + vk_picmip = ri.Cvar_Get("vk_picmip", "0", 0); + vk_skymip = ri.Cvar_Get("vk_skymip", "0", 0); + vk_flashblend = ri.Cvar_Get("vk_flashblend", "0", 0); + vk_finish = ri.Cvar_Get("vk_finish", "0", CVAR_ARCHIVE); + r_clear = ri.Cvar_Get("r_clear", "0", CVAR_ARCHIVE); + r_lockpvs = ri.Cvar_Get("r_lockpvs", "0", 0); + vk_polyblend = ri.Cvar_Get("vk_polyblend", "1", 0); + r_modulate = ri.Cvar_Get("r_modulate", "1", CVAR_ARCHIVE); + vk_shadows = ri.Cvar_Get("r_shadows", "0", CVAR_ARCHIVE); + vk_pixel_size = ri.Cvar_Get("vk_pixel_size", "1", CVAR_ARCHIVE); + vk_particle_size = ri.Cvar_Get("vk_particle_size", "40", CVAR_ARCHIVE); + vk_particle_att_a = ri.Cvar_Get("vk_particle_att_a", "0.01", CVAR_ARCHIVE); + vk_particle_att_b = ri.Cvar_Get("vk_particle_att_b", "0.0", CVAR_ARCHIVE); + vk_particle_att_c = ri.Cvar_Get("vk_particle_att_c", "0.01", CVAR_ARCHIVE); + vk_particle_min_size = ri.Cvar_Get("vk_particle_min_size", "2", CVAR_ARCHIVE); + vk_particle_max_size = ri.Cvar_Get("vk_particle_max_size", "40", CVAR_ARCHIVE); + vk_custom_particles = ri.Cvar_Get("vk_custom_particles", "1", CVAR_ARCHIVE); + vk_postprocess = ri.Cvar_Get("vk_postprocess", "1", CVAR_ARCHIVE); + vk_dynamic = ri.Cvar_Get("vk_dynamic", "1", 0); + vk_msaa = ri.Cvar_Get("r_msaa_samples", "0", CVAR_ARCHIVE); + vk_showtris = ri.Cvar_Get("vk_showtris", "0", 0); + vk_lightmap = ri.Cvar_Get("vk_lightmap", "0", 0); + vk_texturemode = ri.Cvar_Get("vk_texturemode", "VK_MIPMAP_LINEAR", CVAR_ARCHIVE); + vk_lmaptexturemode = ri.Cvar_Get("vk_lmaptexturemode", "VK_MIPMAP_LINEAR", CVAR_ARCHIVE); + vk_aniso = ri.Cvar_Get("r_anisotropic", "0", CVAR_ARCHIVE); + vk_mip_nearfilter = ri.Cvar_Get("vk_mip_nearfilter", "0", CVAR_ARCHIVE); + vk_sampleshading = ri.Cvar_Get("vk_sampleshading", "1", CVAR_ARCHIVE); + vk_device_idx = ri.Cvar_Get("vk_device", "-1", CVAR_ARCHIVE); + vk_retexturing = ri.Cvar_Get("r_retexturing", "1", CVAR_ARCHIVE); + vk_underwater = ri.Cvar_Get("vk_underwater", "1", CVAR_ARCHIVE); + /* don't bilerp characters and crosshairs */ + vk_nolerp_list = ri.Cvar_Get("r_nolerp_list", "pics/conchars.pcx pics/ch1.pcx pics/ch2.pcx pics/ch3.pcx", 0); + r_fixsurfsky = ri.Cvar_Get("r_fixsurfsky", "0", CVAR_ARCHIVE); + + // clamp vk_msaa to accepted range so that video menu doesn't crash on us + if (vk_msaa->value < 0) + ri.Cvar_Set("r_msaa_samples", "0"); + + vid_fullscreen = ri.Cvar_Get("vid_fullscreen", "0", CVAR_ARCHIVE); + vid_gamma = ri.Cvar_Get("vid_gamma", "1.0", CVAR_ARCHIVE); + viewsize = ri.Cvar_Get("viewsize", "100", CVAR_ARCHIVE); + + ri.Cmd_AddCommand("vk_strings", Vk_Strings_f); + ri.Cmd_AddCommand("vk_mem", Vk_Mem_f); + ri.Cmd_AddCommand("imagelist", Vk_ImageList_f); + ri.Cmd_AddCommand("screenshot", Vk_ScreenShot_f); + ri.Cmd_AddCommand("modellist", Mod_Modellist_f); +} + +static int +Vkimp_SetMode(int *pwidth, int *pheight, int mode, int fullscreen) +{ + R_Printf(PRINT_ALL, "Setting mode %d:", mode); + + /* mode -1 is not in the vid mode table - so we keep the values in pwidth + and pheight and don't even try to look up the mode info */ + if ((mode >= 0) && !ri.Vid_GetModeInfo(pwidth, pheight, mode)) + { + R_Printf(PRINT_ALL, " invalid mode\n"); + return rserr_invalid_mode; + } + + /* We trying to get resolution from desktop */ + if (mode == -2) + { + if(!ri.GLimp_GetDesktopMode(pwidth, pheight)) + { + R_Printf( PRINT_ALL, " can't detect mode\n" ); + return rserr_invalid_mode; + } + } + + R_Printf(PRINT_ALL, " %dx%d (vid_fullscreen %i)\n", *pwidth, *pheight, fullscreen); + + if (!ri.GLimp_InitGraphics(fullscreen, pwidth, pheight)) + { + return rserr_invalid_mode; + } + + return rserr_ok; +} + +/* +================== +R_SetMode +================== +*/ +qboolean R_SetMode (void) +{ + rserr_t err; + int fullscreen; + + r_mode->modified = false; + r_vsync->modified = false; + + fullscreen = (int)vid_fullscreen->value; + vid_gamma->modified = false; + + vk_msaa->modified = false; + r_clear->modified = false; + vk_validation->modified = false; + vk_mip_nearfilter->modified = false; + vk_sampleshading->modified = false; + vk_device_idx->modified = false; + vk_picmip->modified = false; + vk_overbrightbits->modified = false; + // refresh texture samplers + vk_texturemode->modified = true; + vk_lmaptexturemode->modified = true; + + /* a bit hackish approach to enable custom resolutions: + Glimp_SetMode needs these values set for mode -1 */ + vid.width = r_customwidth->value; + vid.height = r_customheight->value; + + if ((err = Vkimp_SetMode((int*)&vid.width, (int*)&vid.height, r_mode->value, fullscreen)) == rserr_ok) + { + vk_state.prev_mode = r_mode->value; + } + else + { + if (err == rserr_invalid_fullscreen) + { + ri.Cvar_SetValue("vid_fullscreen", 0); + R_Printf(PRINT_ALL, "%s() - fullscreen unavailable in this mode\n", __func__); + if (Vkimp_SetMode((int*)&vid.width, (int*)&vid.height, r_mode->value, false) == rserr_ok) + return true; + } + else if (err == rserr_invalid_mode) + { + ri.Cvar_SetValue("r_mode", vk_state.prev_mode); + r_mode->modified = false; + R_Printf(PRINT_ALL, "%s() - invalid mode\n", __func__); + } + + // try setting it back to something safe + if (Vkimp_SetMode((int*)&vid.width, (int*)&vid.height, vk_state.prev_mode, false) != rserr_ok) + { + R_Printf(PRINT_ALL, "%s() - could not revert to safe mode\n", __func__); + return false; + } + } + return true; +} + +/* +=============== +RE_Init +=============== +*/ +static qboolean RE_Init( void ) +{ + + R_Printf(PRINT_ALL, "Refresh: " REF_VERSION "\n"); + R_Printf(PRINT_ALL, "Client: " YQ2VERSION "\n\n"); + + R_Register(); + + // set our "safe" modes + vk_state.prev_mode = 4; + // set video mode/screen resolution + if (!R_SetMode()) + { + R_Printf(PRINT_ALL, "%s() - could not R_SetMode()\n", __func__); + return false; + } + ri.Vid_MenuInit(); + + // print device information during startup + Vk_Strings_f(); + + R_Printf(PRINT_ALL, "Successfully initialized Vulkan!\n"); + + return true; +} + +/* +** RE_ShutdownContext +** +** This routine does all OS specific shutdown procedures for the Vulkan +** subsystem. +** +*/ +static void RE_ShutdownContext( void ) +{ + // Shutdown Vulkan subsystem + QVk_WaitAndShutdownAll(); +} + +/* +=============== +RE_Shutdown +=============== +*/ +void RE_Shutdown (void) +{ + ri.Cmd_RemoveCommand("vk_strings"); + ri.Cmd_RemoveCommand("vk_mem"); + ri.Cmd_RemoveCommand("imagelist"); + ri.Cmd_RemoveCommand("screenshot"); + ri.Cmd_RemoveCommand( "modellist" ); + + QVk_WaitAndShutdownAll(); +} + +/* +===================== +RE_BeginFrame +===================== +*/ +static void +RE_BeginFrame( float camera_separation ) +{ + // world has not rendered yet + world_rendered = false; + + /* VK hasn't been initialized yet. I'm pretty sure that + we can't get here without having called QVk_Init(), + but better save than sorry. */ + if (!vk_initialized) + { + vk_frameStarted = false; + return; + } + + /* Some GPU drivers set a maximum extent size of 0x0 when + the window is minimized. But a swapchain with an extent + size of 0x0 is invalid. This leads to the following + problem: + + 1. The window is minimized, the extent size changes and + the Vulkan state get's corrupted. QVk_EndFrame() + above or QVk_BeginFrame() below detect the Vulkan + state as corrupted, set vk_frameStated to false and + request a restart by setting vk_restartNeeded to + true. + 2. RE_EndFrame() triggers the restart. QVk_Shutdown() + is successfull, but QVk_Init() can't create a valid + swapchains and errors out. An incomplete internal + renderer state is left behind. The only way out is + to trigger a full render restart. + 3. The full renderer restart would lead to a restart + loop: Restart -> QVk_Init() fails -> restart -> ... + The only alternative is not to restart. Instead the + renderer could error out, that would print an error + message and quit the client. + + Work around this by not starting the frame or restarting + the renderer, as long as the maximum extent size is 0x0. + This is part 1 (the state corruption war detect in the + last frame), part 2 (the state corruption was detected in + the current frame) is in RE_EndFrame(). */ + if (vk_restartNeeded) + { + if (!QVk_CheckExtent()) + { + vk_frameStarted = false; + return; + } + } + + // if ri.Sys_Error() had been issued mid-frame, we might end up here without properly submitting the image, so call QVk_EndFrame to be safe + if (QVk_EndFrame(true) != VK_SUCCESS) + vk_restartNeeded = true; + + /* + ** change modes if necessary + */ + if (r_mode->modified || vk_msaa->modified || r_clear->modified || vk_picmip->modified || + vk_validation->modified || vk_texturemode->modified || vk_lmaptexturemode->modified || + vk_aniso->modified || vk_mip_nearfilter->modified || vk_sampleshading->modified || + r_vsync->modified || vk_device_idx->modified || vk_overbrightbits->modified) + { + if (vk_texturemode->modified || vk_lmaptexturemode->modified || vk_aniso->modified) + { + if (vk_texturemode->modified || vk_aniso->modified) + { + Vk_TextureMode(vk_texturemode->string); + vk_texturemode->modified = false; + } + if (vk_lmaptexturemode->modified || vk_aniso->modified) + { + Vk_LmapTextureMode(vk_lmaptexturemode->string); + vk_lmaptexturemode->modified = false; + } + + vk_aniso->modified = false; + } + } + + if (QVk_BeginFrame(&vk_viewport, &vk_scissor) != VK_SUCCESS) + vk_restartNeeded = true; + else + QVk_BeginRenderpass(RP_WORLD); +} + +/* +===================== +RE_EndFrame +===================== +*/ +static void +RE_EndFrame( void ) +{ + if (QVk_EndFrame(false) != VK_SUCCESS) + vk_restartNeeded = true; + + // world has not rendered yet + world_rendered = false; + + /* Part two of the 'maximum extent size may be 0x0' + * work around. See the explanation in RE_BeginFrame() + * for details. */ + if (vk_restartNeeded) + { + if (QVk_CheckExtent()) + { + QVk_Restart(); + vk_restartNeeded = false; + } + } +} + +/* +============= +RE_SetPalette +============= +*/ +unsigned r_rawpalette[256]; + +static void +RE_SetPalette (const unsigned char *palette) +{ + int i; + + byte *rp = (byte *)r_rawpalette; + + if (palette) + { + for (i = 0; i < 256; i++) + { + rp[i * 4 + 0] = palette[i * 3 + 0]; + rp[i * 4 + 1] = palette[i * 3 + 1]; + rp[i * 4 + 2] = palette[i * 3 + 2]; + rp[i * 4 + 3] = 0xff; + } + } + else + { + for (i = 0; i < 256; i++) + { + rp[i * 4 + 0] = d_8to24table[i] & 0xff; + rp[i * 4 + 1] = (d_8to24table[i] >> 8) & 0xff; + rp[i * 4 + 2] = (d_8to24table[i] >> 16) & 0xff; + rp[i * 4 + 3] = 0xff; + } + } +} + +/* +** R_DrawBeam +*/ +void R_DrawBeam( entity_t *currententity ) +{ +#define NUM_BEAM_SEGS 6 + + int i; + float r, g, b; + + vec3_t perpvec; + vec3_t direction, normalized_direction; + vec3_t start_points[NUM_BEAM_SEGS], end_points[NUM_BEAM_SEGS]; + vec3_t oldorigin, origin; + + oldorigin[0] = currententity->oldorigin[0]; + oldorigin[1] = currententity->oldorigin[1]; + oldorigin[2] = currententity->oldorigin[2]; + + origin[0] = currententity->origin[0]; + origin[1] = currententity->origin[1]; + origin[2] = currententity->origin[2]; + + normalized_direction[0] = direction[0] = oldorigin[0] - origin[0]; + normalized_direction[1] = direction[1] = oldorigin[1] - origin[1]; + normalized_direction[2] = direction[2] = oldorigin[2] - origin[2]; + + if (VectorNormalize(normalized_direction) == 0) + return; + + PerpendicularVector(perpvec, normalized_direction); + VectorScale(perpvec, currententity->frame / 2, perpvec); + + for (i = 0; i < 6; i++) + { + RotatePointAroundVector(start_points[i], normalized_direction, perpvec, (360.0 / NUM_BEAM_SEGS)*i); + VectorAdd(start_points[i], origin, start_points[i]); + VectorAdd(start_points[i], direction, end_points[i]); + } + + r = (d_8to24table[currententity->skinnum & 0xFF]) & 0xFF; + g = (d_8to24table[currententity->skinnum & 0xFF] >> 8) & 0xFF; + b = (d_8to24table[currententity->skinnum & 0xFF] >> 16) & 0xFF; + + r *= 1 / 255.0F; + g *= 1 / 255.0F; + b *= 1 / 255.0F; + + float color[4] = { r, g, b, currententity->alpha }; + + struct { + float v[3]; + } beamvertex[NUM_BEAM_SEGS*4]; + + for (i = 0; i < NUM_BEAM_SEGS; i++) + { + int idx = i * 4; + beamvertex[idx].v[0] = start_points[i][0]; + beamvertex[idx].v[1] = start_points[i][1]; + beamvertex[idx].v[2] = start_points[i][2]; + + beamvertex[idx + 1].v[0] = end_points[i][0]; + beamvertex[idx + 1].v[1] = end_points[i][1]; + beamvertex[idx + 1].v[2] = end_points[i][2]; + + beamvertex[idx + 2].v[0] = start_points[(i + 1) % NUM_BEAM_SEGS][0]; + beamvertex[idx + 2].v[1] = start_points[(i + 1) % NUM_BEAM_SEGS][1]; + beamvertex[idx + 2].v[2] = start_points[(i + 1) % NUM_BEAM_SEGS][2]; + + beamvertex[idx + 3].v[0] = end_points[(i + 1) % NUM_BEAM_SEGS][0]; + beamvertex[idx + 3].v[1] = end_points[(i + 1) % NUM_BEAM_SEGS][1]; + beamvertex[idx + 3].v[2] = end_points[(i + 1) % NUM_BEAM_SEGS][2]; + } + + QVk_BindPipeline(&vk_drawBeamPipeline); + + VkBuffer vbo; + VkDeviceSize vboOffset; + uint32_t uboOffset; + VkDescriptorSet uboDescriptorSet; + uint8_t *vertData = QVk_GetVertexBuffer(sizeof(beamvertex), &vbo, &vboOffset); + uint8_t *uboData = QVk_GetUniformBuffer(sizeof(color), &uboOffset, &uboDescriptorSet); + memcpy(vertData, beamvertex, sizeof(beamvertex)); + memcpy(uboData, color, sizeof(color)); + + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_drawBeamPipeline.layout, 0, 1, &uboDescriptorSet, 1, &uboOffset); + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + vkCmdDraw(vk_activeCmdbuffer, NUM_BEAM_SEGS * 4, 1, 0, 0); +} + +//=================================================================== + +static int +RE_InitContext(void *win) +{ + char title[40] = {0}; + + SDL_Window *window = (SDL_Window *)win; + + if(window == NULL) + { + R_Printf(PRINT_ALL, "%s() must not be called with NULL argument!", __func__); + return false; + } + + /* Window title - set here so we can display renderer name in it */ + snprintf(title, sizeof(title), "Yamagi Quake II %s - Vulkan Render", YQ2VERSION); + SDL_SetWindowTitle(window, title); + + // window is ready, initialize Vulkan now + QVk_SetWindow(window); + if (!QVk_Init()) + { + R_Printf(PRINT_ALL, "%s() - could not initialize Vulkan!\n", __func__); + return false; + } + + QVk_PostInit(); + + return true; +} + +qboolean Vkimp_CreateSurface(SDL_Window *window) +{ + if (!SDL_Vulkan_CreateSurface(window, vk_instance, &vk_surface)) + { + R_Printf(PRINT_ALL, "%s() SDL_Vulkan_CreateSurface failed: %s", + __func__, SDL_GetError()); + return false; + } + return true; +} + +static qboolean +RE_IsVsyncActive(void) +{ + if (r_vsync->value) + { + return true; + } + else + { + return false; + } +} + +static int RE_PrepareForWindow(void) +{ + if (SDL_Vulkan_LoadLibrary(NULL)) + { + R_Printf(PRINT_ALL, "%s() Loader import failed: %s", __func__, SDL_GetError()); + } + + volkInitializeCustom(SDL_Vulkan_GetVkGetInstanceProcAddr()); + + return SDL_WINDOW_VULKAN; +} + +/* +=============== +GetRefAPI +=============== +*/ +Q2_DLL_EXPORTED refexport_t +GetRefAPI(refimport_t imp) +{ + refexport_t refexport = {0}; + + ri = imp; + + refexport.api_version = API_VERSION; + + refexport.BeginRegistration = RE_BeginRegistration; + refexport.RegisterModel = RE_RegisterModel; + refexport.RegisterSkin = RE_RegisterSkin; + refexport.DrawFindPic = RE_Draw_FindPic; + refexport.SetSky = RE_SetSky; + refexport.EndRegistration = RE_EndRegistration; + + refexport.RenderFrame = RE_RenderFrame; + + refexport.DrawGetPicSize = RE_Draw_GetPicSize; + refexport.DrawPicScaled = RE_Draw_PicScaled; + refexport.DrawStretchPic = RE_Draw_StretchPic; + refexport.DrawCharScaled = RE_Draw_CharScaled; + refexport.DrawTileClear = RE_Draw_TileClear; + refexport.DrawFill = RE_Draw_Fill; + refexport.DrawFadeScreen= RE_Draw_FadeScreen; + + refexport.DrawStretchRaw = RE_Draw_StretchRaw; + + refexport.Init = RE_Init; + refexport.IsVSyncActive = RE_IsVsyncActive; + refexport.Shutdown = RE_Shutdown; + refexport.InitContext = RE_InitContext; + refexport.ShutdownContext = RE_ShutdownContext; + refexport.PrepareForWindow = RE_PrepareForWindow; + + refexport.SetPalette = RE_SetPalette; + refexport.BeginFrame = RE_BeginFrame; + refexport.EndWorldRenderpass = RE_EndWorldRenderpass; + refexport.EndFrame = RE_EndFrame; + + // Tell the client that we're unsing the + // new renderer restart API. + ri.Vid_RequestRestart(RESTART_NO); + + Swap_Init (); + + return refexport; +} + +// this is only here so the functions in q_shared.c and q_shwin.c can link +void R_Printf(int level, const char* msg, ...) +{ + va_list argptr; + va_start(argptr, msg); + ri.Com_VPrintf(level, msg, argptr); + va_end(argptr); +} + +void +Sys_Error (char *error, ...) +{ + va_list argptr; + char text[4096]; // MAXPRINTMSG == 4096 + + va_start(argptr, error); + vsnprintf(text, sizeof(text), error, argptr); + va_end(argptr); + + ri.Sys_Error (ERR_FATAL, "%s", text); +} + +void +Com_Printf (char *msg, ...) +{ + va_list argptr; + va_start(argptr, msg); + ri.Com_VPrintf(PRINT_ALL, msg, argptr); + va_end(argptr); +} diff --git a/src/vk/vk_rmisc.c b/src/vk/vk_rmisc.c new file mode 100644 index 0000000..b9cbf4c --- /dev/null +++ b/src/vk/vk_rmisc.c @@ -0,0 +1,325 @@ +/* +Copyright (C) 1997-2001 Id Software, Inc. +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. +*/ + +// vk_misc.c + +#include "header/local.h" + +/* +================== +RE_InitParticleTexture +================== +*/ +static byte dottexture[8][8] = +{ + {0,0,0,0,0,0,0,0}, + {0,0,1,1,0,0,0,0}, + {0,1,1,1,1,0,0,0}, + {0,1,1,1,1,0,0,0}, + {0,0,1,1,0,0,0,0}, + {0,0,0,0,0,0,0,0}, + {0,0,0,0,0,0,0,0}, + {0,0,0,0,0,0,0,0}, +}; + +void RE_InitParticleTexture (void) +{ + int x,y,i; + byte data[8][8][4]; + + // + // particle texture + // + for (x=0 ; x<8 ; x++) + { + for (y=0 ; y<8 ; y++) + { + data[y][x][0] = 255; + data[y][x][1] = 255; + data[y][x][2] = 255; + data[y][x][3] = dottexture[x][y]*255; + } + } + r_particletexture = Vk_LoadPic("***particle***", (byte *)data, + 8, 8, 8, 8, it_sprite, 32); + + // + // particle texture + // + for (x=0 ; x<8 ; x++) + { + for (y=0 ; y<8 ; y++) + { + for (i=0 ; i<4 ; i++) + { + data[y][x][i] = ((y < 4) && (x < 4)) ? 255 : 0; + } + } + } + r_squaretexture = Vk_LoadPic("***square***", (byte *)data, + 8, 8, 8, 8, it_sprite, 32); + + // + // also use this for bad textures, but without alpha + // + for (x=0 ; x<8 ; x++) + { + for (y=0 ; y<8 ; y++) + { + data[y][x][0] = dottexture[x&3][y&3]*255; + data[y][x][1] = 0; + data[y][x][2] = 0; + data[y][x][3] = 255; + } + } + r_notexture = Vk_LoadPic("***r_notexture***", (byte *)data, + 8, 8, 8, 8, it_wall, 32); +} + + +/* +============================================================================== + + SCREEN SHOTS + +============================================================================== +*/ + +/* +================== +Vk_ScreenShot_f +================== +*/ +void Vk_ScreenShot_f (void) +{ + byte *buffer; + int i; + size_t buffSize = vid.width * vid.height * 4; + + if (!vk_device.screenshotSupported) + { + R_Printf(PRINT_ALL, "%s: Screenshots are not supported by this GPU.\n", __func__); + return; + } + + buffer = malloc(buffSize); + + VkExtent2D extent = { + .width = (uint32_t)(vid.width), + .height = (uint32_t)(vid.height), + }; + VkOffset2D offset = { + .x = (int32_t)((vk_swapchain.extent.width - extent.width) / 2u), + .y = (int32_t)((vk_swapchain.extent.height - extent.height) / 2u), + }; + + QVk_ReadPixels(buffer, &offset, &extent); + + // swap rgb to bgr + if (vk_swapchain.format == VK_FORMAT_R8G8B8A8_UNORM || + vk_swapchain.format == VK_FORMAT_R8G8B8A8_SRGB) + { + for (i = 0; i < buffSize; i += 4) + { + buffer[i + 3] = 255; + } + } + else + { + for (i = 0; i < buffSize; i += 4) + { + int temp; + + temp = buffer[i]; + buffer[i] = buffer[i + 2]; + buffer[i + 2] = temp; + buffer[i + 3] = 255; // alpha component + } + } + + ri.Vid_WriteScreenshot(vid.width, vid.height, 4, buffer); + + free(buffer); +} + +/* +** Vk_Strings_f +*/ +void Vk_Strings_f(void) +{ + int i = 0; + + uint32_t numDevices = 0; + int usedDevice = 0; + VkPhysicalDevice *physicalDevices; + VkPhysicalDeviceProperties deviceProperties; + int preferredDevice = (int)vk_device_idx->value; + int msaa = (int)vk_msaa->value; + uint32_t driverMajor = VK_VERSION_MAJOR(vk_device.properties.driverVersion); + uint32_t driverMinor = VK_VERSION_MINOR(vk_device.properties.driverVersion); + uint32_t driverPatch = VK_VERSION_PATCH(vk_device.properties.driverVersion); + + // NVIDIA driver version decoding scheme + if (vk_device.properties.vendorID == 0x10DE) + { + driverMajor = ((uint32_t)(vk_device.properties.driverVersion) >> 22) & 0x3ff; + driverMinor = ((uint32_t)(vk_device.properties.driverVersion) >> 14) & 0x0ff; + + uint32_t secondary = ((uint32_t)(vk_device.properties.driverVersion) >> 6) & 0x0ff; + uint32_t tertiary = vk_device.properties.driverVersion & 0x03f; + + driverPatch = (secondary << 8) | tertiary; + } + + VK_VERIFY(vkEnumeratePhysicalDevices(vk_instance, &numDevices, NULL)); + if (!numDevices) + { + return; + } + physicalDevices = malloc(sizeof(VkPhysicalDevice) * numDevices); + if (!physicalDevices) + { + return; + } + VK_VERIFY(vkEnumeratePhysicalDevices(vk_instance, &numDevices, physicalDevices)); + + if (preferredDevice >= numDevices) + { + preferredDevice = -1; + } + + R_Printf(PRINT_ALL, "------------------------------------\n"); + R_Printf(PRINT_ALL, "Vulkan API: %d.%d\n", VK_VERSION_MAJOR(vk_config.vk_version), + VK_VERSION_MINOR(vk_config.vk_version)); + R_Printf(PRINT_ALL, "Header version: %d\n", VK_HEADER_VERSION); + R_Printf(PRINT_ALL, "Devices found:\n"); + for (i = 0; i < numDevices; ++i) + { + qboolean isPreferred = false; + + vkGetPhysicalDeviceProperties(physicalDevices[i], &deviceProperties); + isPreferred = (preferredDevice == i) || ( + preferredDevice < 0 && + deviceProperties.deviceType == VK_PHYSICAL_DEVICE_TYPE_DISCRETE_GPU + ); + if (isPreferred) { + usedDevice = i; + } + R_Printf(PRINT_ALL, "%s#%d: %s\n", + isPreferred && numDevices > 1 ? "* " : " ", + i, deviceProperties.deviceName); + } + free(physicalDevices); + + R_Printf(PRINT_ALL, "Using device #%d:\n", usedDevice); + R_Printf(PRINT_ALL, " deviceName: %s\n", vk_device.properties.deviceName); + R_Printf(PRINT_ALL, " resolution: %dx%d", vid.width, vid.height); + if (msaa > 0) + { + R_Printf(PRINT_ALL, " (MSAAx%d)\n", msaa); + } + else + { + R_Printf(PRINT_ALL, "\n"); + } +#ifndef __linux__ + // Intel on Windows and MacOS (Linux uses semver for Mesa drivers) + if (vk_device.properties.vendorID == 0x8086) + { + R_Printf(PRINT_ALL, " driverVersion: %d (0x%X)\n", + vk_device.properties.driverVersion, + vk_device.properties.driverVersion); + } + else +#endif + { + R_Printf(PRINT_ALL, " driverVersion: %d.%d.%d (0x%X)\n", + driverMajor, driverMinor, driverPatch, + vk_device.properties.driverVersion); + } + + R_Printf(PRINT_ALL, " apiVersion: %d.%d.%d\n", + VK_VERSION_MAJOR(vk_device.properties.apiVersion), + VK_VERSION_MINOR(vk_device.properties.apiVersion), + VK_VERSION_PATCH(vk_device.properties.apiVersion)); + R_Printf(PRINT_ALL, " deviceID: %d\n", vk_device.properties.deviceID); + R_Printf(PRINT_ALL, " vendorID: 0x%X (%s)\n", + vk_device.properties.vendorID, vk_config.vendor_name); + R_Printf(PRINT_ALL, " deviceType: %s\n", vk_config.device_type); + R_Printf(PRINT_ALL, " gfx/present/transfer: %d/%d/%d\n", + vk_device.gfxFamilyIndex, + vk_device.presentFamilyIndex, + vk_device.transferFamilyIndex); + R_Printf(PRINT_ALL, "Present mode: %s\n", vk_config.present_mode); + R_Printf(PRINT_ALL, "Swapchain image format: %d\n", vk_swapchain.format); + + R_Printf(PRINT_ALL, "Supported present modes: "); + i = 0; + while(vk_config.supported_present_modes[i]) + { + R_Printf(PRINT_ALL, "%s ", vk_config.supported_present_modes[i++]); + } + R_Printf(PRINT_ALL, "\n"); + + R_Printf(PRINT_ALL, "Enabled extensions: "); + i = 0; + while(vk_config.extensions[i]) + { + R_Printf(PRINT_ALL, "%s ", vk_config.extensions[i++]); + } + R_Printf(PRINT_ALL, "\n"); + + R_Printf(PRINT_ALL, "Enabled layers: "); + i = 0; + while(vk_config.layers[i]) + { + R_Printf(PRINT_ALL, "%s ", vk_config.layers[i++]); + } + R_Printf(PRINT_ALL, "\n"); +} + +/* +** Vk_Mem_f +*/ +void Vk_Mem_f(void) +{ + R_Printf(PRINT_ALL, "\nDynamic buffer stats: \n"); + R_Printf(PRINT_ALL, "Vertex : %u/%ukB (%.1f%% max: %ukB)\n", + vk_config.vertex_buffer_usage / 1024, + vk_config.vertex_buffer_size / 1024, + 100.f * vk_config.vertex_buffer_usage / vk_config.vertex_buffer_size, + vk_config.vertex_buffer_max_usage / 1024); + + R_Printf(PRINT_ALL, "Index : %u/%uB (%.1f%% max: %uB)\n", + vk_config.index_buffer_usage, + vk_config.index_buffer_size, + 100.f * vk_config.index_buffer_usage / vk_config.index_buffer_size, + vk_config.index_buffer_max_usage); + R_Printf(PRINT_ALL, "Uniform: %u/%ukB (%.1f%% max: %ukB)\n", + vk_config.uniform_buffer_usage / 1024, + vk_config.uniform_buffer_size / 1024, + 100.f * vk_config.uniform_buffer_usage / vk_config.uniform_buffer_size, + vk_config.uniform_buffer_max_usage / 1024); + R_Printf(PRINT_ALL, "Triangle index: %u/%u (%.1f%% max: %u)\n", + vk_config.triangle_index_usage, + vk_config.triangle_index_count, + 100.f * vk_config.triangle_index_usage / vk_config.triangle_index_count, + vk_config.triangle_index_max_usage); +} diff --git a/src/vk/vk_rsurf.c b/src/vk/vk_rsurf.c new file mode 100644 index 0000000..97b5679 --- /dev/null +++ b/src/vk/vk_rsurf.c @@ -0,0 +1,1324 @@ +/* +Copyright (C) 1997-2001 Id Software, Inc. +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. +*/ + +// vk_rsurf.c: surface-related refresh code +#include + +#include "header/local.h" + +static vec3_t modelorg; // relative to viewpoint + +msurface_t *r_alpha_surfaces; + +#define DYNAMIC_LIGHT_WIDTH 128 +#define DYNAMIC_LIGHT_HEIGHT 128 + +#define LIGHTMAP_BYTES 4 + +#define BLOCK_WIDTH 128 +#define BLOCK_HEIGHT 128 + +int c_visible_lightmaps; +int c_visible_textures; + +typedef struct +{ + int current_lightmap_texture; + + msurface_t *lightmap_surfaces[MAX_LIGHTMAPS]; + + int allocated[BLOCK_WIDTH]; + + // the lightmap texture data needs to be kept in + // main memory so texsubimage can update properly + byte lightmap_buffer[4*BLOCK_WIDTH*BLOCK_HEIGHT]; +} vklightmapstate_t; + +static vklightmapstate_t vk_lms; + + +static void LM_InitBlock( void ); +static void LM_UploadBlock( qboolean dynamic ); +static qboolean LM_AllocBlock (int w, int h, int *x, int *y); + +extern void R_SetCacheState( msurface_t *surf ); +extern void R_BuildLightMap (msurface_t *surf, byte *dest, int stride); + +/* +============================================================= + + BRUSH MODELS + +============================================================= +*/ + +/* +=============== +R_TextureAnimation + +Returns the proper texture for a given time and base texture +=============== +*/ +static image_t *R_TextureAnimation (mtexinfo_t *tex, entity_t *currententity) +{ + int c; + + if (!tex->next) + return tex->image; + + c = currententity->frame % tex->numframes; + while (c) + { + tex = tex->next; + c--; + } + + return tex->image; +} + +/* +================ +DrawVkPoly +================ +*/ +static void DrawVkPoly (vkpoly_t *p, image_t *texture, float *color) +{ + int i; + float *v; + + if (Mesh_VertsRealloc(p->numverts)) + { + ri.Sys_Error(ERR_FATAL, "%s: can't allocate memory", __func__); + } + + v = p->verts[0]; + for (i = 0; i < p->numverts; i++, v += VERTEXSIZE) + { + verts_buffer[i].vertex[0] = v[0]; + verts_buffer[i].vertex[1] = v[1]; + verts_buffer[i].vertex[2] = v[2]; + verts_buffer[i].texCoord[0] = v[3]; + verts_buffer[i].texCoord[1] = v[4]; + } + + QVk_BindPipeline(&vk_drawPolyPipeline); + + VkBuffer vbo; + VkDeviceSize vboOffset; + uint32_t uboOffset; + VkDescriptorSet uboDescriptorSet; + uint8_t *vertData = QVk_GetVertexBuffer(sizeof(polyvert_t) * p->numverts, &vbo, &vboOffset); + uint8_t *uboData = QVk_GetUniformBuffer(sizeof(float) * 4, &uboOffset, &uboDescriptorSet); + memcpy(vertData, verts_buffer, sizeof(polyvert_t) * p->numverts); + memcpy(uboData, color, sizeof(float) * 4); + + VkDescriptorSet descriptorSets[] = { texture->vk_texture.descriptorSet, uboDescriptorSet }; + + float gamma = 2.1F - vid_gamma->value; + + vkCmdPushConstants(vk_activeCmdbuffer, vk_drawTexQuadPipeline[vk_state.current_renderpass].layout, + VK_SHADER_STAGE_FRAGMENT_BIT, 17 * sizeof(float), sizeof(gamma), &gamma); + + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_drawPolyPipeline.layout, 0, 2, descriptorSets, 1, &uboOffset); + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + vkCmdBindIndexBuffer(vk_activeCmdbuffer, QVk_GetTriangleFanIbo((p->numverts - 2) * 3), 0, VK_INDEX_TYPE_UINT16); + vkCmdDrawIndexed(vk_activeCmdbuffer, (p->numverts - 2) * 3, 1, 0, 0, 0); +} + +//============ +//PGM +/* +================ +DrawVkFlowingPoly -- version of DrawVkPoly that handles scrolling texture +================ +*/ +static void DrawVkFlowingPoly (msurface_t *fa, image_t *texture, float *color) +{ + int i; + float *v; + vkpoly_t *p; + float scroll; + + p = fa->polys; + + scroll = -64 * ((r_newrefdef.time / 40.0) - (int)(r_newrefdef.time / 40.0)); + if (scroll == 0.0) + scroll = -64.0; + + if (Mesh_VertsRealloc(p->numverts)) + { + ri.Sys_Error(ERR_FATAL, "%s: can't allocate memory", __func__); + } + + v = p->verts[0]; + for (i = 0; i < p->numverts; i++, v += VERTEXSIZE) + { + verts_buffer[i].vertex[0] = v[0]; + verts_buffer[i].vertex[1] = v[1]; + verts_buffer[i].vertex[2] = v[2]; + verts_buffer[i].texCoord[0] = v[3] + scroll; + verts_buffer[i].texCoord[1] = v[4]; + } + + QVk_BindPipeline(&vk_drawPolyPipeline); + + VkBuffer vbo; + VkDeviceSize vboOffset; + uint32_t uboOffset; + VkDescriptorSet uboDescriptorSet; + uint8_t *vertData = QVk_GetVertexBuffer(sizeof(polyvert_t) * p->numverts, &vbo, &vboOffset); + uint8_t *uboData = QVk_GetUniformBuffer(sizeof(float) * 4, &uboOffset, &uboDescriptorSet); + memcpy(vertData, verts_buffer, sizeof(polyvert_t) * p->numverts); + memcpy(uboData, color, sizeof(float) * 4); + + VkDescriptorSet descriptorSets[] = { texture->vk_texture.descriptorSet, uboDescriptorSet }; + + float gamma = 2.1F - vid_gamma->value; + + vkCmdPushConstants(vk_activeCmdbuffer, vk_drawTexQuadPipeline[vk_state.current_renderpass].layout, + VK_SHADER_STAGE_FRAGMENT_BIT, 17 * sizeof(float), sizeof(gamma), &gamma); + + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_drawPolyPipeline.layout, 0, 2, descriptorSets, 1, &uboOffset); + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + vkCmdBindIndexBuffer(vk_activeCmdbuffer, QVk_GetTriangleFanIbo((p->numverts - 2) * 3), 0, VK_INDEX_TYPE_UINT16); + vkCmdDrawIndexed(vk_activeCmdbuffer, (p->numverts - 2) * 3, 1, 0, 0, 0); +} +//PGM +//============ + +/* +** R_DrawTriangleOutlines +*/ +static void R_DrawTriangleOutlines (void) +{ + int i, j, k; + vkpoly_t *p; + + if (!vk_showtris->value) + return; + + VkBuffer vbo; + VkDeviceSize vboOffset; + float color[3] = { 1.f, 1.f, 1.f }; + struct { + vec3_t v; + float color[3]; + } triVert[4]; + + QVk_BindPipeline(&vk_showTrisPipeline); + uint32_t uboOffset; + VkDescriptorSet uboDescriptorSet; + uint8_t *uboData = QVk_GetUniformBuffer(sizeof(r_viewproj_matrix), &uboOffset, &uboDescriptorSet); + memcpy(uboData, r_viewproj_matrix, sizeof(r_viewproj_matrix)); + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_showTrisPipeline.layout, 0, 1, &uboDescriptorSet, 1, &uboOffset); + + for (i = 0; i < MAX_LIGHTMAPS; i++) + { + msurface_t *surf; + + for (surf = vk_lms.lightmap_surfaces[i]; surf != 0; surf = surf->lightmapchain) + { + p = surf->polys; + for (; p; p = p->chain) + { + for (j = 2, k = 0; j < p->numverts; j++, k++) + { + triVert[0].v[0] = p->verts[0][0]; + triVert[0].v[1] = p->verts[0][1]; + triVert[0].v[2] = p->verts[0][2]; + memcpy(triVert[0].color, color, sizeof(color)); + + triVert[1].v[0] = p->verts[j - 1][0]; + triVert[1].v[1] = p->verts[j - 1][1]; + triVert[1].v[2] = p->verts[j - 1][2]; + memcpy(triVert[1].color, color, sizeof(color)); + + triVert[2].v[0] = p->verts[j][0]; + triVert[2].v[1] = p->verts[j][1]; + triVert[2].v[2] = p->verts[j][2]; + memcpy(triVert[2].color, color, sizeof(color)); + + triVert[3].v[0] = p->verts[0][0]; + triVert[3].v[1] = p->verts[0][1]; + triVert[3].v[2] = p->verts[0][2]; + memcpy(triVert[3].color, color, sizeof(color)); + + uint8_t *vertData = QVk_GetVertexBuffer(sizeof(triVert), &vbo, &vboOffset); + memcpy(vertData, triVert, sizeof(triVert)); + + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + vkCmdDraw(vk_activeCmdbuffer, 4, 1, 0, 0); + } + } + } + } +} + +/* +================ +R_RenderBrushPoly +================ +*/ +static void R_RenderBrushPoly (msurface_t *fa, float *modelMatrix, float alpha, entity_t *currententity) +{ + int maps; + image_t *image; + qboolean is_dynamic = false; + float color[4] = { 1.f, 1.f, 1.f, alpha }; + c_brush_polys++; + + image = R_TextureAnimation(fa->texinfo, currententity); + + if (fa->flags & SURF_DRAWTURB) + { + color[0] = color[1] = color[2] = vk_state.inverse_intensity; + color[3] = 1.f; + // warp texture, no lightmaps + EmitWaterPolys(fa, image, modelMatrix, color, alpha == 1.f); + return; + } + + //====== + //PGM + if (fa->texinfo->flags & SURF_FLOWING) + DrawVkFlowingPoly(fa, image, color); + else + DrawVkPoly(fa->polys, image, color); + //PGM + //====== + + /* + ** check for lightmap modification + */ + for (maps = 0; maps < MAXLIGHTMAPS && fa->styles[maps] != 255; maps++) + { + if (r_newrefdef.lightstyles[fa->styles[maps]].white != fa->cached_light[maps]) + goto dynamic; + } + + // dynamic this frame or dynamic previously + if (fa->dlightframe == r_framecount) + { + dynamic: + if (vk_dynamic->value) + { + if (!(fa->texinfo->flags & (SURF_SKY | SURF_TRANS33 | SURF_TRANS66 | SURF_WARP))) + { + is_dynamic = true; + } + } + } + + if (is_dynamic) + { + if ((fa->styles[maps] >= 32 || fa->styles[maps] == 0) && (fa->dlightframe != r_framecount)) + { + unsigned temp[34 * 34]; + int smax, tmax; + + smax = (fa->extents[0] >> 4) + 1; + tmax = (fa->extents[1] >> 4) + 1; + + R_BuildLightMap(fa, (void *)temp, smax * 4); + R_SetCacheState(fa); + + QVk_UpdateTextureData(&vk_state.lightmap_textures[fa->lightmaptexturenum], (unsigned char*)temp, fa->light_s, fa->light_t, smax, tmax); + + fa->lightmapchain = vk_lms.lightmap_surfaces[fa->lightmaptexturenum]; + vk_lms.lightmap_surfaces[fa->lightmaptexturenum] = fa; + } + else + { + fa->lightmapchain = vk_lms.lightmap_surfaces[0]; + vk_lms.lightmap_surfaces[0] = fa; + } + } + else + { + fa->lightmapchain = vk_lms.lightmap_surfaces[fa->lightmaptexturenum]; + vk_lms.lightmap_surfaces[fa->lightmaptexturenum] = fa; + } +} + + +/* +================ +R_DrawAlphaSurfaces + +Draw water surfaces and windows. +The BSP tree is waled front to back, so unwinding the chain +of alpha_surfaces will draw back to front, giving proper ordering. +================ +*/ +void R_DrawAlphaSurfaces (void) +{ + msurface_t *s; + float intens; + + // the textures are prescaled up for a better lighting range, + // so scale it back down + intens = vk_state.inverse_intensity; + float color[4] = { intens, intens, intens, 1.f }; + + for (s = r_alpha_surfaces; s; s = s->texturechain) + { + c_brush_polys++; + if (s->texinfo->flags & SURF_TRANS33) + color[3] = 0.33f; + else if (s->texinfo->flags & SURF_TRANS66) + color[3] = 0.66f; + + if (s->flags & SURF_DRAWTURB) + EmitWaterPolys(s, s->texinfo->image, NULL, color, false); + else if (s->texinfo->flags & SURF_FLOWING) // PGM 9/16/98 + DrawVkFlowingPoly(s, s->texinfo->image, color); // PGM + else + DrawVkPoly(s->polys, s->texinfo->image, color); + } + + r_alpha_surfaces = NULL; +} + +/* +================ +DrawTextureChains + +Draw world surfaces (mostly solid with alpha == 1.f) +================ +*/ +static void DrawTextureChains (entity_t *currententity) +{ + int i; + msurface_t *s; + image_t *image; + + c_visible_textures = 0; + + for (i = 0, image = vktextures; i < numvktextures; i++, image++) + { + if (!image->registration_sequence) + continue; + if (!image->texturechain) + continue; + c_visible_textures++; + + for (s = image->texturechain; s; s = s->texturechain) + { + if (!(s->flags & SURF_DRAWTURB)) + R_RenderBrushPoly(s, NULL, 1.f, currententity); + } + } + + for (i = 0, image = vktextures; i < numvktextures; i++, image++) + { + if (!image->registration_sequence) + continue; + s = image->texturechain; + if (!s) + continue; + + for (; s; s = s->texturechain) + { + if (s->flags & SURF_DRAWTURB) + R_RenderBrushPoly(s, NULL, 1.f, currententity); + } + + image->texturechain = NULL; + } +} + + +static void Vk_RenderLightmappedPoly( msurface_t *surf, float *modelMatrix, float alpha, entity_t *currententity ) +{ + int i, nv = surf->polys->numverts; + int map; + float *v; + image_t *image = R_TextureAnimation(surf->texinfo, currententity); + qboolean is_dynamic = false; + unsigned lmtex = surf->lightmaptexturenum; + vkpoly_t *p; + + struct { + float model[16]; + float viewLightmaps; + } lmapPolyUbo; + + lmapPolyUbo.viewLightmaps = vk_lightmap->value ? 1.f : 0.f; + + if (modelMatrix) + { + memcpy(lmapPolyUbo.model, modelMatrix, sizeof(float) * 16); + } + else + { + Mat_Identity(lmapPolyUbo.model); + } + + QVk_BindPipeline(&vk_drawPolyLmapPipeline); + + uint32_t uboOffset; + VkDescriptorSet uboDescriptorSet; + uint8_t *uboData = QVk_GetUniformBuffer(sizeof(lmapPolyUbo), &uboOffset, &uboDescriptorSet); + memcpy(uboData, &lmapPolyUbo, sizeof(lmapPolyUbo)); + + for (map = 0; map < MAXLIGHTMAPS && surf->styles[map] != 255; map++) + { + if (r_newrefdef.lightstyles[surf->styles[map]].white != surf->cached_light[map]) + goto dynamic; + } + + // dynamic this frame or dynamic previously + if (surf->dlightframe == r_framecount) + { + dynamic: + if (vk_dynamic->value) + { + if (!(surf->texinfo->flags & (SURF_SKY | SURF_TRANS33 | SURF_TRANS66 | SURF_WARP))) + { + is_dynamic = true; + } + } + } + + if (Mesh_VertsRealloc(nv)) + { + ri.Sys_Error(ERR_FATAL, "%s: can't allocate memory", __func__); + } + + if (is_dynamic) + { + unsigned temp[128 * 128]; + int smax, tmax; + + if ((surf->styles[map] >= 32 || surf->styles[map] == 0) && (surf->dlightframe != r_framecount)) + { + smax = (surf->extents[0] >> 4) + 1; + tmax = (surf->extents[1] >> 4) + 1; + + R_BuildLightMap(surf, (void *)temp, smax * 4); + R_SetCacheState(surf); + + lmtex = surf->lightmaptexturenum; + QVk_UpdateTextureData(&vk_state.lightmap_textures[surf->lightmaptexturenum], (unsigned char *)temp, surf->light_s, surf->light_t, smax, tmax); + } + else + { + smax = (surf->extents[0] >> 4) + 1; + tmax = (surf->extents[1] >> 4) + 1; + + R_BuildLightMap(surf, (void *)temp, smax * 4); + + lmtex = surf->lightmaptexturenum + DYNLIGHTMAP_OFFSET; + QVk_UpdateTextureData(&vk_state.lightmap_textures[lmtex], (unsigned char *)temp, surf->light_s, surf->light_t, smax, tmax); + } + + c_brush_polys++; + + //========== + //PGM + if (surf->texinfo->flags & SURF_FLOWING) + { + float scroll; + + scroll = -64 * ((r_newrefdef.time / 40.0) - (int)(r_newrefdef.time / 40.0)); + if (scroll == 0.0) + scroll = -64.0; + + VkBuffer vbo; + VkDeviceSize vboOffset; + VkDescriptorSet descriptorSets[] = { image->vk_texture.descriptorSet, uboDescriptorSet, vk_state.lightmap_textures[lmtex].descriptorSet }; + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_drawPolyLmapPipeline.layout, 0, 3, descriptorSets, 1, &uboOffset); + + for (p = surf->polys; p; p = p->chain) + { + v = p->verts[0]; + for (i = 0; i < nv; i++, v += VERTEXSIZE) + { + lmappolyverts_buffer[i].vertex[0] = v[0]; + lmappolyverts_buffer[i].vertex[1] = v[1]; + lmappolyverts_buffer[i].vertex[2] = v[2]; + lmappolyverts_buffer[i].texCoord[0] = v[3] + scroll; + lmappolyverts_buffer[i].texCoord[1] = v[4]; + lmappolyverts_buffer[i].texCoordLmap[0] = v[5]; + lmappolyverts_buffer[i].texCoordLmap[1] = v[6]; + } + + uint8_t *vertData = QVk_GetVertexBuffer(sizeof(lmappolyvert_t) * nv, &vbo, &vboOffset); + memcpy(vertData, lmappolyverts_buffer, sizeof(lmappolyvert_t) * nv); + + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + vkCmdBindIndexBuffer(vk_activeCmdbuffer, QVk_GetTriangleFanIbo((nv - 2) * 3), 0, VK_INDEX_TYPE_UINT16); + vkCmdDrawIndexed(vk_activeCmdbuffer, (nv - 2) * 3, 1, 0, 0, 0); + } + } + else + { + VkBuffer vbo; + VkDeviceSize vboOffset; + VkDescriptorSet descriptorSets[] = { image->vk_texture.descriptorSet, uboDescriptorSet, vk_state.lightmap_textures[lmtex].descriptorSet }; + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_drawPolyLmapPipeline.layout, 0, 3, descriptorSets, 1, &uboOffset); + + for (p = surf->polys; p; p = p->chain) + { + v = p->verts[0]; + for (i = 0; i < nv; i++, v += VERTEXSIZE) + { + lmappolyverts_buffer[i].vertex[0] = v[0]; + lmappolyverts_buffer[i].vertex[1] = v[1]; + lmappolyverts_buffer[i].vertex[2] = v[2]; + lmappolyverts_buffer[i].texCoord[0] = v[3]; + lmappolyverts_buffer[i].texCoord[1] = v[4]; + lmappolyverts_buffer[i].texCoordLmap[0] = v[5]; + lmappolyverts_buffer[i].texCoordLmap[1] = v[6]; + } + + uint8_t *vertData = QVk_GetVertexBuffer(sizeof(lmappolyvert_t) * nv, &vbo, &vboOffset); + memcpy(vertData, lmappolyverts_buffer, sizeof(lmappolyvert_t) * nv); + + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + vkCmdBindIndexBuffer(vk_activeCmdbuffer, QVk_GetTriangleFanIbo((nv - 2) * 3), 0, VK_INDEX_TYPE_UINT16); + vkCmdDrawIndexed(vk_activeCmdbuffer, (nv - 2) * 3, 1, 0, 0, 0); + } + } + //PGM + //========== + } + else + { + c_brush_polys++; + + //========== + //PGM + if (surf->texinfo->flags & SURF_FLOWING) + { + float scroll; + + scroll = -64 * ((r_newrefdef.time / 40.0) - (int)(r_newrefdef.time / 40.0)); + if (scroll == 0.0) + scroll = -64.0; + + for (p = surf->polys; p; p = p->chain) + { + v = p->verts[0]; + for (i = 0; i < nv; i++, v += VERTEXSIZE) + { + lmappolyverts_buffer[i].vertex[0] = v[0]; + lmappolyverts_buffer[i].vertex[1] = v[1]; + lmappolyverts_buffer[i].vertex[2] = v[2]; + lmappolyverts_buffer[i].texCoord[0] = v[3] + scroll; + lmappolyverts_buffer[i].texCoord[1] = v[4]; + lmappolyverts_buffer[i].texCoordLmap[0] = v[5]; + lmappolyverts_buffer[i].texCoordLmap[1] = v[6]; + } + VkBuffer vbo; + VkDeviceSize vboOffset; + uint8_t *vertData = QVk_GetVertexBuffer(sizeof(lmappolyvert_t) * nv, &vbo, &vboOffset); + memcpy(vertData, lmappolyverts_buffer, sizeof(lmappolyvert_t) * nv); + + VkDescriptorSet descriptorSets[] = { image->vk_texture.descriptorSet, uboDescriptorSet, vk_state.lightmap_textures[lmtex].descriptorSet }; + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_drawPolyLmapPipeline.layout, 0, 3, descriptorSets, 1, &uboOffset); + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + vkCmdBindIndexBuffer(vk_activeCmdbuffer, QVk_GetTriangleFanIbo((nv - 2) * 3), 0, VK_INDEX_TYPE_UINT16); + vkCmdDrawIndexed(vk_activeCmdbuffer, (nv - 2) * 3, 1, 0, 0, 0); + } + } + else + { + //PGM + //========== + for (p = surf->polys; p; p = p->chain) + { + v = p->verts[0]; + for (i = 0; i < nv; i++, v += VERTEXSIZE) + { + lmappolyverts_buffer[i].vertex[0] = v[0]; + lmappolyverts_buffer[i].vertex[1] = v[1]; + lmappolyverts_buffer[i].vertex[2] = v[2]; + lmappolyverts_buffer[i].texCoord[0] = v[3]; + lmappolyverts_buffer[i].texCoord[1] = v[4]; + lmappolyverts_buffer[i].texCoordLmap[0] = v[5]; + lmappolyverts_buffer[i].texCoordLmap[1] = v[6]; + } + VkBuffer vbo; + VkDeviceSize vboOffset; + uint8_t *vertData = QVk_GetVertexBuffer(sizeof(lmappolyvert_t) * nv, &vbo, &vboOffset); + memcpy(vertData, lmappolyverts_buffer, sizeof(lmappolyvert_t) * nv); + + VkDescriptorSet descriptorSets[] = { image->vk_texture.descriptorSet, uboDescriptorSet, vk_state.lightmap_textures[lmtex].descriptorSet }; + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, vk_drawPolyLmapPipeline.layout, 0, 3, descriptorSets, 1, &uboOffset); + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + vkCmdBindIndexBuffer(vk_activeCmdbuffer, QVk_GetTriangleFanIbo((nv - 2) * 3), 0, VK_INDEX_TYPE_UINT16); + vkCmdDrawIndexed(vk_activeCmdbuffer, (nv - 2) * 3, 1, 0, 0, 0); + } + //========== + //PGM + } + //PGM + //========== + } +} + +/* +================= +R_DrawInlineBModel +================= +*/ +static void R_DrawInlineBModel (entity_t *currententity, model_t *currentmodel, float *modelMatrix) +{ + int i; + msurface_t *psurf; + float alpha = 1.f; + + // calculate dynamic lighting for bmodel + if (!vk_flashblend->value) + { + dlight_t *lt; + int k; + + lt = r_newrefdef.dlights; + for (k = 0; knodes + currentmodel->firstnode); + } + } + + psurf = ¤tmodel->surfaces[currentmodel->firstmodelsurface]; + + if (currententity->flags & RF_TRANSLUCENT) + { + alpha = .25f; + } + + // + // draw texture + // + for (i = 0; inummodelsurfaces; i++, psurf++) + { + float dot; + cplane_t *pplane; + + // find which side of the node we are on + pplane = psurf->plane; + + dot = DotProduct(modelorg, pplane->normal) - pplane->dist; + + // draw the polygon + if (((psurf->flags & SURF_PLANEBACK) && (dot < -BACKFACE_EPSILON)) || + (!(psurf->flags & SURF_PLANEBACK) && (dot > BACKFACE_EPSILON))) + { + if (psurf->texinfo->flags & (SURF_TRANS33 | SURF_TRANS66)) + { // add to the translucent chain + psurf->texturechain = r_alpha_surfaces; + r_alpha_surfaces = psurf; + } + else if (!(psurf->flags & SURF_DRAWTURB) && !vk_showtris->value) + { + Vk_RenderLightmappedPoly(psurf, modelMatrix, alpha, currententity); + } + else + { + R_RenderBrushPoly(psurf, modelMatrix, alpha, currententity); + } + } + } +} + +/* +================= +R_DrawBrushModel +================= +*/ +void R_DrawBrushModel (entity_t *currententity, model_t *currentmodel) +{ + vec3_t mins, maxs; + qboolean rotated; + + if (currentmodel->nummodelsurfaces == 0) + return; + + if (currententity->angles[0] || currententity->angles[1] || currententity->angles[2]) + { + int i; + + rotated = true; + for (i = 0; i<3; i++) + { + mins[i] = currententity->origin[i] - currentmodel->radius; + maxs[i] = currententity->origin[i] + currentmodel->radius; + } + } + else + { + rotated = false; + VectorAdd(currententity->origin, currentmodel->mins, mins); + VectorAdd(currententity->origin, currentmodel->maxs, maxs); + } + + if (R_CullBox(mins, maxs)) + return; + + memset(vk_lms.lightmap_surfaces, 0, sizeof(vk_lms.lightmap_surfaces)); + + VectorSubtract(r_newrefdef.vieworg, currententity->origin, modelorg); + if (rotated) + { + vec3_t temp; + vec3_t forward, right, up; + + VectorCopy(modelorg, temp); + AngleVectors(currententity->angles, forward, right, up); + modelorg[0] = DotProduct(temp, forward); + modelorg[1] = -DotProduct(temp, right); + modelorg[2] = DotProduct(temp, up); + } + + currententity->angles[0] = -currententity->angles[0]; // stupid quake bug + currententity->angles[2] = -currententity->angles[2]; // stupid quake bug + float model[16]; + Mat_Identity(model); + R_RotateForEntity(currententity, model); + currententity->angles[0] = -currententity->angles[0]; // stupid quake bug + currententity->angles[2] = -currententity->angles[2]; // stupid quake bug + + R_DrawInlineBModel(currententity, currentmodel, model); +} + +/* +============================================================= + + WORLD MODEL + +============================================================= +*/ + +/* +================ +R_RecursiveWorldNode +================ +*/ +static void R_RecursiveWorldNode (mnode_t *node, entity_t *currententity) +{ + int c, side, sidebit; + cplane_t *plane; + msurface_t *surf; + mleaf_t *pleaf; + float dot; + image_t *image; + + if (node->contents == CONTENTS_SOLID) + return; // solid + + if (node->visframe != r_visframecount) + return; + if (R_CullBox (node->minmaxs, node->minmaxs+3)) + return; + + // if a leaf node, draw stuff + if (node->contents != -1) + { + msurface_t **mark; + + pleaf = (mleaf_t *)node; + + // check for door connected areas + if (r_newrefdef.areabits) + { + if (! (r_newrefdef.areabits[pleaf->area>>3] & (1<<(pleaf->area&7)) ) ) + return; // not visible + } + + mark = pleaf->firstmarksurface; + c = pleaf->nummarksurfaces; + + if (c) + { + do + { + (*mark)->visframe = r_framecount; + mark++; + } while (--c); + } + + return; + } + + // node is just a decision point, so go down the apropriate sides + // find which side of the node we are on + plane = node->plane; + + switch (plane->type) + { + case PLANE_X: + dot = modelorg[0] - plane->dist; + break; + case PLANE_Y: + dot = modelorg[1] - plane->dist; + break; + case PLANE_Z: + dot = modelorg[2] - plane->dist; + break; + default: + dot = DotProduct (modelorg, plane->normal) - plane->dist; + break; + } + + if (dot >= 0) + { + side = 0; + sidebit = 0; + } + else + { + side = 1; + sidebit = SURF_PLANEBACK; + } + + // recurse down the children, front side first + R_RecursiveWorldNode (node->children[side], currententity); + + // draw stuff + for ( c = node->numsurfaces, surf = r_worldmodel->surfaces + node->firstsurface; c ; c--, surf++) + { + if (surf->visframe != r_framecount) + continue; + + if ( (surf->flags & SURF_PLANEBACK) != sidebit ) + continue; // wrong side + + if (surf->texinfo->flags & SURF_SKY) + { // just adds to visible sky bounds + R_AddSkySurface (surf); + } + else if (surf->texinfo->flags & (SURF_TRANS33|SURF_TRANS66)) + { // add to the translucent chain + surf->texturechain = r_alpha_surfaces; + r_alpha_surfaces = surf; + } + else + { + if (!(surf->flags & SURF_DRAWTURB) && !vk_showtris->value) + { + Vk_RenderLightmappedPoly(surf, NULL, 1.f, currententity); + } + else + { + // the polygon is visible, so add it to the texture + // sorted chain + // FIXME: this is a hack for animation + image = R_TextureAnimation(surf->texinfo, currententity); + surf->texturechain = image->texturechain; + image->texturechain = surf; + } + } + } + + // recurse down the back side + R_RecursiveWorldNode (node->children[!side], currententity); +} + + +/* +============= +R_DrawWorld +============= +*/ +void R_DrawWorld (void) +{ + entity_t ent; + + if (!r_drawworld->value) + return; + + if ( r_newrefdef.rdflags & RDF_NOWORLDMODEL ) + return; + + VectorCopy (r_newrefdef.vieworg, modelorg); + + // auto cycle the world frame for texture animation + memset (&ent, 0, sizeof(ent)); + ent.frame = (int)(r_newrefdef.time*2); + + memset (vk_lms.lightmap_surfaces, 0, sizeof(vk_lms.lightmap_surfaces)); + R_ClearSkyBox (); + + R_RecursiveWorldNode (r_worldmodel->nodes, &ent); + + /* + ** theoretically nothing should happen in the next two functions + ** if multitexture is enabled - in practice, this code renders non-transparent liquids! + */ + DrawTextureChains (&ent); + + R_DrawSkyBox (); + + R_DrawTriangleOutlines (); +} + + +/* +=============== +R_MarkLeaves + +Mark the leaves and nodes that are in the PVS for the current +cluster +=============== +*/ +void R_MarkLeaves (void) +{ + byte *vis; + byte fatvis[MAX_MAP_LEAFS/8]; + mnode_t *node; + int i; + mleaf_t *leaf; + + if (r_oldviewcluster == r_viewcluster && r_oldviewcluster2 == r_viewcluster2 && !r_novis->value && r_viewcluster != -1) + return; + + // development aid to let you run around and see exactly where + // the pvs ends + if (r_lockpvs->value) + return; + + r_visframecount++; + r_oldviewcluster = r_viewcluster; + r_oldviewcluster2 = r_viewcluster2; + + if (r_novis->value || r_viewcluster == -1 || !r_worldmodel->vis) + { + // mark everything + for (i=0 ; inumleafs ; i++) + r_worldmodel->leafs[i].visframe = r_visframecount; + for (i=0 ; inumnodes ; i++) + r_worldmodel->nodes[i].visframe = r_visframecount; + return; + } + + vis = Mod_ClusterPVS (r_viewcluster, r_worldmodel); + // may have to combine two clusters because of solid water boundaries + if (r_viewcluster2 != r_viewcluster) + { + int c; + + memcpy (fatvis, vis, (r_worldmodel->numleafs+7)/8); + vis = Mod_ClusterPVS (r_viewcluster2, r_worldmodel); + c = (r_worldmodel->numleafs+31)/32; + for (i=0 ; ileafs ; inumleafs ; i++, leaf++) + { + int cluster; + + cluster = leaf->cluster; + if (cluster == -1) + continue; + if (vis[cluster>>3] & (1<<(cluster&7))) + { + node = (mnode_t *)leaf; + do + { + if (node->visframe == r_visframecount) + break; + node->visframe = r_visframecount; + node = node->parent; + } while (node); + } + } +} + + + +/* +============================================================================= + + LIGHTMAP ALLOCATION + +============================================================================= +*/ + +static void LM_InitBlock( void ) +{ + memset( vk_lms.allocated, 0, sizeof( vk_lms.allocated ) ); +} + +static void LM_UploadBlock( qboolean dynamic ) +{ + int texture; + + if ( dynamic ) + { + texture = 0; + } + else + { + texture = vk_lms.current_lightmap_texture; + } + + if ( dynamic ) + { + int i; + int height = 0; + + for ( i = 0; i < BLOCK_WIDTH; i++ ) + { + if ( vk_lms.allocated[i] > height ) + height = vk_lms.allocated[i]; + } + QVk_UpdateTextureData(&vk_state.lightmap_textures[texture], vk_lms.lightmap_buffer, 0, 0, BLOCK_WIDTH, height); + } + else + { + if (vk_state.lightmap_textures[texture].resource.image != VK_NULL_HANDLE) + QVk_UpdateTextureData(&vk_state.lightmap_textures[texture], vk_lms.lightmap_buffer, 0, 0, BLOCK_WIDTH, BLOCK_HEIGHT); + else + { + QVVKTEXTURE_CLEAR(vk_state.lightmap_textures[texture]); + QVk_CreateTexture(&vk_state.lightmap_textures[texture], vk_lms.lightmap_buffer, + BLOCK_WIDTH, BLOCK_HEIGHT, vk_current_lmap_sampler, false); + QVk_DebugSetObjectName((uint64_t)vk_state.lightmap_textures[texture].resource.image, + VK_OBJECT_TYPE_IMAGE, va("Image: dynamic lightmap #%d", texture)); + QVk_DebugSetObjectName((uint64_t)vk_state.lightmap_textures[texture].imageView, + VK_OBJECT_TYPE_IMAGE_VIEW, va("Image View: dynamic lightmap #%d", texture)); + QVk_DebugSetObjectName((uint64_t)vk_state.lightmap_textures[texture].descriptorSet, + VK_OBJECT_TYPE_DESCRIPTOR_SET, va("Descriptor Set: dynamic lightmap #%d", texture)); + QVk_DebugSetObjectName((uint64_t)vk_state.lightmap_textures[texture].resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, va("Memory: dynamic lightmap #%d", texture)); + } + if ( ++vk_lms.current_lightmap_texture == MAX_LIGHTMAPS ) + ri.Sys_Error( ERR_DROP, "%s() - MAX_LIGHTMAPS exceeded\n", __func__); + } +} + +// returns a texture number and the position inside it +static qboolean LM_AllocBlock (int w, int h, int *x, int *y) +{ + int i, best; + + best = BLOCK_HEIGHT; + + for (i=0 ; i= best) + break; + if (vk_lms.allocated[i+j] > best2) + best2 = vk_lms.allocated[i+j]; + } + if (j == w) + { // this is a valid spot + *x = i; + *y = best = best2; + } + } + + if (best + h > BLOCK_HEIGHT) + return false; + + for (i=0 ; iedges; + lnumverts = fa->numedges; + + VectorClear (total); + // + // draw texture + // + poly = Hunk_Alloc (sizeof(vkpoly_t) + (lnumverts-4) * VERTEXSIZE*sizeof(float)); + poly->next = fa->polys; + poly->flags = fa->flags; + fa->polys = poly; + poly->numverts = lnumverts; + + for (i=0 ; isurfedges[fa->firstedge + i]; + + if (lindex > 0) + { + r_pedge = &pedges[lindex]; + vec = currentmodel->vertexes[r_pedge->v[0]].position; + } + else + { + r_pedge = &pedges[-lindex]; + vec = currentmodel->vertexes[r_pedge->v[1]].position; + } + s = DotProduct (vec, fa->texinfo->vecs[0]) + fa->texinfo->vecs[0][3]; + s /= fa->texinfo->image->width; + + t = DotProduct (vec, fa->texinfo->vecs[1]) + fa->texinfo->vecs[1][3]; + t /= fa->texinfo->image->height; + + VectorAdd (total, vec, total); + VectorCopy (vec, poly->verts[i]); + poly->verts[i][3] = s; + poly->verts[i][4] = t; + + // + // lightmap texture coordinates + // + s = DotProduct (vec, fa->texinfo->vecs[0]) + fa->texinfo->vecs[0][3]; + s -= fa->texturemins[0]; + s += fa->light_s*16; + s += 8; + s /= BLOCK_WIDTH*16; //fa->texinfo->texture->width; + + t = DotProduct (vec, fa->texinfo->vecs[1]) + fa->texinfo->vecs[1][3]; + t -= fa->texturemins[1]; + t += fa->light_t*16; + t += 8; + t /= BLOCK_HEIGHT*16; //fa->texinfo->texture->height; + + poly->verts[i][5] = s; + poly->verts[i][6] = t; + } + + poly->numverts = lnumverts; + +} + +/* +======================== +Vk_CreateSurfaceLightmap +======================== +*/ +void Vk_CreateSurfaceLightmap (msurface_t *surf) +{ + int smax, tmax; + byte *base; + + if (surf->flags & (SURF_DRAWSKY|SURF_DRAWTURB)) + return; + + smax = (surf->extents[0]>>4)+1; + tmax = (surf->extents[1]>>4)+1; + + if ( !LM_AllocBlock( smax, tmax, &surf->light_s, &surf->light_t ) ) + { + LM_UploadBlock( false ); + LM_InitBlock(); + if ( !LM_AllocBlock( smax, tmax, &surf->light_s, &surf->light_t ) ) + { + ri.Sys_Error( ERR_FATAL, "%s: Consecutive calls to LM_AllocBlock(%d,%d) failed\n", __func__, smax, tmax ); + } + } + + surf->lightmaptexturenum = vk_lms.current_lightmap_texture; + + base = vk_lms.lightmap_buffer; + base += (surf->light_t * BLOCK_WIDTH + surf->light_s) * LIGHTMAP_BYTES; + + R_SetCacheState( surf ); + R_BuildLightMap (surf, base, BLOCK_WIDTH*LIGHTMAP_BYTES); +} + + +/* +================== +Vk_BeginBuildingLightmaps + +================== +*/ +void Vk_BeginBuildingLightmaps (model_t *m) +{ + static lightstyle_t lightstyles[MAX_LIGHTSTYLES]; + int i; + + memset(vk_lms.allocated, 0, sizeof(vk_lms.allocated)); + + r_framecount = 1; // no dlightcache + + /* + ** setup the base lightstyles so the lightmaps won't have to be regenerated + ** the first time they're seen + */ + for (i = 0; i < MAX_LIGHTSTYLES; i++) + { + lightstyles[i].rgb[0] = 1; + lightstyles[i].rgb[1] = 1; + lightstyles[i].rgb[2] = 1; + lightstyles[i].white = 3; + } + r_newrefdef.lightstyles = lightstyles; + + vk_lms.current_lightmap_texture = 0; + + /* + ** initialize the dynamic lightmap textures + */ + if (vk_state.lightmap_textures[DYNLIGHTMAP_OFFSET].resource.image == VK_NULL_HANDLE) + { + for (i = DYNLIGHTMAP_OFFSET; i < MAX_LIGHTMAPS*2; i++) + { + unsigned dummy[BLOCK_WIDTH * BLOCK_HEIGHT]; + + QVVKTEXTURE_CLEAR(vk_state.lightmap_textures[i]); + QVk_CreateTexture(&vk_state.lightmap_textures[i], (unsigned char*)dummy, + BLOCK_WIDTH, BLOCK_HEIGHT, vk_current_lmap_sampler, false); + QVk_DebugSetObjectName((uint64_t)vk_state.lightmap_textures[i].resource.image, + VK_OBJECT_TYPE_IMAGE, va("Image: dynamic lightmap #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_state.lightmap_textures[i].imageView, + VK_OBJECT_TYPE_IMAGE_VIEW, va("Image View: dynamic lightmap #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_state.lightmap_textures[i].descriptorSet, + VK_OBJECT_TYPE_DESCRIPTOR_SET, va("Descriptor Set: dynamic lightmap #%d", i)); + QVk_DebugSetObjectName((uint64_t)vk_state.lightmap_textures[i].resource.memory, + VK_OBJECT_TYPE_DEVICE_MEMORY, va("Memory: dynamic lightmap #%d", i)); + } + } +} + +/* +======================= +Vk_EndBuildingLightmaps +======================= +*/ +void Vk_EndBuildingLightmaps (void) +{ + LM_UploadBlock( false ); +} + diff --git a/src/vk/vk_shaders.c b/src/vk/vk_shaders.c new file mode 100644 index 0000000..4de4b53 --- /dev/null +++ b/src/vk/vk_shaders.c @@ -0,0 +1,85 @@ +/* +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +*/ + +#include "header/shaders.h" + +// ingame shaders stored as uint32_t arrays (autogenerated by glslangValidator) +#include "spirv/basic_vert.c" +#include "spirv/basic_frag.c" +#include "spirv/basic_color_quad_vert.c" +#include "spirv/basic_color_quad_frag.c" +#include "spirv/model_vert.c" +#include "spirv/model_frag.c" +#include "spirv/nullmodel_vert.c" +#include "spirv/particle_vert.c" +#include "spirv/point_particle_vert.c" +#include "spirv/point_particle_frag.c" +#include "spirv/sprite_vert.c" +#include "spirv/beam_vert.c" +#include "spirv/skybox_vert.c" +#include "spirv/d_light_vert.c" +#include "spirv/polygon_vert.c" +#include "spirv/polygon_lmap_vert.c" +#include "spirv/polygon_lmap_frag.c" +#include "spirv/polygon_warp_vert.c" +#include "spirv/shadows_vert.c" +#include "spirv/postprocess_vert.c" +#include "spirv/postprocess_frag.c" +#include "spirv/world_warp_vert.c" +#include "spirv/world_warp_frag.c" + +const size_t basic_vert_size = sizeof(basic_vert_spv); +const size_t basic_frag_size = sizeof(basic_frag_spv); + +const size_t basic_color_quad_vert_size = sizeof(basic_color_quad_vert_spv); +const size_t basic_color_quad_frag_size = sizeof(basic_color_quad_frag_spv); + +const size_t model_vert_size = sizeof(model_vert_spv); +const size_t model_frag_size = sizeof(model_frag_spv); + +const size_t nullmodel_vert_size = sizeof(nullmodel_vert_spv); + +const size_t particle_vert_size = sizeof(particle_vert_spv); + +const size_t point_particle_vert_size = sizeof(point_particle_vert_spv); +const size_t point_particle_frag_size = sizeof(point_particle_frag_spv); + +const size_t sprite_vert_size = sizeof(sprite_vert_spv); + +const size_t beam_vert_size = sizeof(beam_vert_spv); + +const size_t skybox_vert_size = sizeof(skybox_vert_spv); + +const size_t d_light_vert_size = sizeof(d_light_vert_spv); + +const size_t polygon_vert_size = sizeof(polygon_vert_spv); + +const size_t polygon_lmap_vert_size = sizeof(polygon_lmap_vert_spv); +const size_t polygon_lmap_frag_size = sizeof(polygon_lmap_frag_spv); + +const size_t polygon_warp_vert_size = sizeof(polygon_warp_vert_spv); + +const size_t shadows_vert_size = sizeof(shadows_vert_spv); + +const size_t postprocess_vert_size = sizeof(postprocess_vert_spv); +const size_t postprocess_frag_size = sizeof(postprocess_frag_spv); + +const size_t world_warp_vert_size = sizeof(world_warp_vert_spv); +const size_t world_warp_frag_size = sizeof(world_warp_frag_spv); diff --git a/src/vk/vk_swapchain.c b/src/vk/vk_swapchain.c new file mode 100644 index 0000000..2ee91de --- /dev/null +++ b/src/vk/vk_swapchain.c @@ -0,0 +1,250 @@ +/* +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +*/ + +#include "header/local.h" + +// internal helper +static const char *presentModeString(VkPresentModeKHR presentMode) +{ +#define PMSTR(r) case VK_ ##r: return "VK_"#r + switch (presentMode) + { + PMSTR(PRESENT_MODE_IMMEDIATE_KHR); + PMSTR(PRESENT_MODE_MAILBOX_KHR); + PMSTR(PRESENT_MODE_FIFO_KHR); + PMSTR(PRESENT_MODE_FIFO_RELAXED_KHR); + default: return ""; + } +#undef PMSTR + return "UNKNOWN PRESENT MODE"; +} + +// internal helper +static VkSurfaceFormatKHR getSwapSurfaceFormat(const VkSurfaceFormatKHR *surfaceFormats, uint32_t formatCount) +{ + VkSurfaceFormatKHR swapSurfaceFormat; + memset(&swapSurfaceFormat, 0, sizeof(swapSurfaceFormat)); + if (!surfaceFormats || !formatCount) + { + return swapSurfaceFormat; + } + + for (size_t i = 0; i < formatCount; ++i) + { + if (surfaceFormats[i].colorSpace == VK_COLOR_SPACE_SRGB_NONLINEAR_KHR && + surfaceFormats[i].format == VK_FORMAT_B8G8R8A8_UNORM) + { + swapSurfaceFormat.colorSpace = surfaceFormats[i].colorSpace; + swapSurfaceFormat.format = surfaceFormats[i].format; + return swapSurfaceFormat; + } + } + // no preferred format, so get the first one from list + swapSurfaceFormat.colorSpace = surfaceFormats[0].colorSpace; + swapSurfaceFormat.format = surfaceFormats[0].format; + + return swapSurfaceFormat; +} + +// internal helper +// look to https://www.khronos.org/registry/vulkan/specs/1.2-extensions/man/html/VkPresentModeKHR.html for more information +static VkPresentModeKHR getSwapPresentMode(const VkPresentModeKHR *presentModes, uint32_t presentModesCount, VkPresentModeKHR desiredMode) +{ + // PRESENT_MODE_FIFO_KHR is guaranteed to exist due to spec requirements + VkPresentModeKHR usedPresentMode = VK_PRESENT_MODE_FIFO_KHR; + + if (!presentModes) + { + return usedPresentMode; + } + + // check if the desired present mode is supported + for (uint32_t i = 0; i < presentModesCount; ++i) + { + // mode supported, nothing to do here + if (presentModes[i] == desiredMode) + { + vk_config.present_mode = presentModeString(desiredMode); + R_Printf(PRINT_ALL, "...using present mode: %s\n", vk_config.present_mode); + return desiredMode; + } + } + + // preferred present mode not found - choose the next best thing + for (uint32_t i = 0; i < presentModesCount; ++i) + { + // always prefer mailbox for triple buffering with whole image replace + if (presentModes[i] == VK_PRESENT_MODE_MAILBOX_KHR) + { + usedPresentMode = presentModes[i]; + break; + } + // prefer immediate update with tearing + else if (presentModes[i] == VK_PRESENT_MODE_IMMEDIATE_KHR) + { + usedPresentMode = presentModes[i]; + } + } + + vk_config.present_mode = presentModeString(usedPresentMode); + R_Printf(PRINT_ALL, "...present mode %s not supported, using present mode: %s\n", presentModeString(desiredMode), vk_config.present_mode); + return usedPresentMode; +} + +// internal helper +static VkCompositeAlphaFlagBitsKHR getSupportedCompositeAlpha(VkCompositeAlphaFlagsKHR supportedFlags) +{ + VkCompositeAlphaFlagBitsKHR compositeAlphaFlags[] = { + VK_COMPOSITE_ALPHA_OPAQUE_BIT_KHR, + VK_COMPOSITE_ALPHA_PRE_MULTIPLIED_BIT_KHR, + VK_COMPOSITE_ALPHA_POST_MULTIPLIED_BIT_KHR, + VK_COMPOSITE_ALPHA_INHERIT_BIT_KHR + }; + + for (int i = 0; i < 4; ++i) + { + if (supportedFlags & compositeAlphaFlags[i]) + return compositeAlphaFlags[i]; + } + + return VK_COMPOSITE_ALPHA_INHERIT_BIT_KHR; +} + +qboolean QVk_CheckExtent(void) +{ + VkSurfaceCapabilitiesKHR surfaceCaps; + VK_VERIFY(vkGetPhysicalDeviceSurfaceCapabilitiesKHR(vk_device.physical, vk_surface, &surfaceCaps)); + + if (surfaceCaps.currentExtent.width == 0 || surfaceCaps.currentExtent.height == 0) + { + return false; + } + + return true; +} + +VkResult QVk_CreateSwapchain() +{ + VkSurfaceCapabilitiesKHR surfaceCaps; + VkSurfaceFormatKHR *surfaceFormats = NULL; + VkPresentModeKHR *presentModes = NULL; + uint32_t formatCount, presentModesCount; + VK_VERIFY(vkGetPhysicalDeviceSurfaceCapabilitiesKHR(vk_device.physical, vk_surface, &surfaceCaps)); + VK_VERIFY(vkGetPhysicalDeviceSurfaceFormatsKHR(vk_device.physical, vk_surface, &formatCount, NULL)); + VK_VERIFY(vkGetPhysicalDeviceSurfacePresentModesKHR(vk_device.physical, vk_surface, &presentModesCount, NULL)); + + if (formatCount > 0) + { + surfaceFormats = (VkSurfaceFormatKHR *)malloc(formatCount * sizeof(VkSurfaceFormatKHR)); + VK_VERIFY(vkGetPhysicalDeviceSurfaceFormatsKHR(vk_device.physical, vk_surface, &formatCount, surfaceFormats)); + } + + if (presentModesCount > 0) + { + presentModes = (VkPresentModeKHR *)malloc(presentModesCount * sizeof(VkPresentModeKHR)); + VK_VERIFY(vkGetPhysicalDeviceSurfacePresentModesKHR(vk_device.physical, vk_surface, &presentModesCount, presentModes)); + + R_Printf(PRINT_ALL, "Supported present modes: "); + for (int i = 0; i < presentModesCount; i++) + { + R_Printf(PRINT_ALL, "%s ", presentModeString(presentModes[i])); + vk_config.supported_present_modes[i] = presentModeString(presentModes[i]); + } + R_Printf(PRINT_ALL, "\n"); + } + + VkSurfaceFormatKHR swapSurfaceFormat = getSwapSurfaceFormat(surfaceFormats, formatCount); + VkPresentModeKHR swapPresentMode = getSwapPresentMode(presentModes, presentModesCount, r_vsync->value > 0 ? VK_PRESENT_MODE_FIFO_KHR : VK_PRESENT_MODE_MAILBOX_KHR); + free(surfaceFormats); + free(presentModes); + + VkExtent2D extent = surfaceCaps.currentExtent; + if(extent.width == UINT32_MAX || extent.height == UINT32_MAX) + { + extent.width = max(surfaceCaps.minImageExtent.width, min(surfaceCaps.maxImageExtent.width, vid.width)); + extent.height = max(surfaceCaps.minImageExtent.height, min(surfaceCaps.maxImageExtent.height, vid.height)); + } + + // request at least 2 images - this fixes fullscreen crashes on AMD when launching the game in fullscreen + uint32_t imageCount = max(2, surfaceCaps.minImageCount); + if (swapPresentMode != VK_PRESENT_MODE_FIFO_KHR) + imageCount = max(3, surfaceCaps.minImageCount); + + if (surfaceCaps.maxImageCount > 0) + imageCount = min(imageCount, surfaceCaps.maxImageCount); + + VkImageUsageFlags imgUsage = VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT; + + // TRANSFER_SRC_BIT is required for taking screenshots + if (surfaceCaps.supportedUsageFlags & VK_IMAGE_USAGE_TRANSFER_SRC_BIT) + { + imgUsage |= VK_IMAGE_USAGE_TRANSFER_SRC_BIT; + vk_device.screenshotSupported = true; + } + + VkSwapchainKHR oldSwapchain = vk_swapchain.sc; + VkSwapchainCreateInfoKHR scCreateInfo = { + .sType = VK_STRUCTURE_TYPE_SWAPCHAIN_CREATE_INFO_KHR, + .pNext = NULL, + .flags = 0, + .surface = vk_surface, + .minImageCount = imageCount, + .imageFormat = swapSurfaceFormat.format, + .imageColorSpace = swapSurfaceFormat.colorSpace, + .imageExtent = extent, + .imageArrayLayers = 1, + .imageUsage = imgUsage, + .imageSharingMode = VK_SHARING_MODE_EXCLUSIVE, + .queueFamilyIndexCount = 0, + .pQueueFamilyIndices = NULL, + .preTransform = (surfaceCaps.supportedTransforms & VK_SURFACE_TRANSFORM_IDENTITY_BIT_KHR) ? VK_SURFACE_TRANSFORM_IDENTITY_BIT_KHR : surfaceCaps.currentTransform, + .compositeAlpha = getSupportedCompositeAlpha(surfaceCaps.supportedCompositeAlpha), + .presentMode = swapPresentMode, + .clipped = VK_TRUE, + .oldSwapchain = oldSwapchain + }; + + uint32_t queueFamilyIndices[] = { (uint32_t)vk_device.gfxFamilyIndex, (uint32_t)vk_device.presentFamilyIndex }; + if (vk_device.presentFamilyIndex != vk_device.gfxFamilyIndex) + { + scCreateInfo.imageSharingMode = VK_SHARING_MODE_CONCURRENT; + scCreateInfo.queueFamilyIndexCount = 2; + scCreateInfo.pQueueFamilyIndices = queueFamilyIndices; + } + + vk_swapchain.format = swapSurfaceFormat.format; + vk_swapchain.extent = extent; + R_Printf(PRINT_ALL, "...trying swapchain extent: %dx%d\n", vk_swapchain.extent.width, vk_swapchain.extent.height); + R_Printf(PRINT_ALL, "...trying swapchain image format: %d\n", vk_swapchain.format); + + VkResult res = vkCreateSwapchainKHR(vk_device.logical, &scCreateInfo, NULL, &vk_swapchain.sc); + if (res != VK_SUCCESS) + return res; + + VK_VERIFY(vkGetSwapchainImagesKHR(vk_device.logical, vk_swapchain.sc, &imageCount, NULL)); + vk_swapchain.images = (VkImage *)malloc(imageCount * sizeof(VkImage)); + vk_swapchain.imageCount = imageCount; + res = vkGetSwapchainImagesKHR(vk_device.logical, vk_swapchain.sc, &imageCount, vk_swapchain.images); + + if (oldSwapchain != VK_NULL_HANDLE) + vkDestroySwapchainKHR(vk_device.logical, oldSwapchain, NULL); + + return res; +} diff --git a/src/vk/vk_util.c b/src/vk/vk_util.c new file mode 100644 index 0000000..3126bf5 --- /dev/null +++ b/src/vk/vk_util.c @@ -0,0 +1,620 @@ +/* +Copyright (C) 2018 Christoph Schied +Copyright (C) 2020 Denis Pauk + +This program is free software; you can redistribute it and/or modify +it under the terms of the GNU General Public License as published by +the Free Software Foundation; either version 2 of the License, or +(at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +GNU General Public License for more details. + +You should have received a copy of the GNU General Public License along +with this program; if not, write to the Free Software Foundation, Inc., +51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. +*/ + +#include "header/util.h" +#include "header/local.h" + +#include + +static uint32_t +get_memory_type(uint32_t mem_req_type_bits, VkMemoryPropertyFlags mem_prop) +{ + for(uint32_t i = 0; i < VK_MAX_MEMORY_TYPES; i++) { + if(mem_req_type_bits & (1 << i)) { + if((vk_device.mem_properties.memoryTypes[i].propertyFlags & mem_prop) == mem_prop) + return i; + } + } + return -1; +} + +typedef struct MemoryResource_s { + // type of memory + uint32_t memory_type; + // offset step + VkDeviceSize alignment; + // id memory used + VkBool32 used; + // suballocate + VkBool32 suballocate; + // shared memory used for image + VkDeviceMemory memory; + // image size + VkDeviceSize size; + // posision in shared memory + VkDeviceSize offset; +} MemoryResource_t; + +// 1MB buffers / 512 x 512 * RGBA +#define MEMORY_THRESHOLD (512 * 512 * 4) + +static VkDeviceSize memory_block_threshold; +static MemoryResource_t *used_memory; +static VkDeviceSize used_memory_size; + +void +vulkan_memory_init(void) +{ + memory_block_threshold = MEMORY_THRESHOLD; + used_memory_size = 1024; // Size of buffers history + used_memory = malloc(used_memory_size * sizeof(MemoryResource_t)); + memset(used_memory, 0, used_memory_size * sizeof(MemoryResource_t)); +} + +static void +memory_type_print(VkMemoryPropertyFlags mem_prop) +{ +#define MPSTR(r, prop) \ + if((prop & VK_ ##r) != 0) \ + { R_Printf(PRINT_ALL, " %s", "VK_"#r); }; + + MPSTR(MEMORY_PROPERTY_DEVICE_LOCAL_BIT, mem_prop); + MPSTR(MEMORY_PROPERTY_HOST_VISIBLE_BIT, mem_prop); + MPSTR(MEMORY_PROPERTY_HOST_COHERENT_BIT, mem_prop); + MPSTR(MEMORY_PROPERTY_HOST_CACHED_BIT, mem_prop); + MPSTR(MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT, mem_prop); + MPSTR(MEMORY_PROPERTY_PROTECTED_BIT, mem_prop); + MPSTR(MEMORY_PROPERTY_DEVICE_COHERENT_BIT_AMD, mem_prop); + MPSTR(MEMORY_PROPERTY_DEVICE_UNCACHED_BIT_AMD, mem_prop); + +#undef PMSTR +} + +void +vulkan_memory_types_show(void) +{ + R_Printf(PRINT_ALL, "\nMemory blocks:"); + + for(uint32_t i = 0; i < VK_MAX_MEMORY_TYPES; i++) + { + if (vk_device.mem_properties.memoryTypes[i].propertyFlags) + { + R_Printf(PRINT_ALL, "\n #%d:", i); + memory_type_print(vk_device.mem_properties.memoryTypes[i].propertyFlags); + } + } + R_Printf(PRINT_ALL, "\n"); +} + +static VkBool32 +vulkan_memory_is_used(int start_pos, int end_pos, VkDeviceMemory memory) +{ + int pos; + + for (pos = start_pos; pos < end_pos; pos++) + { + if (used_memory[pos].memory == memory && used_memory[pos].used) + return VK_TRUE; + } + + return VK_FALSE; +} + +void +vulkan_memory_free_unused(void) +{ + int pos_global; + + for (pos_global = 0; pos_global < used_memory_size; pos_global ++) + { + VkDeviceMemory memory = used_memory[pos_global].memory; + if (memory != VK_NULL_HANDLE && !used_memory[pos_global].used) + { + int pos_local; + + // is used somewhere else after + if (vulkan_memory_is_used(pos_global, used_memory_size, memory)) + continue; + + // is used somewhere else before + if (vulkan_memory_is_used(0, pos_global, memory)) + continue; + + // free current memory block + vkFreeMemory(vk_device.logical, memory, NULL); + memset(&used_memory[pos_global], 0, sizeof(MemoryResource_t)); + + // cleanup same block + for (pos_local = pos_global + 1; pos_local < used_memory_size; pos_local++) + { + if (used_memory[pos_local].memory == memory) + { + memset(&used_memory[pos_local], 0, sizeof(MemoryResource_t)); + } + } + + } + } +} + +void +vulkan_memory_delete(void) +{ + int pos_global; + for (pos_global = 0; pos_global < used_memory_size; pos_global ++) + { + VkDeviceMemory memory = used_memory[pos_global].memory; + if (memory != VK_NULL_HANDLE) + { + int pos_local; + + // free current memory block + vkFreeMemory(vk_device.logical, memory, NULL); + memset(&used_memory[pos_global], 0, sizeof(MemoryResource_t)); + + // cleanup same block + for (pos_local = pos_global + 1; pos_local < used_memory_size; pos_local++) + { + if (used_memory[pos_local].memory == memory) + { + memset(&used_memory[pos_local], 0, sizeof(MemoryResource_t)); + } + } + } + } + free(used_memory); +} + +static VkResult +memory_block_min(VkDeviceSize size, + uint32_t memory_type, + VkDeviceSize alignment, + VkBool32 suballocate, + int* block_pos) +{ + int pos; + VkDeviceSize min_size = memory_block_threshold; + VkResult result = VK_ERROR_OUT_OF_DEVICE_MEMORY; + + // update max_size + if (min_size < size) + { + *block_pos = -1; + return result; + } + + // search minimal posible size + for (pos = 0; pos < used_memory_size; pos ++) + { + if (used_memory[pos].memory_type == memory_type && + used_memory[pos].suballocate == suballocate && + used_memory[pos].alignment == alignment && + used_memory[pos].memory != VK_NULL_HANDLE && + used_memory[pos].used == VK_FALSE && + used_memory[pos].size < min_size && + used_memory[pos].size >= size) + { + // save minimal size + min_size = used_memory[pos].size; + *block_pos = pos; + result = VK_SUCCESS; + } + } + + return result; +} + +static VkResult +memory_block_empty(int *block_pos) +{ + int pos; + MemoryResource_t *memory; + + // search empty memory + for (pos = *block_pos; pos < used_memory_size; pos ++) + { + if (used_memory[pos].memory == VK_NULL_HANDLE) + { + *block_pos = pos; + return VK_SUCCESS; + } + } + + memory = realloc(used_memory, (used_memory_size * 2) * sizeof(MemoryResource_t)); + if (!memory) + return VK_ERROR_OUT_OF_DEVICE_MEMORY; + + // use previous end + *block_pos = used_memory_size; + + // update old struct + memset(memory + used_memory_size, 0, used_memory_size * sizeof(MemoryResource_t)); + used_memory_size *= 2; + used_memory = memory; + + return VK_SUCCESS; +} + +static VkResult +memory_block_allocate(VkDeviceSize size, + uint32_t memory_type, + VkDeviceSize alignment, + VkBool32 suballocate, + int *block_pos) +{ + int pos = 0; + if (memory_block_empty(&pos) == VK_SUCCESS) + { + VkResult result; + VkDeviceMemory memory; + + if (size < MEMORY_THRESHOLD) + size = MEMORY_THRESHOLD; + + // allocate only aligned + size = ROUNDUP(size, alignment); + + // Need to split only buffers with suballocate support + if (suballocate) + { + // requested bigger then usual + if (size > memory_block_threshold) + { + size *= 2; + // up threshold for next allocations + memory_block_threshold = size; + } + // allcate bigger memory for reuse + else if (size < memory_block_threshold) + { + size = memory_block_threshold; + } + } + + VkMemoryAllocateInfo mem_alloc_info = { + .sType = VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_INFO, + .allocationSize = size, + .memoryTypeIndex = memory_type + }; + + result = vkAllocateMemory(vk_device.logical, &mem_alloc_info, NULL, &memory); + if (result == VK_SUCCESS) + { + used_memory[pos].memory = memory; + used_memory[pos].memory_type = memory_type; + used_memory[pos].alignment = alignment; + used_memory[pos].offset = 0; + used_memory[pos].size = size; + used_memory[pos].suballocate = suballocate; + used_memory[pos].used = VK_FALSE; + *block_pos = pos; + } + return result; + } + + return VK_ERROR_OUT_OF_DEVICE_MEMORY; +} + +static VkResult +memory_create(VkDeviceSize size, + uint32_t memory_type, + VkBool32 suballocate, + VkDeviceSize alignment, + VkDeviceMemory *memory, + VkDeviceSize *offset) +{ + int pos = -1; + VkResult result; + result = memory_block_min(size, memory_type, alignment, suballocate, &pos); + + if (result != VK_SUCCESS) + { + result = memory_block_allocate(size, memory_type, alignment, suballocate, &pos); + } + if (result == VK_SUCCESS) + { + // check size of block, + // new block should be at least same size as current + // and beger than double minimal offset + // and marked as not for mmap + if (used_memory[pos].size > (size * 2) && + (used_memory[pos].size > (used_memory[pos].alignment * 2)) && + used_memory[pos].suballocate) + { + // search from next slot + int new_pos = pos + 1; + result = memory_block_empty(&new_pos); + if (result == VK_SUCCESS) + { + VkDeviceSize new_size = ROUNDUP(size, used_memory[pos].alignment); + + // split to several blocks + memcpy(&used_memory[new_pos], &used_memory[pos], sizeof(MemoryResource_t)); + used_memory[new_pos].offset = used_memory[pos].offset + new_size; + used_memory[new_pos].size = used_memory[pos].size - new_size; + + // save new size to block, it can be bigger than required + used_memory[pos].size = used_memory[new_pos].offset - used_memory[pos].offset; + assert(used_memory[pos].size > 0); + } + } + + used_memory[pos].used = VK_TRUE; + *offset = used_memory[pos].offset; + *memory = used_memory[pos].memory; + return result; + } + + return VK_ERROR_OUT_OF_DEVICE_MEMORY; +} + +static void +memory_destroy(VkDeviceMemory memory, VkDeviceSize offset) +{ + int pos; + for (pos = 0; pos < used_memory_size; pos ++) + { + if (used_memory[pos].memory == memory && used_memory[pos].offset == offset) + { + used_memory[pos].used = VK_FALSE; + return; + } + } + // looks as no such memory registered + vkFreeMemory(vk_device.logical, memory, NULL); +} + +static VkResult +memory_create_by_property(VkMemoryRequirements* mem_reqs, + VkMemoryPropertyFlags mem_properties, + VkMemoryPropertyFlags mem_preferences, + VkDeviceMemory *memory, + VkDeviceSize *offset) +{ + uint32_t memory_index; + VkMemoryPropertyFlags host_visible; + + memory_index = get_memory_type(mem_reqs->memoryTypeBits, + mem_properties | mem_preferences); + // prefered memory allocation + if (memory_index == -1) + { + memory_index = get_memory_type(mem_reqs->memoryTypeBits, + mem_properties); + } + // strictly required + if (memory_index == -1) + { + R_Printf(PRINT_ALL, "%s:%d: Have not found required memory type.\n", + __func__, __LINE__); + return VK_ERROR_OUT_OF_DEVICE_MEMORY; + } + + host_visible = mem_properties & VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT; + + return memory_create(mem_reqs->size, memory_index, + // suballocate allowed + host_visible != VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT, + mem_reqs->alignment, memory, offset); +} + +VkResult +buffer_create(BufferResource_t *buf, + VkBufferCreateInfo buf_create_info, + VkMemoryPropertyFlags mem_properties, + VkMemoryPropertyFlags mem_preferences) +{ + assert(buf_create_info.size > 0); + assert(buf); + VkResult result = VK_SUCCESS; + + buf->size = buf_create_info.size; + buf->is_mapped = VK_FALSE; + + result = vkCreateBuffer(vk_device.logical, &buf_create_info, NULL, &buf->buffer); + if(result != VK_SUCCESS) { + R_Printf(PRINT_ALL, "%s:%d: VkResult verification: %s\n", + __func__, __LINE__, QVk_GetError(result)); + goto fail_buffer; + } + assert(buf->buffer != VK_NULL_HANDLE); + + VkMemoryRequirements mem_reqs; + vkGetBufferMemoryRequirements(vk_device.logical, buf->buffer, &mem_reqs); + + result = memory_create_by_property(&mem_reqs, mem_properties, mem_preferences, + &buf->memory, &buf->offset); + if(result != VK_SUCCESS) { + R_Printf(PRINT_ALL, "%s:%d: VkResult verification: %s\n", + __func__, __LINE__, QVk_GetError(result)); + goto fail_mem_alloc; + } + + assert(buf->memory != VK_NULL_HANDLE); + + result = vkBindBufferMemory(vk_device.logical, buf->buffer, buf->memory, buf->offset); + if(result != VK_SUCCESS) { + R_Printf(PRINT_ALL, "%s:%d: VkResult verification: %s\n", + __func__, __LINE__, QVk_GetError(result)); + goto fail_bind_buf_memory; + } + + return VK_SUCCESS; + +fail_bind_buf_memory: + memory_destroy(buf->memory, buf->offset); +fail_mem_alloc: + vkDestroyBuffer(vk_device.logical, buf->buffer, NULL); +fail_buffer: + buf->buffer = VK_NULL_HANDLE; + buf->memory = VK_NULL_HANDLE; + buf->size = 0; + return result; +} + +VkResult +image_create(ImageResource_t *img, + VkImageCreateInfo img_create_info, + VkMemoryPropertyFlags mem_properties, + VkMemoryPropertyFlags mem_preferences) +{ + assert(img); + VkResult result = VK_SUCCESS; + + result = vkCreateImage(vk_device.logical, &img_create_info, NULL, &img->image); + if(result != VK_SUCCESS) { + R_Printf(PRINT_ALL, "%s:%d: VkResult verification: %s\n", + __func__, __LINE__, QVk_GetError(result)); + goto fail_buffer; + } + assert(img->image != VK_NULL_HANDLE); + + VkMemoryRequirements mem_reqs; + vkGetImageMemoryRequirements(vk_device.logical, img->image, &mem_reqs); + img->size = mem_reqs.size; + + result = memory_create_by_property(&mem_reqs, mem_properties, mem_preferences, + &img->memory, &img->offset); + if(result != VK_SUCCESS) { + R_Printf(PRINT_ALL, "%s:%d: VkResult verification: %s\n", + __func__, __LINE__, QVk_GetError(result)); + goto fail_mem_alloc; + } + + assert(img->memory != VK_NULL_HANDLE); + + result = vkBindImageMemory(vk_device.logical, img->image, img->memory, img->offset); + if(result != VK_SUCCESS) { + R_Printf(PRINT_ALL, "%s:%d: VkResult verification: %s\n", + __func__, __LINE__, QVk_GetError(result)); + goto fail_bind_buf_memory; + } + + return VK_SUCCESS; + +fail_bind_buf_memory: + memory_destroy(img->memory, img->offset); +fail_mem_alloc: + vkDestroyImage(vk_device.logical, img->image, NULL); +fail_buffer: + img->image = VK_NULL_HANDLE; + img->memory = VK_NULL_HANDLE; + img->size = 0; + return result; +} + +VkResult +buffer_destroy(BufferResource_t *buf) +{ + assert(!buf->is_mapped); + + // buffer should be destroed before bound memory + if(buf->buffer != VK_NULL_HANDLE) + { + vkDestroyBuffer(vk_device.logical, buf->buffer, NULL); + buf->buffer = VK_NULL_HANDLE; + } + + // buffer desroed, we can free up memory + if(buf->memory != VK_NULL_HANDLE) + { + memory_destroy(buf->memory, buf->offset); + buf->memory = VK_NULL_HANDLE; + } + + memset(buf, 0, sizeof(BufferResource_t)); + + return VK_SUCCESS; +} + +VkResult +image_destroy(ImageResource_t *img) +{ + // image should be destroed before bound memory + if(img->image != VK_NULL_HANDLE) + { + vkDestroyImage(vk_device.logical, img->image, NULL); + img->image = VK_NULL_HANDLE; + } + + // image destroed, we can free up memory + if(img->memory != VK_NULL_HANDLE) + { + memory_destroy(img->memory, img->offset); + img->memory = VK_NULL_HANDLE; + } + + memset(img, 0, sizeof(ImageResource_t)); + + return VK_SUCCESS; +} + +VkResult +buffer_flush(BufferResource_t *buf) +{ + VkResult result = VK_SUCCESS; + + VkMappedMemoryRange ranges[1] = {{ + .sType = VK_STRUCTURE_TYPE_MAPPED_MEMORY_RANGE, + .memory = buf->memory, + .offset = buf->offset, + .size = buf->size + }}; + result = vkFlushMappedMemoryRanges(vk_device.logical, 1, ranges); + return result; +} + +VkResult +buffer_invalidate(BufferResource_t *buf) +{ + VkResult result = VK_SUCCESS; + + VkMappedMemoryRange ranges[1] = {{ + .sType = VK_STRUCTURE_TYPE_MAPPED_MEMORY_RANGE, + .memory = buf->memory, + .offset = buf->offset, + .size = buf->size + }}; + result = vkInvalidateMappedMemoryRanges(vk_device.logical, 1, ranges); + return result; + +} + +void * +buffer_map(BufferResource_t *buf) +{ + assert(buf->memory); + assert(!buf->is_mapped); + buf->is_mapped = VK_TRUE; + void *ret = NULL; + assert(buf->memory != VK_NULL_HANDLE); + assert(buf->size > 0); + VK_VERIFY(vkMapMemory(vk_device.logical, buf->memory, + buf->offset/*offset*/, buf->size, 0 /*flags*/, &ret)); + return ret; +} + +void +buffer_unmap(BufferResource_t *buf) +{ + assert(buf->memory); + assert(buf->is_mapped); + buf->is_mapped = VK_FALSE; + vkUnmapMemory(vk_device.logical, buf->memory); +} diff --git a/src/vk/vk_validation.c b/src/vk/vk_validation.c new file mode 100644 index 0000000..2710ba8 --- /dev/null +++ b/src/vk/vk_validation.c @@ -0,0 +1,99 @@ +/* +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +*/ + +#include "header/local.h" + +static VkDebugUtilsMessengerEXT validationMessenger = VK_NULL_HANDLE; + +// layer message to string +static const char* msgToString(VkDebugUtilsMessageTypeFlagsEXT type) +{ + int g = (type & VK_DEBUG_UTILS_MESSAGE_TYPE_GENERAL_BIT_EXT) != 0 ? 1 : 0; + int p = (type & VK_DEBUG_UTILS_MESSAGE_TYPE_PERFORMANCE_BIT_EXT) != 0 ? 1 : 0; + int v = (type & VK_DEBUG_UTILS_MESSAGE_TYPE_VALIDATION_BIT_EXT) != 0 ? 1 : 0; + + if (g) return ""; + if (p && !v) return "(performance)"; + if (p && v) return "(performance and validation)"; + if (v) return "(validation)"; + + return "?"; +} + +// validation layer callback function (VK_EXT_debug_utils) +static VKAPI_ATTR VkBool32 VKAPI_CALL debugUtilsCallback(VkDebugUtilsMessageSeverityFlagBitsEXT msgSeverity, + VkDebugUtilsMessageTypeFlagsEXT msgType, + const VkDebugUtilsMessengerCallbackDataEXT *callbackData, + void* userData) +{ + switch (msgSeverity) + { + case VK_DEBUG_UTILS_MESSAGE_SEVERITY_INFO_BIT_EXT: + R_Printf(PRINT_ALL, "VK_INFO: %s %s\n", callbackData->pMessage, msgToString(msgType)); + break; + case VK_DEBUG_UTILS_MESSAGE_SEVERITY_VERBOSE_BIT_EXT: + R_Printf(PRINT_ALL, "VK_VERBOSE: %s %s\n", callbackData->pMessage, msgToString(msgType)); + break; + case VK_DEBUG_UTILS_MESSAGE_SEVERITY_WARNING_BIT_EXT: + R_Printf(PRINT_ALL, "VK_WARNING: %s %s\n", callbackData->pMessage, msgToString(msgType)); + break; + default: + R_Printf(PRINT_ALL, "VK_ERROR: %s %s\n", callbackData->pMessage, msgToString(msgType)); + assert(!"Vulkan error occured!"); + } + return VK_FALSE; +} + +void QVk_CreateValidationLayers() +{ + VkDebugUtilsMessengerCreateInfoEXT callbackInfo = { + .sType = VK_STRUCTURE_TYPE_DEBUG_UTILS_MESSENGER_CREATE_INFO_EXT, + .pNext = NULL, + .flags = 0, + .messageSeverity = VK_DEBUG_UTILS_MESSAGE_SEVERITY_WARNING_BIT_EXT | + VK_DEBUG_UTILS_MESSAGE_SEVERITY_ERROR_BIT_EXT, + .messageType = VK_DEBUG_UTILS_MESSAGE_TYPE_GENERAL_BIT_EXT | + VK_DEBUG_UTILS_MESSAGE_TYPE_VALIDATION_BIT_EXT | + VK_DEBUG_UTILS_MESSAGE_TYPE_PERFORMANCE_BIT_EXT, + .pfnUserCallback = debugUtilsCallback, + .pUserData = NULL + }; + + if(vk_validation->value) + { + callbackInfo.messageSeverity |= VK_DEBUG_UTILS_MESSAGE_SEVERITY_INFO_BIT_EXT | + VK_DEBUG_UTILS_MESSAGE_SEVERITY_VERBOSE_BIT_EXT; + } + + if (qvkCreateDebugUtilsMessengerEXT) + { + VK_VERIFY(qvkCreateDebugUtilsMessengerEXT(vk_instance, &callbackInfo, NULL, &validationMessenger)); + R_Printf(PRINT_ALL, "...Vulkan validation layers enabled\n"); + } +} + +void QVk_DestroyValidationLayers() +{ + if( validationMessenger != VK_NULL_HANDLE && qvkDestroyDebugUtilsMessengerEXT) + { + qvkDestroyDebugUtilsMessengerEXT( vk_instance, validationMessenger, NULL ); + validationMessenger = VK_NULL_HANDLE; + } +} diff --git a/src/vk/vk_warp.c b/src/vk/vk_warp.c new file mode 100644 index 0000000..208f99a --- /dev/null +++ b/src/vk/vk_warp.c @@ -0,0 +1,732 @@ +/* +Copyright (C) 1997-2001 Id Software, Inc. +Copyright (C) 2018-2019 Krzysztof Kondrak + +This program is free software; you can redistribute it and/or +modify it under the terms of the GNU General Public License +as published by the Free Software Foundation; either version 2 +of the License, or (at your option) any later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + +See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software +Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +*/ +// vk_warp.c -- sky and water polygons +#include "header/local.h" + +static char skyname[MAX_QPATH]; +static float skyrotate; +static vec3_t skyaxis; +static image_t *sky_images[6]; + +static msurface_t *warpface; + +#define SUBDIVIDE_SIZE 64 + +static void BoundPoly (int numverts, float *verts, vec3_t mins, vec3_t maxs) +{ + int i, j; + float *v; + + mins[0] = mins[1] = mins[2] = 9999; + maxs[0] = maxs[1] = maxs[2] = -9999; + v = verts; + for (i=0 ; i maxs[j]) + maxs[j] = *v; + } +} + +static void SubdividePolygon (int numverts, float *verts) +{ + int i, j, k; + vec3_t mins, maxs; + float *v; + vec3_t front[64], back[64]; + int f, b; + float dist[64]; + float frac; + vkpoly_t *poly; + vec3_t total; + float total_s, total_t; + + if (numverts > 60) + ri.Sys_Error (ERR_DROP, "%s: numverts = %i", __func__, numverts); + + BoundPoly (numverts, verts, mins, maxs); + + for (i=0 ; i<3 ; i++) + { + float m; + + m = (mins[i] + maxs[i]) * 0.5; + m = SUBDIVIDE_SIZE * floor (m/SUBDIVIDE_SIZE + 0.5); + if (maxs[i] - m < 8) + continue; + if (m - mins[i] < 8) + continue; + + // cut it + v = verts + i; + for (j=0 ; j= 0) + { + VectorCopy (v, front[f]); + f++; + } + if (dist[j] <= 0) + { + VectorCopy (v, back[b]); + b++; + } + if (dist[j] == 0 || dist[j+1] == 0) + continue; + if ( (dist[j] > 0) != (dist[j+1] > 0) ) + { + // clip point + frac = dist[j] / (dist[j] - dist[j+1]); + for (k=0 ; k<3 ; k++) + front[f][k] = back[b][k] = v[k] + frac*(v[3+k] - v[k]); + f++; + b++; + } + } + + SubdividePolygon (f, front[0]); + SubdividePolygon (b, back[0]); + return; + } + + // add a point in the center to help keep warp valid + poly = Hunk_Alloc (sizeof(vkpoly_t) + ((numverts-4)+2) * VERTEXSIZE*sizeof(float)); + poly->next = warpface->polys; + warpface->polys = poly; + poly->numverts = numverts+2; + VectorClear (total); + total_s = 0; + total_t = 0; + for (i=0 ; iverts[i+1]); + s = DotProduct (verts, warpface->texinfo->vecs[0]); + t = DotProduct (verts, warpface->texinfo->vecs[1]); + + total_s += s; + total_t += t; + VectorAdd (total, verts, total); + + poly->verts[i+1][3] = s; + poly->verts[i+1][4] = t; + } + + VectorScale (total, (1.0/numverts), poly->verts[0]); + poly->verts[0][3] = total_s/numverts; + poly->verts[0][4] = total_t/numverts; + + // copy first vertex to last + memcpy (poly->verts[i+1], poly->verts[1], sizeof(poly->verts[0])); +} + +/* +================ +Vk_SubdivideSurface + +Breaks a polygon up along axial 64 unit +boundaries so that turbulent and sky warps +can be done reasonably. +================ +*/ +void Vk_SubdivideSurface (msurface_t *fa, model_t *loadmodel) +{ + vec3_t verts[64]; + int numverts; + int i; + float *vec; + + warpface = fa; + + // + // convert edges back to a normal polygon + // + numverts = 0; + for (i=0 ; inumedges ; i++) + { + int lindex; + + lindex = loadmodel->surfedges[fa->firstedge + i]; + + if (lindex > 0) + vec = loadmodel->vertexes[loadmodel->edges[lindex].v[0]].position; + else + vec = loadmodel->vertexes[loadmodel->edges[-lindex].v[1]].position; + VectorCopy (vec, verts[numverts]); + numverts++; + } + + SubdividePolygon (numverts, verts[0]); +} + +//========================================================= + +/* +============= +EmitWaterPolys + +Does a water warp on the pre-fragmented vkpoly_t chain +============= +*/ +void +EmitWaterPolys (msurface_t *fa, image_t *texture, float *modelMatrix, + float *color, qboolean solid_surface) +{ + vkpoly_t *p, *bp; + float *v; + int i; + + struct { + float model[16]; + float color[4]; + float time; + float scroll; + } polyUbo; + + polyUbo.color[0] = color[0]; + polyUbo.color[1] = color[1]; + polyUbo.color[2] = color[2]; + polyUbo.color[3] = color[3]; + polyUbo.time = r_newrefdef.time; + + if (fa->texinfo->flags & SURF_FLOWING) + polyUbo.scroll = (-64 * ((r_newrefdef.time*0.5) - (int)(r_newrefdef.time*0.5))) / 64.f; + else + polyUbo.scroll = 0; + + if (modelMatrix) + { + memcpy(polyUbo.model, modelMatrix, sizeof(float) * 16); + } + else + { + Mat_Identity(polyUbo.model); + } + + if (solid_surface) + { + // Solid surface + QVk_BindPipeline(&vk_drawPolySolidWarpPipeline); + } + else + { + // Blend surface + QVk_BindPipeline(&vk_drawPolyWarpPipeline); + } + + uint32_t uboOffset; + VkDescriptorSet uboDescriptorSet; + uint8_t *uboData = QVk_GetUniformBuffer(sizeof(polyUbo), &uboOffset, &uboDescriptorSet); + memcpy(uboData, &polyUbo, sizeof(polyUbo)); + + VkBuffer vbo; + VkDeviceSize vboOffset; + VkDescriptorSet descriptorSets[] = { texture->vk_texture.descriptorSet, uboDescriptorSet }; + + float gamma = 2.1F - vid_gamma->value; + + vkCmdPushConstants(vk_activeCmdbuffer, vk_drawTexQuadPipeline[vk_state.current_renderpass].layout, + VK_SHADER_STAGE_FRAGMENT_BIT, 17 * sizeof(float), sizeof(gamma), &gamma); + + if (solid_surface) + { + // Solid surface + vkCmdBindDescriptorSets( + vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, + vk_drawPolySolidWarpPipeline.layout, 0, 2, + descriptorSets, 1, &uboOffset); + } + else + { + // Blend surface + vkCmdBindDescriptorSets( + vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, + vk_drawPolyWarpPipeline.layout, 0, 2, + descriptorSets, 1, &uboOffset); + } + + for (bp = fa->polys; bp; bp = bp->next) + { + p = bp; + + if (Mesh_VertsRealloc(p->numverts)) + { + ri.Sys_Error(ERR_FATAL, "%s: can't allocate memory", __func__); + } + + for (i = 0, v = p->verts[0]; i < p->numverts; i++, v += VERTEXSIZE) + { + verts_buffer[i].vertex[0] = v[0]; + verts_buffer[i].vertex[1] = v[1]; + verts_buffer[i].vertex[2] = v[2]; + verts_buffer[i].texCoord[0] = v[3] / 64.f; + verts_buffer[i].texCoord[1] = v[4] / 64.f; + } + + uint8_t *vertData = QVk_GetVertexBuffer(sizeof(polyvert_t) * p->numverts, &vbo, &vboOffset); + memcpy(vertData, verts_buffer, sizeof(polyvert_t) * p->numverts); + + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + vkCmdBindIndexBuffer(vk_activeCmdbuffer, QVk_GetTriangleFanIbo((p->numverts - 2) * 3), 0, VK_INDEX_TYPE_UINT16); + vkCmdDrawIndexed(vk_activeCmdbuffer, (p->numverts - 2) * 3, 1, 0, 0, 0); + } +} + + +//=================================================================== + + +static vec3_t skyclip[6] = { + {1,1,0}, + {1,-1,0}, + {0,-1,1}, + {0,1,1}, + {1,0,1}, + {-1,0,1} +}; + +// 1 = s, 2 = t, 3 = 2048 +static int st_to_vec[6][3] = +{ + {3,-1,2}, + {-3,1,2}, + + {1,3,2}, + {-1,-3,2}, + + {-2,-1,3}, // 0 degrees yaw, look straight up + {2,-1,-3} // look straight down + +// {-1,2,3}, +// {1,2,-3} +}; + +// s = [0]/[2], t = [1]/[2] +static int vec_to_st[6][3] = +{ + {-2,3,1}, + {2,3,-1}, + + {1,3,2}, + {-1,3,-2}, + + {-2,-1,3}, + {-2,1,-3} + +// {-1,2,3}, +// {1,2,-3} +}; + +static float skymins[2][6], skymaxs[2][6]; +static float sky_min, sky_max; + +static void DrawSkyPolygon (int nump, vec3_t vecs) +{ + int i; + vec3_t v, av; + float s, t, dv; + int axis; + float *vp; + + // decide which face it maps to + VectorCopy (vec3_origin, v); + for (i=0, vp=vecs ; i av[1] && av[0] > av[2]) + { + if (v[0] < 0) + axis = 1; + else + axis = 0; + } + else if (av[1] > av[2] && av[1] > av[0]) + { + if (v[1] < 0) + axis = 3; + else + axis = 2; + } + else + { + if (v[2] < 0) + axis = 5; + else + axis = 4; + } + + // project new texture coords + for (i=0 ; i 0) + dv = vecs[j - 1]; + else + dv = -vecs[-j - 1]; + if (dv < 0.001) + continue; // don't divide by zero + j = vec_to_st[axis][0]; + if (j < 0) + s = -vecs[-j -1] / dv; + else + s = vecs[j-1] / dv; + j = vec_to_st[axis][1]; + if (j < 0) + t = -vecs[-j -1] / dv; + else + t = vecs[j-1] / dv; + + if (s < skymins[0][axis]) + skymins[0][axis] = s; + if (t < skymins[1][axis]) + skymins[1][axis] = t; + if (s > skymaxs[0][axis]) + skymaxs[0][axis] = s; + if (t > skymaxs[1][axis]) + skymaxs[1][axis] = t; + } +} + +#define ON_EPSILON 0.1 // point on plane side epsilon +#define MAX_CLIP_VERTS 64 +static void ClipSkyPolygon (int nump, vec3_t vecs, int stage) +{ + float *norm; + float *v; + qboolean front, back; + float d, e; + float dists[MAX_CLIP_VERTS]; + int sides[MAX_CLIP_VERTS]; + vec3_t newv[2][MAX_CLIP_VERTS]; + int newc[2]; + int i, j; + + if (nump > MAX_CLIP_VERTS-2) + ri.Sys_Error (ERR_DROP, "%s: MAX_CLIP_VERTS", __func__); + if (stage == 6) + { // fully clipped, so draw it + DrawSkyPolygon (nump, vecs); + return; + } + + front = back = false; + norm = skyclip[stage]; + for (i=0, v = vecs ; i ON_EPSILON) + { + front = true; + sides[i] = SIDE_FRONT; + } + else if (d < -ON_EPSILON) + { + back = true; + sides[i] = SIDE_BACK; + } + else + sides[i] = SIDE_ON; + dists[i] = d; + } + + if (!front || !back) + { // not clipped + ClipSkyPolygon (nump, vecs, stage+1); + return; + } + + // clip it + sides[i] = sides[0]; + dists[i] = dists[0]; + VectorCopy (vecs, (vecs+(i*3)) ); + newc[0] = newc[1] = 0; + + for (i=0, v = vecs ; ipolys ; p ; p=p->next) + { + for (i=0 ; inumverts ; i++) + { + VectorSubtract (p->verts[i], r_origin, verts[i]); + } + ClipSkyPolygon (p->numverts, verts[0], 0); + } +} + + +/* +============== +R_ClearSkyBox +============== +*/ +void R_ClearSkyBox (void) +{ + int i; + + for (i=0 ; i<6 ; i++) + { + skymins[0][i] = skymins[1][i] = 9999; + skymaxs[0][i] = skymaxs[1][i] = -9999; + } +} + + +static void MakeSkyVec (float s, float t, int axis, float *vertexData) +{ + vec3_t v, b; + int j; + + float dist = (r_farsee->value == 0) ? 2300.0f : 4096.0f; + + b[0] = s * dist; + b[1] = t * dist; + b[2] = dist; + + for (j = 0; j<3; j++) + { + int k; + + k = st_to_vec[axis][j]; + if (k < 0) + v[j] = -b[-k - 1]; + else + v[j] = b[k - 1]; + } + + // avoid bilerp seam + s = (s + 1)*0.5; + t = (t + 1)*0.5; + + if (s < sky_min) + s = sky_min; + else if (s > sky_max) + s = sky_max; + if (t < sky_min) + t = sky_min; + else if (t > sky_max) + t = sky_max; + + t = 1.0 - t; + + vertexData[0] = v[0]; + vertexData[1] = v[1]; + vertexData[2] = v[2]; + vertexData[3] = s; + vertexData[4] = t; +} + +/* +============== +R_DrawSkyBox +============== +*/ +static int skytexorder[6] = {0,2,1,3,4,5}; +void R_DrawSkyBox (void) +{ + int i; + + if (skyrotate) + { // check for no sky at all + for (i = 0; i<6; i++) + if (skymins[0][i] < skymaxs[0][i] + && skymins[1][i] < skymaxs[1][i]) + break; + if (i == 6) + return; // nothing visible + } + + float model[16]; + Mat_Identity(model); + Mat_Rotate(model, r_newrefdef.time * skyrotate, skyaxis[0], skyaxis[1], skyaxis[2]); + Mat_Translate(model, r_origin[0], r_origin[1], r_origin[2]); + + struct { + float data[5]; + } skyVerts[4]; + + QVk_BindPipeline(&vk_drawSkyboxPipeline); + uint32_t uboOffset; + VkDescriptorSet uboDescriptorSet; + uint8_t *uboData = QVk_GetUniformBuffer(sizeof(model), &uboOffset, &uboDescriptorSet); + memcpy(uboData, model, sizeof(model)); + + for (i = 0; i<6; i++) + { + if (skyrotate) + { // hack, forces full sky to draw when rotating + skymins[0][i] = -1; + skymins[1][i] = -1; + skymaxs[0][i] = 1; + skymaxs[1][i] = 1; + } + + if (skymins[0][i] >= skymaxs[0][i] + || skymins[1][i] >= skymaxs[1][i]) + continue; + + MakeSkyVec(skymins[0][i], skymins[1][i], i, skyVerts[0].data); + MakeSkyVec(skymins[0][i], skymaxs[1][i], i, skyVerts[1].data); + MakeSkyVec(skymaxs[0][i], skymaxs[1][i], i, skyVerts[2].data); + MakeSkyVec(skymaxs[0][i], skymins[1][i], i, skyVerts[3].data); + + float verts[] = { + skyVerts[0].data[0], skyVerts[0].data[1], skyVerts[0].data[2], skyVerts[0].data[3], skyVerts[0].data[4], + skyVerts[1].data[0], skyVerts[1].data[1], skyVerts[1].data[2], skyVerts[1].data[3], skyVerts[1].data[4], + skyVerts[2].data[0], skyVerts[2].data[1], skyVerts[2].data[2], skyVerts[2].data[3], skyVerts[2].data[4], + skyVerts[0].data[0], skyVerts[0].data[1], skyVerts[0].data[2], skyVerts[0].data[3], skyVerts[0].data[4], + skyVerts[2].data[0], skyVerts[2].data[1], skyVerts[2].data[2], skyVerts[2].data[3], skyVerts[2].data[4], + skyVerts[3].data[0], skyVerts[3].data[1], skyVerts[3].data[2], skyVerts[3].data[3], skyVerts[3].data[4] + }; + + VkBuffer vbo; + VkDeviceSize vboOffset; + uint8_t *vertData = QVk_GetVertexBuffer(sizeof(verts), &vbo, &vboOffset); + memcpy(vertData, verts, sizeof(verts)); + + VkDescriptorSet descriptorSets[] = { sky_images[skytexorder[i]]->vk_texture.descriptorSet, uboDescriptorSet }; + + float gamma = 2.1F - vid_gamma->value; + + vkCmdPushConstants(vk_activeCmdbuffer, vk_drawTexQuadPipeline[vk_state.current_renderpass].layout, + VK_SHADER_STAGE_FRAGMENT_BIT, 17 * sizeof(float), sizeof(gamma), &gamma); + + vkCmdBindDescriptorSets(vk_activeCmdbuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, + vk_drawSkyboxPipeline.layout, 0, 2, descriptorSets, 1, &uboOffset); + vkCmdBindVertexBuffers(vk_activeCmdbuffer, 0, 1, &vbo, &vboOffset); + vkCmdDraw(vk_activeCmdbuffer, 6, 1, 0, 0); + } +} + + +/* +============ +RE_SetSky +============ +*/ +// 3dstudio environment map names +static char *suf[6] = {"rt", "bk", "lf", "ft", "up", "dn"}; +void RE_SetSky (char *name, float rotate, vec3_t axis) +{ + int i; + char pathname[MAX_QPATH]; + + strncpy(skyname, name, sizeof(skyname) - 1); + skyrotate = rotate; + VectorCopy(axis, skyaxis); + + for (i = 0; i<6; i++) + { + // chop down rotating skies for less memory + if (vk_skymip->value || skyrotate) + vk_picmip->value++; + + Com_sprintf(pathname, sizeof(pathname), "env/%s%s.tga", skyname, suf[i]); + + sky_images[i] = Vk_FindImage(pathname, it_sky); + if (!sky_images[i]) { + Com_sprintf(pathname, sizeof(pathname), "pics/Skies/%s%s.m8", skyname, suf[i]); + sky_images[i] = Vk_FindImage(pathname, it_sky); + } + + if (!sky_images[i]) + sky_images[i] = r_notexture; + + if (vk_skymip->value || skyrotate) + { // take less memory + vk_picmip->value--; + sky_min = 1.0 / 256; + sky_max = 255.0 / 256; + } + else + { + sky_min = 1.0 / 512; + sky_max = 511.0 / 512; + } + } +} diff --git a/src/vk/volk/volk.c b/src/vk/volk/volk.c new file mode 100644 index 0000000..3926871 --- /dev/null +++ b/src/vk/volk/volk.c @@ -0,0 +1,1905 @@ +/* This file is part of volk library; see volk.h for version/license details */ +/* clang-format off */ +#include "volk.h" + +#ifdef _WIN32 + typedef const char* LPCSTR; + typedef struct HINSTANCE__* HINSTANCE; + typedef HINSTANCE HMODULE; + #ifdef _WIN64 + typedef __int64 (__stdcall* FARPROC)(void); + #else + typedef int (__stdcall* FARPROC)(void); + #endif +#else +# include +#endif + +#ifdef __cplusplus +extern "C" { +#endif + +#ifdef _WIN32 +__declspec(dllimport) HMODULE __stdcall LoadLibraryA(LPCSTR); +__declspec(dllimport) FARPROC __stdcall GetProcAddress(HMODULE, LPCSTR); +#endif + +static VkInstance loadedInstance = VK_NULL_HANDLE; +static VkDevice loadedDevice = VK_NULL_HANDLE; + +static void volkGenLoadLoader(void* context, PFN_vkVoidFunction (*load)(void*, const char*)); +static void volkGenLoadInstance(void* context, PFN_vkVoidFunction (*load)(void*, const char*)); +static void volkGenLoadDevice(void* context, PFN_vkVoidFunction (*load)(void*, const char*)); +static void volkGenLoadDeviceTable(struct VolkDeviceTable* table, void* context, PFN_vkVoidFunction (*load)(void*, const char*)); + +static PFN_vkVoidFunction vkGetInstanceProcAddrStub(void* context, const char* name) +{ + return vkGetInstanceProcAddr((VkInstance)context, name); +} + +static PFN_vkVoidFunction vkGetDeviceProcAddrStub(void* context, const char* name) +{ + return vkGetDeviceProcAddr((VkDevice)context, name); +} + +VkResult volkInitialize(void) +{ +#if defined(_WIN32) + HMODULE module = LoadLibraryA("vulkan-1.dll"); + if (!module) + return VK_ERROR_INITIALIZATION_FAILED; + + // note: function pointer is cast through void function pointer to silence cast-function-type warning on gcc8 + vkGetInstanceProcAddr = (PFN_vkGetInstanceProcAddr)(void(*)(void))GetProcAddress(module, "vkGetInstanceProcAddr"); +#elif defined(__APPLE__) + void* module = dlopen("libvulkan.dylib", RTLD_NOW | RTLD_LOCAL); + if (!module) + module = dlopen("libvulkan.1.dylib", RTLD_NOW | RTLD_LOCAL); + if (!module) + module = dlopen("libMoltenVK.dylib", RTLD_NOW | RTLD_LOCAL); + if (!module) + return VK_ERROR_INITIALIZATION_FAILED; + + vkGetInstanceProcAddr = (PFN_vkGetInstanceProcAddr)dlsym(module, "vkGetInstanceProcAddr"); +#else + void* module = dlopen("libvulkan.so.1", RTLD_NOW | RTLD_LOCAL); + if (!module) + module = dlopen("libvulkan.so", RTLD_NOW | RTLD_LOCAL); + if (!module) + return VK_ERROR_INITIALIZATION_FAILED; + + vkGetInstanceProcAddr = (PFN_vkGetInstanceProcAddr)dlsym(module, "vkGetInstanceProcAddr"); +#endif + + volkGenLoadLoader(NULL, vkGetInstanceProcAddrStub); + + return VK_SUCCESS; +} + +void volkInitializeCustom(PFN_vkGetInstanceProcAddr handler) +{ + vkGetInstanceProcAddr = handler; + + volkGenLoadLoader(NULL, vkGetInstanceProcAddrStub); +} + +uint32_t volkGetInstanceVersion(void) +{ +#if defined(VK_VERSION_1_1) + uint32_t apiVersion = 0; + if (vkEnumerateInstanceVersion && vkEnumerateInstanceVersion(&apiVersion) == VK_SUCCESS) + return apiVersion; +#endif + + if (vkCreateInstance) + return VK_API_VERSION_1_0; + + return 0; +} + +void volkLoadInstance(VkInstance instance) +{ + loadedInstance = instance; + volkGenLoadInstance(instance, vkGetInstanceProcAddrStub); + volkGenLoadDevice(instance, vkGetInstanceProcAddrStub); +} + +void volkLoadInstanceOnly(VkInstance instance) +{ + loadedInstance = instance; + volkGenLoadInstance(instance, vkGetInstanceProcAddrStub); +} + +VkInstance volkGetLoadedInstance() +{ + return loadedInstance; +} + +void volkLoadDevice(VkDevice device) +{ + loadedDevice = device; + volkGenLoadDevice(device, vkGetDeviceProcAddrStub); +} + +VkDevice volkGetLoadedDevice() +{ + return loadedDevice; +} + +void volkLoadDeviceTable(struct VolkDeviceTable* table, VkDevice device) +{ + volkGenLoadDeviceTable(table, device, vkGetDeviceProcAddrStub); +} + +static void volkGenLoadLoader(void* context, PFN_vkVoidFunction (*load)(void*, const char*)) +{ + /* VOLK_GENERATE_LOAD_LOADER */ +#if defined(VK_VERSION_1_0) + vkCreateInstance = (PFN_vkCreateInstance)load(context, "vkCreateInstance"); + vkEnumerateInstanceExtensionProperties = (PFN_vkEnumerateInstanceExtensionProperties)load(context, "vkEnumerateInstanceExtensionProperties"); + vkEnumerateInstanceLayerProperties = (PFN_vkEnumerateInstanceLayerProperties)load(context, "vkEnumerateInstanceLayerProperties"); +#endif /* defined(VK_VERSION_1_0) */ +#if defined(VK_VERSION_1_1) + vkEnumerateInstanceVersion = (PFN_vkEnumerateInstanceVersion)load(context, "vkEnumerateInstanceVersion"); +#endif /* defined(VK_VERSION_1_1) */ + /* VOLK_GENERATE_LOAD_LOADER */ +} + +static void volkGenLoadInstance(void* context, PFN_vkVoidFunction (*load)(void*, const char*)) +{ + /* VOLK_GENERATE_LOAD_INSTANCE */ +#if defined(VK_VERSION_1_0) + vkCreateDevice = (PFN_vkCreateDevice)load(context, "vkCreateDevice"); + vkDestroyInstance = (PFN_vkDestroyInstance)load(context, "vkDestroyInstance"); + vkEnumerateDeviceExtensionProperties = (PFN_vkEnumerateDeviceExtensionProperties)load(context, "vkEnumerateDeviceExtensionProperties"); + vkEnumerateDeviceLayerProperties = (PFN_vkEnumerateDeviceLayerProperties)load(context, "vkEnumerateDeviceLayerProperties"); + vkEnumeratePhysicalDevices = (PFN_vkEnumeratePhysicalDevices)load(context, "vkEnumeratePhysicalDevices"); + vkGetDeviceProcAddr = (PFN_vkGetDeviceProcAddr)load(context, "vkGetDeviceProcAddr"); + vkGetPhysicalDeviceFeatures = (PFN_vkGetPhysicalDeviceFeatures)load(context, "vkGetPhysicalDeviceFeatures"); + vkGetPhysicalDeviceFormatProperties = (PFN_vkGetPhysicalDeviceFormatProperties)load(context, "vkGetPhysicalDeviceFormatProperties"); + vkGetPhysicalDeviceImageFormatProperties = (PFN_vkGetPhysicalDeviceImageFormatProperties)load(context, "vkGetPhysicalDeviceImageFormatProperties"); + vkGetPhysicalDeviceMemoryProperties = (PFN_vkGetPhysicalDeviceMemoryProperties)load(context, "vkGetPhysicalDeviceMemoryProperties"); + vkGetPhysicalDeviceProperties = (PFN_vkGetPhysicalDeviceProperties)load(context, "vkGetPhysicalDeviceProperties"); + vkGetPhysicalDeviceQueueFamilyProperties = (PFN_vkGetPhysicalDeviceQueueFamilyProperties)load(context, "vkGetPhysicalDeviceQueueFamilyProperties"); + vkGetPhysicalDeviceSparseImageFormatProperties = (PFN_vkGetPhysicalDeviceSparseImageFormatProperties)load(context, "vkGetPhysicalDeviceSparseImageFormatProperties"); +#endif /* defined(VK_VERSION_1_0) */ +#if defined(VK_VERSION_1_1) + vkEnumeratePhysicalDeviceGroups = (PFN_vkEnumeratePhysicalDeviceGroups)load(context, "vkEnumeratePhysicalDeviceGroups"); + vkGetPhysicalDeviceExternalBufferProperties = (PFN_vkGetPhysicalDeviceExternalBufferProperties)load(context, "vkGetPhysicalDeviceExternalBufferProperties"); + vkGetPhysicalDeviceExternalFenceProperties = (PFN_vkGetPhysicalDeviceExternalFenceProperties)load(context, "vkGetPhysicalDeviceExternalFenceProperties"); + vkGetPhysicalDeviceExternalSemaphoreProperties = (PFN_vkGetPhysicalDeviceExternalSemaphoreProperties)load(context, "vkGetPhysicalDeviceExternalSemaphoreProperties"); + vkGetPhysicalDeviceFeatures2 = (PFN_vkGetPhysicalDeviceFeatures2)load(context, "vkGetPhysicalDeviceFeatures2"); + vkGetPhysicalDeviceFormatProperties2 = (PFN_vkGetPhysicalDeviceFormatProperties2)load(context, "vkGetPhysicalDeviceFormatProperties2"); + vkGetPhysicalDeviceImageFormatProperties2 = (PFN_vkGetPhysicalDeviceImageFormatProperties2)load(context, "vkGetPhysicalDeviceImageFormatProperties2"); + vkGetPhysicalDeviceMemoryProperties2 = (PFN_vkGetPhysicalDeviceMemoryProperties2)load(context, "vkGetPhysicalDeviceMemoryProperties2"); + vkGetPhysicalDeviceProperties2 = (PFN_vkGetPhysicalDeviceProperties2)load(context, "vkGetPhysicalDeviceProperties2"); + vkGetPhysicalDeviceQueueFamilyProperties2 = (PFN_vkGetPhysicalDeviceQueueFamilyProperties2)load(context, "vkGetPhysicalDeviceQueueFamilyProperties2"); + vkGetPhysicalDeviceSparseImageFormatProperties2 = (PFN_vkGetPhysicalDeviceSparseImageFormatProperties2)load(context, "vkGetPhysicalDeviceSparseImageFormatProperties2"); +#endif /* defined(VK_VERSION_1_1) */ +#if defined(VK_EXT_acquire_xlib_display) + vkAcquireXlibDisplayEXT = (PFN_vkAcquireXlibDisplayEXT)load(context, "vkAcquireXlibDisplayEXT"); + vkGetRandROutputDisplayEXT = (PFN_vkGetRandROutputDisplayEXT)load(context, "vkGetRandROutputDisplayEXT"); +#endif /* defined(VK_EXT_acquire_xlib_display) */ +#if defined(VK_EXT_calibrated_timestamps) + vkGetPhysicalDeviceCalibrateableTimeDomainsEXT = (PFN_vkGetPhysicalDeviceCalibrateableTimeDomainsEXT)load(context, "vkGetPhysicalDeviceCalibrateableTimeDomainsEXT"); +#endif /* defined(VK_EXT_calibrated_timestamps) */ +#if defined(VK_EXT_debug_report) + vkCreateDebugReportCallbackEXT = (PFN_vkCreateDebugReportCallbackEXT)load(context, "vkCreateDebugReportCallbackEXT"); + vkDebugReportMessageEXT = (PFN_vkDebugReportMessageEXT)load(context, "vkDebugReportMessageEXT"); + vkDestroyDebugReportCallbackEXT = (PFN_vkDestroyDebugReportCallbackEXT)load(context, "vkDestroyDebugReportCallbackEXT"); +#endif /* defined(VK_EXT_debug_report) */ +#if defined(VK_EXT_debug_utils) + vkCmdBeginDebugUtilsLabelEXT = (PFN_vkCmdBeginDebugUtilsLabelEXT)load(context, "vkCmdBeginDebugUtilsLabelEXT"); + vkCmdEndDebugUtilsLabelEXT = (PFN_vkCmdEndDebugUtilsLabelEXT)load(context, "vkCmdEndDebugUtilsLabelEXT"); + vkCmdInsertDebugUtilsLabelEXT = (PFN_vkCmdInsertDebugUtilsLabelEXT)load(context, "vkCmdInsertDebugUtilsLabelEXT"); + vkCreateDebugUtilsMessengerEXT = (PFN_vkCreateDebugUtilsMessengerEXT)load(context, "vkCreateDebugUtilsMessengerEXT"); + vkDestroyDebugUtilsMessengerEXT = (PFN_vkDestroyDebugUtilsMessengerEXT)load(context, "vkDestroyDebugUtilsMessengerEXT"); + vkQueueBeginDebugUtilsLabelEXT = (PFN_vkQueueBeginDebugUtilsLabelEXT)load(context, "vkQueueBeginDebugUtilsLabelEXT"); + vkQueueEndDebugUtilsLabelEXT = (PFN_vkQueueEndDebugUtilsLabelEXT)load(context, "vkQueueEndDebugUtilsLabelEXT"); + vkQueueInsertDebugUtilsLabelEXT = (PFN_vkQueueInsertDebugUtilsLabelEXT)load(context, "vkQueueInsertDebugUtilsLabelEXT"); + vkSetDebugUtilsObjectNameEXT = (PFN_vkSetDebugUtilsObjectNameEXT)load(context, "vkSetDebugUtilsObjectNameEXT"); + vkSetDebugUtilsObjectTagEXT = (PFN_vkSetDebugUtilsObjectTagEXT)load(context, "vkSetDebugUtilsObjectTagEXT"); + vkSubmitDebugUtilsMessageEXT = (PFN_vkSubmitDebugUtilsMessageEXT)load(context, "vkSubmitDebugUtilsMessageEXT"); +#endif /* defined(VK_EXT_debug_utils) */ +#if defined(VK_EXT_direct_mode_display) + vkReleaseDisplayEXT = (PFN_vkReleaseDisplayEXT)load(context, "vkReleaseDisplayEXT"); +#endif /* defined(VK_EXT_direct_mode_display) */ +#if defined(VK_EXT_directfb_surface) + vkCreateDirectFBSurfaceEXT = (PFN_vkCreateDirectFBSurfaceEXT)load(context, "vkCreateDirectFBSurfaceEXT"); + vkGetPhysicalDeviceDirectFBPresentationSupportEXT = (PFN_vkGetPhysicalDeviceDirectFBPresentationSupportEXT)load(context, "vkGetPhysicalDeviceDirectFBPresentationSupportEXT"); +#endif /* defined(VK_EXT_directfb_surface) */ +#if defined(VK_EXT_display_surface_counter) + vkGetPhysicalDeviceSurfaceCapabilities2EXT = (PFN_vkGetPhysicalDeviceSurfaceCapabilities2EXT)load(context, "vkGetPhysicalDeviceSurfaceCapabilities2EXT"); +#endif /* defined(VK_EXT_display_surface_counter) */ +#if defined(VK_EXT_full_screen_exclusive) + vkGetPhysicalDeviceSurfacePresentModes2EXT = (PFN_vkGetPhysicalDeviceSurfacePresentModes2EXT)load(context, "vkGetPhysicalDeviceSurfacePresentModes2EXT"); +#endif /* defined(VK_EXT_full_screen_exclusive) */ +#if defined(VK_EXT_headless_surface) + vkCreateHeadlessSurfaceEXT = (PFN_vkCreateHeadlessSurfaceEXT)load(context, "vkCreateHeadlessSurfaceEXT"); +#endif /* defined(VK_EXT_headless_surface) */ +#if defined(VK_EXT_metal_surface) + vkCreateMetalSurfaceEXT = (PFN_vkCreateMetalSurfaceEXT)load(context, "vkCreateMetalSurfaceEXT"); +#endif /* defined(VK_EXT_metal_surface) */ +#if defined(VK_EXT_sample_locations) + vkGetPhysicalDeviceMultisamplePropertiesEXT = (PFN_vkGetPhysicalDeviceMultisamplePropertiesEXT)load(context, "vkGetPhysicalDeviceMultisamplePropertiesEXT"); +#endif /* defined(VK_EXT_sample_locations) */ +#if defined(VK_EXT_tooling_info) + vkGetPhysicalDeviceToolPropertiesEXT = (PFN_vkGetPhysicalDeviceToolPropertiesEXT)load(context, "vkGetPhysicalDeviceToolPropertiesEXT"); +#endif /* defined(VK_EXT_tooling_info) */ +#if defined(VK_FUCHSIA_imagepipe_surface) + vkCreateImagePipeSurfaceFUCHSIA = (PFN_vkCreateImagePipeSurfaceFUCHSIA)load(context, "vkCreateImagePipeSurfaceFUCHSIA"); +#endif /* defined(VK_FUCHSIA_imagepipe_surface) */ +#if defined(VK_GGP_stream_descriptor_surface) + vkCreateStreamDescriptorSurfaceGGP = (PFN_vkCreateStreamDescriptorSurfaceGGP)load(context, "vkCreateStreamDescriptorSurfaceGGP"); +#endif /* defined(VK_GGP_stream_descriptor_surface) */ +#if defined(VK_KHR_android_surface) + vkCreateAndroidSurfaceKHR = (PFN_vkCreateAndroidSurfaceKHR)load(context, "vkCreateAndroidSurfaceKHR"); +#endif /* defined(VK_KHR_android_surface) */ +#if defined(VK_KHR_device_group_creation) + vkEnumeratePhysicalDeviceGroupsKHR = (PFN_vkEnumeratePhysicalDeviceGroupsKHR)load(context, "vkEnumeratePhysicalDeviceGroupsKHR"); +#endif /* defined(VK_KHR_device_group_creation) */ +#if defined(VK_KHR_display) + vkCreateDisplayModeKHR = (PFN_vkCreateDisplayModeKHR)load(context, "vkCreateDisplayModeKHR"); + vkCreateDisplayPlaneSurfaceKHR = (PFN_vkCreateDisplayPlaneSurfaceKHR)load(context, "vkCreateDisplayPlaneSurfaceKHR"); + vkGetDisplayModePropertiesKHR = (PFN_vkGetDisplayModePropertiesKHR)load(context, "vkGetDisplayModePropertiesKHR"); + vkGetDisplayPlaneCapabilitiesKHR = (PFN_vkGetDisplayPlaneCapabilitiesKHR)load(context, "vkGetDisplayPlaneCapabilitiesKHR"); + vkGetDisplayPlaneSupportedDisplaysKHR = (PFN_vkGetDisplayPlaneSupportedDisplaysKHR)load(context, "vkGetDisplayPlaneSupportedDisplaysKHR"); + vkGetPhysicalDeviceDisplayPlanePropertiesKHR = (PFN_vkGetPhysicalDeviceDisplayPlanePropertiesKHR)load(context, "vkGetPhysicalDeviceDisplayPlanePropertiesKHR"); + vkGetPhysicalDeviceDisplayPropertiesKHR = (PFN_vkGetPhysicalDeviceDisplayPropertiesKHR)load(context, "vkGetPhysicalDeviceDisplayPropertiesKHR"); +#endif /* defined(VK_KHR_display) */ +#if defined(VK_KHR_external_fence_capabilities) + vkGetPhysicalDeviceExternalFencePropertiesKHR = (PFN_vkGetPhysicalDeviceExternalFencePropertiesKHR)load(context, "vkGetPhysicalDeviceExternalFencePropertiesKHR"); +#endif /* defined(VK_KHR_external_fence_capabilities) */ +#if defined(VK_KHR_external_memory_capabilities) + vkGetPhysicalDeviceExternalBufferPropertiesKHR = (PFN_vkGetPhysicalDeviceExternalBufferPropertiesKHR)load(context, "vkGetPhysicalDeviceExternalBufferPropertiesKHR"); +#endif /* defined(VK_KHR_external_memory_capabilities) */ +#if defined(VK_KHR_external_semaphore_capabilities) + vkGetPhysicalDeviceExternalSemaphorePropertiesKHR = (PFN_vkGetPhysicalDeviceExternalSemaphorePropertiesKHR)load(context, "vkGetPhysicalDeviceExternalSemaphorePropertiesKHR"); +#endif /* defined(VK_KHR_external_semaphore_capabilities) */ +#if defined(VK_KHR_get_display_properties2) + vkGetDisplayModeProperties2KHR = (PFN_vkGetDisplayModeProperties2KHR)load(context, "vkGetDisplayModeProperties2KHR"); + vkGetDisplayPlaneCapabilities2KHR = (PFN_vkGetDisplayPlaneCapabilities2KHR)load(context, "vkGetDisplayPlaneCapabilities2KHR"); + vkGetPhysicalDeviceDisplayPlaneProperties2KHR = (PFN_vkGetPhysicalDeviceDisplayPlaneProperties2KHR)load(context, "vkGetPhysicalDeviceDisplayPlaneProperties2KHR"); + vkGetPhysicalDeviceDisplayProperties2KHR = (PFN_vkGetPhysicalDeviceDisplayProperties2KHR)load(context, "vkGetPhysicalDeviceDisplayProperties2KHR"); +#endif /* defined(VK_KHR_get_display_properties2) */ +#if defined(VK_KHR_get_physical_device_properties2) + vkGetPhysicalDeviceFeatures2KHR = (PFN_vkGetPhysicalDeviceFeatures2KHR)load(context, "vkGetPhysicalDeviceFeatures2KHR"); + vkGetPhysicalDeviceFormatProperties2KHR = (PFN_vkGetPhysicalDeviceFormatProperties2KHR)load(context, "vkGetPhysicalDeviceFormatProperties2KHR"); + vkGetPhysicalDeviceImageFormatProperties2KHR = (PFN_vkGetPhysicalDeviceImageFormatProperties2KHR)load(context, "vkGetPhysicalDeviceImageFormatProperties2KHR"); + vkGetPhysicalDeviceMemoryProperties2KHR = (PFN_vkGetPhysicalDeviceMemoryProperties2KHR)load(context, "vkGetPhysicalDeviceMemoryProperties2KHR"); + vkGetPhysicalDeviceProperties2KHR = (PFN_vkGetPhysicalDeviceProperties2KHR)load(context, "vkGetPhysicalDeviceProperties2KHR"); + vkGetPhysicalDeviceQueueFamilyProperties2KHR = (PFN_vkGetPhysicalDeviceQueueFamilyProperties2KHR)load(context, "vkGetPhysicalDeviceQueueFamilyProperties2KHR"); + vkGetPhysicalDeviceSparseImageFormatProperties2KHR = (PFN_vkGetPhysicalDeviceSparseImageFormatProperties2KHR)load(context, "vkGetPhysicalDeviceSparseImageFormatProperties2KHR"); +#endif /* defined(VK_KHR_get_physical_device_properties2) */ +#if defined(VK_KHR_get_surface_capabilities2) + vkGetPhysicalDeviceSurfaceCapabilities2KHR = (PFN_vkGetPhysicalDeviceSurfaceCapabilities2KHR)load(context, "vkGetPhysicalDeviceSurfaceCapabilities2KHR"); + vkGetPhysicalDeviceSurfaceFormats2KHR = (PFN_vkGetPhysicalDeviceSurfaceFormats2KHR)load(context, "vkGetPhysicalDeviceSurfaceFormats2KHR"); +#endif /* defined(VK_KHR_get_surface_capabilities2) */ +#if defined(VK_KHR_performance_query) + vkEnumeratePhysicalDeviceQueueFamilyPerformanceQueryCountersKHR = (PFN_vkEnumeratePhysicalDeviceQueueFamilyPerformanceQueryCountersKHR)load(context, "vkEnumeratePhysicalDeviceQueueFamilyPerformanceQueryCountersKHR"); + vkGetPhysicalDeviceQueueFamilyPerformanceQueryPassesKHR = (PFN_vkGetPhysicalDeviceQueueFamilyPerformanceQueryPassesKHR)load(context, "vkGetPhysicalDeviceQueueFamilyPerformanceQueryPassesKHR"); +#endif /* defined(VK_KHR_performance_query) */ +#if defined(VK_KHR_surface) + vkDestroySurfaceKHR = (PFN_vkDestroySurfaceKHR)load(context, "vkDestroySurfaceKHR"); + vkGetPhysicalDeviceSurfaceCapabilitiesKHR = (PFN_vkGetPhysicalDeviceSurfaceCapabilitiesKHR)load(context, "vkGetPhysicalDeviceSurfaceCapabilitiesKHR"); + vkGetPhysicalDeviceSurfaceFormatsKHR = (PFN_vkGetPhysicalDeviceSurfaceFormatsKHR)load(context, "vkGetPhysicalDeviceSurfaceFormatsKHR"); + vkGetPhysicalDeviceSurfacePresentModesKHR = (PFN_vkGetPhysicalDeviceSurfacePresentModesKHR)load(context, "vkGetPhysicalDeviceSurfacePresentModesKHR"); + vkGetPhysicalDeviceSurfaceSupportKHR = (PFN_vkGetPhysicalDeviceSurfaceSupportKHR)load(context, "vkGetPhysicalDeviceSurfaceSupportKHR"); +#endif /* defined(VK_KHR_surface) */ +#if defined(VK_KHR_wayland_surface) + vkCreateWaylandSurfaceKHR = (PFN_vkCreateWaylandSurfaceKHR)load(context, "vkCreateWaylandSurfaceKHR"); + vkGetPhysicalDeviceWaylandPresentationSupportKHR = (PFN_vkGetPhysicalDeviceWaylandPresentationSupportKHR)load(context, "vkGetPhysicalDeviceWaylandPresentationSupportKHR"); +#endif /* defined(VK_KHR_wayland_surface) */ +#if defined(VK_KHR_win32_surface) + vkCreateWin32SurfaceKHR = (PFN_vkCreateWin32SurfaceKHR)load(context, "vkCreateWin32SurfaceKHR"); + vkGetPhysicalDeviceWin32PresentationSupportKHR = (PFN_vkGetPhysicalDeviceWin32PresentationSupportKHR)load(context, "vkGetPhysicalDeviceWin32PresentationSupportKHR"); +#endif /* defined(VK_KHR_win32_surface) */ +#if defined(VK_KHR_xcb_surface) + vkCreateXcbSurfaceKHR = (PFN_vkCreateXcbSurfaceKHR)load(context, "vkCreateXcbSurfaceKHR"); + vkGetPhysicalDeviceXcbPresentationSupportKHR = (PFN_vkGetPhysicalDeviceXcbPresentationSupportKHR)load(context, "vkGetPhysicalDeviceXcbPresentationSupportKHR"); +#endif /* defined(VK_KHR_xcb_surface) */ +#if defined(VK_KHR_xlib_surface) + vkCreateXlibSurfaceKHR = (PFN_vkCreateXlibSurfaceKHR)load(context, "vkCreateXlibSurfaceKHR"); + vkGetPhysicalDeviceXlibPresentationSupportKHR = (PFN_vkGetPhysicalDeviceXlibPresentationSupportKHR)load(context, "vkGetPhysicalDeviceXlibPresentationSupportKHR"); +#endif /* defined(VK_KHR_xlib_surface) */ +#if defined(VK_MVK_ios_surface) + vkCreateIOSSurfaceMVK = (PFN_vkCreateIOSSurfaceMVK)load(context, "vkCreateIOSSurfaceMVK"); +#endif /* defined(VK_MVK_ios_surface) */ +#if defined(VK_MVK_macos_surface) + vkCreateMacOSSurfaceMVK = (PFN_vkCreateMacOSSurfaceMVK)load(context, "vkCreateMacOSSurfaceMVK"); +#endif /* defined(VK_MVK_macos_surface) */ +#if defined(VK_NN_vi_surface) + vkCreateViSurfaceNN = (PFN_vkCreateViSurfaceNN)load(context, "vkCreateViSurfaceNN"); +#endif /* defined(VK_NN_vi_surface) */ +#if defined(VK_NV_cooperative_matrix) + vkGetPhysicalDeviceCooperativeMatrixPropertiesNV = (PFN_vkGetPhysicalDeviceCooperativeMatrixPropertiesNV)load(context, "vkGetPhysicalDeviceCooperativeMatrixPropertiesNV"); +#endif /* defined(VK_NV_cooperative_matrix) */ +#if defined(VK_NV_coverage_reduction_mode) + vkGetPhysicalDeviceSupportedFramebufferMixedSamplesCombinationsNV = (PFN_vkGetPhysicalDeviceSupportedFramebufferMixedSamplesCombinationsNV)load(context, "vkGetPhysicalDeviceSupportedFramebufferMixedSamplesCombinationsNV"); +#endif /* defined(VK_NV_coverage_reduction_mode) */ +#if defined(VK_NV_external_memory_capabilities) + vkGetPhysicalDeviceExternalImageFormatPropertiesNV = (PFN_vkGetPhysicalDeviceExternalImageFormatPropertiesNV)load(context, "vkGetPhysicalDeviceExternalImageFormatPropertiesNV"); +#endif /* defined(VK_NV_external_memory_capabilities) */ +#if (defined(VK_KHR_device_group) && defined(VK_KHR_surface)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) + vkGetPhysicalDevicePresentRectanglesKHR = (PFN_vkGetPhysicalDevicePresentRectanglesKHR)load(context, "vkGetPhysicalDevicePresentRectanglesKHR"); +#endif /* (defined(VK_KHR_device_group) && defined(VK_KHR_surface)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) */ + /* VOLK_GENERATE_LOAD_INSTANCE */ +} + +static void volkGenLoadDevice(void* context, PFN_vkVoidFunction (*load)(void*, const char*)) +{ + /* VOLK_GENERATE_LOAD_DEVICE */ +#if defined(VK_VERSION_1_0) + vkAllocateCommandBuffers = (PFN_vkAllocateCommandBuffers)load(context, "vkAllocateCommandBuffers"); + vkAllocateDescriptorSets = (PFN_vkAllocateDescriptorSets)load(context, "vkAllocateDescriptorSets"); + vkAllocateMemory = (PFN_vkAllocateMemory)load(context, "vkAllocateMemory"); + vkBeginCommandBuffer = (PFN_vkBeginCommandBuffer)load(context, "vkBeginCommandBuffer"); + vkBindBufferMemory = (PFN_vkBindBufferMemory)load(context, "vkBindBufferMemory"); + vkBindImageMemory = (PFN_vkBindImageMemory)load(context, "vkBindImageMemory"); + vkCmdBeginQuery = (PFN_vkCmdBeginQuery)load(context, "vkCmdBeginQuery"); + vkCmdBeginRenderPass = (PFN_vkCmdBeginRenderPass)load(context, "vkCmdBeginRenderPass"); + vkCmdBindDescriptorSets = (PFN_vkCmdBindDescriptorSets)load(context, "vkCmdBindDescriptorSets"); + vkCmdBindIndexBuffer = (PFN_vkCmdBindIndexBuffer)load(context, "vkCmdBindIndexBuffer"); + vkCmdBindPipeline = (PFN_vkCmdBindPipeline)load(context, "vkCmdBindPipeline"); + vkCmdBindVertexBuffers = (PFN_vkCmdBindVertexBuffers)load(context, "vkCmdBindVertexBuffers"); + vkCmdBlitImage = (PFN_vkCmdBlitImage)load(context, "vkCmdBlitImage"); + vkCmdClearAttachments = (PFN_vkCmdClearAttachments)load(context, "vkCmdClearAttachments"); + vkCmdClearColorImage = (PFN_vkCmdClearColorImage)load(context, "vkCmdClearColorImage"); + vkCmdClearDepthStencilImage = (PFN_vkCmdClearDepthStencilImage)load(context, "vkCmdClearDepthStencilImage"); + vkCmdCopyBuffer = (PFN_vkCmdCopyBuffer)load(context, "vkCmdCopyBuffer"); + vkCmdCopyBufferToImage = (PFN_vkCmdCopyBufferToImage)load(context, "vkCmdCopyBufferToImage"); + vkCmdCopyImage = (PFN_vkCmdCopyImage)load(context, "vkCmdCopyImage"); + vkCmdCopyImageToBuffer = (PFN_vkCmdCopyImageToBuffer)load(context, "vkCmdCopyImageToBuffer"); + vkCmdCopyQueryPoolResults = (PFN_vkCmdCopyQueryPoolResults)load(context, "vkCmdCopyQueryPoolResults"); + vkCmdDispatch = (PFN_vkCmdDispatch)load(context, "vkCmdDispatch"); + vkCmdDispatchIndirect = (PFN_vkCmdDispatchIndirect)load(context, "vkCmdDispatchIndirect"); + vkCmdDraw = (PFN_vkCmdDraw)load(context, "vkCmdDraw"); + vkCmdDrawIndexed = (PFN_vkCmdDrawIndexed)load(context, "vkCmdDrawIndexed"); + vkCmdDrawIndexedIndirect = (PFN_vkCmdDrawIndexedIndirect)load(context, "vkCmdDrawIndexedIndirect"); + vkCmdDrawIndirect = (PFN_vkCmdDrawIndirect)load(context, "vkCmdDrawIndirect"); + vkCmdEndQuery = (PFN_vkCmdEndQuery)load(context, "vkCmdEndQuery"); + vkCmdEndRenderPass = (PFN_vkCmdEndRenderPass)load(context, "vkCmdEndRenderPass"); + vkCmdExecuteCommands = (PFN_vkCmdExecuteCommands)load(context, "vkCmdExecuteCommands"); + vkCmdFillBuffer = (PFN_vkCmdFillBuffer)load(context, "vkCmdFillBuffer"); + vkCmdNextSubpass = (PFN_vkCmdNextSubpass)load(context, "vkCmdNextSubpass"); + vkCmdPipelineBarrier = (PFN_vkCmdPipelineBarrier)load(context, "vkCmdPipelineBarrier"); + vkCmdPushConstants = (PFN_vkCmdPushConstants)load(context, "vkCmdPushConstants"); + vkCmdResetEvent = (PFN_vkCmdResetEvent)load(context, "vkCmdResetEvent"); + vkCmdResetQueryPool = (PFN_vkCmdResetQueryPool)load(context, "vkCmdResetQueryPool"); + vkCmdResolveImage = (PFN_vkCmdResolveImage)load(context, "vkCmdResolveImage"); + vkCmdSetBlendConstants = (PFN_vkCmdSetBlendConstants)load(context, "vkCmdSetBlendConstants"); + vkCmdSetDepthBias = (PFN_vkCmdSetDepthBias)load(context, "vkCmdSetDepthBias"); + vkCmdSetDepthBounds = (PFN_vkCmdSetDepthBounds)load(context, "vkCmdSetDepthBounds"); + vkCmdSetEvent = (PFN_vkCmdSetEvent)load(context, "vkCmdSetEvent"); + vkCmdSetLineWidth = (PFN_vkCmdSetLineWidth)load(context, "vkCmdSetLineWidth"); + vkCmdSetScissor = (PFN_vkCmdSetScissor)load(context, "vkCmdSetScissor"); + vkCmdSetStencilCompareMask = (PFN_vkCmdSetStencilCompareMask)load(context, "vkCmdSetStencilCompareMask"); + vkCmdSetStencilReference = (PFN_vkCmdSetStencilReference)load(context, "vkCmdSetStencilReference"); + vkCmdSetStencilWriteMask = (PFN_vkCmdSetStencilWriteMask)load(context, "vkCmdSetStencilWriteMask"); + vkCmdSetViewport = (PFN_vkCmdSetViewport)load(context, "vkCmdSetViewport"); + vkCmdUpdateBuffer = (PFN_vkCmdUpdateBuffer)load(context, "vkCmdUpdateBuffer"); + vkCmdWaitEvents = (PFN_vkCmdWaitEvents)load(context, "vkCmdWaitEvents"); + vkCmdWriteTimestamp = (PFN_vkCmdWriteTimestamp)load(context, "vkCmdWriteTimestamp"); + vkCreateBuffer = (PFN_vkCreateBuffer)load(context, "vkCreateBuffer"); + vkCreateBufferView = (PFN_vkCreateBufferView)load(context, "vkCreateBufferView"); + vkCreateCommandPool = (PFN_vkCreateCommandPool)load(context, "vkCreateCommandPool"); + vkCreateComputePipelines = (PFN_vkCreateComputePipelines)load(context, "vkCreateComputePipelines"); + vkCreateDescriptorPool = (PFN_vkCreateDescriptorPool)load(context, "vkCreateDescriptorPool"); + vkCreateDescriptorSetLayout = (PFN_vkCreateDescriptorSetLayout)load(context, "vkCreateDescriptorSetLayout"); + vkCreateEvent = (PFN_vkCreateEvent)load(context, "vkCreateEvent"); + vkCreateFence = (PFN_vkCreateFence)load(context, "vkCreateFence"); + vkCreateFramebuffer = (PFN_vkCreateFramebuffer)load(context, "vkCreateFramebuffer"); + vkCreateGraphicsPipelines = (PFN_vkCreateGraphicsPipelines)load(context, "vkCreateGraphicsPipelines"); + vkCreateImage = (PFN_vkCreateImage)load(context, "vkCreateImage"); + vkCreateImageView = (PFN_vkCreateImageView)load(context, "vkCreateImageView"); + vkCreatePipelineCache = (PFN_vkCreatePipelineCache)load(context, "vkCreatePipelineCache"); + vkCreatePipelineLayout = (PFN_vkCreatePipelineLayout)load(context, "vkCreatePipelineLayout"); + vkCreateQueryPool = (PFN_vkCreateQueryPool)load(context, "vkCreateQueryPool"); + vkCreateRenderPass = (PFN_vkCreateRenderPass)load(context, "vkCreateRenderPass"); + vkCreateSampler = (PFN_vkCreateSampler)load(context, "vkCreateSampler"); + vkCreateSemaphore = (PFN_vkCreateSemaphore)load(context, "vkCreateSemaphore"); + vkCreateShaderModule = (PFN_vkCreateShaderModule)load(context, "vkCreateShaderModule"); + vkDestroyBuffer = (PFN_vkDestroyBuffer)load(context, "vkDestroyBuffer"); + vkDestroyBufferView = (PFN_vkDestroyBufferView)load(context, "vkDestroyBufferView"); + vkDestroyCommandPool = (PFN_vkDestroyCommandPool)load(context, "vkDestroyCommandPool"); + vkDestroyDescriptorPool = (PFN_vkDestroyDescriptorPool)load(context, "vkDestroyDescriptorPool"); + vkDestroyDescriptorSetLayout = (PFN_vkDestroyDescriptorSetLayout)load(context, "vkDestroyDescriptorSetLayout"); + vkDestroyDevice = (PFN_vkDestroyDevice)load(context, "vkDestroyDevice"); + vkDestroyEvent = (PFN_vkDestroyEvent)load(context, "vkDestroyEvent"); + vkDestroyFence = (PFN_vkDestroyFence)load(context, "vkDestroyFence"); + vkDestroyFramebuffer = (PFN_vkDestroyFramebuffer)load(context, "vkDestroyFramebuffer"); + vkDestroyImage = (PFN_vkDestroyImage)load(context, "vkDestroyImage"); + vkDestroyImageView = (PFN_vkDestroyImageView)load(context, "vkDestroyImageView"); + vkDestroyPipeline = (PFN_vkDestroyPipeline)load(context, "vkDestroyPipeline"); + vkDestroyPipelineCache = (PFN_vkDestroyPipelineCache)load(context, "vkDestroyPipelineCache"); + vkDestroyPipelineLayout = (PFN_vkDestroyPipelineLayout)load(context, "vkDestroyPipelineLayout"); + vkDestroyQueryPool = (PFN_vkDestroyQueryPool)load(context, "vkDestroyQueryPool"); + vkDestroyRenderPass = (PFN_vkDestroyRenderPass)load(context, "vkDestroyRenderPass"); + vkDestroySampler = (PFN_vkDestroySampler)load(context, "vkDestroySampler"); + vkDestroySemaphore = (PFN_vkDestroySemaphore)load(context, "vkDestroySemaphore"); + vkDestroyShaderModule = (PFN_vkDestroyShaderModule)load(context, "vkDestroyShaderModule"); + vkDeviceWaitIdle = (PFN_vkDeviceWaitIdle)load(context, "vkDeviceWaitIdle"); + vkEndCommandBuffer = (PFN_vkEndCommandBuffer)load(context, "vkEndCommandBuffer"); + vkFlushMappedMemoryRanges = (PFN_vkFlushMappedMemoryRanges)load(context, "vkFlushMappedMemoryRanges"); + vkFreeCommandBuffers = (PFN_vkFreeCommandBuffers)load(context, "vkFreeCommandBuffers"); + vkFreeDescriptorSets = (PFN_vkFreeDescriptorSets)load(context, "vkFreeDescriptorSets"); + vkFreeMemory = (PFN_vkFreeMemory)load(context, "vkFreeMemory"); + vkGetBufferMemoryRequirements = (PFN_vkGetBufferMemoryRequirements)load(context, "vkGetBufferMemoryRequirements"); + vkGetDeviceMemoryCommitment = (PFN_vkGetDeviceMemoryCommitment)load(context, "vkGetDeviceMemoryCommitment"); + vkGetDeviceQueue = (PFN_vkGetDeviceQueue)load(context, "vkGetDeviceQueue"); + vkGetEventStatus = (PFN_vkGetEventStatus)load(context, "vkGetEventStatus"); + vkGetFenceStatus = (PFN_vkGetFenceStatus)load(context, "vkGetFenceStatus"); + vkGetImageMemoryRequirements = (PFN_vkGetImageMemoryRequirements)load(context, "vkGetImageMemoryRequirements"); + vkGetImageSparseMemoryRequirements = (PFN_vkGetImageSparseMemoryRequirements)load(context, "vkGetImageSparseMemoryRequirements"); + vkGetImageSubresourceLayout = (PFN_vkGetImageSubresourceLayout)load(context, "vkGetImageSubresourceLayout"); + vkGetPipelineCacheData = (PFN_vkGetPipelineCacheData)load(context, "vkGetPipelineCacheData"); + vkGetQueryPoolResults = (PFN_vkGetQueryPoolResults)load(context, "vkGetQueryPoolResults"); + vkGetRenderAreaGranularity = (PFN_vkGetRenderAreaGranularity)load(context, "vkGetRenderAreaGranularity"); + vkInvalidateMappedMemoryRanges = (PFN_vkInvalidateMappedMemoryRanges)load(context, "vkInvalidateMappedMemoryRanges"); + vkMapMemory = (PFN_vkMapMemory)load(context, "vkMapMemory"); + vkMergePipelineCaches = (PFN_vkMergePipelineCaches)load(context, "vkMergePipelineCaches"); + vkQueueBindSparse = (PFN_vkQueueBindSparse)load(context, "vkQueueBindSparse"); + vkQueueSubmit = (PFN_vkQueueSubmit)load(context, "vkQueueSubmit"); + vkQueueWaitIdle = (PFN_vkQueueWaitIdle)load(context, "vkQueueWaitIdle"); + vkResetCommandBuffer = (PFN_vkResetCommandBuffer)load(context, "vkResetCommandBuffer"); + vkResetCommandPool = (PFN_vkResetCommandPool)load(context, "vkResetCommandPool"); + vkResetDescriptorPool = (PFN_vkResetDescriptorPool)load(context, "vkResetDescriptorPool"); + vkResetEvent = (PFN_vkResetEvent)load(context, "vkResetEvent"); + vkResetFences = (PFN_vkResetFences)load(context, "vkResetFences"); + vkSetEvent = (PFN_vkSetEvent)load(context, "vkSetEvent"); + vkUnmapMemory = (PFN_vkUnmapMemory)load(context, "vkUnmapMemory"); + vkUpdateDescriptorSets = (PFN_vkUpdateDescriptorSets)load(context, "vkUpdateDescriptorSets"); + vkWaitForFences = (PFN_vkWaitForFences)load(context, "vkWaitForFences"); +#endif /* defined(VK_VERSION_1_0) */ +#if defined(VK_VERSION_1_1) + vkBindBufferMemory2 = (PFN_vkBindBufferMemory2)load(context, "vkBindBufferMemory2"); + vkBindImageMemory2 = (PFN_vkBindImageMemory2)load(context, "vkBindImageMemory2"); + vkCmdDispatchBase = (PFN_vkCmdDispatchBase)load(context, "vkCmdDispatchBase"); + vkCmdSetDeviceMask = (PFN_vkCmdSetDeviceMask)load(context, "vkCmdSetDeviceMask"); + vkCreateDescriptorUpdateTemplate = (PFN_vkCreateDescriptorUpdateTemplate)load(context, "vkCreateDescriptorUpdateTemplate"); + vkCreateSamplerYcbcrConversion = (PFN_vkCreateSamplerYcbcrConversion)load(context, "vkCreateSamplerYcbcrConversion"); + vkDestroyDescriptorUpdateTemplate = (PFN_vkDestroyDescriptorUpdateTemplate)load(context, "vkDestroyDescriptorUpdateTemplate"); + vkDestroySamplerYcbcrConversion = (PFN_vkDestroySamplerYcbcrConversion)load(context, "vkDestroySamplerYcbcrConversion"); + vkGetBufferMemoryRequirements2 = (PFN_vkGetBufferMemoryRequirements2)load(context, "vkGetBufferMemoryRequirements2"); + vkGetDescriptorSetLayoutSupport = (PFN_vkGetDescriptorSetLayoutSupport)load(context, "vkGetDescriptorSetLayoutSupport"); + vkGetDeviceGroupPeerMemoryFeatures = (PFN_vkGetDeviceGroupPeerMemoryFeatures)load(context, "vkGetDeviceGroupPeerMemoryFeatures"); + vkGetDeviceQueue2 = (PFN_vkGetDeviceQueue2)load(context, "vkGetDeviceQueue2"); + vkGetImageMemoryRequirements2 = (PFN_vkGetImageMemoryRequirements2)load(context, "vkGetImageMemoryRequirements2"); + vkGetImageSparseMemoryRequirements2 = (PFN_vkGetImageSparseMemoryRequirements2)load(context, "vkGetImageSparseMemoryRequirements2"); + vkTrimCommandPool = (PFN_vkTrimCommandPool)load(context, "vkTrimCommandPool"); + vkUpdateDescriptorSetWithTemplate = (PFN_vkUpdateDescriptorSetWithTemplate)load(context, "vkUpdateDescriptorSetWithTemplate"); +#endif /* defined(VK_VERSION_1_1) */ +#if defined(VK_VERSION_1_2) + vkCmdBeginRenderPass2 = (PFN_vkCmdBeginRenderPass2)load(context, "vkCmdBeginRenderPass2"); + vkCmdDrawIndexedIndirectCount = (PFN_vkCmdDrawIndexedIndirectCount)load(context, "vkCmdDrawIndexedIndirectCount"); + vkCmdDrawIndirectCount = (PFN_vkCmdDrawIndirectCount)load(context, "vkCmdDrawIndirectCount"); + vkCmdEndRenderPass2 = (PFN_vkCmdEndRenderPass2)load(context, "vkCmdEndRenderPass2"); + vkCmdNextSubpass2 = (PFN_vkCmdNextSubpass2)load(context, "vkCmdNextSubpass2"); + vkCreateRenderPass2 = (PFN_vkCreateRenderPass2)load(context, "vkCreateRenderPass2"); + vkGetBufferDeviceAddress = (PFN_vkGetBufferDeviceAddress)load(context, "vkGetBufferDeviceAddress"); + vkGetBufferOpaqueCaptureAddress = (PFN_vkGetBufferOpaqueCaptureAddress)load(context, "vkGetBufferOpaqueCaptureAddress"); + vkGetDeviceMemoryOpaqueCaptureAddress = (PFN_vkGetDeviceMemoryOpaqueCaptureAddress)load(context, "vkGetDeviceMemoryOpaqueCaptureAddress"); + vkGetSemaphoreCounterValue = (PFN_vkGetSemaphoreCounterValue)load(context, "vkGetSemaphoreCounterValue"); + vkResetQueryPool = (PFN_vkResetQueryPool)load(context, "vkResetQueryPool"); + vkSignalSemaphore = (PFN_vkSignalSemaphore)load(context, "vkSignalSemaphore"); + vkWaitSemaphores = (PFN_vkWaitSemaphores)load(context, "vkWaitSemaphores"); +#endif /* defined(VK_VERSION_1_2) */ +#if defined(VK_AMD_buffer_marker) + vkCmdWriteBufferMarkerAMD = (PFN_vkCmdWriteBufferMarkerAMD)load(context, "vkCmdWriteBufferMarkerAMD"); +#endif /* defined(VK_AMD_buffer_marker) */ +#if defined(VK_AMD_display_native_hdr) + vkSetLocalDimmingAMD = (PFN_vkSetLocalDimmingAMD)load(context, "vkSetLocalDimmingAMD"); +#endif /* defined(VK_AMD_display_native_hdr) */ +#if defined(VK_AMD_draw_indirect_count) + vkCmdDrawIndexedIndirectCountAMD = (PFN_vkCmdDrawIndexedIndirectCountAMD)load(context, "vkCmdDrawIndexedIndirectCountAMD"); + vkCmdDrawIndirectCountAMD = (PFN_vkCmdDrawIndirectCountAMD)load(context, "vkCmdDrawIndirectCountAMD"); +#endif /* defined(VK_AMD_draw_indirect_count) */ +#if defined(VK_AMD_shader_info) + vkGetShaderInfoAMD = (PFN_vkGetShaderInfoAMD)load(context, "vkGetShaderInfoAMD"); +#endif /* defined(VK_AMD_shader_info) */ +#if defined(VK_ANDROID_external_memory_android_hardware_buffer) + vkGetAndroidHardwareBufferPropertiesANDROID = (PFN_vkGetAndroidHardwareBufferPropertiesANDROID)load(context, "vkGetAndroidHardwareBufferPropertiesANDROID"); + vkGetMemoryAndroidHardwareBufferANDROID = (PFN_vkGetMemoryAndroidHardwareBufferANDROID)load(context, "vkGetMemoryAndroidHardwareBufferANDROID"); +#endif /* defined(VK_ANDROID_external_memory_android_hardware_buffer) */ +#if defined(VK_EXT_buffer_device_address) + vkGetBufferDeviceAddressEXT = (PFN_vkGetBufferDeviceAddressEXT)load(context, "vkGetBufferDeviceAddressEXT"); +#endif /* defined(VK_EXT_buffer_device_address) */ +#if defined(VK_EXT_calibrated_timestamps) + vkGetCalibratedTimestampsEXT = (PFN_vkGetCalibratedTimestampsEXT)load(context, "vkGetCalibratedTimestampsEXT"); +#endif /* defined(VK_EXT_calibrated_timestamps) */ +#if defined(VK_EXT_conditional_rendering) + vkCmdBeginConditionalRenderingEXT = (PFN_vkCmdBeginConditionalRenderingEXT)load(context, "vkCmdBeginConditionalRenderingEXT"); + vkCmdEndConditionalRenderingEXT = (PFN_vkCmdEndConditionalRenderingEXT)load(context, "vkCmdEndConditionalRenderingEXT"); +#endif /* defined(VK_EXT_conditional_rendering) */ +#if defined(VK_EXT_debug_marker) + vkCmdDebugMarkerBeginEXT = (PFN_vkCmdDebugMarkerBeginEXT)load(context, "vkCmdDebugMarkerBeginEXT"); + vkCmdDebugMarkerEndEXT = (PFN_vkCmdDebugMarkerEndEXT)load(context, "vkCmdDebugMarkerEndEXT"); + vkCmdDebugMarkerInsertEXT = (PFN_vkCmdDebugMarkerInsertEXT)load(context, "vkCmdDebugMarkerInsertEXT"); + vkDebugMarkerSetObjectNameEXT = (PFN_vkDebugMarkerSetObjectNameEXT)load(context, "vkDebugMarkerSetObjectNameEXT"); + vkDebugMarkerSetObjectTagEXT = (PFN_vkDebugMarkerSetObjectTagEXT)load(context, "vkDebugMarkerSetObjectTagEXT"); +#endif /* defined(VK_EXT_debug_marker) */ +#if defined(VK_EXT_discard_rectangles) + vkCmdSetDiscardRectangleEXT = (PFN_vkCmdSetDiscardRectangleEXT)load(context, "vkCmdSetDiscardRectangleEXT"); +#endif /* defined(VK_EXT_discard_rectangles) */ +#if defined(VK_EXT_display_control) + vkDisplayPowerControlEXT = (PFN_vkDisplayPowerControlEXT)load(context, "vkDisplayPowerControlEXT"); + vkGetSwapchainCounterEXT = (PFN_vkGetSwapchainCounterEXT)load(context, "vkGetSwapchainCounterEXT"); + vkRegisterDeviceEventEXT = (PFN_vkRegisterDeviceEventEXT)load(context, "vkRegisterDeviceEventEXT"); + vkRegisterDisplayEventEXT = (PFN_vkRegisterDisplayEventEXT)load(context, "vkRegisterDisplayEventEXT"); +#endif /* defined(VK_EXT_display_control) */ +#if defined(VK_EXT_extended_dynamic_state) + vkCmdBindVertexBuffers2EXT = (PFN_vkCmdBindVertexBuffers2EXT)load(context, "vkCmdBindVertexBuffers2EXT"); + vkCmdSetCullModeEXT = (PFN_vkCmdSetCullModeEXT)load(context, "vkCmdSetCullModeEXT"); + vkCmdSetDepthBoundsTestEnableEXT = (PFN_vkCmdSetDepthBoundsTestEnableEXT)load(context, "vkCmdSetDepthBoundsTestEnableEXT"); + vkCmdSetDepthCompareOpEXT = (PFN_vkCmdSetDepthCompareOpEXT)load(context, "vkCmdSetDepthCompareOpEXT"); + vkCmdSetDepthTestEnableEXT = (PFN_vkCmdSetDepthTestEnableEXT)load(context, "vkCmdSetDepthTestEnableEXT"); + vkCmdSetDepthWriteEnableEXT = (PFN_vkCmdSetDepthWriteEnableEXT)load(context, "vkCmdSetDepthWriteEnableEXT"); + vkCmdSetFrontFaceEXT = (PFN_vkCmdSetFrontFaceEXT)load(context, "vkCmdSetFrontFaceEXT"); + vkCmdSetPrimitiveTopologyEXT = (PFN_vkCmdSetPrimitiveTopologyEXT)load(context, "vkCmdSetPrimitiveTopologyEXT"); + vkCmdSetScissorWithCountEXT = (PFN_vkCmdSetScissorWithCountEXT)load(context, "vkCmdSetScissorWithCountEXT"); + vkCmdSetStencilOpEXT = (PFN_vkCmdSetStencilOpEXT)load(context, "vkCmdSetStencilOpEXT"); + vkCmdSetStencilTestEnableEXT = (PFN_vkCmdSetStencilTestEnableEXT)load(context, "vkCmdSetStencilTestEnableEXT"); + vkCmdSetViewportWithCountEXT = (PFN_vkCmdSetViewportWithCountEXT)load(context, "vkCmdSetViewportWithCountEXT"); +#endif /* defined(VK_EXT_extended_dynamic_state) */ +#if defined(VK_EXT_external_memory_host) + vkGetMemoryHostPointerPropertiesEXT = (PFN_vkGetMemoryHostPointerPropertiesEXT)load(context, "vkGetMemoryHostPointerPropertiesEXT"); +#endif /* defined(VK_EXT_external_memory_host) */ +#if defined(VK_EXT_full_screen_exclusive) + vkAcquireFullScreenExclusiveModeEXT = (PFN_vkAcquireFullScreenExclusiveModeEXT)load(context, "vkAcquireFullScreenExclusiveModeEXT"); + vkReleaseFullScreenExclusiveModeEXT = (PFN_vkReleaseFullScreenExclusiveModeEXT)load(context, "vkReleaseFullScreenExclusiveModeEXT"); +#endif /* defined(VK_EXT_full_screen_exclusive) */ +#if defined(VK_EXT_hdr_metadata) + vkSetHdrMetadataEXT = (PFN_vkSetHdrMetadataEXT)load(context, "vkSetHdrMetadataEXT"); +#endif /* defined(VK_EXT_hdr_metadata) */ +#if defined(VK_EXT_host_query_reset) + vkResetQueryPoolEXT = (PFN_vkResetQueryPoolEXT)load(context, "vkResetQueryPoolEXT"); +#endif /* defined(VK_EXT_host_query_reset) */ +#if defined(VK_EXT_image_drm_format_modifier) + vkGetImageDrmFormatModifierPropertiesEXT = (PFN_vkGetImageDrmFormatModifierPropertiesEXT)load(context, "vkGetImageDrmFormatModifierPropertiesEXT"); +#endif /* defined(VK_EXT_image_drm_format_modifier) */ +#if defined(VK_EXT_line_rasterization) + vkCmdSetLineStippleEXT = (PFN_vkCmdSetLineStippleEXT)load(context, "vkCmdSetLineStippleEXT"); +#endif /* defined(VK_EXT_line_rasterization) */ +#if defined(VK_EXT_private_data) + vkCreatePrivateDataSlotEXT = (PFN_vkCreatePrivateDataSlotEXT)load(context, "vkCreatePrivateDataSlotEXT"); + vkDestroyPrivateDataSlotEXT = (PFN_vkDestroyPrivateDataSlotEXT)load(context, "vkDestroyPrivateDataSlotEXT"); + vkGetPrivateDataEXT = (PFN_vkGetPrivateDataEXT)load(context, "vkGetPrivateDataEXT"); + vkSetPrivateDataEXT = (PFN_vkSetPrivateDataEXT)load(context, "vkSetPrivateDataEXT"); +#endif /* defined(VK_EXT_private_data) */ +#if defined(VK_EXT_sample_locations) + vkCmdSetSampleLocationsEXT = (PFN_vkCmdSetSampleLocationsEXT)load(context, "vkCmdSetSampleLocationsEXT"); +#endif /* defined(VK_EXT_sample_locations) */ +#if defined(VK_EXT_transform_feedback) + vkCmdBeginQueryIndexedEXT = (PFN_vkCmdBeginQueryIndexedEXT)load(context, "vkCmdBeginQueryIndexedEXT"); + vkCmdBeginTransformFeedbackEXT = (PFN_vkCmdBeginTransformFeedbackEXT)load(context, "vkCmdBeginTransformFeedbackEXT"); + vkCmdBindTransformFeedbackBuffersEXT = (PFN_vkCmdBindTransformFeedbackBuffersEXT)load(context, "vkCmdBindTransformFeedbackBuffersEXT"); + vkCmdDrawIndirectByteCountEXT = (PFN_vkCmdDrawIndirectByteCountEXT)load(context, "vkCmdDrawIndirectByteCountEXT"); + vkCmdEndQueryIndexedEXT = (PFN_vkCmdEndQueryIndexedEXT)load(context, "vkCmdEndQueryIndexedEXT"); + vkCmdEndTransformFeedbackEXT = (PFN_vkCmdEndTransformFeedbackEXT)load(context, "vkCmdEndTransformFeedbackEXT"); +#endif /* defined(VK_EXT_transform_feedback) */ +#if defined(VK_EXT_validation_cache) + vkCreateValidationCacheEXT = (PFN_vkCreateValidationCacheEXT)load(context, "vkCreateValidationCacheEXT"); + vkDestroyValidationCacheEXT = (PFN_vkDestroyValidationCacheEXT)load(context, "vkDestroyValidationCacheEXT"); + vkGetValidationCacheDataEXT = (PFN_vkGetValidationCacheDataEXT)load(context, "vkGetValidationCacheDataEXT"); + vkMergeValidationCachesEXT = (PFN_vkMergeValidationCachesEXT)load(context, "vkMergeValidationCachesEXT"); +#endif /* defined(VK_EXT_validation_cache) */ +#if defined(VK_GOOGLE_display_timing) + vkGetPastPresentationTimingGOOGLE = (PFN_vkGetPastPresentationTimingGOOGLE)load(context, "vkGetPastPresentationTimingGOOGLE"); + vkGetRefreshCycleDurationGOOGLE = (PFN_vkGetRefreshCycleDurationGOOGLE)load(context, "vkGetRefreshCycleDurationGOOGLE"); +#endif /* defined(VK_GOOGLE_display_timing) */ +#if defined(VK_INTEL_performance_query) + vkAcquirePerformanceConfigurationINTEL = (PFN_vkAcquirePerformanceConfigurationINTEL)load(context, "vkAcquirePerformanceConfigurationINTEL"); + vkCmdSetPerformanceMarkerINTEL = (PFN_vkCmdSetPerformanceMarkerINTEL)load(context, "vkCmdSetPerformanceMarkerINTEL"); + vkCmdSetPerformanceOverrideINTEL = (PFN_vkCmdSetPerformanceOverrideINTEL)load(context, "vkCmdSetPerformanceOverrideINTEL"); + vkCmdSetPerformanceStreamMarkerINTEL = (PFN_vkCmdSetPerformanceStreamMarkerINTEL)load(context, "vkCmdSetPerformanceStreamMarkerINTEL"); + vkGetPerformanceParameterINTEL = (PFN_vkGetPerformanceParameterINTEL)load(context, "vkGetPerformanceParameterINTEL"); + vkInitializePerformanceApiINTEL = (PFN_vkInitializePerformanceApiINTEL)load(context, "vkInitializePerformanceApiINTEL"); + vkQueueSetPerformanceConfigurationINTEL = (PFN_vkQueueSetPerformanceConfigurationINTEL)load(context, "vkQueueSetPerformanceConfigurationINTEL"); + vkReleasePerformanceConfigurationINTEL = (PFN_vkReleasePerformanceConfigurationINTEL)load(context, "vkReleasePerformanceConfigurationINTEL"); + vkUninitializePerformanceApiINTEL = (PFN_vkUninitializePerformanceApiINTEL)load(context, "vkUninitializePerformanceApiINTEL"); +#endif /* defined(VK_INTEL_performance_query) */ +#if defined(VK_KHR_bind_memory2) + vkBindBufferMemory2KHR = (PFN_vkBindBufferMemory2KHR)load(context, "vkBindBufferMemory2KHR"); + vkBindImageMemory2KHR = (PFN_vkBindImageMemory2KHR)load(context, "vkBindImageMemory2KHR"); +#endif /* defined(VK_KHR_bind_memory2) */ +#if defined(VK_KHR_buffer_device_address) + vkGetBufferDeviceAddressKHR = (PFN_vkGetBufferDeviceAddressKHR)load(context, "vkGetBufferDeviceAddressKHR"); + vkGetBufferOpaqueCaptureAddressKHR = (PFN_vkGetBufferOpaqueCaptureAddressKHR)load(context, "vkGetBufferOpaqueCaptureAddressKHR"); + vkGetDeviceMemoryOpaqueCaptureAddressKHR = (PFN_vkGetDeviceMemoryOpaqueCaptureAddressKHR)load(context, "vkGetDeviceMemoryOpaqueCaptureAddressKHR"); +#endif /* defined(VK_KHR_buffer_device_address) */ +#if defined(VK_KHR_create_renderpass2) + vkCmdBeginRenderPass2KHR = (PFN_vkCmdBeginRenderPass2KHR)load(context, "vkCmdBeginRenderPass2KHR"); + vkCmdEndRenderPass2KHR = (PFN_vkCmdEndRenderPass2KHR)load(context, "vkCmdEndRenderPass2KHR"); + vkCmdNextSubpass2KHR = (PFN_vkCmdNextSubpass2KHR)load(context, "vkCmdNextSubpass2KHR"); + vkCreateRenderPass2KHR = (PFN_vkCreateRenderPass2KHR)load(context, "vkCreateRenderPass2KHR"); +#endif /* defined(VK_KHR_create_renderpass2) */ +#if defined(VK_KHR_deferred_host_operations) + vkCreateDeferredOperationKHR = (PFN_vkCreateDeferredOperationKHR)load(context, "vkCreateDeferredOperationKHR"); + vkDeferredOperationJoinKHR = (PFN_vkDeferredOperationJoinKHR)load(context, "vkDeferredOperationJoinKHR"); + vkDestroyDeferredOperationKHR = (PFN_vkDestroyDeferredOperationKHR)load(context, "vkDestroyDeferredOperationKHR"); + vkGetDeferredOperationMaxConcurrencyKHR = (PFN_vkGetDeferredOperationMaxConcurrencyKHR)load(context, "vkGetDeferredOperationMaxConcurrencyKHR"); + vkGetDeferredOperationResultKHR = (PFN_vkGetDeferredOperationResultKHR)load(context, "vkGetDeferredOperationResultKHR"); +#endif /* defined(VK_KHR_deferred_host_operations) */ +#if defined(VK_KHR_descriptor_update_template) + vkCreateDescriptorUpdateTemplateKHR = (PFN_vkCreateDescriptorUpdateTemplateKHR)load(context, "vkCreateDescriptorUpdateTemplateKHR"); + vkDestroyDescriptorUpdateTemplateKHR = (PFN_vkDestroyDescriptorUpdateTemplateKHR)load(context, "vkDestroyDescriptorUpdateTemplateKHR"); + vkUpdateDescriptorSetWithTemplateKHR = (PFN_vkUpdateDescriptorSetWithTemplateKHR)load(context, "vkUpdateDescriptorSetWithTemplateKHR"); +#endif /* defined(VK_KHR_descriptor_update_template) */ +#if defined(VK_KHR_device_group) + vkCmdDispatchBaseKHR = (PFN_vkCmdDispatchBaseKHR)load(context, "vkCmdDispatchBaseKHR"); + vkCmdSetDeviceMaskKHR = (PFN_vkCmdSetDeviceMaskKHR)load(context, "vkCmdSetDeviceMaskKHR"); + vkGetDeviceGroupPeerMemoryFeaturesKHR = (PFN_vkGetDeviceGroupPeerMemoryFeaturesKHR)load(context, "vkGetDeviceGroupPeerMemoryFeaturesKHR"); +#endif /* defined(VK_KHR_device_group) */ +#if defined(VK_KHR_display_swapchain) + vkCreateSharedSwapchainsKHR = (PFN_vkCreateSharedSwapchainsKHR)load(context, "vkCreateSharedSwapchainsKHR"); +#endif /* defined(VK_KHR_display_swapchain) */ +#if defined(VK_KHR_draw_indirect_count) + vkCmdDrawIndexedIndirectCountKHR = (PFN_vkCmdDrawIndexedIndirectCountKHR)load(context, "vkCmdDrawIndexedIndirectCountKHR"); + vkCmdDrawIndirectCountKHR = (PFN_vkCmdDrawIndirectCountKHR)load(context, "vkCmdDrawIndirectCountKHR"); +#endif /* defined(VK_KHR_draw_indirect_count) */ +#if defined(VK_KHR_external_fence_fd) + vkGetFenceFdKHR = (PFN_vkGetFenceFdKHR)load(context, "vkGetFenceFdKHR"); + vkImportFenceFdKHR = (PFN_vkImportFenceFdKHR)load(context, "vkImportFenceFdKHR"); +#endif /* defined(VK_KHR_external_fence_fd) */ +#if defined(VK_KHR_external_fence_win32) + vkGetFenceWin32HandleKHR = (PFN_vkGetFenceWin32HandleKHR)load(context, "vkGetFenceWin32HandleKHR"); + vkImportFenceWin32HandleKHR = (PFN_vkImportFenceWin32HandleKHR)load(context, "vkImportFenceWin32HandleKHR"); +#endif /* defined(VK_KHR_external_fence_win32) */ +#if defined(VK_KHR_external_memory_fd) + vkGetMemoryFdKHR = (PFN_vkGetMemoryFdKHR)load(context, "vkGetMemoryFdKHR"); + vkGetMemoryFdPropertiesKHR = (PFN_vkGetMemoryFdPropertiesKHR)load(context, "vkGetMemoryFdPropertiesKHR"); +#endif /* defined(VK_KHR_external_memory_fd) */ +#if defined(VK_KHR_external_memory_win32) + vkGetMemoryWin32HandleKHR = (PFN_vkGetMemoryWin32HandleKHR)load(context, "vkGetMemoryWin32HandleKHR"); + vkGetMemoryWin32HandlePropertiesKHR = (PFN_vkGetMemoryWin32HandlePropertiesKHR)load(context, "vkGetMemoryWin32HandlePropertiesKHR"); +#endif /* defined(VK_KHR_external_memory_win32) */ +#if defined(VK_KHR_external_semaphore_fd) + vkGetSemaphoreFdKHR = (PFN_vkGetSemaphoreFdKHR)load(context, "vkGetSemaphoreFdKHR"); + vkImportSemaphoreFdKHR = (PFN_vkImportSemaphoreFdKHR)load(context, "vkImportSemaphoreFdKHR"); +#endif /* defined(VK_KHR_external_semaphore_fd) */ +#if defined(VK_KHR_external_semaphore_win32) + vkGetSemaphoreWin32HandleKHR = (PFN_vkGetSemaphoreWin32HandleKHR)load(context, "vkGetSemaphoreWin32HandleKHR"); + vkImportSemaphoreWin32HandleKHR = (PFN_vkImportSemaphoreWin32HandleKHR)load(context, "vkImportSemaphoreWin32HandleKHR"); +#endif /* defined(VK_KHR_external_semaphore_win32) */ +#if defined(VK_KHR_get_memory_requirements2) + vkGetBufferMemoryRequirements2KHR = (PFN_vkGetBufferMemoryRequirements2KHR)load(context, "vkGetBufferMemoryRequirements2KHR"); + vkGetImageMemoryRequirements2KHR = (PFN_vkGetImageMemoryRequirements2KHR)load(context, "vkGetImageMemoryRequirements2KHR"); + vkGetImageSparseMemoryRequirements2KHR = (PFN_vkGetImageSparseMemoryRequirements2KHR)load(context, "vkGetImageSparseMemoryRequirements2KHR"); +#endif /* defined(VK_KHR_get_memory_requirements2) */ +#if defined(VK_KHR_maintenance1) + vkTrimCommandPoolKHR = (PFN_vkTrimCommandPoolKHR)load(context, "vkTrimCommandPoolKHR"); +#endif /* defined(VK_KHR_maintenance1) */ +#if defined(VK_KHR_maintenance3) + vkGetDescriptorSetLayoutSupportKHR = (PFN_vkGetDescriptorSetLayoutSupportKHR)load(context, "vkGetDescriptorSetLayoutSupportKHR"); +#endif /* defined(VK_KHR_maintenance3) */ +#if defined(VK_KHR_performance_query) + vkAcquireProfilingLockKHR = (PFN_vkAcquireProfilingLockKHR)load(context, "vkAcquireProfilingLockKHR"); + vkReleaseProfilingLockKHR = (PFN_vkReleaseProfilingLockKHR)load(context, "vkReleaseProfilingLockKHR"); +#endif /* defined(VK_KHR_performance_query) */ +#if defined(VK_KHR_pipeline_executable_properties) + vkGetPipelineExecutableInternalRepresentationsKHR = (PFN_vkGetPipelineExecutableInternalRepresentationsKHR)load(context, "vkGetPipelineExecutableInternalRepresentationsKHR"); + vkGetPipelineExecutablePropertiesKHR = (PFN_vkGetPipelineExecutablePropertiesKHR)load(context, "vkGetPipelineExecutablePropertiesKHR"); + vkGetPipelineExecutableStatisticsKHR = (PFN_vkGetPipelineExecutableStatisticsKHR)load(context, "vkGetPipelineExecutableStatisticsKHR"); +#endif /* defined(VK_KHR_pipeline_executable_properties) */ +#if defined(VK_KHR_push_descriptor) + vkCmdPushDescriptorSetKHR = (PFN_vkCmdPushDescriptorSetKHR)load(context, "vkCmdPushDescriptorSetKHR"); +#endif /* defined(VK_KHR_push_descriptor) */ +#if defined(VK_KHR_ray_tracing) + vkBindAccelerationStructureMemoryKHR = (PFN_vkBindAccelerationStructureMemoryKHR)load(context, "vkBindAccelerationStructureMemoryKHR"); + vkBuildAccelerationStructureKHR = (PFN_vkBuildAccelerationStructureKHR)load(context, "vkBuildAccelerationStructureKHR"); + vkCmdBuildAccelerationStructureIndirectKHR = (PFN_vkCmdBuildAccelerationStructureIndirectKHR)load(context, "vkCmdBuildAccelerationStructureIndirectKHR"); + vkCmdBuildAccelerationStructureKHR = (PFN_vkCmdBuildAccelerationStructureKHR)load(context, "vkCmdBuildAccelerationStructureKHR"); + vkCmdCopyAccelerationStructureKHR = (PFN_vkCmdCopyAccelerationStructureKHR)load(context, "vkCmdCopyAccelerationStructureKHR"); + vkCmdCopyAccelerationStructureToMemoryKHR = (PFN_vkCmdCopyAccelerationStructureToMemoryKHR)load(context, "vkCmdCopyAccelerationStructureToMemoryKHR"); + vkCmdCopyMemoryToAccelerationStructureKHR = (PFN_vkCmdCopyMemoryToAccelerationStructureKHR)load(context, "vkCmdCopyMemoryToAccelerationStructureKHR"); + vkCmdTraceRaysIndirectKHR = (PFN_vkCmdTraceRaysIndirectKHR)load(context, "vkCmdTraceRaysIndirectKHR"); + vkCmdTraceRaysKHR = (PFN_vkCmdTraceRaysKHR)load(context, "vkCmdTraceRaysKHR"); + vkCmdWriteAccelerationStructuresPropertiesKHR = (PFN_vkCmdWriteAccelerationStructuresPropertiesKHR)load(context, "vkCmdWriteAccelerationStructuresPropertiesKHR"); + vkCopyAccelerationStructureKHR = (PFN_vkCopyAccelerationStructureKHR)load(context, "vkCopyAccelerationStructureKHR"); + vkCopyAccelerationStructureToMemoryKHR = (PFN_vkCopyAccelerationStructureToMemoryKHR)load(context, "vkCopyAccelerationStructureToMemoryKHR"); + vkCopyMemoryToAccelerationStructureKHR = (PFN_vkCopyMemoryToAccelerationStructureKHR)load(context, "vkCopyMemoryToAccelerationStructureKHR"); + vkCreateAccelerationStructureKHR = (PFN_vkCreateAccelerationStructureKHR)load(context, "vkCreateAccelerationStructureKHR"); + vkCreateRayTracingPipelinesKHR = (PFN_vkCreateRayTracingPipelinesKHR)load(context, "vkCreateRayTracingPipelinesKHR"); + vkDestroyAccelerationStructureKHR = (PFN_vkDestroyAccelerationStructureKHR)load(context, "vkDestroyAccelerationStructureKHR"); + vkGetAccelerationStructureDeviceAddressKHR = (PFN_vkGetAccelerationStructureDeviceAddressKHR)load(context, "vkGetAccelerationStructureDeviceAddressKHR"); + vkGetAccelerationStructureMemoryRequirementsKHR = (PFN_vkGetAccelerationStructureMemoryRequirementsKHR)load(context, "vkGetAccelerationStructureMemoryRequirementsKHR"); + vkGetDeviceAccelerationStructureCompatibilityKHR = (PFN_vkGetDeviceAccelerationStructureCompatibilityKHR)load(context, "vkGetDeviceAccelerationStructureCompatibilityKHR"); + vkGetRayTracingCaptureReplayShaderGroupHandlesKHR = (PFN_vkGetRayTracingCaptureReplayShaderGroupHandlesKHR)load(context, "vkGetRayTracingCaptureReplayShaderGroupHandlesKHR"); + vkGetRayTracingShaderGroupHandlesKHR = (PFN_vkGetRayTracingShaderGroupHandlesKHR)load(context, "vkGetRayTracingShaderGroupHandlesKHR"); + vkWriteAccelerationStructuresPropertiesKHR = (PFN_vkWriteAccelerationStructuresPropertiesKHR)load(context, "vkWriteAccelerationStructuresPropertiesKHR"); +#endif /* defined(VK_KHR_ray_tracing) */ +#if defined(VK_KHR_sampler_ycbcr_conversion) + vkCreateSamplerYcbcrConversionKHR = (PFN_vkCreateSamplerYcbcrConversionKHR)load(context, "vkCreateSamplerYcbcrConversionKHR"); + vkDestroySamplerYcbcrConversionKHR = (PFN_vkDestroySamplerYcbcrConversionKHR)load(context, "vkDestroySamplerYcbcrConversionKHR"); +#endif /* defined(VK_KHR_sampler_ycbcr_conversion) */ +#if defined(VK_KHR_shared_presentable_image) + vkGetSwapchainStatusKHR = (PFN_vkGetSwapchainStatusKHR)load(context, "vkGetSwapchainStatusKHR"); +#endif /* defined(VK_KHR_shared_presentable_image) */ +#if defined(VK_KHR_swapchain) + vkAcquireNextImageKHR = (PFN_vkAcquireNextImageKHR)load(context, "vkAcquireNextImageKHR"); + vkCreateSwapchainKHR = (PFN_vkCreateSwapchainKHR)load(context, "vkCreateSwapchainKHR"); + vkDestroySwapchainKHR = (PFN_vkDestroySwapchainKHR)load(context, "vkDestroySwapchainKHR"); + vkGetSwapchainImagesKHR = (PFN_vkGetSwapchainImagesKHR)load(context, "vkGetSwapchainImagesKHR"); + vkQueuePresentKHR = (PFN_vkQueuePresentKHR)load(context, "vkQueuePresentKHR"); +#endif /* defined(VK_KHR_swapchain) */ +#if defined(VK_KHR_timeline_semaphore) + vkGetSemaphoreCounterValueKHR = (PFN_vkGetSemaphoreCounterValueKHR)load(context, "vkGetSemaphoreCounterValueKHR"); + vkSignalSemaphoreKHR = (PFN_vkSignalSemaphoreKHR)load(context, "vkSignalSemaphoreKHR"); + vkWaitSemaphoresKHR = (PFN_vkWaitSemaphoresKHR)load(context, "vkWaitSemaphoresKHR"); +#endif /* defined(VK_KHR_timeline_semaphore) */ +#if defined(VK_NVX_image_view_handle) + vkGetImageViewAddressNVX = (PFN_vkGetImageViewAddressNVX)load(context, "vkGetImageViewAddressNVX"); + vkGetImageViewHandleNVX = (PFN_vkGetImageViewHandleNVX)load(context, "vkGetImageViewHandleNVX"); +#endif /* defined(VK_NVX_image_view_handle) */ +#if defined(VK_NV_clip_space_w_scaling) + vkCmdSetViewportWScalingNV = (PFN_vkCmdSetViewportWScalingNV)load(context, "vkCmdSetViewportWScalingNV"); +#endif /* defined(VK_NV_clip_space_w_scaling) */ +#if defined(VK_NV_device_diagnostic_checkpoints) + vkCmdSetCheckpointNV = (PFN_vkCmdSetCheckpointNV)load(context, "vkCmdSetCheckpointNV"); + vkGetQueueCheckpointDataNV = (PFN_vkGetQueueCheckpointDataNV)load(context, "vkGetQueueCheckpointDataNV"); +#endif /* defined(VK_NV_device_diagnostic_checkpoints) */ +#if defined(VK_NV_device_generated_commands) + vkCmdBindPipelineShaderGroupNV = (PFN_vkCmdBindPipelineShaderGroupNV)load(context, "vkCmdBindPipelineShaderGroupNV"); + vkCmdExecuteGeneratedCommandsNV = (PFN_vkCmdExecuteGeneratedCommandsNV)load(context, "vkCmdExecuteGeneratedCommandsNV"); + vkCmdPreprocessGeneratedCommandsNV = (PFN_vkCmdPreprocessGeneratedCommandsNV)load(context, "vkCmdPreprocessGeneratedCommandsNV"); + vkCreateIndirectCommandsLayoutNV = (PFN_vkCreateIndirectCommandsLayoutNV)load(context, "vkCreateIndirectCommandsLayoutNV"); + vkDestroyIndirectCommandsLayoutNV = (PFN_vkDestroyIndirectCommandsLayoutNV)load(context, "vkDestroyIndirectCommandsLayoutNV"); + vkGetGeneratedCommandsMemoryRequirementsNV = (PFN_vkGetGeneratedCommandsMemoryRequirementsNV)load(context, "vkGetGeneratedCommandsMemoryRequirementsNV"); +#endif /* defined(VK_NV_device_generated_commands) */ +#if defined(VK_NV_external_memory_win32) + vkGetMemoryWin32HandleNV = (PFN_vkGetMemoryWin32HandleNV)load(context, "vkGetMemoryWin32HandleNV"); +#endif /* defined(VK_NV_external_memory_win32) */ +#if defined(VK_NV_mesh_shader) + vkCmdDrawMeshTasksIndirectCountNV = (PFN_vkCmdDrawMeshTasksIndirectCountNV)load(context, "vkCmdDrawMeshTasksIndirectCountNV"); + vkCmdDrawMeshTasksIndirectNV = (PFN_vkCmdDrawMeshTasksIndirectNV)load(context, "vkCmdDrawMeshTasksIndirectNV"); + vkCmdDrawMeshTasksNV = (PFN_vkCmdDrawMeshTasksNV)load(context, "vkCmdDrawMeshTasksNV"); +#endif /* defined(VK_NV_mesh_shader) */ +#if defined(VK_NV_ray_tracing) + vkBindAccelerationStructureMemoryNV = (PFN_vkBindAccelerationStructureMemoryNV)load(context, "vkBindAccelerationStructureMemoryNV"); + vkCmdBuildAccelerationStructureNV = (PFN_vkCmdBuildAccelerationStructureNV)load(context, "vkCmdBuildAccelerationStructureNV"); + vkCmdCopyAccelerationStructureNV = (PFN_vkCmdCopyAccelerationStructureNV)load(context, "vkCmdCopyAccelerationStructureNV"); + vkCmdTraceRaysNV = (PFN_vkCmdTraceRaysNV)load(context, "vkCmdTraceRaysNV"); + vkCmdWriteAccelerationStructuresPropertiesNV = (PFN_vkCmdWriteAccelerationStructuresPropertiesNV)load(context, "vkCmdWriteAccelerationStructuresPropertiesNV"); + vkCompileDeferredNV = (PFN_vkCompileDeferredNV)load(context, "vkCompileDeferredNV"); + vkCreateAccelerationStructureNV = (PFN_vkCreateAccelerationStructureNV)load(context, "vkCreateAccelerationStructureNV"); + vkCreateRayTracingPipelinesNV = (PFN_vkCreateRayTracingPipelinesNV)load(context, "vkCreateRayTracingPipelinesNV"); + vkDestroyAccelerationStructureNV = (PFN_vkDestroyAccelerationStructureNV)load(context, "vkDestroyAccelerationStructureNV"); + vkGetAccelerationStructureHandleNV = (PFN_vkGetAccelerationStructureHandleNV)load(context, "vkGetAccelerationStructureHandleNV"); + vkGetAccelerationStructureMemoryRequirementsNV = (PFN_vkGetAccelerationStructureMemoryRequirementsNV)load(context, "vkGetAccelerationStructureMemoryRequirementsNV"); + vkGetRayTracingShaderGroupHandlesNV = (PFN_vkGetRayTracingShaderGroupHandlesNV)load(context, "vkGetRayTracingShaderGroupHandlesNV"); +#endif /* defined(VK_NV_ray_tracing) */ +#if defined(VK_NV_scissor_exclusive) + vkCmdSetExclusiveScissorNV = (PFN_vkCmdSetExclusiveScissorNV)load(context, "vkCmdSetExclusiveScissorNV"); +#endif /* defined(VK_NV_scissor_exclusive) */ +#if defined(VK_NV_shading_rate_image) + vkCmdBindShadingRateImageNV = (PFN_vkCmdBindShadingRateImageNV)load(context, "vkCmdBindShadingRateImageNV"); + vkCmdSetCoarseSampleOrderNV = (PFN_vkCmdSetCoarseSampleOrderNV)load(context, "vkCmdSetCoarseSampleOrderNV"); + vkCmdSetViewportShadingRatePaletteNV = (PFN_vkCmdSetViewportShadingRatePaletteNV)load(context, "vkCmdSetViewportShadingRatePaletteNV"); +#endif /* defined(VK_NV_shading_rate_image) */ +#if (defined(VK_EXT_full_screen_exclusive) && defined(VK_KHR_device_group)) || (defined(VK_EXT_full_screen_exclusive) && defined(VK_VERSION_1_1)) + vkGetDeviceGroupSurfacePresentModes2EXT = (PFN_vkGetDeviceGroupSurfacePresentModes2EXT)load(context, "vkGetDeviceGroupSurfacePresentModes2EXT"); +#endif /* (defined(VK_EXT_full_screen_exclusive) && defined(VK_KHR_device_group)) || (defined(VK_EXT_full_screen_exclusive) && defined(VK_VERSION_1_1)) */ +#if (defined(VK_KHR_descriptor_update_template) && defined(VK_KHR_push_descriptor)) || (defined(VK_KHR_push_descriptor) && defined(VK_VERSION_1_1)) || (defined(VK_KHR_push_descriptor) && defined(VK_KHR_descriptor_update_template)) + vkCmdPushDescriptorSetWithTemplateKHR = (PFN_vkCmdPushDescriptorSetWithTemplateKHR)load(context, "vkCmdPushDescriptorSetWithTemplateKHR"); +#endif /* (defined(VK_KHR_descriptor_update_template) && defined(VK_KHR_push_descriptor)) || (defined(VK_KHR_push_descriptor) && defined(VK_VERSION_1_1)) || (defined(VK_KHR_push_descriptor) && defined(VK_KHR_descriptor_update_template)) */ +#if (defined(VK_KHR_device_group) && defined(VK_KHR_surface)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) + vkGetDeviceGroupPresentCapabilitiesKHR = (PFN_vkGetDeviceGroupPresentCapabilitiesKHR)load(context, "vkGetDeviceGroupPresentCapabilitiesKHR"); + vkGetDeviceGroupSurfacePresentModesKHR = (PFN_vkGetDeviceGroupSurfacePresentModesKHR)load(context, "vkGetDeviceGroupSurfacePresentModesKHR"); +#endif /* (defined(VK_KHR_device_group) && defined(VK_KHR_surface)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) */ +#if (defined(VK_KHR_device_group) && defined(VK_KHR_swapchain)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) + vkAcquireNextImage2KHR = (PFN_vkAcquireNextImage2KHR)load(context, "vkAcquireNextImage2KHR"); +#endif /* (defined(VK_KHR_device_group) && defined(VK_KHR_swapchain)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) */ + /* VOLK_GENERATE_LOAD_DEVICE */ +} + +static void volkGenLoadDeviceTable(struct VolkDeviceTable* table, void* context, PFN_vkVoidFunction (*load)(void*, const char*)) +{ + /* VOLK_GENERATE_LOAD_DEVICE_TABLE */ +#if defined(VK_VERSION_1_0) + table->vkAllocateCommandBuffers = (PFN_vkAllocateCommandBuffers)load(context, "vkAllocateCommandBuffers"); + table->vkAllocateDescriptorSets = (PFN_vkAllocateDescriptorSets)load(context, "vkAllocateDescriptorSets"); + table->vkAllocateMemory = (PFN_vkAllocateMemory)load(context, "vkAllocateMemory"); + table->vkBeginCommandBuffer = (PFN_vkBeginCommandBuffer)load(context, "vkBeginCommandBuffer"); + table->vkBindBufferMemory = (PFN_vkBindBufferMemory)load(context, "vkBindBufferMemory"); + table->vkBindImageMemory = (PFN_vkBindImageMemory)load(context, "vkBindImageMemory"); + table->vkCmdBeginQuery = (PFN_vkCmdBeginQuery)load(context, "vkCmdBeginQuery"); + table->vkCmdBeginRenderPass = (PFN_vkCmdBeginRenderPass)load(context, "vkCmdBeginRenderPass"); + table->vkCmdBindDescriptorSets = (PFN_vkCmdBindDescriptorSets)load(context, "vkCmdBindDescriptorSets"); + table->vkCmdBindIndexBuffer = (PFN_vkCmdBindIndexBuffer)load(context, "vkCmdBindIndexBuffer"); + table->vkCmdBindPipeline = (PFN_vkCmdBindPipeline)load(context, "vkCmdBindPipeline"); + table->vkCmdBindVertexBuffers = (PFN_vkCmdBindVertexBuffers)load(context, "vkCmdBindVertexBuffers"); + table->vkCmdBlitImage = (PFN_vkCmdBlitImage)load(context, "vkCmdBlitImage"); + table->vkCmdClearAttachments = (PFN_vkCmdClearAttachments)load(context, "vkCmdClearAttachments"); + table->vkCmdClearColorImage = (PFN_vkCmdClearColorImage)load(context, "vkCmdClearColorImage"); + table->vkCmdClearDepthStencilImage = (PFN_vkCmdClearDepthStencilImage)load(context, "vkCmdClearDepthStencilImage"); + table->vkCmdCopyBuffer = (PFN_vkCmdCopyBuffer)load(context, "vkCmdCopyBuffer"); + table->vkCmdCopyBufferToImage = (PFN_vkCmdCopyBufferToImage)load(context, "vkCmdCopyBufferToImage"); + table->vkCmdCopyImage = (PFN_vkCmdCopyImage)load(context, "vkCmdCopyImage"); + table->vkCmdCopyImageToBuffer = (PFN_vkCmdCopyImageToBuffer)load(context, "vkCmdCopyImageToBuffer"); + table->vkCmdCopyQueryPoolResults = (PFN_vkCmdCopyQueryPoolResults)load(context, "vkCmdCopyQueryPoolResults"); + table->vkCmdDispatch = (PFN_vkCmdDispatch)load(context, "vkCmdDispatch"); + table->vkCmdDispatchIndirect = (PFN_vkCmdDispatchIndirect)load(context, "vkCmdDispatchIndirect"); + table->vkCmdDraw = (PFN_vkCmdDraw)load(context, "vkCmdDraw"); + table->vkCmdDrawIndexed = (PFN_vkCmdDrawIndexed)load(context, "vkCmdDrawIndexed"); + table->vkCmdDrawIndexedIndirect = (PFN_vkCmdDrawIndexedIndirect)load(context, "vkCmdDrawIndexedIndirect"); + table->vkCmdDrawIndirect = (PFN_vkCmdDrawIndirect)load(context, "vkCmdDrawIndirect"); + table->vkCmdEndQuery = (PFN_vkCmdEndQuery)load(context, "vkCmdEndQuery"); + table->vkCmdEndRenderPass = (PFN_vkCmdEndRenderPass)load(context, "vkCmdEndRenderPass"); + table->vkCmdExecuteCommands = (PFN_vkCmdExecuteCommands)load(context, "vkCmdExecuteCommands"); + table->vkCmdFillBuffer = (PFN_vkCmdFillBuffer)load(context, "vkCmdFillBuffer"); + table->vkCmdNextSubpass = (PFN_vkCmdNextSubpass)load(context, "vkCmdNextSubpass"); + table->vkCmdPipelineBarrier = (PFN_vkCmdPipelineBarrier)load(context, "vkCmdPipelineBarrier"); + table->vkCmdPushConstants = (PFN_vkCmdPushConstants)load(context, "vkCmdPushConstants"); + table->vkCmdResetEvent = (PFN_vkCmdResetEvent)load(context, "vkCmdResetEvent"); + table->vkCmdResetQueryPool = (PFN_vkCmdResetQueryPool)load(context, "vkCmdResetQueryPool"); + table->vkCmdResolveImage = (PFN_vkCmdResolveImage)load(context, "vkCmdResolveImage"); + table->vkCmdSetBlendConstants = (PFN_vkCmdSetBlendConstants)load(context, "vkCmdSetBlendConstants"); + table->vkCmdSetDepthBias = (PFN_vkCmdSetDepthBias)load(context, "vkCmdSetDepthBias"); + table->vkCmdSetDepthBounds = (PFN_vkCmdSetDepthBounds)load(context, "vkCmdSetDepthBounds"); + table->vkCmdSetEvent = (PFN_vkCmdSetEvent)load(context, "vkCmdSetEvent"); + table->vkCmdSetLineWidth = (PFN_vkCmdSetLineWidth)load(context, "vkCmdSetLineWidth"); + table->vkCmdSetScissor = (PFN_vkCmdSetScissor)load(context, "vkCmdSetScissor"); + table->vkCmdSetStencilCompareMask = (PFN_vkCmdSetStencilCompareMask)load(context, "vkCmdSetStencilCompareMask"); + table->vkCmdSetStencilReference = (PFN_vkCmdSetStencilReference)load(context, "vkCmdSetStencilReference"); + table->vkCmdSetStencilWriteMask = (PFN_vkCmdSetStencilWriteMask)load(context, "vkCmdSetStencilWriteMask"); + table->vkCmdSetViewport = (PFN_vkCmdSetViewport)load(context, "vkCmdSetViewport"); + table->vkCmdUpdateBuffer = (PFN_vkCmdUpdateBuffer)load(context, "vkCmdUpdateBuffer"); + table->vkCmdWaitEvents = (PFN_vkCmdWaitEvents)load(context, "vkCmdWaitEvents"); + table->vkCmdWriteTimestamp = (PFN_vkCmdWriteTimestamp)load(context, "vkCmdWriteTimestamp"); + table->vkCreateBuffer = (PFN_vkCreateBuffer)load(context, "vkCreateBuffer"); + table->vkCreateBufferView = (PFN_vkCreateBufferView)load(context, "vkCreateBufferView"); + table->vkCreateCommandPool = (PFN_vkCreateCommandPool)load(context, "vkCreateCommandPool"); + table->vkCreateComputePipelines = (PFN_vkCreateComputePipelines)load(context, "vkCreateComputePipelines"); + table->vkCreateDescriptorPool = (PFN_vkCreateDescriptorPool)load(context, "vkCreateDescriptorPool"); + table->vkCreateDescriptorSetLayout = (PFN_vkCreateDescriptorSetLayout)load(context, "vkCreateDescriptorSetLayout"); + table->vkCreateEvent = (PFN_vkCreateEvent)load(context, "vkCreateEvent"); + table->vkCreateFence = (PFN_vkCreateFence)load(context, "vkCreateFence"); + table->vkCreateFramebuffer = (PFN_vkCreateFramebuffer)load(context, "vkCreateFramebuffer"); + table->vkCreateGraphicsPipelines = (PFN_vkCreateGraphicsPipelines)load(context, "vkCreateGraphicsPipelines"); + table->vkCreateImage = (PFN_vkCreateImage)load(context, "vkCreateImage"); + table->vkCreateImageView = (PFN_vkCreateImageView)load(context, "vkCreateImageView"); + table->vkCreatePipelineCache = (PFN_vkCreatePipelineCache)load(context, "vkCreatePipelineCache"); + table->vkCreatePipelineLayout = (PFN_vkCreatePipelineLayout)load(context, "vkCreatePipelineLayout"); + table->vkCreateQueryPool = (PFN_vkCreateQueryPool)load(context, "vkCreateQueryPool"); + table->vkCreateRenderPass = (PFN_vkCreateRenderPass)load(context, "vkCreateRenderPass"); + table->vkCreateSampler = (PFN_vkCreateSampler)load(context, "vkCreateSampler"); + table->vkCreateSemaphore = (PFN_vkCreateSemaphore)load(context, "vkCreateSemaphore"); + table->vkCreateShaderModule = (PFN_vkCreateShaderModule)load(context, "vkCreateShaderModule"); + table->vkDestroyBuffer = (PFN_vkDestroyBuffer)load(context, "vkDestroyBuffer"); + table->vkDestroyBufferView = (PFN_vkDestroyBufferView)load(context, "vkDestroyBufferView"); + table->vkDestroyCommandPool = (PFN_vkDestroyCommandPool)load(context, "vkDestroyCommandPool"); + table->vkDestroyDescriptorPool = (PFN_vkDestroyDescriptorPool)load(context, "vkDestroyDescriptorPool"); + table->vkDestroyDescriptorSetLayout = (PFN_vkDestroyDescriptorSetLayout)load(context, "vkDestroyDescriptorSetLayout"); + table->vkDestroyDevice = (PFN_vkDestroyDevice)load(context, "vkDestroyDevice"); + table->vkDestroyEvent = (PFN_vkDestroyEvent)load(context, "vkDestroyEvent"); + table->vkDestroyFence = (PFN_vkDestroyFence)load(context, "vkDestroyFence"); + table->vkDestroyFramebuffer = (PFN_vkDestroyFramebuffer)load(context, "vkDestroyFramebuffer"); + table->vkDestroyImage = (PFN_vkDestroyImage)load(context, "vkDestroyImage"); + table->vkDestroyImageView = (PFN_vkDestroyImageView)load(context, "vkDestroyImageView"); + table->vkDestroyPipeline = (PFN_vkDestroyPipeline)load(context, "vkDestroyPipeline"); + table->vkDestroyPipelineCache = (PFN_vkDestroyPipelineCache)load(context, "vkDestroyPipelineCache"); + table->vkDestroyPipelineLayout = (PFN_vkDestroyPipelineLayout)load(context, "vkDestroyPipelineLayout"); + table->vkDestroyQueryPool = (PFN_vkDestroyQueryPool)load(context, "vkDestroyQueryPool"); + table->vkDestroyRenderPass = (PFN_vkDestroyRenderPass)load(context, "vkDestroyRenderPass"); + table->vkDestroySampler = (PFN_vkDestroySampler)load(context, "vkDestroySampler"); + table->vkDestroySemaphore = (PFN_vkDestroySemaphore)load(context, "vkDestroySemaphore"); + table->vkDestroyShaderModule = (PFN_vkDestroyShaderModule)load(context, "vkDestroyShaderModule"); + table->vkDeviceWaitIdle = (PFN_vkDeviceWaitIdle)load(context, "vkDeviceWaitIdle"); + table->vkEndCommandBuffer = (PFN_vkEndCommandBuffer)load(context, "vkEndCommandBuffer"); + table->vkFlushMappedMemoryRanges = (PFN_vkFlushMappedMemoryRanges)load(context, "vkFlushMappedMemoryRanges"); + table->vkFreeCommandBuffers = (PFN_vkFreeCommandBuffers)load(context, "vkFreeCommandBuffers"); + table->vkFreeDescriptorSets = (PFN_vkFreeDescriptorSets)load(context, "vkFreeDescriptorSets"); + table->vkFreeMemory = (PFN_vkFreeMemory)load(context, "vkFreeMemory"); + table->vkGetBufferMemoryRequirements = (PFN_vkGetBufferMemoryRequirements)load(context, "vkGetBufferMemoryRequirements"); + table->vkGetDeviceMemoryCommitment = (PFN_vkGetDeviceMemoryCommitment)load(context, "vkGetDeviceMemoryCommitment"); + table->vkGetDeviceQueue = (PFN_vkGetDeviceQueue)load(context, "vkGetDeviceQueue"); + table->vkGetEventStatus = (PFN_vkGetEventStatus)load(context, "vkGetEventStatus"); + table->vkGetFenceStatus = (PFN_vkGetFenceStatus)load(context, "vkGetFenceStatus"); + table->vkGetImageMemoryRequirements = (PFN_vkGetImageMemoryRequirements)load(context, "vkGetImageMemoryRequirements"); + table->vkGetImageSparseMemoryRequirements = (PFN_vkGetImageSparseMemoryRequirements)load(context, "vkGetImageSparseMemoryRequirements"); + table->vkGetImageSubresourceLayout = (PFN_vkGetImageSubresourceLayout)load(context, "vkGetImageSubresourceLayout"); + table->vkGetPipelineCacheData = (PFN_vkGetPipelineCacheData)load(context, "vkGetPipelineCacheData"); + table->vkGetQueryPoolResults = (PFN_vkGetQueryPoolResults)load(context, "vkGetQueryPoolResults"); + table->vkGetRenderAreaGranularity = (PFN_vkGetRenderAreaGranularity)load(context, "vkGetRenderAreaGranularity"); + table->vkInvalidateMappedMemoryRanges = (PFN_vkInvalidateMappedMemoryRanges)load(context, "vkInvalidateMappedMemoryRanges"); + table->vkMapMemory = (PFN_vkMapMemory)load(context, "vkMapMemory"); + table->vkMergePipelineCaches = (PFN_vkMergePipelineCaches)load(context, "vkMergePipelineCaches"); + table->vkQueueBindSparse = (PFN_vkQueueBindSparse)load(context, "vkQueueBindSparse"); + table->vkQueueSubmit = (PFN_vkQueueSubmit)load(context, "vkQueueSubmit"); + table->vkQueueWaitIdle = (PFN_vkQueueWaitIdle)load(context, "vkQueueWaitIdle"); + table->vkResetCommandBuffer = (PFN_vkResetCommandBuffer)load(context, "vkResetCommandBuffer"); + table->vkResetCommandPool = (PFN_vkResetCommandPool)load(context, "vkResetCommandPool"); + table->vkResetDescriptorPool = (PFN_vkResetDescriptorPool)load(context, "vkResetDescriptorPool"); + table->vkResetEvent = (PFN_vkResetEvent)load(context, "vkResetEvent"); + table->vkResetFences = (PFN_vkResetFences)load(context, "vkResetFences"); + table->vkSetEvent = (PFN_vkSetEvent)load(context, "vkSetEvent"); + table->vkUnmapMemory = (PFN_vkUnmapMemory)load(context, "vkUnmapMemory"); + table->vkUpdateDescriptorSets = (PFN_vkUpdateDescriptorSets)load(context, "vkUpdateDescriptorSets"); + table->vkWaitForFences = (PFN_vkWaitForFences)load(context, "vkWaitForFences"); +#endif /* defined(VK_VERSION_1_0) */ +#if defined(VK_VERSION_1_1) + table->vkBindBufferMemory2 = (PFN_vkBindBufferMemory2)load(context, "vkBindBufferMemory2"); + table->vkBindImageMemory2 = (PFN_vkBindImageMemory2)load(context, "vkBindImageMemory2"); + table->vkCmdDispatchBase = (PFN_vkCmdDispatchBase)load(context, "vkCmdDispatchBase"); + table->vkCmdSetDeviceMask = (PFN_vkCmdSetDeviceMask)load(context, "vkCmdSetDeviceMask"); + table->vkCreateDescriptorUpdateTemplate = (PFN_vkCreateDescriptorUpdateTemplate)load(context, "vkCreateDescriptorUpdateTemplate"); + table->vkCreateSamplerYcbcrConversion = (PFN_vkCreateSamplerYcbcrConversion)load(context, "vkCreateSamplerYcbcrConversion"); + table->vkDestroyDescriptorUpdateTemplate = (PFN_vkDestroyDescriptorUpdateTemplate)load(context, "vkDestroyDescriptorUpdateTemplate"); + table->vkDestroySamplerYcbcrConversion = (PFN_vkDestroySamplerYcbcrConversion)load(context, "vkDestroySamplerYcbcrConversion"); + table->vkGetBufferMemoryRequirements2 = (PFN_vkGetBufferMemoryRequirements2)load(context, "vkGetBufferMemoryRequirements2"); + table->vkGetDescriptorSetLayoutSupport = (PFN_vkGetDescriptorSetLayoutSupport)load(context, "vkGetDescriptorSetLayoutSupport"); + table->vkGetDeviceGroupPeerMemoryFeatures = (PFN_vkGetDeviceGroupPeerMemoryFeatures)load(context, "vkGetDeviceGroupPeerMemoryFeatures"); + table->vkGetDeviceQueue2 = (PFN_vkGetDeviceQueue2)load(context, "vkGetDeviceQueue2"); + table->vkGetImageMemoryRequirements2 = (PFN_vkGetImageMemoryRequirements2)load(context, "vkGetImageMemoryRequirements2"); + table->vkGetImageSparseMemoryRequirements2 = (PFN_vkGetImageSparseMemoryRequirements2)load(context, "vkGetImageSparseMemoryRequirements2"); + table->vkTrimCommandPool = (PFN_vkTrimCommandPool)load(context, "vkTrimCommandPool"); + table->vkUpdateDescriptorSetWithTemplate = (PFN_vkUpdateDescriptorSetWithTemplate)load(context, "vkUpdateDescriptorSetWithTemplate"); +#endif /* defined(VK_VERSION_1_1) */ +#if defined(VK_VERSION_1_2) + table->vkCmdBeginRenderPass2 = (PFN_vkCmdBeginRenderPass2)load(context, "vkCmdBeginRenderPass2"); + table->vkCmdDrawIndexedIndirectCount = (PFN_vkCmdDrawIndexedIndirectCount)load(context, "vkCmdDrawIndexedIndirectCount"); + table->vkCmdDrawIndirectCount = (PFN_vkCmdDrawIndirectCount)load(context, "vkCmdDrawIndirectCount"); + table->vkCmdEndRenderPass2 = (PFN_vkCmdEndRenderPass2)load(context, "vkCmdEndRenderPass2"); + table->vkCmdNextSubpass2 = (PFN_vkCmdNextSubpass2)load(context, "vkCmdNextSubpass2"); + table->vkCreateRenderPass2 = (PFN_vkCreateRenderPass2)load(context, "vkCreateRenderPass2"); + table->vkGetBufferDeviceAddress = (PFN_vkGetBufferDeviceAddress)load(context, "vkGetBufferDeviceAddress"); + table->vkGetBufferOpaqueCaptureAddress = (PFN_vkGetBufferOpaqueCaptureAddress)load(context, "vkGetBufferOpaqueCaptureAddress"); + table->vkGetDeviceMemoryOpaqueCaptureAddress = (PFN_vkGetDeviceMemoryOpaqueCaptureAddress)load(context, "vkGetDeviceMemoryOpaqueCaptureAddress"); + table->vkGetSemaphoreCounterValue = (PFN_vkGetSemaphoreCounterValue)load(context, "vkGetSemaphoreCounterValue"); + table->vkResetQueryPool = (PFN_vkResetQueryPool)load(context, "vkResetQueryPool"); + table->vkSignalSemaphore = (PFN_vkSignalSemaphore)load(context, "vkSignalSemaphore"); + table->vkWaitSemaphores = (PFN_vkWaitSemaphores)load(context, "vkWaitSemaphores"); +#endif /* defined(VK_VERSION_1_2) */ +#if defined(VK_AMD_buffer_marker) + table->vkCmdWriteBufferMarkerAMD = (PFN_vkCmdWriteBufferMarkerAMD)load(context, "vkCmdWriteBufferMarkerAMD"); +#endif /* defined(VK_AMD_buffer_marker) */ +#if defined(VK_AMD_display_native_hdr) + table->vkSetLocalDimmingAMD = (PFN_vkSetLocalDimmingAMD)load(context, "vkSetLocalDimmingAMD"); +#endif /* defined(VK_AMD_display_native_hdr) */ +#if defined(VK_AMD_draw_indirect_count) + table->vkCmdDrawIndexedIndirectCountAMD = (PFN_vkCmdDrawIndexedIndirectCountAMD)load(context, "vkCmdDrawIndexedIndirectCountAMD"); + table->vkCmdDrawIndirectCountAMD = (PFN_vkCmdDrawIndirectCountAMD)load(context, "vkCmdDrawIndirectCountAMD"); +#endif /* defined(VK_AMD_draw_indirect_count) */ +#if defined(VK_AMD_shader_info) + table->vkGetShaderInfoAMD = (PFN_vkGetShaderInfoAMD)load(context, "vkGetShaderInfoAMD"); +#endif /* defined(VK_AMD_shader_info) */ +#if defined(VK_ANDROID_external_memory_android_hardware_buffer) + table->vkGetAndroidHardwareBufferPropertiesANDROID = (PFN_vkGetAndroidHardwareBufferPropertiesANDROID)load(context, "vkGetAndroidHardwareBufferPropertiesANDROID"); + table->vkGetMemoryAndroidHardwareBufferANDROID = (PFN_vkGetMemoryAndroidHardwareBufferANDROID)load(context, "vkGetMemoryAndroidHardwareBufferANDROID"); +#endif /* defined(VK_ANDROID_external_memory_android_hardware_buffer) */ +#if defined(VK_EXT_buffer_device_address) + table->vkGetBufferDeviceAddressEXT = (PFN_vkGetBufferDeviceAddressEXT)load(context, "vkGetBufferDeviceAddressEXT"); +#endif /* defined(VK_EXT_buffer_device_address) */ +#if defined(VK_EXT_calibrated_timestamps) + table->vkGetCalibratedTimestampsEXT = (PFN_vkGetCalibratedTimestampsEXT)load(context, "vkGetCalibratedTimestampsEXT"); +#endif /* defined(VK_EXT_calibrated_timestamps) */ +#if defined(VK_EXT_conditional_rendering) + table->vkCmdBeginConditionalRenderingEXT = (PFN_vkCmdBeginConditionalRenderingEXT)load(context, "vkCmdBeginConditionalRenderingEXT"); + table->vkCmdEndConditionalRenderingEXT = (PFN_vkCmdEndConditionalRenderingEXT)load(context, "vkCmdEndConditionalRenderingEXT"); +#endif /* defined(VK_EXT_conditional_rendering) */ +#if defined(VK_EXT_debug_marker) + table->vkCmdDebugMarkerBeginEXT = (PFN_vkCmdDebugMarkerBeginEXT)load(context, "vkCmdDebugMarkerBeginEXT"); + table->vkCmdDebugMarkerEndEXT = (PFN_vkCmdDebugMarkerEndEXT)load(context, "vkCmdDebugMarkerEndEXT"); + table->vkCmdDebugMarkerInsertEXT = (PFN_vkCmdDebugMarkerInsertEXT)load(context, "vkCmdDebugMarkerInsertEXT"); + table->vkDebugMarkerSetObjectNameEXT = (PFN_vkDebugMarkerSetObjectNameEXT)load(context, "vkDebugMarkerSetObjectNameEXT"); + table->vkDebugMarkerSetObjectTagEXT = (PFN_vkDebugMarkerSetObjectTagEXT)load(context, "vkDebugMarkerSetObjectTagEXT"); +#endif /* defined(VK_EXT_debug_marker) */ +#if defined(VK_EXT_discard_rectangles) + table->vkCmdSetDiscardRectangleEXT = (PFN_vkCmdSetDiscardRectangleEXT)load(context, "vkCmdSetDiscardRectangleEXT"); +#endif /* defined(VK_EXT_discard_rectangles) */ +#if defined(VK_EXT_display_control) + table->vkDisplayPowerControlEXT = (PFN_vkDisplayPowerControlEXT)load(context, "vkDisplayPowerControlEXT"); + table->vkGetSwapchainCounterEXT = (PFN_vkGetSwapchainCounterEXT)load(context, "vkGetSwapchainCounterEXT"); + table->vkRegisterDeviceEventEXT = (PFN_vkRegisterDeviceEventEXT)load(context, "vkRegisterDeviceEventEXT"); + table->vkRegisterDisplayEventEXT = (PFN_vkRegisterDisplayEventEXT)load(context, "vkRegisterDisplayEventEXT"); +#endif /* defined(VK_EXT_display_control) */ +#if defined(VK_EXT_extended_dynamic_state) + table->vkCmdBindVertexBuffers2EXT = (PFN_vkCmdBindVertexBuffers2EXT)load(context, "vkCmdBindVertexBuffers2EXT"); + table->vkCmdSetCullModeEXT = (PFN_vkCmdSetCullModeEXT)load(context, "vkCmdSetCullModeEXT"); + table->vkCmdSetDepthBoundsTestEnableEXT = (PFN_vkCmdSetDepthBoundsTestEnableEXT)load(context, "vkCmdSetDepthBoundsTestEnableEXT"); + table->vkCmdSetDepthCompareOpEXT = (PFN_vkCmdSetDepthCompareOpEXT)load(context, "vkCmdSetDepthCompareOpEXT"); + table->vkCmdSetDepthTestEnableEXT = (PFN_vkCmdSetDepthTestEnableEXT)load(context, "vkCmdSetDepthTestEnableEXT"); + table->vkCmdSetDepthWriteEnableEXT = (PFN_vkCmdSetDepthWriteEnableEXT)load(context, "vkCmdSetDepthWriteEnableEXT"); + table->vkCmdSetFrontFaceEXT = (PFN_vkCmdSetFrontFaceEXT)load(context, "vkCmdSetFrontFaceEXT"); + table->vkCmdSetPrimitiveTopologyEXT = (PFN_vkCmdSetPrimitiveTopologyEXT)load(context, "vkCmdSetPrimitiveTopologyEXT"); + table->vkCmdSetScissorWithCountEXT = (PFN_vkCmdSetScissorWithCountEXT)load(context, "vkCmdSetScissorWithCountEXT"); + table->vkCmdSetStencilOpEXT = (PFN_vkCmdSetStencilOpEXT)load(context, "vkCmdSetStencilOpEXT"); + table->vkCmdSetStencilTestEnableEXT = (PFN_vkCmdSetStencilTestEnableEXT)load(context, "vkCmdSetStencilTestEnableEXT"); + table->vkCmdSetViewportWithCountEXT = (PFN_vkCmdSetViewportWithCountEXT)load(context, "vkCmdSetViewportWithCountEXT"); +#endif /* defined(VK_EXT_extended_dynamic_state) */ +#if defined(VK_EXT_external_memory_host) + table->vkGetMemoryHostPointerPropertiesEXT = (PFN_vkGetMemoryHostPointerPropertiesEXT)load(context, "vkGetMemoryHostPointerPropertiesEXT"); +#endif /* defined(VK_EXT_external_memory_host) */ +#if defined(VK_EXT_full_screen_exclusive) + table->vkAcquireFullScreenExclusiveModeEXT = (PFN_vkAcquireFullScreenExclusiveModeEXT)load(context, "vkAcquireFullScreenExclusiveModeEXT"); + table->vkReleaseFullScreenExclusiveModeEXT = (PFN_vkReleaseFullScreenExclusiveModeEXT)load(context, "vkReleaseFullScreenExclusiveModeEXT"); +#endif /* defined(VK_EXT_full_screen_exclusive) */ +#if defined(VK_EXT_hdr_metadata) + table->vkSetHdrMetadataEXT = (PFN_vkSetHdrMetadataEXT)load(context, "vkSetHdrMetadataEXT"); +#endif /* defined(VK_EXT_hdr_metadata) */ +#if defined(VK_EXT_host_query_reset) + table->vkResetQueryPoolEXT = (PFN_vkResetQueryPoolEXT)load(context, "vkResetQueryPoolEXT"); +#endif /* defined(VK_EXT_host_query_reset) */ +#if defined(VK_EXT_image_drm_format_modifier) + table->vkGetImageDrmFormatModifierPropertiesEXT = (PFN_vkGetImageDrmFormatModifierPropertiesEXT)load(context, "vkGetImageDrmFormatModifierPropertiesEXT"); +#endif /* defined(VK_EXT_image_drm_format_modifier) */ +#if defined(VK_EXT_line_rasterization) + table->vkCmdSetLineStippleEXT = (PFN_vkCmdSetLineStippleEXT)load(context, "vkCmdSetLineStippleEXT"); +#endif /* defined(VK_EXT_line_rasterization) */ +#if defined(VK_EXT_private_data) + table->vkCreatePrivateDataSlotEXT = (PFN_vkCreatePrivateDataSlotEXT)load(context, "vkCreatePrivateDataSlotEXT"); + table->vkDestroyPrivateDataSlotEXT = (PFN_vkDestroyPrivateDataSlotEXT)load(context, "vkDestroyPrivateDataSlotEXT"); + table->vkGetPrivateDataEXT = (PFN_vkGetPrivateDataEXT)load(context, "vkGetPrivateDataEXT"); + table->vkSetPrivateDataEXT = (PFN_vkSetPrivateDataEXT)load(context, "vkSetPrivateDataEXT"); +#endif /* defined(VK_EXT_private_data) */ +#if defined(VK_EXT_sample_locations) + table->vkCmdSetSampleLocationsEXT = (PFN_vkCmdSetSampleLocationsEXT)load(context, "vkCmdSetSampleLocationsEXT"); +#endif /* defined(VK_EXT_sample_locations) */ +#if defined(VK_EXT_transform_feedback) + table->vkCmdBeginQueryIndexedEXT = (PFN_vkCmdBeginQueryIndexedEXT)load(context, "vkCmdBeginQueryIndexedEXT"); + table->vkCmdBeginTransformFeedbackEXT = (PFN_vkCmdBeginTransformFeedbackEXT)load(context, "vkCmdBeginTransformFeedbackEXT"); + table->vkCmdBindTransformFeedbackBuffersEXT = (PFN_vkCmdBindTransformFeedbackBuffersEXT)load(context, "vkCmdBindTransformFeedbackBuffersEXT"); + table->vkCmdDrawIndirectByteCountEXT = (PFN_vkCmdDrawIndirectByteCountEXT)load(context, "vkCmdDrawIndirectByteCountEXT"); + table->vkCmdEndQueryIndexedEXT = (PFN_vkCmdEndQueryIndexedEXT)load(context, "vkCmdEndQueryIndexedEXT"); + table->vkCmdEndTransformFeedbackEXT = (PFN_vkCmdEndTransformFeedbackEXT)load(context, "vkCmdEndTransformFeedbackEXT"); +#endif /* defined(VK_EXT_transform_feedback) */ +#if defined(VK_EXT_validation_cache) + table->vkCreateValidationCacheEXT = (PFN_vkCreateValidationCacheEXT)load(context, "vkCreateValidationCacheEXT"); + table->vkDestroyValidationCacheEXT = (PFN_vkDestroyValidationCacheEXT)load(context, "vkDestroyValidationCacheEXT"); + table->vkGetValidationCacheDataEXT = (PFN_vkGetValidationCacheDataEXT)load(context, "vkGetValidationCacheDataEXT"); + table->vkMergeValidationCachesEXT = (PFN_vkMergeValidationCachesEXT)load(context, "vkMergeValidationCachesEXT"); +#endif /* defined(VK_EXT_validation_cache) */ +#if defined(VK_GOOGLE_display_timing) + table->vkGetPastPresentationTimingGOOGLE = (PFN_vkGetPastPresentationTimingGOOGLE)load(context, "vkGetPastPresentationTimingGOOGLE"); + table->vkGetRefreshCycleDurationGOOGLE = (PFN_vkGetRefreshCycleDurationGOOGLE)load(context, "vkGetRefreshCycleDurationGOOGLE"); +#endif /* defined(VK_GOOGLE_display_timing) */ +#if defined(VK_INTEL_performance_query) + table->vkAcquirePerformanceConfigurationINTEL = (PFN_vkAcquirePerformanceConfigurationINTEL)load(context, "vkAcquirePerformanceConfigurationINTEL"); + table->vkCmdSetPerformanceMarkerINTEL = (PFN_vkCmdSetPerformanceMarkerINTEL)load(context, "vkCmdSetPerformanceMarkerINTEL"); + table->vkCmdSetPerformanceOverrideINTEL = (PFN_vkCmdSetPerformanceOverrideINTEL)load(context, "vkCmdSetPerformanceOverrideINTEL"); + table->vkCmdSetPerformanceStreamMarkerINTEL = (PFN_vkCmdSetPerformanceStreamMarkerINTEL)load(context, "vkCmdSetPerformanceStreamMarkerINTEL"); + table->vkGetPerformanceParameterINTEL = (PFN_vkGetPerformanceParameterINTEL)load(context, "vkGetPerformanceParameterINTEL"); + table->vkInitializePerformanceApiINTEL = (PFN_vkInitializePerformanceApiINTEL)load(context, "vkInitializePerformanceApiINTEL"); + table->vkQueueSetPerformanceConfigurationINTEL = (PFN_vkQueueSetPerformanceConfigurationINTEL)load(context, "vkQueueSetPerformanceConfigurationINTEL"); + table->vkReleasePerformanceConfigurationINTEL = (PFN_vkReleasePerformanceConfigurationINTEL)load(context, "vkReleasePerformanceConfigurationINTEL"); + table->vkUninitializePerformanceApiINTEL = (PFN_vkUninitializePerformanceApiINTEL)load(context, "vkUninitializePerformanceApiINTEL"); +#endif /* defined(VK_INTEL_performance_query) */ +#if defined(VK_KHR_bind_memory2) + table->vkBindBufferMemory2KHR = (PFN_vkBindBufferMemory2KHR)load(context, "vkBindBufferMemory2KHR"); + table->vkBindImageMemory2KHR = (PFN_vkBindImageMemory2KHR)load(context, "vkBindImageMemory2KHR"); +#endif /* defined(VK_KHR_bind_memory2) */ +#if defined(VK_KHR_buffer_device_address) + table->vkGetBufferDeviceAddressKHR = (PFN_vkGetBufferDeviceAddressKHR)load(context, "vkGetBufferDeviceAddressKHR"); + table->vkGetBufferOpaqueCaptureAddressKHR = (PFN_vkGetBufferOpaqueCaptureAddressKHR)load(context, "vkGetBufferOpaqueCaptureAddressKHR"); + table->vkGetDeviceMemoryOpaqueCaptureAddressKHR = (PFN_vkGetDeviceMemoryOpaqueCaptureAddressKHR)load(context, "vkGetDeviceMemoryOpaqueCaptureAddressKHR"); +#endif /* defined(VK_KHR_buffer_device_address) */ +#if defined(VK_KHR_create_renderpass2) + table->vkCmdBeginRenderPass2KHR = (PFN_vkCmdBeginRenderPass2KHR)load(context, "vkCmdBeginRenderPass2KHR"); + table->vkCmdEndRenderPass2KHR = (PFN_vkCmdEndRenderPass2KHR)load(context, "vkCmdEndRenderPass2KHR"); + table->vkCmdNextSubpass2KHR = (PFN_vkCmdNextSubpass2KHR)load(context, "vkCmdNextSubpass2KHR"); + table->vkCreateRenderPass2KHR = (PFN_vkCreateRenderPass2KHR)load(context, "vkCreateRenderPass2KHR"); +#endif /* defined(VK_KHR_create_renderpass2) */ +#if defined(VK_KHR_deferred_host_operations) + table->vkCreateDeferredOperationKHR = (PFN_vkCreateDeferredOperationKHR)load(context, "vkCreateDeferredOperationKHR"); + table->vkDeferredOperationJoinKHR = (PFN_vkDeferredOperationJoinKHR)load(context, "vkDeferredOperationJoinKHR"); + table->vkDestroyDeferredOperationKHR = (PFN_vkDestroyDeferredOperationKHR)load(context, "vkDestroyDeferredOperationKHR"); + table->vkGetDeferredOperationMaxConcurrencyKHR = (PFN_vkGetDeferredOperationMaxConcurrencyKHR)load(context, "vkGetDeferredOperationMaxConcurrencyKHR"); + table->vkGetDeferredOperationResultKHR = (PFN_vkGetDeferredOperationResultKHR)load(context, "vkGetDeferredOperationResultKHR"); +#endif /* defined(VK_KHR_deferred_host_operations) */ +#if defined(VK_KHR_descriptor_update_template) + table->vkCreateDescriptorUpdateTemplateKHR = (PFN_vkCreateDescriptorUpdateTemplateKHR)load(context, "vkCreateDescriptorUpdateTemplateKHR"); + table->vkDestroyDescriptorUpdateTemplateKHR = (PFN_vkDestroyDescriptorUpdateTemplateKHR)load(context, "vkDestroyDescriptorUpdateTemplateKHR"); + table->vkUpdateDescriptorSetWithTemplateKHR = (PFN_vkUpdateDescriptorSetWithTemplateKHR)load(context, "vkUpdateDescriptorSetWithTemplateKHR"); +#endif /* defined(VK_KHR_descriptor_update_template) */ +#if defined(VK_KHR_device_group) + table->vkCmdDispatchBaseKHR = (PFN_vkCmdDispatchBaseKHR)load(context, "vkCmdDispatchBaseKHR"); + table->vkCmdSetDeviceMaskKHR = (PFN_vkCmdSetDeviceMaskKHR)load(context, "vkCmdSetDeviceMaskKHR"); + table->vkGetDeviceGroupPeerMemoryFeaturesKHR = (PFN_vkGetDeviceGroupPeerMemoryFeaturesKHR)load(context, "vkGetDeviceGroupPeerMemoryFeaturesKHR"); +#endif /* defined(VK_KHR_device_group) */ +#if defined(VK_KHR_display_swapchain) + table->vkCreateSharedSwapchainsKHR = (PFN_vkCreateSharedSwapchainsKHR)load(context, "vkCreateSharedSwapchainsKHR"); +#endif /* defined(VK_KHR_display_swapchain) */ +#if defined(VK_KHR_draw_indirect_count) + table->vkCmdDrawIndexedIndirectCountKHR = (PFN_vkCmdDrawIndexedIndirectCountKHR)load(context, "vkCmdDrawIndexedIndirectCountKHR"); + table->vkCmdDrawIndirectCountKHR = (PFN_vkCmdDrawIndirectCountKHR)load(context, "vkCmdDrawIndirectCountKHR"); +#endif /* defined(VK_KHR_draw_indirect_count) */ +#if defined(VK_KHR_external_fence_fd) + table->vkGetFenceFdKHR = (PFN_vkGetFenceFdKHR)load(context, "vkGetFenceFdKHR"); + table->vkImportFenceFdKHR = (PFN_vkImportFenceFdKHR)load(context, "vkImportFenceFdKHR"); +#endif /* defined(VK_KHR_external_fence_fd) */ +#if defined(VK_KHR_external_fence_win32) + table->vkGetFenceWin32HandleKHR = (PFN_vkGetFenceWin32HandleKHR)load(context, "vkGetFenceWin32HandleKHR"); + table->vkImportFenceWin32HandleKHR = (PFN_vkImportFenceWin32HandleKHR)load(context, "vkImportFenceWin32HandleKHR"); +#endif /* defined(VK_KHR_external_fence_win32) */ +#if defined(VK_KHR_external_memory_fd) + table->vkGetMemoryFdKHR = (PFN_vkGetMemoryFdKHR)load(context, "vkGetMemoryFdKHR"); + table->vkGetMemoryFdPropertiesKHR = (PFN_vkGetMemoryFdPropertiesKHR)load(context, "vkGetMemoryFdPropertiesKHR"); +#endif /* defined(VK_KHR_external_memory_fd) */ +#if defined(VK_KHR_external_memory_win32) + table->vkGetMemoryWin32HandleKHR = (PFN_vkGetMemoryWin32HandleKHR)load(context, "vkGetMemoryWin32HandleKHR"); + table->vkGetMemoryWin32HandlePropertiesKHR = (PFN_vkGetMemoryWin32HandlePropertiesKHR)load(context, "vkGetMemoryWin32HandlePropertiesKHR"); +#endif /* defined(VK_KHR_external_memory_win32) */ +#if defined(VK_KHR_external_semaphore_fd) + table->vkGetSemaphoreFdKHR = (PFN_vkGetSemaphoreFdKHR)load(context, "vkGetSemaphoreFdKHR"); + table->vkImportSemaphoreFdKHR = (PFN_vkImportSemaphoreFdKHR)load(context, "vkImportSemaphoreFdKHR"); +#endif /* defined(VK_KHR_external_semaphore_fd) */ +#if defined(VK_KHR_external_semaphore_win32) + table->vkGetSemaphoreWin32HandleKHR = (PFN_vkGetSemaphoreWin32HandleKHR)load(context, "vkGetSemaphoreWin32HandleKHR"); + table->vkImportSemaphoreWin32HandleKHR = (PFN_vkImportSemaphoreWin32HandleKHR)load(context, "vkImportSemaphoreWin32HandleKHR"); +#endif /* defined(VK_KHR_external_semaphore_win32) */ +#if defined(VK_KHR_get_memory_requirements2) + table->vkGetBufferMemoryRequirements2KHR = (PFN_vkGetBufferMemoryRequirements2KHR)load(context, "vkGetBufferMemoryRequirements2KHR"); + table->vkGetImageMemoryRequirements2KHR = (PFN_vkGetImageMemoryRequirements2KHR)load(context, "vkGetImageMemoryRequirements2KHR"); + table->vkGetImageSparseMemoryRequirements2KHR = (PFN_vkGetImageSparseMemoryRequirements2KHR)load(context, "vkGetImageSparseMemoryRequirements2KHR"); +#endif /* defined(VK_KHR_get_memory_requirements2) */ +#if defined(VK_KHR_maintenance1) + table->vkTrimCommandPoolKHR = (PFN_vkTrimCommandPoolKHR)load(context, "vkTrimCommandPoolKHR"); +#endif /* defined(VK_KHR_maintenance1) */ +#if defined(VK_KHR_maintenance3) + table->vkGetDescriptorSetLayoutSupportKHR = (PFN_vkGetDescriptorSetLayoutSupportKHR)load(context, "vkGetDescriptorSetLayoutSupportKHR"); +#endif /* defined(VK_KHR_maintenance3) */ +#if defined(VK_KHR_performance_query) + table->vkAcquireProfilingLockKHR = (PFN_vkAcquireProfilingLockKHR)load(context, "vkAcquireProfilingLockKHR"); + table->vkReleaseProfilingLockKHR = (PFN_vkReleaseProfilingLockKHR)load(context, "vkReleaseProfilingLockKHR"); +#endif /* defined(VK_KHR_performance_query) */ +#if defined(VK_KHR_pipeline_executable_properties) + table->vkGetPipelineExecutableInternalRepresentationsKHR = (PFN_vkGetPipelineExecutableInternalRepresentationsKHR)load(context, "vkGetPipelineExecutableInternalRepresentationsKHR"); + table->vkGetPipelineExecutablePropertiesKHR = (PFN_vkGetPipelineExecutablePropertiesKHR)load(context, "vkGetPipelineExecutablePropertiesKHR"); + table->vkGetPipelineExecutableStatisticsKHR = (PFN_vkGetPipelineExecutableStatisticsKHR)load(context, "vkGetPipelineExecutableStatisticsKHR"); +#endif /* defined(VK_KHR_pipeline_executable_properties) */ +#if defined(VK_KHR_push_descriptor) + table->vkCmdPushDescriptorSetKHR = (PFN_vkCmdPushDescriptorSetKHR)load(context, "vkCmdPushDescriptorSetKHR"); +#endif /* defined(VK_KHR_push_descriptor) */ +#if defined(VK_KHR_ray_tracing) + table->vkBindAccelerationStructureMemoryKHR = (PFN_vkBindAccelerationStructureMemoryKHR)load(context, "vkBindAccelerationStructureMemoryKHR"); + table->vkBuildAccelerationStructureKHR = (PFN_vkBuildAccelerationStructureKHR)load(context, "vkBuildAccelerationStructureKHR"); + table->vkCmdBuildAccelerationStructureIndirectKHR = (PFN_vkCmdBuildAccelerationStructureIndirectKHR)load(context, "vkCmdBuildAccelerationStructureIndirectKHR"); + table->vkCmdBuildAccelerationStructureKHR = (PFN_vkCmdBuildAccelerationStructureKHR)load(context, "vkCmdBuildAccelerationStructureKHR"); + table->vkCmdCopyAccelerationStructureKHR = (PFN_vkCmdCopyAccelerationStructureKHR)load(context, "vkCmdCopyAccelerationStructureKHR"); + table->vkCmdCopyAccelerationStructureToMemoryKHR = (PFN_vkCmdCopyAccelerationStructureToMemoryKHR)load(context, "vkCmdCopyAccelerationStructureToMemoryKHR"); + table->vkCmdCopyMemoryToAccelerationStructureKHR = (PFN_vkCmdCopyMemoryToAccelerationStructureKHR)load(context, "vkCmdCopyMemoryToAccelerationStructureKHR"); + table->vkCmdTraceRaysIndirectKHR = (PFN_vkCmdTraceRaysIndirectKHR)load(context, "vkCmdTraceRaysIndirectKHR"); + table->vkCmdTraceRaysKHR = (PFN_vkCmdTraceRaysKHR)load(context, "vkCmdTraceRaysKHR"); + table->vkCmdWriteAccelerationStructuresPropertiesKHR = (PFN_vkCmdWriteAccelerationStructuresPropertiesKHR)load(context, "vkCmdWriteAccelerationStructuresPropertiesKHR"); + table->vkCopyAccelerationStructureKHR = (PFN_vkCopyAccelerationStructureKHR)load(context, "vkCopyAccelerationStructureKHR"); + table->vkCopyAccelerationStructureToMemoryKHR = (PFN_vkCopyAccelerationStructureToMemoryKHR)load(context, "vkCopyAccelerationStructureToMemoryKHR"); + table->vkCopyMemoryToAccelerationStructureKHR = (PFN_vkCopyMemoryToAccelerationStructureKHR)load(context, "vkCopyMemoryToAccelerationStructureKHR"); + table->vkCreateAccelerationStructureKHR = (PFN_vkCreateAccelerationStructureKHR)load(context, "vkCreateAccelerationStructureKHR"); + table->vkCreateRayTracingPipelinesKHR = (PFN_vkCreateRayTracingPipelinesKHR)load(context, "vkCreateRayTracingPipelinesKHR"); + table->vkDestroyAccelerationStructureKHR = (PFN_vkDestroyAccelerationStructureKHR)load(context, "vkDestroyAccelerationStructureKHR"); + table->vkGetAccelerationStructureDeviceAddressKHR = (PFN_vkGetAccelerationStructureDeviceAddressKHR)load(context, "vkGetAccelerationStructureDeviceAddressKHR"); + table->vkGetAccelerationStructureMemoryRequirementsKHR = (PFN_vkGetAccelerationStructureMemoryRequirementsKHR)load(context, "vkGetAccelerationStructureMemoryRequirementsKHR"); + table->vkGetDeviceAccelerationStructureCompatibilityKHR = (PFN_vkGetDeviceAccelerationStructureCompatibilityKHR)load(context, "vkGetDeviceAccelerationStructureCompatibilityKHR"); + table->vkGetRayTracingCaptureReplayShaderGroupHandlesKHR = (PFN_vkGetRayTracingCaptureReplayShaderGroupHandlesKHR)load(context, "vkGetRayTracingCaptureReplayShaderGroupHandlesKHR"); + table->vkGetRayTracingShaderGroupHandlesKHR = (PFN_vkGetRayTracingShaderGroupHandlesKHR)load(context, "vkGetRayTracingShaderGroupHandlesKHR"); + table->vkWriteAccelerationStructuresPropertiesKHR = (PFN_vkWriteAccelerationStructuresPropertiesKHR)load(context, "vkWriteAccelerationStructuresPropertiesKHR"); +#endif /* defined(VK_KHR_ray_tracing) */ +#if defined(VK_KHR_sampler_ycbcr_conversion) + table->vkCreateSamplerYcbcrConversionKHR = (PFN_vkCreateSamplerYcbcrConversionKHR)load(context, "vkCreateSamplerYcbcrConversionKHR"); + table->vkDestroySamplerYcbcrConversionKHR = (PFN_vkDestroySamplerYcbcrConversionKHR)load(context, "vkDestroySamplerYcbcrConversionKHR"); +#endif /* defined(VK_KHR_sampler_ycbcr_conversion) */ +#if defined(VK_KHR_shared_presentable_image) + table->vkGetSwapchainStatusKHR = (PFN_vkGetSwapchainStatusKHR)load(context, "vkGetSwapchainStatusKHR"); +#endif /* defined(VK_KHR_shared_presentable_image) */ +#if defined(VK_KHR_swapchain) + table->vkAcquireNextImageKHR = (PFN_vkAcquireNextImageKHR)load(context, "vkAcquireNextImageKHR"); + table->vkCreateSwapchainKHR = (PFN_vkCreateSwapchainKHR)load(context, "vkCreateSwapchainKHR"); + table->vkDestroySwapchainKHR = (PFN_vkDestroySwapchainKHR)load(context, "vkDestroySwapchainKHR"); + table->vkGetSwapchainImagesKHR = (PFN_vkGetSwapchainImagesKHR)load(context, "vkGetSwapchainImagesKHR"); + table->vkQueuePresentKHR = (PFN_vkQueuePresentKHR)load(context, "vkQueuePresentKHR"); +#endif /* defined(VK_KHR_swapchain) */ +#if defined(VK_KHR_timeline_semaphore) + table->vkGetSemaphoreCounterValueKHR = (PFN_vkGetSemaphoreCounterValueKHR)load(context, "vkGetSemaphoreCounterValueKHR"); + table->vkSignalSemaphoreKHR = (PFN_vkSignalSemaphoreKHR)load(context, "vkSignalSemaphoreKHR"); + table->vkWaitSemaphoresKHR = (PFN_vkWaitSemaphoresKHR)load(context, "vkWaitSemaphoresKHR"); +#endif /* defined(VK_KHR_timeline_semaphore) */ +#if defined(VK_NVX_image_view_handle) + table->vkGetImageViewAddressNVX = (PFN_vkGetImageViewAddressNVX)load(context, "vkGetImageViewAddressNVX"); + table->vkGetImageViewHandleNVX = (PFN_vkGetImageViewHandleNVX)load(context, "vkGetImageViewHandleNVX"); +#endif /* defined(VK_NVX_image_view_handle) */ +#if defined(VK_NV_clip_space_w_scaling) + table->vkCmdSetViewportWScalingNV = (PFN_vkCmdSetViewportWScalingNV)load(context, "vkCmdSetViewportWScalingNV"); +#endif /* defined(VK_NV_clip_space_w_scaling) */ +#if defined(VK_NV_device_diagnostic_checkpoints) + table->vkCmdSetCheckpointNV = (PFN_vkCmdSetCheckpointNV)load(context, "vkCmdSetCheckpointNV"); + table->vkGetQueueCheckpointDataNV = (PFN_vkGetQueueCheckpointDataNV)load(context, "vkGetQueueCheckpointDataNV"); +#endif /* defined(VK_NV_device_diagnostic_checkpoints) */ +#if defined(VK_NV_device_generated_commands) + table->vkCmdBindPipelineShaderGroupNV = (PFN_vkCmdBindPipelineShaderGroupNV)load(context, "vkCmdBindPipelineShaderGroupNV"); + table->vkCmdExecuteGeneratedCommandsNV = (PFN_vkCmdExecuteGeneratedCommandsNV)load(context, "vkCmdExecuteGeneratedCommandsNV"); + table->vkCmdPreprocessGeneratedCommandsNV = (PFN_vkCmdPreprocessGeneratedCommandsNV)load(context, "vkCmdPreprocessGeneratedCommandsNV"); + table->vkCreateIndirectCommandsLayoutNV = (PFN_vkCreateIndirectCommandsLayoutNV)load(context, "vkCreateIndirectCommandsLayoutNV"); + table->vkDestroyIndirectCommandsLayoutNV = (PFN_vkDestroyIndirectCommandsLayoutNV)load(context, "vkDestroyIndirectCommandsLayoutNV"); + table->vkGetGeneratedCommandsMemoryRequirementsNV = (PFN_vkGetGeneratedCommandsMemoryRequirementsNV)load(context, "vkGetGeneratedCommandsMemoryRequirementsNV"); +#endif /* defined(VK_NV_device_generated_commands) */ +#if defined(VK_NV_external_memory_win32) + table->vkGetMemoryWin32HandleNV = (PFN_vkGetMemoryWin32HandleNV)load(context, "vkGetMemoryWin32HandleNV"); +#endif /* defined(VK_NV_external_memory_win32) */ +#if defined(VK_NV_mesh_shader) + table->vkCmdDrawMeshTasksIndirectCountNV = (PFN_vkCmdDrawMeshTasksIndirectCountNV)load(context, "vkCmdDrawMeshTasksIndirectCountNV"); + table->vkCmdDrawMeshTasksIndirectNV = (PFN_vkCmdDrawMeshTasksIndirectNV)load(context, "vkCmdDrawMeshTasksIndirectNV"); + table->vkCmdDrawMeshTasksNV = (PFN_vkCmdDrawMeshTasksNV)load(context, "vkCmdDrawMeshTasksNV"); +#endif /* defined(VK_NV_mesh_shader) */ +#if defined(VK_NV_ray_tracing) + table->vkBindAccelerationStructureMemoryNV = (PFN_vkBindAccelerationStructureMemoryNV)load(context, "vkBindAccelerationStructureMemoryNV"); + table->vkCmdBuildAccelerationStructureNV = (PFN_vkCmdBuildAccelerationStructureNV)load(context, "vkCmdBuildAccelerationStructureNV"); + table->vkCmdCopyAccelerationStructureNV = (PFN_vkCmdCopyAccelerationStructureNV)load(context, "vkCmdCopyAccelerationStructureNV"); + table->vkCmdTraceRaysNV = (PFN_vkCmdTraceRaysNV)load(context, "vkCmdTraceRaysNV"); + table->vkCmdWriteAccelerationStructuresPropertiesNV = (PFN_vkCmdWriteAccelerationStructuresPropertiesNV)load(context, "vkCmdWriteAccelerationStructuresPropertiesNV"); + table->vkCompileDeferredNV = (PFN_vkCompileDeferredNV)load(context, "vkCompileDeferredNV"); + table->vkCreateAccelerationStructureNV = (PFN_vkCreateAccelerationStructureNV)load(context, "vkCreateAccelerationStructureNV"); + table->vkCreateRayTracingPipelinesNV = (PFN_vkCreateRayTracingPipelinesNV)load(context, "vkCreateRayTracingPipelinesNV"); + table->vkDestroyAccelerationStructureNV = (PFN_vkDestroyAccelerationStructureNV)load(context, "vkDestroyAccelerationStructureNV"); + table->vkGetAccelerationStructureHandleNV = (PFN_vkGetAccelerationStructureHandleNV)load(context, "vkGetAccelerationStructureHandleNV"); + table->vkGetAccelerationStructureMemoryRequirementsNV = (PFN_vkGetAccelerationStructureMemoryRequirementsNV)load(context, "vkGetAccelerationStructureMemoryRequirementsNV"); + table->vkGetRayTracingShaderGroupHandlesNV = (PFN_vkGetRayTracingShaderGroupHandlesNV)load(context, "vkGetRayTracingShaderGroupHandlesNV"); +#endif /* defined(VK_NV_ray_tracing) */ +#if defined(VK_NV_scissor_exclusive) + table->vkCmdSetExclusiveScissorNV = (PFN_vkCmdSetExclusiveScissorNV)load(context, "vkCmdSetExclusiveScissorNV"); +#endif /* defined(VK_NV_scissor_exclusive) */ +#if defined(VK_NV_shading_rate_image) + table->vkCmdBindShadingRateImageNV = (PFN_vkCmdBindShadingRateImageNV)load(context, "vkCmdBindShadingRateImageNV"); + table->vkCmdSetCoarseSampleOrderNV = (PFN_vkCmdSetCoarseSampleOrderNV)load(context, "vkCmdSetCoarseSampleOrderNV"); + table->vkCmdSetViewportShadingRatePaletteNV = (PFN_vkCmdSetViewportShadingRatePaletteNV)load(context, "vkCmdSetViewportShadingRatePaletteNV"); +#endif /* defined(VK_NV_shading_rate_image) */ +#if (defined(VK_EXT_full_screen_exclusive) && defined(VK_KHR_device_group)) || (defined(VK_EXT_full_screen_exclusive) && defined(VK_VERSION_1_1)) + table->vkGetDeviceGroupSurfacePresentModes2EXT = (PFN_vkGetDeviceGroupSurfacePresentModes2EXT)load(context, "vkGetDeviceGroupSurfacePresentModes2EXT"); +#endif /* (defined(VK_EXT_full_screen_exclusive) && defined(VK_KHR_device_group)) || (defined(VK_EXT_full_screen_exclusive) && defined(VK_VERSION_1_1)) */ +#if (defined(VK_KHR_descriptor_update_template) && defined(VK_KHR_push_descriptor)) || (defined(VK_KHR_push_descriptor) && defined(VK_VERSION_1_1)) || (defined(VK_KHR_push_descriptor) && defined(VK_KHR_descriptor_update_template)) + table->vkCmdPushDescriptorSetWithTemplateKHR = (PFN_vkCmdPushDescriptorSetWithTemplateKHR)load(context, "vkCmdPushDescriptorSetWithTemplateKHR"); +#endif /* (defined(VK_KHR_descriptor_update_template) && defined(VK_KHR_push_descriptor)) || (defined(VK_KHR_push_descriptor) && defined(VK_VERSION_1_1)) || (defined(VK_KHR_push_descriptor) && defined(VK_KHR_descriptor_update_template)) */ +#if (defined(VK_KHR_device_group) && defined(VK_KHR_surface)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) + table->vkGetDeviceGroupPresentCapabilitiesKHR = (PFN_vkGetDeviceGroupPresentCapabilitiesKHR)load(context, "vkGetDeviceGroupPresentCapabilitiesKHR"); + table->vkGetDeviceGroupSurfacePresentModesKHR = (PFN_vkGetDeviceGroupSurfacePresentModesKHR)load(context, "vkGetDeviceGroupSurfacePresentModesKHR"); +#endif /* (defined(VK_KHR_device_group) && defined(VK_KHR_surface)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) */ +#if (defined(VK_KHR_device_group) && defined(VK_KHR_swapchain)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) + table->vkAcquireNextImage2KHR = (PFN_vkAcquireNextImage2KHR)load(context, "vkAcquireNextImage2KHR"); +#endif /* (defined(VK_KHR_device_group) && defined(VK_KHR_swapchain)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) */ + /* VOLK_GENERATE_LOAD_DEVICE_TABLE */ +} + +#ifdef __GNUC__ +#ifdef VOLK_DEFAULT_VISIBILITY +# pragma GCC visibility push(default) +#else +# pragma GCC visibility push(hidden) +#endif +#endif + +/* VOLK_GENERATE_PROTOTYPES_C */ +#if defined(VK_VERSION_1_0) +PFN_vkAllocateCommandBuffers vkAllocateCommandBuffers; +PFN_vkAllocateDescriptorSets vkAllocateDescriptorSets; +PFN_vkAllocateMemory vkAllocateMemory; +PFN_vkBeginCommandBuffer vkBeginCommandBuffer; +PFN_vkBindBufferMemory vkBindBufferMemory; +PFN_vkBindImageMemory vkBindImageMemory; +PFN_vkCmdBeginQuery vkCmdBeginQuery; +PFN_vkCmdBeginRenderPass vkCmdBeginRenderPass; +PFN_vkCmdBindDescriptorSets vkCmdBindDescriptorSets; +PFN_vkCmdBindIndexBuffer vkCmdBindIndexBuffer; +PFN_vkCmdBindPipeline vkCmdBindPipeline; +PFN_vkCmdBindVertexBuffers vkCmdBindVertexBuffers; +PFN_vkCmdBlitImage vkCmdBlitImage; +PFN_vkCmdClearAttachments vkCmdClearAttachments; +PFN_vkCmdClearColorImage vkCmdClearColorImage; +PFN_vkCmdClearDepthStencilImage vkCmdClearDepthStencilImage; +PFN_vkCmdCopyBuffer vkCmdCopyBuffer; +PFN_vkCmdCopyBufferToImage vkCmdCopyBufferToImage; +PFN_vkCmdCopyImage vkCmdCopyImage; +PFN_vkCmdCopyImageToBuffer vkCmdCopyImageToBuffer; +PFN_vkCmdCopyQueryPoolResults vkCmdCopyQueryPoolResults; +PFN_vkCmdDispatch vkCmdDispatch; +PFN_vkCmdDispatchIndirect vkCmdDispatchIndirect; +PFN_vkCmdDraw vkCmdDraw; +PFN_vkCmdDrawIndexed vkCmdDrawIndexed; +PFN_vkCmdDrawIndexedIndirect vkCmdDrawIndexedIndirect; +PFN_vkCmdDrawIndirect vkCmdDrawIndirect; +PFN_vkCmdEndQuery vkCmdEndQuery; +PFN_vkCmdEndRenderPass vkCmdEndRenderPass; +PFN_vkCmdExecuteCommands vkCmdExecuteCommands; +PFN_vkCmdFillBuffer vkCmdFillBuffer; +PFN_vkCmdNextSubpass vkCmdNextSubpass; +PFN_vkCmdPipelineBarrier vkCmdPipelineBarrier; +PFN_vkCmdPushConstants vkCmdPushConstants; +PFN_vkCmdResetEvent vkCmdResetEvent; +PFN_vkCmdResetQueryPool vkCmdResetQueryPool; +PFN_vkCmdResolveImage vkCmdResolveImage; +PFN_vkCmdSetBlendConstants vkCmdSetBlendConstants; +PFN_vkCmdSetDepthBias vkCmdSetDepthBias; +PFN_vkCmdSetDepthBounds vkCmdSetDepthBounds; +PFN_vkCmdSetEvent vkCmdSetEvent; +PFN_vkCmdSetLineWidth vkCmdSetLineWidth; +PFN_vkCmdSetScissor vkCmdSetScissor; +PFN_vkCmdSetStencilCompareMask vkCmdSetStencilCompareMask; +PFN_vkCmdSetStencilReference vkCmdSetStencilReference; +PFN_vkCmdSetStencilWriteMask vkCmdSetStencilWriteMask; +PFN_vkCmdSetViewport vkCmdSetViewport; +PFN_vkCmdUpdateBuffer vkCmdUpdateBuffer; +PFN_vkCmdWaitEvents vkCmdWaitEvents; +PFN_vkCmdWriteTimestamp vkCmdWriteTimestamp; +PFN_vkCreateBuffer vkCreateBuffer; +PFN_vkCreateBufferView vkCreateBufferView; +PFN_vkCreateCommandPool vkCreateCommandPool; +PFN_vkCreateComputePipelines vkCreateComputePipelines; +PFN_vkCreateDescriptorPool vkCreateDescriptorPool; +PFN_vkCreateDescriptorSetLayout vkCreateDescriptorSetLayout; +PFN_vkCreateDevice vkCreateDevice; +PFN_vkCreateEvent vkCreateEvent; +PFN_vkCreateFence vkCreateFence; +PFN_vkCreateFramebuffer vkCreateFramebuffer; +PFN_vkCreateGraphicsPipelines vkCreateGraphicsPipelines; +PFN_vkCreateImage vkCreateImage; +PFN_vkCreateImageView vkCreateImageView; +PFN_vkCreateInstance vkCreateInstance; +PFN_vkCreatePipelineCache vkCreatePipelineCache; +PFN_vkCreatePipelineLayout vkCreatePipelineLayout; +PFN_vkCreateQueryPool vkCreateQueryPool; +PFN_vkCreateRenderPass vkCreateRenderPass; +PFN_vkCreateSampler vkCreateSampler; +PFN_vkCreateSemaphore vkCreateSemaphore; +PFN_vkCreateShaderModule vkCreateShaderModule; +PFN_vkDestroyBuffer vkDestroyBuffer; +PFN_vkDestroyBufferView vkDestroyBufferView; +PFN_vkDestroyCommandPool vkDestroyCommandPool; +PFN_vkDestroyDescriptorPool vkDestroyDescriptorPool; +PFN_vkDestroyDescriptorSetLayout vkDestroyDescriptorSetLayout; +PFN_vkDestroyDevice vkDestroyDevice; +PFN_vkDestroyEvent vkDestroyEvent; +PFN_vkDestroyFence vkDestroyFence; +PFN_vkDestroyFramebuffer vkDestroyFramebuffer; +PFN_vkDestroyImage vkDestroyImage; +PFN_vkDestroyImageView vkDestroyImageView; +PFN_vkDestroyInstance vkDestroyInstance; +PFN_vkDestroyPipeline vkDestroyPipeline; +PFN_vkDestroyPipelineCache vkDestroyPipelineCache; +PFN_vkDestroyPipelineLayout vkDestroyPipelineLayout; +PFN_vkDestroyQueryPool vkDestroyQueryPool; +PFN_vkDestroyRenderPass vkDestroyRenderPass; +PFN_vkDestroySampler vkDestroySampler; +PFN_vkDestroySemaphore vkDestroySemaphore; +PFN_vkDestroyShaderModule vkDestroyShaderModule; +PFN_vkDeviceWaitIdle vkDeviceWaitIdle; +PFN_vkEndCommandBuffer vkEndCommandBuffer; +PFN_vkEnumerateDeviceExtensionProperties vkEnumerateDeviceExtensionProperties; +PFN_vkEnumerateDeviceLayerProperties vkEnumerateDeviceLayerProperties; +PFN_vkEnumerateInstanceExtensionProperties vkEnumerateInstanceExtensionProperties; +PFN_vkEnumerateInstanceLayerProperties vkEnumerateInstanceLayerProperties; +PFN_vkEnumeratePhysicalDevices vkEnumeratePhysicalDevices; +PFN_vkFlushMappedMemoryRanges vkFlushMappedMemoryRanges; +PFN_vkFreeCommandBuffers vkFreeCommandBuffers; +PFN_vkFreeDescriptorSets vkFreeDescriptorSets; +PFN_vkFreeMemory vkFreeMemory; +PFN_vkGetBufferMemoryRequirements vkGetBufferMemoryRequirements; +PFN_vkGetDeviceMemoryCommitment vkGetDeviceMemoryCommitment; +PFN_vkGetDeviceProcAddr vkGetDeviceProcAddr; +PFN_vkGetDeviceQueue vkGetDeviceQueue; +PFN_vkGetEventStatus vkGetEventStatus; +PFN_vkGetFenceStatus vkGetFenceStatus; +PFN_vkGetImageMemoryRequirements vkGetImageMemoryRequirements; +PFN_vkGetImageSparseMemoryRequirements vkGetImageSparseMemoryRequirements; +PFN_vkGetImageSubresourceLayout vkGetImageSubresourceLayout; +PFN_vkGetInstanceProcAddr vkGetInstanceProcAddr; +PFN_vkGetPhysicalDeviceFeatures vkGetPhysicalDeviceFeatures; +PFN_vkGetPhysicalDeviceFormatProperties vkGetPhysicalDeviceFormatProperties; +PFN_vkGetPhysicalDeviceImageFormatProperties vkGetPhysicalDeviceImageFormatProperties; +PFN_vkGetPhysicalDeviceMemoryProperties vkGetPhysicalDeviceMemoryProperties; +PFN_vkGetPhysicalDeviceProperties vkGetPhysicalDeviceProperties; +PFN_vkGetPhysicalDeviceQueueFamilyProperties vkGetPhysicalDeviceQueueFamilyProperties; +PFN_vkGetPhysicalDeviceSparseImageFormatProperties vkGetPhysicalDeviceSparseImageFormatProperties; +PFN_vkGetPipelineCacheData vkGetPipelineCacheData; +PFN_vkGetQueryPoolResults vkGetQueryPoolResults; +PFN_vkGetRenderAreaGranularity vkGetRenderAreaGranularity; +PFN_vkInvalidateMappedMemoryRanges vkInvalidateMappedMemoryRanges; +PFN_vkMapMemory vkMapMemory; +PFN_vkMergePipelineCaches vkMergePipelineCaches; +PFN_vkQueueBindSparse vkQueueBindSparse; +PFN_vkQueueSubmit vkQueueSubmit; +PFN_vkQueueWaitIdle vkQueueWaitIdle; +PFN_vkResetCommandBuffer vkResetCommandBuffer; +PFN_vkResetCommandPool vkResetCommandPool; +PFN_vkResetDescriptorPool vkResetDescriptorPool; +PFN_vkResetEvent vkResetEvent; +PFN_vkResetFences vkResetFences; +PFN_vkSetEvent vkSetEvent; +PFN_vkUnmapMemory vkUnmapMemory; +PFN_vkUpdateDescriptorSets vkUpdateDescriptorSets; +PFN_vkWaitForFences vkWaitForFences; +#endif /* defined(VK_VERSION_1_0) */ +#if defined(VK_VERSION_1_1) +PFN_vkBindBufferMemory2 vkBindBufferMemory2; +PFN_vkBindImageMemory2 vkBindImageMemory2; +PFN_vkCmdDispatchBase vkCmdDispatchBase; +PFN_vkCmdSetDeviceMask vkCmdSetDeviceMask; +PFN_vkCreateDescriptorUpdateTemplate vkCreateDescriptorUpdateTemplate; +PFN_vkCreateSamplerYcbcrConversion vkCreateSamplerYcbcrConversion; +PFN_vkDestroyDescriptorUpdateTemplate vkDestroyDescriptorUpdateTemplate; +PFN_vkDestroySamplerYcbcrConversion vkDestroySamplerYcbcrConversion; +PFN_vkEnumerateInstanceVersion vkEnumerateInstanceVersion; +PFN_vkEnumeratePhysicalDeviceGroups vkEnumeratePhysicalDeviceGroups; +PFN_vkGetBufferMemoryRequirements2 vkGetBufferMemoryRequirements2; +PFN_vkGetDescriptorSetLayoutSupport vkGetDescriptorSetLayoutSupport; +PFN_vkGetDeviceGroupPeerMemoryFeatures vkGetDeviceGroupPeerMemoryFeatures; +PFN_vkGetDeviceQueue2 vkGetDeviceQueue2; +PFN_vkGetImageMemoryRequirements2 vkGetImageMemoryRequirements2; +PFN_vkGetImageSparseMemoryRequirements2 vkGetImageSparseMemoryRequirements2; +PFN_vkGetPhysicalDeviceExternalBufferProperties vkGetPhysicalDeviceExternalBufferProperties; +PFN_vkGetPhysicalDeviceExternalFenceProperties vkGetPhysicalDeviceExternalFenceProperties; +PFN_vkGetPhysicalDeviceExternalSemaphoreProperties vkGetPhysicalDeviceExternalSemaphoreProperties; +PFN_vkGetPhysicalDeviceFeatures2 vkGetPhysicalDeviceFeatures2; +PFN_vkGetPhysicalDeviceFormatProperties2 vkGetPhysicalDeviceFormatProperties2; +PFN_vkGetPhysicalDeviceImageFormatProperties2 vkGetPhysicalDeviceImageFormatProperties2; +PFN_vkGetPhysicalDeviceMemoryProperties2 vkGetPhysicalDeviceMemoryProperties2; +PFN_vkGetPhysicalDeviceProperties2 vkGetPhysicalDeviceProperties2; +PFN_vkGetPhysicalDeviceQueueFamilyProperties2 vkGetPhysicalDeviceQueueFamilyProperties2; +PFN_vkGetPhysicalDeviceSparseImageFormatProperties2 vkGetPhysicalDeviceSparseImageFormatProperties2; +PFN_vkTrimCommandPool vkTrimCommandPool; +PFN_vkUpdateDescriptorSetWithTemplate vkUpdateDescriptorSetWithTemplate; +#endif /* defined(VK_VERSION_1_1) */ +#if defined(VK_VERSION_1_2) +PFN_vkCmdBeginRenderPass2 vkCmdBeginRenderPass2; +PFN_vkCmdDrawIndexedIndirectCount vkCmdDrawIndexedIndirectCount; +PFN_vkCmdDrawIndirectCount vkCmdDrawIndirectCount; +PFN_vkCmdEndRenderPass2 vkCmdEndRenderPass2; +PFN_vkCmdNextSubpass2 vkCmdNextSubpass2; +PFN_vkCreateRenderPass2 vkCreateRenderPass2; +PFN_vkGetBufferDeviceAddress vkGetBufferDeviceAddress; +PFN_vkGetBufferOpaqueCaptureAddress vkGetBufferOpaqueCaptureAddress; +PFN_vkGetDeviceMemoryOpaqueCaptureAddress vkGetDeviceMemoryOpaqueCaptureAddress; +PFN_vkGetSemaphoreCounterValue vkGetSemaphoreCounterValue; +PFN_vkResetQueryPool vkResetQueryPool; +PFN_vkSignalSemaphore vkSignalSemaphore; +PFN_vkWaitSemaphores vkWaitSemaphores; +#endif /* defined(VK_VERSION_1_2) */ +#if defined(VK_AMD_buffer_marker) +PFN_vkCmdWriteBufferMarkerAMD vkCmdWriteBufferMarkerAMD; +#endif /* defined(VK_AMD_buffer_marker) */ +#if defined(VK_AMD_display_native_hdr) +PFN_vkSetLocalDimmingAMD vkSetLocalDimmingAMD; +#endif /* defined(VK_AMD_display_native_hdr) */ +#if defined(VK_AMD_draw_indirect_count) +PFN_vkCmdDrawIndexedIndirectCountAMD vkCmdDrawIndexedIndirectCountAMD; +PFN_vkCmdDrawIndirectCountAMD vkCmdDrawIndirectCountAMD; +#endif /* defined(VK_AMD_draw_indirect_count) */ +#if defined(VK_AMD_shader_info) +PFN_vkGetShaderInfoAMD vkGetShaderInfoAMD; +#endif /* defined(VK_AMD_shader_info) */ +#if defined(VK_ANDROID_external_memory_android_hardware_buffer) +PFN_vkGetAndroidHardwareBufferPropertiesANDROID vkGetAndroidHardwareBufferPropertiesANDROID; +PFN_vkGetMemoryAndroidHardwareBufferANDROID vkGetMemoryAndroidHardwareBufferANDROID; +#endif /* defined(VK_ANDROID_external_memory_android_hardware_buffer) */ +#if defined(VK_EXT_acquire_xlib_display) +PFN_vkAcquireXlibDisplayEXT vkAcquireXlibDisplayEXT; +PFN_vkGetRandROutputDisplayEXT vkGetRandROutputDisplayEXT; +#endif /* defined(VK_EXT_acquire_xlib_display) */ +#if defined(VK_EXT_buffer_device_address) +PFN_vkGetBufferDeviceAddressEXT vkGetBufferDeviceAddressEXT; +#endif /* defined(VK_EXT_buffer_device_address) */ +#if defined(VK_EXT_calibrated_timestamps) +PFN_vkGetCalibratedTimestampsEXT vkGetCalibratedTimestampsEXT; +PFN_vkGetPhysicalDeviceCalibrateableTimeDomainsEXT vkGetPhysicalDeviceCalibrateableTimeDomainsEXT; +#endif /* defined(VK_EXT_calibrated_timestamps) */ +#if defined(VK_EXT_conditional_rendering) +PFN_vkCmdBeginConditionalRenderingEXT vkCmdBeginConditionalRenderingEXT; +PFN_vkCmdEndConditionalRenderingEXT vkCmdEndConditionalRenderingEXT; +#endif /* defined(VK_EXT_conditional_rendering) */ +#if defined(VK_EXT_debug_marker) +PFN_vkCmdDebugMarkerBeginEXT vkCmdDebugMarkerBeginEXT; +PFN_vkCmdDebugMarkerEndEXT vkCmdDebugMarkerEndEXT; +PFN_vkCmdDebugMarkerInsertEXT vkCmdDebugMarkerInsertEXT; +PFN_vkDebugMarkerSetObjectNameEXT vkDebugMarkerSetObjectNameEXT; +PFN_vkDebugMarkerSetObjectTagEXT vkDebugMarkerSetObjectTagEXT; +#endif /* defined(VK_EXT_debug_marker) */ +#if defined(VK_EXT_debug_report) +PFN_vkCreateDebugReportCallbackEXT vkCreateDebugReportCallbackEXT; +PFN_vkDebugReportMessageEXT vkDebugReportMessageEXT; +PFN_vkDestroyDebugReportCallbackEXT vkDestroyDebugReportCallbackEXT; +#endif /* defined(VK_EXT_debug_report) */ +#if defined(VK_EXT_debug_utils) +PFN_vkCmdBeginDebugUtilsLabelEXT vkCmdBeginDebugUtilsLabelEXT; +PFN_vkCmdEndDebugUtilsLabelEXT vkCmdEndDebugUtilsLabelEXT; +PFN_vkCmdInsertDebugUtilsLabelEXT vkCmdInsertDebugUtilsLabelEXT; +PFN_vkCreateDebugUtilsMessengerEXT vkCreateDebugUtilsMessengerEXT; +PFN_vkDestroyDebugUtilsMessengerEXT vkDestroyDebugUtilsMessengerEXT; +PFN_vkQueueBeginDebugUtilsLabelEXT vkQueueBeginDebugUtilsLabelEXT; +PFN_vkQueueEndDebugUtilsLabelEXT vkQueueEndDebugUtilsLabelEXT; +PFN_vkQueueInsertDebugUtilsLabelEXT vkQueueInsertDebugUtilsLabelEXT; +PFN_vkSetDebugUtilsObjectNameEXT vkSetDebugUtilsObjectNameEXT; +PFN_vkSetDebugUtilsObjectTagEXT vkSetDebugUtilsObjectTagEXT; +PFN_vkSubmitDebugUtilsMessageEXT vkSubmitDebugUtilsMessageEXT; +#endif /* defined(VK_EXT_debug_utils) */ +#if defined(VK_EXT_direct_mode_display) +PFN_vkReleaseDisplayEXT vkReleaseDisplayEXT; +#endif /* defined(VK_EXT_direct_mode_display) */ +#if defined(VK_EXT_directfb_surface) +PFN_vkCreateDirectFBSurfaceEXT vkCreateDirectFBSurfaceEXT; +PFN_vkGetPhysicalDeviceDirectFBPresentationSupportEXT vkGetPhysicalDeviceDirectFBPresentationSupportEXT; +#endif /* defined(VK_EXT_directfb_surface) */ +#if defined(VK_EXT_discard_rectangles) +PFN_vkCmdSetDiscardRectangleEXT vkCmdSetDiscardRectangleEXT; +#endif /* defined(VK_EXT_discard_rectangles) */ +#if defined(VK_EXT_display_control) +PFN_vkDisplayPowerControlEXT vkDisplayPowerControlEXT; +PFN_vkGetSwapchainCounterEXT vkGetSwapchainCounterEXT; +PFN_vkRegisterDeviceEventEXT vkRegisterDeviceEventEXT; +PFN_vkRegisterDisplayEventEXT vkRegisterDisplayEventEXT; +#endif /* defined(VK_EXT_display_control) */ +#if defined(VK_EXT_display_surface_counter) +PFN_vkGetPhysicalDeviceSurfaceCapabilities2EXT vkGetPhysicalDeviceSurfaceCapabilities2EXT; +#endif /* defined(VK_EXT_display_surface_counter) */ +#if defined(VK_EXT_extended_dynamic_state) +PFN_vkCmdBindVertexBuffers2EXT vkCmdBindVertexBuffers2EXT; +PFN_vkCmdSetCullModeEXT vkCmdSetCullModeEXT; +PFN_vkCmdSetDepthBoundsTestEnableEXT vkCmdSetDepthBoundsTestEnableEXT; +PFN_vkCmdSetDepthCompareOpEXT vkCmdSetDepthCompareOpEXT; +PFN_vkCmdSetDepthTestEnableEXT vkCmdSetDepthTestEnableEXT; +PFN_vkCmdSetDepthWriteEnableEXT vkCmdSetDepthWriteEnableEXT; +PFN_vkCmdSetFrontFaceEXT vkCmdSetFrontFaceEXT; +PFN_vkCmdSetPrimitiveTopologyEXT vkCmdSetPrimitiveTopologyEXT; +PFN_vkCmdSetScissorWithCountEXT vkCmdSetScissorWithCountEXT; +PFN_vkCmdSetStencilOpEXT vkCmdSetStencilOpEXT; +PFN_vkCmdSetStencilTestEnableEXT vkCmdSetStencilTestEnableEXT; +PFN_vkCmdSetViewportWithCountEXT vkCmdSetViewportWithCountEXT; +#endif /* defined(VK_EXT_extended_dynamic_state) */ +#if defined(VK_EXT_external_memory_host) +PFN_vkGetMemoryHostPointerPropertiesEXT vkGetMemoryHostPointerPropertiesEXT; +#endif /* defined(VK_EXT_external_memory_host) */ +#if defined(VK_EXT_full_screen_exclusive) +PFN_vkAcquireFullScreenExclusiveModeEXT vkAcquireFullScreenExclusiveModeEXT; +PFN_vkGetPhysicalDeviceSurfacePresentModes2EXT vkGetPhysicalDeviceSurfacePresentModes2EXT; +PFN_vkReleaseFullScreenExclusiveModeEXT vkReleaseFullScreenExclusiveModeEXT; +#endif /* defined(VK_EXT_full_screen_exclusive) */ +#if defined(VK_EXT_hdr_metadata) +PFN_vkSetHdrMetadataEXT vkSetHdrMetadataEXT; +#endif /* defined(VK_EXT_hdr_metadata) */ +#if defined(VK_EXT_headless_surface) +PFN_vkCreateHeadlessSurfaceEXT vkCreateHeadlessSurfaceEXT; +#endif /* defined(VK_EXT_headless_surface) */ +#if defined(VK_EXT_host_query_reset) +PFN_vkResetQueryPoolEXT vkResetQueryPoolEXT; +#endif /* defined(VK_EXT_host_query_reset) */ +#if defined(VK_EXT_image_drm_format_modifier) +PFN_vkGetImageDrmFormatModifierPropertiesEXT vkGetImageDrmFormatModifierPropertiesEXT; +#endif /* defined(VK_EXT_image_drm_format_modifier) */ +#if defined(VK_EXT_line_rasterization) +PFN_vkCmdSetLineStippleEXT vkCmdSetLineStippleEXT; +#endif /* defined(VK_EXT_line_rasterization) */ +#if defined(VK_EXT_metal_surface) +PFN_vkCreateMetalSurfaceEXT vkCreateMetalSurfaceEXT; +#endif /* defined(VK_EXT_metal_surface) */ +#if defined(VK_EXT_private_data) +PFN_vkCreatePrivateDataSlotEXT vkCreatePrivateDataSlotEXT; +PFN_vkDestroyPrivateDataSlotEXT vkDestroyPrivateDataSlotEXT; +PFN_vkGetPrivateDataEXT vkGetPrivateDataEXT; +PFN_vkSetPrivateDataEXT vkSetPrivateDataEXT; +#endif /* defined(VK_EXT_private_data) */ +#if defined(VK_EXT_sample_locations) +PFN_vkCmdSetSampleLocationsEXT vkCmdSetSampleLocationsEXT; +PFN_vkGetPhysicalDeviceMultisamplePropertiesEXT vkGetPhysicalDeviceMultisamplePropertiesEXT; +#endif /* defined(VK_EXT_sample_locations) */ +#if defined(VK_EXT_tooling_info) +PFN_vkGetPhysicalDeviceToolPropertiesEXT vkGetPhysicalDeviceToolPropertiesEXT; +#endif /* defined(VK_EXT_tooling_info) */ +#if defined(VK_EXT_transform_feedback) +PFN_vkCmdBeginQueryIndexedEXT vkCmdBeginQueryIndexedEXT; +PFN_vkCmdBeginTransformFeedbackEXT vkCmdBeginTransformFeedbackEXT; +PFN_vkCmdBindTransformFeedbackBuffersEXT vkCmdBindTransformFeedbackBuffersEXT; +PFN_vkCmdDrawIndirectByteCountEXT vkCmdDrawIndirectByteCountEXT; +PFN_vkCmdEndQueryIndexedEXT vkCmdEndQueryIndexedEXT; +PFN_vkCmdEndTransformFeedbackEXT vkCmdEndTransformFeedbackEXT; +#endif /* defined(VK_EXT_transform_feedback) */ +#if defined(VK_EXT_validation_cache) +PFN_vkCreateValidationCacheEXT vkCreateValidationCacheEXT; +PFN_vkDestroyValidationCacheEXT vkDestroyValidationCacheEXT; +PFN_vkGetValidationCacheDataEXT vkGetValidationCacheDataEXT; +PFN_vkMergeValidationCachesEXT vkMergeValidationCachesEXT; +#endif /* defined(VK_EXT_validation_cache) */ +#if defined(VK_FUCHSIA_imagepipe_surface) +PFN_vkCreateImagePipeSurfaceFUCHSIA vkCreateImagePipeSurfaceFUCHSIA; +#endif /* defined(VK_FUCHSIA_imagepipe_surface) */ +#if defined(VK_GGP_stream_descriptor_surface) +PFN_vkCreateStreamDescriptorSurfaceGGP vkCreateStreamDescriptorSurfaceGGP; +#endif /* defined(VK_GGP_stream_descriptor_surface) */ +#if defined(VK_GOOGLE_display_timing) +PFN_vkGetPastPresentationTimingGOOGLE vkGetPastPresentationTimingGOOGLE; +PFN_vkGetRefreshCycleDurationGOOGLE vkGetRefreshCycleDurationGOOGLE; +#endif /* defined(VK_GOOGLE_display_timing) */ +#if defined(VK_INTEL_performance_query) +PFN_vkAcquirePerformanceConfigurationINTEL vkAcquirePerformanceConfigurationINTEL; +PFN_vkCmdSetPerformanceMarkerINTEL vkCmdSetPerformanceMarkerINTEL; +PFN_vkCmdSetPerformanceOverrideINTEL vkCmdSetPerformanceOverrideINTEL; +PFN_vkCmdSetPerformanceStreamMarkerINTEL vkCmdSetPerformanceStreamMarkerINTEL; +PFN_vkGetPerformanceParameterINTEL vkGetPerformanceParameterINTEL; +PFN_vkInitializePerformanceApiINTEL vkInitializePerformanceApiINTEL; +PFN_vkQueueSetPerformanceConfigurationINTEL vkQueueSetPerformanceConfigurationINTEL; +PFN_vkReleasePerformanceConfigurationINTEL vkReleasePerformanceConfigurationINTEL; +PFN_vkUninitializePerformanceApiINTEL vkUninitializePerformanceApiINTEL; +#endif /* defined(VK_INTEL_performance_query) */ +#if defined(VK_KHR_android_surface) +PFN_vkCreateAndroidSurfaceKHR vkCreateAndroidSurfaceKHR; +#endif /* defined(VK_KHR_android_surface) */ +#if defined(VK_KHR_bind_memory2) +PFN_vkBindBufferMemory2KHR vkBindBufferMemory2KHR; +PFN_vkBindImageMemory2KHR vkBindImageMemory2KHR; +#endif /* defined(VK_KHR_bind_memory2) */ +#if defined(VK_KHR_buffer_device_address) +PFN_vkGetBufferDeviceAddressKHR vkGetBufferDeviceAddressKHR; +PFN_vkGetBufferOpaqueCaptureAddressKHR vkGetBufferOpaqueCaptureAddressKHR; +PFN_vkGetDeviceMemoryOpaqueCaptureAddressKHR vkGetDeviceMemoryOpaqueCaptureAddressKHR; +#endif /* defined(VK_KHR_buffer_device_address) */ +#if defined(VK_KHR_create_renderpass2) +PFN_vkCmdBeginRenderPass2KHR vkCmdBeginRenderPass2KHR; +PFN_vkCmdEndRenderPass2KHR vkCmdEndRenderPass2KHR; +PFN_vkCmdNextSubpass2KHR vkCmdNextSubpass2KHR; +PFN_vkCreateRenderPass2KHR vkCreateRenderPass2KHR; +#endif /* defined(VK_KHR_create_renderpass2) */ +#if defined(VK_KHR_deferred_host_operations) +PFN_vkCreateDeferredOperationKHR vkCreateDeferredOperationKHR; +PFN_vkDeferredOperationJoinKHR vkDeferredOperationJoinKHR; +PFN_vkDestroyDeferredOperationKHR vkDestroyDeferredOperationKHR; +PFN_vkGetDeferredOperationMaxConcurrencyKHR vkGetDeferredOperationMaxConcurrencyKHR; +PFN_vkGetDeferredOperationResultKHR vkGetDeferredOperationResultKHR; +#endif /* defined(VK_KHR_deferred_host_operations) */ +#if defined(VK_KHR_descriptor_update_template) +PFN_vkCreateDescriptorUpdateTemplateKHR vkCreateDescriptorUpdateTemplateKHR; +PFN_vkDestroyDescriptorUpdateTemplateKHR vkDestroyDescriptorUpdateTemplateKHR; +PFN_vkUpdateDescriptorSetWithTemplateKHR vkUpdateDescriptorSetWithTemplateKHR; +#endif /* defined(VK_KHR_descriptor_update_template) */ +#if defined(VK_KHR_device_group) +PFN_vkCmdDispatchBaseKHR vkCmdDispatchBaseKHR; +PFN_vkCmdSetDeviceMaskKHR vkCmdSetDeviceMaskKHR; +PFN_vkGetDeviceGroupPeerMemoryFeaturesKHR vkGetDeviceGroupPeerMemoryFeaturesKHR; +#endif /* defined(VK_KHR_device_group) */ +#if defined(VK_KHR_device_group_creation) +PFN_vkEnumeratePhysicalDeviceGroupsKHR vkEnumeratePhysicalDeviceGroupsKHR; +#endif /* defined(VK_KHR_device_group_creation) */ +#if defined(VK_KHR_display) +PFN_vkCreateDisplayModeKHR vkCreateDisplayModeKHR; +PFN_vkCreateDisplayPlaneSurfaceKHR vkCreateDisplayPlaneSurfaceKHR; +PFN_vkGetDisplayModePropertiesKHR vkGetDisplayModePropertiesKHR; +PFN_vkGetDisplayPlaneCapabilitiesKHR vkGetDisplayPlaneCapabilitiesKHR; +PFN_vkGetDisplayPlaneSupportedDisplaysKHR vkGetDisplayPlaneSupportedDisplaysKHR; +PFN_vkGetPhysicalDeviceDisplayPlanePropertiesKHR vkGetPhysicalDeviceDisplayPlanePropertiesKHR; +PFN_vkGetPhysicalDeviceDisplayPropertiesKHR vkGetPhysicalDeviceDisplayPropertiesKHR; +#endif /* defined(VK_KHR_display) */ +#if defined(VK_KHR_display_swapchain) +PFN_vkCreateSharedSwapchainsKHR vkCreateSharedSwapchainsKHR; +#endif /* defined(VK_KHR_display_swapchain) */ +#if defined(VK_KHR_draw_indirect_count) +PFN_vkCmdDrawIndexedIndirectCountKHR vkCmdDrawIndexedIndirectCountKHR; +PFN_vkCmdDrawIndirectCountKHR vkCmdDrawIndirectCountKHR; +#endif /* defined(VK_KHR_draw_indirect_count) */ +#if defined(VK_KHR_external_fence_capabilities) +PFN_vkGetPhysicalDeviceExternalFencePropertiesKHR vkGetPhysicalDeviceExternalFencePropertiesKHR; +#endif /* defined(VK_KHR_external_fence_capabilities) */ +#if defined(VK_KHR_external_fence_fd) +PFN_vkGetFenceFdKHR vkGetFenceFdKHR; +PFN_vkImportFenceFdKHR vkImportFenceFdKHR; +#endif /* defined(VK_KHR_external_fence_fd) */ +#if defined(VK_KHR_external_fence_win32) +PFN_vkGetFenceWin32HandleKHR vkGetFenceWin32HandleKHR; +PFN_vkImportFenceWin32HandleKHR vkImportFenceWin32HandleKHR; +#endif /* defined(VK_KHR_external_fence_win32) */ +#if defined(VK_KHR_external_memory_capabilities) +PFN_vkGetPhysicalDeviceExternalBufferPropertiesKHR vkGetPhysicalDeviceExternalBufferPropertiesKHR; +#endif /* defined(VK_KHR_external_memory_capabilities) */ +#if defined(VK_KHR_external_memory_fd) +PFN_vkGetMemoryFdKHR vkGetMemoryFdKHR; +PFN_vkGetMemoryFdPropertiesKHR vkGetMemoryFdPropertiesKHR; +#endif /* defined(VK_KHR_external_memory_fd) */ +#if defined(VK_KHR_external_memory_win32) +PFN_vkGetMemoryWin32HandleKHR vkGetMemoryWin32HandleKHR; +PFN_vkGetMemoryWin32HandlePropertiesKHR vkGetMemoryWin32HandlePropertiesKHR; +#endif /* defined(VK_KHR_external_memory_win32) */ +#if defined(VK_KHR_external_semaphore_capabilities) +PFN_vkGetPhysicalDeviceExternalSemaphorePropertiesKHR vkGetPhysicalDeviceExternalSemaphorePropertiesKHR; +#endif /* defined(VK_KHR_external_semaphore_capabilities) */ +#if defined(VK_KHR_external_semaphore_fd) +PFN_vkGetSemaphoreFdKHR vkGetSemaphoreFdKHR; +PFN_vkImportSemaphoreFdKHR vkImportSemaphoreFdKHR; +#endif /* defined(VK_KHR_external_semaphore_fd) */ +#if defined(VK_KHR_external_semaphore_win32) +PFN_vkGetSemaphoreWin32HandleKHR vkGetSemaphoreWin32HandleKHR; +PFN_vkImportSemaphoreWin32HandleKHR vkImportSemaphoreWin32HandleKHR; +#endif /* defined(VK_KHR_external_semaphore_win32) */ +#if defined(VK_KHR_get_display_properties2) +PFN_vkGetDisplayModeProperties2KHR vkGetDisplayModeProperties2KHR; +PFN_vkGetDisplayPlaneCapabilities2KHR vkGetDisplayPlaneCapabilities2KHR; +PFN_vkGetPhysicalDeviceDisplayPlaneProperties2KHR vkGetPhysicalDeviceDisplayPlaneProperties2KHR; +PFN_vkGetPhysicalDeviceDisplayProperties2KHR vkGetPhysicalDeviceDisplayProperties2KHR; +#endif /* defined(VK_KHR_get_display_properties2) */ +#if defined(VK_KHR_get_memory_requirements2) +PFN_vkGetBufferMemoryRequirements2KHR vkGetBufferMemoryRequirements2KHR; +PFN_vkGetImageMemoryRequirements2KHR vkGetImageMemoryRequirements2KHR; +PFN_vkGetImageSparseMemoryRequirements2KHR vkGetImageSparseMemoryRequirements2KHR; +#endif /* defined(VK_KHR_get_memory_requirements2) */ +#if defined(VK_KHR_get_physical_device_properties2) +PFN_vkGetPhysicalDeviceFeatures2KHR vkGetPhysicalDeviceFeatures2KHR; +PFN_vkGetPhysicalDeviceFormatProperties2KHR vkGetPhysicalDeviceFormatProperties2KHR; +PFN_vkGetPhysicalDeviceImageFormatProperties2KHR vkGetPhysicalDeviceImageFormatProperties2KHR; +PFN_vkGetPhysicalDeviceMemoryProperties2KHR vkGetPhysicalDeviceMemoryProperties2KHR; +PFN_vkGetPhysicalDeviceProperties2KHR vkGetPhysicalDeviceProperties2KHR; +PFN_vkGetPhysicalDeviceQueueFamilyProperties2KHR vkGetPhysicalDeviceQueueFamilyProperties2KHR; +PFN_vkGetPhysicalDeviceSparseImageFormatProperties2KHR vkGetPhysicalDeviceSparseImageFormatProperties2KHR; +#endif /* defined(VK_KHR_get_physical_device_properties2) */ +#if defined(VK_KHR_get_surface_capabilities2) +PFN_vkGetPhysicalDeviceSurfaceCapabilities2KHR vkGetPhysicalDeviceSurfaceCapabilities2KHR; +PFN_vkGetPhysicalDeviceSurfaceFormats2KHR vkGetPhysicalDeviceSurfaceFormats2KHR; +#endif /* defined(VK_KHR_get_surface_capabilities2) */ +#if defined(VK_KHR_maintenance1) +PFN_vkTrimCommandPoolKHR vkTrimCommandPoolKHR; +#endif /* defined(VK_KHR_maintenance1) */ +#if defined(VK_KHR_maintenance3) +PFN_vkGetDescriptorSetLayoutSupportKHR vkGetDescriptorSetLayoutSupportKHR; +#endif /* defined(VK_KHR_maintenance3) */ +#if defined(VK_KHR_performance_query) +PFN_vkAcquireProfilingLockKHR vkAcquireProfilingLockKHR; +PFN_vkEnumeratePhysicalDeviceQueueFamilyPerformanceQueryCountersKHR vkEnumeratePhysicalDeviceQueueFamilyPerformanceQueryCountersKHR; +PFN_vkGetPhysicalDeviceQueueFamilyPerformanceQueryPassesKHR vkGetPhysicalDeviceQueueFamilyPerformanceQueryPassesKHR; +PFN_vkReleaseProfilingLockKHR vkReleaseProfilingLockKHR; +#endif /* defined(VK_KHR_performance_query) */ +#if defined(VK_KHR_pipeline_executable_properties) +PFN_vkGetPipelineExecutableInternalRepresentationsKHR vkGetPipelineExecutableInternalRepresentationsKHR; +PFN_vkGetPipelineExecutablePropertiesKHR vkGetPipelineExecutablePropertiesKHR; +PFN_vkGetPipelineExecutableStatisticsKHR vkGetPipelineExecutableStatisticsKHR; +#endif /* defined(VK_KHR_pipeline_executable_properties) */ +#if defined(VK_KHR_push_descriptor) +PFN_vkCmdPushDescriptorSetKHR vkCmdPushDescriptorSetKHR; +#endif /* defined(VK_KHR_push_descriptor) */ +#if defined(VK_KHR_ray_tracing) +PFN_vkBindAccelerationStructureMemoryKHR vkBindAccelerationStructureMemoryKHR; +PFN_vkBuildAccelerationStructureKHR vkBuildAccelerationStructureKHR; +PFN_vkCmdBuildAccelerationStructureIndirectKHR vkCmdBuildAccelerationStructureIndirectKHR; +PFN_vkCmdBuildAccelerationStructureKHR vkCmdBuildAccelerationStructureKHR; +PFN_vkCmdCopyAccelerationStructureKHR vkCmdCopyAccelerationStructureKHR; +PFN_vkCmdCopyAccelerationStructureToMemoryKHR vkCmdCopyAccelerationStructureToMemoryKHR; +PFN_vkCmdCopyMemoryToAccelerationStructureKHR vkCmdCopyMemoryToAccelerationStructureKHR; +PFN_vkCmdTraceRaysIndirectKHR vkCmdTraceRaysIndirectKHR; +PFN_vkCmdTraceRaysKHR vkCmdTraceRaysKHR; +PFN_vkCmdWriteAccelerationStructuresPropertiesKHR vkCmdWriteAccelerationStructuresPropertiesKHR; +PFN_vkCopyAccelerationStructureKHR vkCopyAccelerationStructureKHR; +PFN_vkCopyAccelerationStructureToMemoryKHR vkCopyAccelerationStructureToMemoryKHR; +PFN_vkCopyMemoryToAccelerationStructureKHR vkCopyMemoryToAccelerationStructureKHR; +PFN_vkCreateAccelerationStructureKHR vkCreateAccelerationStructureKHR; +PFN_vkCreateRayTracingPipelinesKHR vkCreateRayTracingPipelinesKHR; +PFN_vkDestroyAccelerationStructureKHR vkDestroyAccelerationStructureKHR; +PFN_vkGetAccelerationStructureDeviceAddressKHR vkGetAccelerationStructureDeviceAddressKHR; +PFN_vkGetAccelerationStructureMemoryRequirementsKHR vkGetAccelerationStructureMemoryRequirementsKHR; +PFN_vkGetDeviceAccelerationStructureCompatibilityKHR vkGetDeviceAccelerationStructureCompatibilityKHR; +PFN_vkGetRayTracingCaptureReplayShaderGroupHandlesKHR vkGetRayTracingCaptureReplayShaderGroupHandlesKHR; +PFN_vkGetRayTracingShaderGroupHandlesKHR vkGetRayTracingShaderGroupHandlesKHR; +PFN_vkWriteAccelerationStructuresPropertiesKHR vkWriteAccelerationStructuresPropertiesKHR; +#endif /* defined(VK_KHR_ray_tracing) */ +#if defined(VK_KHR_sampler_ycbcr_conversion) +PFN_vkCreateSamplerYcbcrConversionKHR vkCreateSamplerYcbcrConversionKHR; +PFN_vkDestroySamplerYcbcrConversionKHR vkDestroySamplerYcbcrConversionKHR; +#endif /* defined(VK_KHR_sampler_ycbcr_conversion) */ +#if defined(VK_KHR_shared_presentable_image) +PFN_vkGetSwapchainStatusKHR vkGetSwapchainStatusKHR; +#endif /* defined(VK_KHR_shared_presentable_image) */ +#if defined(VK_KHR_surface) +PFN_vkDestroySurfaceKHR vkDestroySurfaceKHR; +PFN_vkGetPhysicalDeviceSurfaceCapabilitiesKHR vkGetPhysicalDeviceSurfaceCapabilitiesKHR; +PFN_vkGetPhysicalDeviceSurfaceFormatsKHR vkGetPhysicalDeviceSurfaceFormatsKHR; +PFN_vkGetPhysicalDeviceSurfacePresentModesKHR vkGetPhysicalDeviceSurfacePresentModesKHR; +PFN_vkGetPhysicalDeviceSurfaceSupportKHR vkGetPhysicalDeviceSurfaceSupportKHR; +#endif /* defined(VK_KHR_surface) */ +#if defined(VK_KHR_swapchain) +PFN_vkAcquireNextImageKHR vkAcquireNextImageKHR; +PFN_vkCreateSwapchainKHR vkCreateSwapchainKHR; +PFN_vkDestroySwapchainKHR vkDestroySwapchainKHR; +PFN_vkGetSwapchainImagesKHR vkGetSwapchainImagesKHR; +PFN_vkQueuePresentKHR vkQueuePresentKHR; +#endif /* defined(VK_KHR_swapchain) */ +#if defined(VK_KHR_timeline_semaphore) +PFN_vkGetSemaphoreCounterValueKHR vkGetSemaphoreCounterValueKHR; +PFN_vkSignalSemaphoreKHR vkSignalSemaphoreKHR; +PFN_vkWaitSemaphoresKHR vkWaitSemaphoresKHR; +#endif /* defined(VK_KHR_timeline_semaphore) */ +#if defined(VK_KHR_wayland_surface) +PFN_vkCreateWaylandSurfaceKHR vkCreateWaylandSurfaceKHR; +PFN_vkGetPhysicalDeviceWaylandPresentationSupportKHR vkGetPhysicalDeviceWaylandPresentationSupportKHR; +#endif /* defined(VK_KHR_wayland_surface) */ +#if defined(VK_KHR_win32_surface) +PFN_vkCreateWin32SurfaceKHR vkCreateWin32SurfaceKHR; +PFN_vkGetPhysicalDeviceWin32PresentationSupportKHR vkGetPhysicalDeviceWin32PresentationSupportKHR; +#endif /* defined(VK_KHR_win32_surface) */ +#if defined(VK_KHR_xcb_surface) +PFN_vkCreateXcbSurfaceKHR vkCreateXcbSurfaceKHR; +PFN_vkGetPhysicalDeviceXcbPresentationSupportKHR vkGetPhysicalDeviceXcbPresentationSupportKHR; +#endif /* defined(VK_KHR_xcb_surface) */ +#if defined(VK_KHR_xlib_surface) +PFN_vkCreateXlibSurfaceKHR vkCreateXlibSurfaceKHR; +PFN_vkGetPhysicalDeviceXlibPresentationSupportKHR vkGetPhysicalDeviceXlibPresentationSupportKHR; +#endif /* defined(VK_KHR_xlib_surface) */ +#if defined(VK_MVK_ios_surface) +PFN_vkCreateIOSSurfaceMVK vkCreateIOSSurfaceMVK; +#endif /* defined(VK_MVK_ios_surface) */ +#if defined(VK_MVK_macos_surface) +PFN_vkCreateMacOSSurfaceMVK vkCreateMacOSSurfaceMVK; +#endif /* defined(VK_MVK_macos_surface) */ +#if defined(VK_NN_vi_surface) +PFN_vkCreateViSurfaceNN vkCreateViSurfaceNN; +#endif /* defined(VK_NN_vi_surface) */ +#if defined(VK_NVX_image_view_handle) +PFN_vkGetImageViewAddressNVX vkGetImageViewAddressNVX; +PFN_vkGetImageViewHandleNVX vkGetImageViewHandleNVX; +#endif /* defined(VK_NVX_image_view_handle) */ +#if defined(VK_NV_clip_space_w_scaling) +PFN_vkCmdSetViewportWScalingNV vkCmdSetViewportWScalingNV; +#endif /* defined(VK_NV_clip_space_w_scaling) */ +#if defined(VK_NV_cooperative_matrix) +PFN_vkGetPhysicalDeviceCooperativeMatrixPropertiesNV vkGetPhysicalDeviceCooperativeMatrixPropertiesNV; +#endif /* defined(VK_NV_cooperative_matrix) */ +#if defined(VK_NV_coverage_reduction_mode) +PFN_vkGetPhysicalDeviceSupportedFramebufferMixedSamplesCombinationsNV vkGetPhysicalDeviceSupportedFramebufferMixedSamplesCombinationsNV; +#endif /* defined(VK_NV_coverage_reduction_mode) */ +#if defined(VK_NV_device_diagnostic_checkpoints) +PFN_vkCmdSetCheckpointNV vkCmdSetCheckpointNV; +PFN_vkGetQueueCheckpointDataNV vkGetQueueCheckpointDataNV; +#endif /* defined(VK_NV_device_diagnostic_checkpoints) */ +#if defined(VK_NV_device_generated_commands) +PFN_vkCmdBindPipelineShaderGroupNV vkCmdBindPipelineShaderGroupNV; +PFN_vkCmdExecuteGeneratedCommandsNV vkCmdExecuteGeneratedCommandsNV; +PFN_vkCmdPreprocessGeneratedCommandsNV vkCmdPreprocessGeneratedCommandsNV; +PFN_vkCreateIndirectCommandsLayoutNV vkCreateIndirectCommandsLayoutNV; +PFN_vkDestroyIndirectCommandsLayoutNV vkDestroyIndirectCommandsLayoutNV; +PFN_vkGetGeneratedCommandsMemoryRequirementsNV vkGetGeneratedCommandsMemoryRequirementsNV; +#endif /* defined(VK_NV_device_generated_commands) */ +#if defined(VK_NV_external_memory_capabilities) +PFN_vkGetPhysicalDeviceExternalImageFormatPropertiesNV vkGetPhysicalDeviceExternalImageFormatPropertiesNV; +#endif /* defined(VK_NV_external_memory_capabilities) */ +#if defined(VK_NV_external_memory_win32) +PFN_vkGetMemoryWin32HandleNV vkGetMemoryWin32HandleNV; +#endif /* defined(VK_NV_external_memory_win32) */ +#if defined(VK_NV_mesh_shader) +PFN_vkCmdDrawMeshTasksIndirectCountNV vkCmdDrawMeshTasksIndirectCountNV; +PFN_vkCmdDrawMeshTasksIndirectNV vkCmdDrawMeshTasksIndirectNV; +PFN_vkCmdDrawMeshTasksNV vkCmdDrawMeshTasksNV; +#endif /* defined(VK_NV_mesh_shader) */ +#if defined(VK_NV_ray_tracing) +PFN_vkBindAccelerationStructureMemoryNV vkBindAccelerationStructureMemoryNV; +PFN_vkCmdBuildAccelerationStructureNV vkCmdBuildAccelerationStructureNV; +PFN_vkCmdCopyAccelerationStructureNV vkCmdCopyAccelerationStructureNV; +PFN_vkCmdTraceRaysNV vkCmdTraceRaysNV; +PFN_vkCmdWriteAccelerationStructuresPropertiesNV vkCmdWriteAccelerationStructuresPropertiesNV; +PFN_vkCompileDeferredNV vkCompileDeferredNV; +PFN_vkCreateAccelerationStructureNV vkCreateAccelerationStructureNV; +PFN_vkCreateRayTracingPipelinesNV vkCreateRayTracingPipelinesNV; +PFN_vkDestroyAccelerationStructureNV vkDestroyAccelerationStructureNV; +PFN_vkGetAccelerationStructureHandleNV vkGetAccelerationStructureHandleNV; +PFN_vkGetAccelerationStructureMemoryRequirementsNV vkGetAccelerationStructureMemoryRequirementsNV; +PFN_vkGetRayTracingShaderGroupHandlesNV vkGetRayTracingShaderGroupHandlesNV; +#endif /* defined(VK_NV_ray_tracing) */ +#if defined(VK_NV_scissor_exclusive) +PFN_vkCmdSetExclusiveScissorNV vkCmdSetExclusiveScissorNV; +#endif /* defined(VK_NV_scissor_exclusive) */ +#if defined(VK_NV_shading_rate_image) +PFN_vkCmdBindShadingRateImageNV vkCmdBindShadingRateImageNV; +PFN_vkCmdSetCoarseSampleOrderNV vkCmdSetCoarseSampleOrderNV; +PFN_vkCmdSetViewportShadingRatePaletteNV vkCmdSetViewportShadingRatePaletteNV; +#endif /* defined(VK_NV_shading_rate_image) */ +#if (defined(VK_EXT_full_screen_exclusive) && defined(VK_KHR_device_group)) || (defined(VK_EXT_full_screen_exclusive) && defined(VK_VERSION_1_1)) +PFN_vkGetDeviceGroupSurfacePresentModes2EXT vkGetDeviceGroupSurfacePresentModes2EXT; +#endif /* (defined(VK_EXT_full_screen_exclusive) && defined(VK_KHR_device_group)) || (defined(VK_EXT_full_screen_exclusive) && defined(VK_VERSION_1_1)) */ +#if (defined(VK_KHR_descriptor_update_template) && defined(VK_KHR_push_descriptor)) || (defined(VK_KHR_push_descriptor) && defined(VK_VERSION_1_1)) || (defined(VK_KHR_push_descriptor) && defined(VK_KHR_descriptor_update_template)) +PFN_vkCmdPushDescriptorSetWithTemplateKHR vkCmdPushDescriptorSetWithTemplateKHR; +#endif /* (defined(VK_KHR_descriptor_update_template) && defined(VK_KHR_push_descriptor)) || (defined(VK_KHR_push_descriptor) && defined(VK_VERSION_1_1)) || (defined(VK_KHR_push_descriptor) && defined(VK_KHR_descriptor_update_template)) */ +#if (defined(VK_KHR_device_group) && defined(VK_KHR_surface)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) +PFN_vkGetDeviceGroupPresentCapabilitiesKHR vkGetDeviceGroupPresentCapabilitiesKHR; +PFN_vkGetDeviceGroupSurfacePresentModesKHR vkGetDeviceGroupSurfacePresentModesKHR; +PFN_vkGetPhysicalDevicePresentRectanglesKHR vkGetPhysicalDevicePresentRectanglesKHR; +#endif /* (defined(VK_KHR_device_group) && defined(VK_KHR_surface)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) */ +#if (defined(VK_KHR_device_group) && defined(VK_KHR_swapchain)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) +PFN_vkAcquireNextImage2KHR vkAcquireNextImage2KHR; +#endif /* (defined(VK_KHR_device_group) && defined(VK_KHR_swapchain)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) */ +/* VOLK_GENERATE_PROTOTYPES_C */ + +#ifdef __GNUC__ +# pragma GCC visibility pop +#endif + +#ifdef __cplusplus +} +#endif +/* clang-format on */ diff --git a/src/vk/volk/volk.h b/src/vk/volk/volk.h new file mode 100644 index 0000000..f05de67 --- /dev/null +++ b/src/vk/volk/volk.h @@ -0,0 +1,1257 @@ +/** + * volk + * + * Copyright (C) 2018-2019, by Arseny Kapoulkine (arseny.kapoulkine@gmail.com) + * Report bugs and download new versions at https://github.com/zeux/volk + * + * This library is distributed under the MIT License. See notice at the end of this file. + */ +/* clang-format off */ +#ifndef VOLK_H_ +#define VOLK_H_ + +#if defined(VULKAN_H_) && !defined(VK_NO_PROTOTYPES) +# error To use volk, you need to define VK_NO_PROTOTYPES before including vulkan.h +#endif + +/* VOLK_GENERATE_VERSION_DEFINE */ +#define VOLK_HEADER_VERSION 153 +/* VOLK_GENERATE_VERSION_DEFINE */ + +#ifndef VK_NO_PROTOTYPES +# define VK_NO_PROTOTYPES +#endif + +#ifndef VULKAN_H_ +# ifdef VOLK_VULKAN_H_PATH +# include VOLK_VULKAN_H_PATH +# elif defined(VK_USE_PLATFORM_WIN32_KHR) +# include +# include + + /* When VK_USE_PLATFORM_WIN32_KHR is defined, instead of including vulkan.h directly, we include individual parts of the SDK + * This is necessary to avoid including which is very heavy - it takes 200ms to parse without WIN32_LEAN_AND_MEAN + * and 100ms to parse with it. vulkan_win32.h only needs a few symbols that are easy to redefine ourselves. + */ + typedef unsigned long DWORD; + typedef const wchar_t* LPCWSTR; + typedef void* HANDLE; + typedef struct HINSTANCE__* HINSTANCE; + typedef struct HWND__* HWND; + typedef struct HMONITOR__* HMONITOR; + typedef struct _SECURITY_ATTRIBUTES SECURITY_ATTRIBUTES; + +# include + +# ifdef VK_ENABLE_BETA_EXTENSIONS +# include +# endif +# else +# include +# endif +#endif + +/* Disable VK_NVX_image_view_handle because SDK 140 introduced a change that can't be used with prior versions */ +#if VK_HEADER_VERSION < 140 +# undef VK_NVX_image_view_handle +#endif + +#ifdef __cplusplus +extern "C" { +#endif + +struct VolkDeviceTable; + +/** + * Initialize library by loading Vulkan loader; call this function before creating the Vulkan instance. + * + * Returns VK_SUCCESS on success and VK_ERROR_INITIALIZATION_FAILED otherwise. + */ +VkResult volkInitialize(void); + +/** + * Initialize library by providing a custom handler to load global symbols. + * + * This function can be used instead of volkInitialize. + * The handler function pointer will be asked to load global Vulkan symbols which require no instance + * (such as vkCreateInstance, vkEnumerateInstance* and vkEnumerateInstanceVersion if available). + */ +void volkInitializeCustom(PFN_vkGetInstanceProcAddr handler); + +/** + * Get Vulkan instance version supported by the Vulkan loader, or 0 if Vulkan isn't supported + * + * Returns 0 if volkInitialize wasn't called or failed. + */ +uint32_t volkGetInstanceVersion(void); + +/** + * Load global function pointers using application-created VkInstance; call this function after creating the Vulkan instance. + */ +void volkLoadInstance(VkInstance instance); + +/** + * Load global function pointers using application-created VkInstance; call this function after creating the Vulkan instance. + * Skips loading device-based function pointers, requires usage of volkLoadDevice afterwards. + */ +void volkLoadInstanceOnly(VkInstance instance); + +/** + * Load global function pointers using application-created VkDevice; call this function after creating the Vulkan device. + * + * Note: this is not suitable for applications that want to use multiple VkDevice objects concurrently. + */ +void volkLoadDevice(VkDevice device); + +/** + * Return last VkInstance for which global function pointers have been loaded via volkLoadInstance(), + * or VK_NULL_HANDLE if volkLoadInstance() has not been called. + */ +VkInstance volkGetLoadedInstance(void); + +/** + * Return last VkDevice for which global function pointers have been loaded via volkLoadDevice(), + * or VK_NULL_HANDLE if volkLoadDevice() has not been called. + */ +VkDevice volkGetLoadedDevice(void); + +/** + * Load function pointers using application-created VkDevice into a table. + * Application should use function pointers from that table instead of using global function pointers. + */ +void volkLoadDeviceTable(struct VolkDeviceTable* table, VkDevice device); + +/** + * Device-specific function pointer table + */ +struct VolkDeviceTable +{ + /* VOLK_GENERATE_DEVICE_TABLE */ +#if defined(VK_VERSION_1_0) + PFN_vkAllocateCommandBuffers vkAllocateCommandBuffers; + PFN_vkAllocateDescriptorSets vkAllocateDescriptorSets; + PFN_vkAllocateMemory vkAllocateMemory; + PFN_vkBeginCommandBuffer vkBeginCommandBuffer; + PFN_vkBindBufferMemory vkBindBufferMemory; + PFN_vkBindImageMemory vkBindImageMemory; + PFN_vkCmdBeginQuery vkCmdBeginQuery; + PFN_vkCmdBeginRenderPass vkCmdBeginRenderPass; + PFN_vkCmdBindDescriptorSets vkCmdBindDescriptorSets; + PFN_vkCmdBindIndexBuffer vkCmdBindIndexBuffer; + PFN_vkCmdBindPipeline vkCmdBindPipeline; + PFN_vkCmdBindVertexBuffers vkCmdBindVertexBuffers; + PFN_vkCmdBlitImage vkCmdBlitImage; + PFN_vkCmdClearAttachments vkCmdClearAttachments; + PFN_vkCmdClearColorImage vkCmdClearColorImage; + PFN_vkCmdClearDepthStencilImage vkCmdClearDepthStencilImage; + PFN_vkCmdCopyBuffer vkCmdCopyBuffer; + PFN_vkCmdCopyBufferToImage vkCmdCopyBufferToImage; + PFN_vkCmdCopyImage vkCmdCopyImage; + PFN_vkCmdCopyImageToBuffer vkCmdCopyImageToBuffer; + PFN_vkCmdCopyQueryPoolResults vkCmdCopyQueryPoolResults; + PFN_vkCmdDispatch vkCmdDispatch; + PFN_vkCmdDispatchIndirect vkCmdDispatchIndirect; + PFN_vkCmdDraw vkCmdDraw; + PFN_vkCmdDrawIndexed vkCmdDrawIndexed; + PFN_vkCmdDrawIndexedIndirect vkCmdDrawIndexedIndirect; + PFN_vkCmdDrawIndirect vkCmdDrawIndirect; + PFN_vkCmdEndQuery vkCmdEndQuery; + PFN_vkCmdEndRenderPass vkCmdEndRenderPass; + PFN_vkCmdExecuteCommands vkCmdExecuteCommands; + PFN_vkCmdFillBuffer vkCmdFillBuffer; + PFN_vkCmdNextSubpass vkCmdNextSubpass; + PFN_vkCmdPipelineBarrier vkCmdPipelineBarrier; + PFN_vkCmdPushConstants vkCmdPushConstants; + PFN_vkCmdResetEvent vkCmdResetEvent; + PFN_vkCmdResetQueryPool vkCmdResetQueryPool; + PFN_vkCmdResolveImage vkCmdResolveImage; + PFN_vkCmdSetBlendConstants vkCmdSetBlendConstants; + PFN_vkCmdSetDepthBias vkCmdSetDepthBias; + PFN_vkCmdSetDepthBounds vkCmdSetDepthBounds; + PFN_vkCmdSetEvent vkCmdSetEvent; + PFN_vkCmdSetLineWidth vkCmdSetLineWidth; + PFN_vkCmdSetScissor vkCmdSetScissor; + PFN_vkCmdSetStencilCompareMask vkCmdSetStencilCompareMask; + PFN_vkCmdSetStencilReference vkCmdSetStencilReference; + PFN_vkCmdSetStencilWriteMask vkCmdSetStencilWriteMask; + PFN_vkCmdSetViewport vkCmdSetViewport; + PFN_vkCmdUpdateBuffer vkCmdUpdateBuffer; + PFN_vkCmdWaitEvents vkCmdWaitEvents; + PFN_vkCmdWriteTimestamp vkCmdWriteTimestamp; + PFN_vkCreateBuffer vkCreateBuffer; + PFN_vkCreateBufferView vkCreateBufferView; + PFN_vkCreateCommandPool vkCreateCommandPool; + PFN_vkCreateComputePipelines vkCreateComputePipelines; + PFN_vkCreateDescriptorPool vkCreateDescriptorPool; + PFN_vkCreateDescriptorSetLayout vkCreateDescriptorSetLayout; + PFN_vkCreateEvent vkCreateEvent; + PFN_vkCreateFence vkCreateFence; + PFN_vkCreateFramebuffer vkCreateFramebuffer; + PFN_vkCreateGraphicsPipelines vkCreateGraphicsPipelines; + PFN_vkCreateImage vkCreateImage; + PFN_vkCreateImageView vkCreateImageView; + PFN_vkCreatePipelineCache vkCreatePipelineCache; + PFN_vkCreatePipelineLayout vkCreatePipelineLayout; + PFN_vkCreateQueryPool vkCreateQueryPool; + PFN_vkCreateRenderPass vkCreateRenderPass; + PFN_vkCreateSampler vkCreateSampler; + PFN_vkCreateSemaphore vkCreateSemaphore; + PFN_vkCreateShaderModule vkCreateShaderModule; + PFN_vkDestroyBuffer vkDestroyBuffer; + PFN_vkDestroyBufferView vkDestroyBufferView; + PFN_vkDestroyCommandPool vkDestroyCommandPool; + PFN_vkDestroyDescriptorPool vkDestroyDescriptorPool; + PFN_vkDestroyDescriptorSetLayout vkDestroyDescriptorSetLayout; + PFN_vkDestroyDevice vkDestroyDevice; + PFN_vkDestroyEvent vkDestroyEvent; + PFN_vkDestroyFence vkDestroyFence; + PFN_vkDestroyFramebuffer vkDestroyFramebuffer; + PFN_vkDestroyImage vkDestroyImage; + PFN_vkDestroyImageView vkDestroyImageView; + PFN_vkDestroyPipeline vkDestroyPipeline; + PFN_vkDestroyPipelineCache vkDestroyPipelineCache; + PFN_vkDestroyPipelineLayout vkDestroyPipelineLayout; + PFN_vkDestroyQueryPool vkDestroyQueryPool; + PFN_vkDestroyRenderPass vkDestroyRenderPass; + PFN_vkDestroySampler vkDestroySampler; + PFN_vkDestroySemaphore vkDestroySemaphore; + PFN_vkDestroyShaderModule vkDestroyShaderModule; + PFN_vkDeviceWaitIdle vkDeviceWaitIdle; + PFN_vkEndCommandBuffer vkEndCommandBuffer; + PFN_vkFlushMappedMemoryRanges vkFlushMappedMemoryRanges; + PFN_vkFreeCommandBuffers vkFreeCommandBuffers; + PFN_vkFreeDescriptorSets vkFreeDescriptorSets; + PFN_vkFreeMemory vkFreeMemory; + PFN_vkGetBufferMemoryRequirements vkGetBufferMemoryRequirements; + PFN_vkGetDeviceMemoryCommitment vkGetDeviceMemoryCommitment; + PFN_vkGetDeviceQueue vkGetDeviceQueue; + PFN_vkGetEventStatus vkGetEventStatus; + PFN_vkGetFenceStatus vkGetFenceStatus; + PFN_vkGetImageMemoryRequirements vkGetImageMemoryRequirements; + PFN_vkGetImageSparseMemoryRequirements vkGetImageSparseMemoryRequirements; + PFN_vkGetImageSubresourceLayout vkGetImageSubresourceLayout; + PFN_vkGetPipelineCacheData vkGetPipelineCacheData; + PFN_vkGetQueryPoolResults vkGetQueryPoolResults; + PFN_vkGetRenderAreaGranularity vkGetRenderAreaGranularity; + PFN_vkInvalidateMappedMemoryRanges vkInvalidateMappedMemoryRanges; + PFN_vkMapMemory vkMapMemory; + PFN_vkMergePipelineCaches vkMergePipelineCaches; + PFN_vkQueueBindSparse vkQueueBindSparse; + PFN_vkQueueSubmit vkQueueSubmit; + PFN_vkQueueWaitIdle vkQueueWaitIdle; + PFN_vkResetCommandBuffer vkResetCommandBuffer; + PFN_vkResetCommandPool vkResetCommandPool; + PFN_vkResetDescriptorPool vkResetDescriptorPool; + PFN_vkResetEvent vkResetEvent; + PFN_vkResetFences vkResetFences; + PFN_vkSetEvent vkSetEvent; + PFN_vkUnmapMemory vkUnmapMemory; + PFN_vkUpdateDescriptorSets vkUpdateDescriptorSets; + PFN_vkWaitForFences vkWaitForFences; +#endif /* defined(VK_VERSION_1_0) */ +#if defined(VK_VERSION_1_1) + PFN_vkBindBufferMemory2 vkBindBufferMemory2; + PFN_vkBindImageMemory2 vkBindImageMemory2; + PFN_vkCmdDispatchBase vkCmdDispatchBase; + PFN_vkCmdSetDeviceMask vkCmdSetDeviceMask; + PFN_vkCreateDescriptorUpdateTemplate vkCreateDescriptorUpdateTemplate; + PFN_vkCreateSamplerYcbcrConversion vkCreateSamplerYcbcrConversion; + PFN_vkDestroyDescriptorUpdateTemplate vkDestroyDescriptorUpdateTemplate; + PFN_vkDestroySamplerYcbcrConversion vkDestroySamplerYcbcrConversion; + PFN_vkGetBufferMemoryRequirements2 vkGetBufferMemoryRequirements2; + PFN_vkGetDescriptorSetLayoutSupport vkGetDescriptorSetLayoutSupport; + PFN_vkGetDeviceGroupPeerMemoryFeatures vkGetDeviceGroupPeerMemoryFeatures; + PFN_vkGetDeviceQueue2 vkGetDeviceQueue2; + PFN_vkGetImageMemoryRequirements2 vkGetImageMemoryRequirements2; + PFN_vkGetImageSparseMemoryRequirements2 vkGetImageSparseMemoryRequirements2; + PFN_vkTrimCommandPool vkTrimCommandPool; + PFN_vkUpdateDescriptorSetWithTemplate vkUpdateDescriptorSetWithTemplate; +#endif /* defined(VK_VERSION_1_1) */ +#if defined(VK_VERSION_1_2) + PFN_vkCmdBeginRenderPass2 vkCmdBeginRenderPass2; + PFN_vkCmdDrawIndexedIndirectCount vkCmdDrawIndexedIndirectCount; + PFN_vkCmdDrawIndirectCount vkCmdDrawIndirectCount; + PFN_vkCmdEndRenderPass2 vkCmdEndRenderPass2; + PFN_vkCmdNextSubpass2 vkCmdNextSubpass2; + PFN_vkCreateRenderPass2 vkCreateRenderPass2; + PFN_vkGetBufferDeviceAddress vkGetBufferDeviceAddress; + PFN_vkGetBufferOpaqueCaptureAddress vkGetBufferOpaqueCaptureAddress; + PFN_vkGetDeviceMemoryOpaqueCaptureAddress vkGetDeviceMemoryOpaqueCaptureAddress; + PFN_vkGetSemaphoreCounterValue vkGetSemaphoreCounterValue; + PFN_vkResetQueryPool vkResetQueryPool; + PFN_vkSignalSemaphore vkSignalSemaphore; + PFN_vkWaitSemaphores vkWaitSemaphores; +#endif /* defined(VK_VERSION_1_2) */ +#if defined(VK_AMD_buffer_marker) + PFN_vkCmdWriteBufferMarkerAMD vkCmdWriteBufferMarkerAMD; +#endif /* defined(VK_AMD_buffer_marker) */ +#if defined(VK_AMD_display_native_hdr) + PFN_vkSetLocalDimmingAMD vkSetLocalDimmingAMD; +#endif /* defined(VK_AMD_display_native_hdr) */ +#if defined(VK_AMD_draw_indirect_count) + PFN_vkCmdDrawIndexedIndirectCountAMD vkCmdDrawIndexedIndirectCountAMD; + PFN_vkCmdDrawIndirectCountAMD vkCmdDrawIndirectCountAMD; +#endif /* defined(VK_AMD_draw_indirect_count) */ +#if defined(VK_AMD_shader_info) + PFN_vkGetShaderInfoAMD vkGetShaderInfoAMD; +#endif /* defined(VK_AMD_shader_info) */ +#if defined(VK_ANDROID_external_memory_android_hardware_buffer) + PFN_vkGetAndroidHardwareBufferPropertiesANDROID vkGetAndroidHardwareBufferPropertiesANDROID; + PFN_vkGetMemoryAndroidHardwareBufferANDROID vkGetMemoryAndroidHardwareBufferANDROID; +#endif /* defined(VK_ANDROID_external_memory_android_hardware_buffer) */ +#if defined(VK_EXT_buffer_device_address) + PFN_vkGetBufferDeviceAddressEXT vkGetBufferDeviceAddressEXT; +#endif /* defined(VK_EXT_buffer_device_address) */ +#if defined(VK_EXT_calibrated_timestamps) + PFN_vkGetCalibratedTimestampsEXT vkGetCalibratedTimestampsEXT; +#endif /* defined(VK_EXT_calibrated_timestamps) */ +#if defined(VK_EXT_conditional_rendering) + PFN_vkCmdBeginConditionalRenderingEXT vkCmdBeginConditionalRenderingEXT; + PFN_vkCmdEndConditionalRenderingEXT vkCmdEndConditionalRenderingEXT; +#endif /* defined(VK_EXT_conditional_rendering) */ +#if defined(VK_EXT_debug_marker) + PFN_vkCmdDebugMarkerBeginEXT vkCmdDebugMarkerBeginEXT; + PFN_vkCmdDebugMarkerEndEXT vkCmdDebugMarkerEndEXT; + PFN_vkCmdDebugMarkerInsertEXT vkCmdDebugMarkerInsertEXT; + PFN_vkDebugMarkerSetObjectNameEXT vkDebugMarkerSetObjectNameEXT; + PFN_vkDebugMarkerSetObjectTagEXT vkDebugMarkerSetObjectTagEXT; +#endif /* defined(VK_EXT_debug_marker) */ +#if defined(VK_EXT_discard_rectangles) + PFN_vkCmdSetDiscardRectangleEXT vkCmdSetDiscardRectangleEXT; +#endif /* defined(VK_EXT_discard_rectangles) */ +#if defined(VK_EXT_display_control) + PFN_vkDisplayPowerControlEXT vkDisplayPowerControlEXT; + PFN_vkGetSwapchainCounterEXT vkGetSwapchainCounterEXT; + PFN_vkRegisterDeviceEventEXT vkRegisterDeviceEventEXT; + PFN_vkRegisterDisplayEventEXT vkRegisterDisplayEventEXT; +#endif /* defined(VK_EXT_display_control) */ +#if defined(VK_EXT_extended_dynamic_state) + PFN_vkCmdBindVertexBuffers2EXT vkCmdBindVertexBuffers2EXT; + PFN_vkCmdSetCullModeEXT vkCmdSetCullModeEXT; + PFN_vkCmdSetDepthBoundsTestEnableEXT vkCmdSetDepthBoundsTestEnableEXT; + PFN_vkCmdSetDepthCompareOpEXT vkCmdSetDepthCompareOpEXT; + PFN_vkCmdSetDepthTestEnableEXT vkCmdSetDepthTestEnableEXT; + PFN_vkCmdSetDepthWriteEnableEXT vkCmdSetDepthWriteEnableEXT; + PFN_vkCmdSetFrontFaceEXT vkCmdSetFrontFaceEXT; + PFN_vkCmdSetPrimitiveTopologyEXT vkCmdSetPrimitiveTopologyEXT; + PFN_vkCmdSetScissorWithCountEXT vkCmdSetScissorWithCountEXT; + PFN_vkCmdSetStencilOpEXT vkCmdSetStencilOpEXT; + PFN_vkCmdSetStencilTestEnableEXT vkCmdSetStencilTestEnableEXT; + PFN_vkCmdSetViewportWithCountEXT vkCmdSetViewportWithCountEXT; +#endif /* defined(VK_EXT_extended_dynamic_state) */ +#if defined(VK_EXT_external_memory_host) + PFN_vkGetMemoryHostPointerPropertiesEXT vkGetMemoryHostPointerPropertiesEXT; +#endif /* defined(VK_EXT_external_memory_host) */ +#if defined(VK_EXT_full_screen_exclusive) + PFN_vkAcquireFullScreenExclusiveModeEXT vkAcquireFullScreenExclusiveModeEXT; + PFN_vkReleaseFullScreenExclusiveModeEXT vkReleaseFullScreenExclusiveModeEXT; +#endif /* defined(VK_EXT_full_screen_exclusive) */ +#if defined(VK_EXT_hdr_metadata) + PFN_vkSetHdrMetadataEXT vkSetHdrMetadataEXT; +#endif /* defined(VK_EXT_hdr_metadata) */ +#if defined(VK_EXT_host_query_reset) + PFN_vkResetQueryPoolEXT vkResetQueryPoolEXT; +#endif /* defined(VK_EXT_host_query_reset) */ +#if defined(VK_EXT_image_drm_format_modifier) + PFN_vkGetImageDrmFormatModifierPropertiesEXT vkGetImageDrmFormatModifierPropertiesEXT; +#endif /* defined(VK_EXT_image_drm_format_modifier) */ +#if defined(VK_EXT_line_rasterization) + PFN_vkCmdSetLineStippleEXT vkCmdSetLineStippleEXT; +#endif /* defined(VK_EXT_line_rasterization) */ +#if defined(VK_EXT_private_data) + PFN_vkCreatePrivateDataSlotEXT vkCreatePrivateDataSlotEXT; + PFN_vkDestroyPrivateDataSlotEXT vkDestroyPrivateDataSlotEXT; + PFN_vkGetPrivateDataEXT vkGetPrivateDataEXT; + PFN_vkSetPrivateDataEXT vkSetPrivateDataEXT; +#endif /* defined(VK_EXT_private_data) */ +#if defined(VK_EXT_sample_locations) + PFN_vkCmdSetSampleLocationsEXT vkCmdSetSampleLocationsEXT; +#endif /* defined(VK_EXT_sample_locations) */ +#if defined(VK_EXT_transform_feedback) + PFN_vkCmdBeginQueryIndexedEXT vkCmdBeginQueryIndexedEXT; + PFN_vkCmdBeginTransformFeedbackEXT vkCmdBeginTransformFeedbackEXT; + PFN_vkCmdBindTransformFeedbackBuffersEXT vkCmdBindTransformFeedbackBuffersEXT; + PFN_vkCmdDrawIndirectByteCountEXT vkCmdDrawIndirectByteCountEXT; + PFN_vkCmdEndQueryIndexedEXT vkCmdEndQueryIndexedEXT; + PFN_vkCmdEndTransformFeedbackEXT vkCmdEndTransformFeedbackEXT; +#endif /* defined(VK_EXT_transform_feedback) */ +#if defined(VK_EXT_validation_cache) + PFN_vkCreateValidationCacheEXT vkCreateValidationCacheEXT; + PFN_vkDestroyValidationCacheEXT vkDestroyValidationCacheEXT; + PFN_vkGetValidationCacheDataEXT vkGetValidationCacheDataEXT; + PFN_vkMergeValidationCachesEXT vkMergeValidationCachesEXT; +#endif /* defined(VK_EXT_validation_cache) */ +#if defined(VK_GOOGLE_display_timing) + PFN_vkGetPastPresentationTimingGOOGLE vkGetPastPresentationTimingGOOGLE; + PFN_vkGetRefreshCycleDurationGOOGLE vkGetRefreshCycleDurationGOOGLE; +#endif /* defined(VK_GOOGLE_display_timing) */ +#if defined(VK_INTEL_performance_query) + PFN_vkAcquirePerformanceConfigurationINTEL vkAcquirePerformanceConfigurationINTEL; + PFN_vkCmdSetPerformanceMarkerINTEL vkCmdSetPerformanceMarkerINTEL; + PFN_vkCmdSetPerformanceOverrideINTEL vkCmdSetPerformanceOverrideINTEL; + PFN_vkCmdSetPerformanceStreamMarkerINTEL vkCmdSetPerformanceStreamMarkerINTEL; + PFN_vkGetPerformanceParameterINTEL vkGetPerformanceParameterINTEL; + PFN_vkInitializePerformanceApiINTEL vkInitializePerformanceApiINTEL; + PFN_vkQueueSetPerformanceConfigurationINTEL vkQueueSetPerformanceConfigurationINTEL; + PFN_vkReleasePerformanceConfigurationINTEL vkReleasePerformanceConfigurationINTEL; + PFN_vkUninitializePerformanceApiINTEL vkUninitializePerformanceApiINTEL; +#endif /* defined(VK_INTEL_performance_query) */ +#if defined(VK_KHR_bind_memory2) + PFN_vkBindBufferMemory2KHR vkBindBufferMemory2KHR; + PFN_vkBindImageMemory2KHR vkBindImageMemory2KHR; +#endif /* defined(VK_KHR_bind_memory2) */ +#if defined(VK_KHR_buffer_device_address) + PFN_vkGetBufferDeviceAddressKHR vkGetBufferDeviceAddressKHR; + PFN_vkGetBufferOpaqueCaptureAddressKHR vkGetBufferOpaqueCaptureAddressKHR; + PFN_vkGetDeviceMemoryOpaqueCaptureAddressKHR vkGetDeviceMemoryOpaqueCaptureAddressKHR; +#endif /* defined(VK_KHR_buffer_device_address) */ +#if defined(VK_KHR_create_renderpass2) + PFN_vkCmdBeginRenderPass2KHR vkCmdBeginRenderPass2KHR; + PFN_vkCmdEndRenderPass2KHR vkCmdEndRenderPass2KHR; + PFN_vkCmdNextSubpass2KHR vkCmdNextSubpass2KHR; + PFN_vkCreateRenderPass2KHR vkCreateRenderPass2KHR; +#endif /* defined(VK_KHR_create_renderpass2) */ +#if defined(VK_KHR_deferred_host_operations) + PFN_vkCreateDeferredOperationKHR vkCreateDeferredOperationKHR; + PFN_vkDeferredOperationJoinKHR vkDeferredOperationJoinKHR; + PFN_vkDestroyDeferredOperationKHR vkDestroyDeferredOperationKHR; + PFN_vkGetDeferredOperationMaxConcurrencyKHR vkGetDeferredOperationMaxConcurrencyKHR; + PFN_vkGetDeferredOperationResultKHR vkGetDeferredOperationResultKHR; +#endif /* defined(VK_KHR_deferred_host_operations) */ +#if defined(VK_KHR_descriptor_update_template) + PFN_vkCreateDescriptorUpdateTemplateKHR vkCreateDescriptorUpdateTemplateKHR; + PFN_vkDestroyDescriptorUpdateTemplateKHR vkDestroyDescriptorUpdateTemplateKHR; + PFN_vkUpdateDescriptorSetWithTemplateKHR vkUpdateDescriptorSetWithTemplateKHR; +#endif /* defined(VK_KHR_descriptor_update_template) */ +#if defined(VK_KHR_device_group) + PFN_vkCmdDispatchBaseKHR vkCmdDispatchBaseKHR; + PFN_vkCmdSetDeviceMaskKHR vkCmdSetDeviceMaskKHR; + PFN_vkGetDeviceGroupPeerMemoryFeaturesKHR vkGetDeviceGroupPeerMemoryFeaturesKHR; +#endif /* defined(VK_KHR_device_group) */ +#if defined(VK_KHR_display_swapchain) + PFN_vkCreateSharedSwapchainsKHR vkCreateSharedSwapchainsKHR; +#endif /* defined(VK_KHR_display_swapchain) */ +#if defined(VK_KHR_draw_indirect_count) + PFN_vkCmdDrawIndexedIndirectCountKHR vkCmdDrawIndexedIndirectCountKHR; + PFN_vkCmdDrawIndirectCountKHR vkCmdDrawIndirectCountKHR; +#endif /* defined(VK_KHR_draw_indirect_count) */ +#if defined(VK_KHR_external_fence_fd) + PFN_vkGetFenceFdKHR vkGetFenceFdKHR; + PFN_vkImportFenceFdKHR vkImportFenceFdKHR; +#endif /* defined(VK_KHR_external_fence_fd) */ +#if defined(VK_KHR_external_fence_win32) + PFN_vkGetFenceWin32HandleKHR vkGetFenceWin32HandleKHR; + PFN_vkImportFenceWin32HandleKHR vkImportFenceWin32HandleKHR; +#endif /* defined(VK_KHR_external_fence_win32) */ +#if defined(VK_KHR_external_memory_fd) + PFN_vkGetMemoryFdKHR vkGetMemoryFdKHR; + PFN_vkGetMemoryFdPropertiesKHR vkGetMemoryFdPropertiesKHR; +#endif /* defined(VK_KHR_external_memory_fd) */ +#if defined(VK_KHR_external_memory_win32) + PFN_vkGetMemoryWin32HandleKHR vkGetMemoryWin32HandleKHR; + PFN_vkGetMemoryWin32HandlePropertiesKHR vkGetMemoryWin32HandlePropertiesKHR; +#endif /* defined(VK_KHR_external_memory_win32) */ +#if defined(VK_KHR_external_semaphore_fd) + PFN_vkGetSemaphoreFdKHR vkGetSemaphoreFdKHR; + PFN_vkImportSemaphoreFdKHR vkImportSemaphoreFdKHR; +#endif /* defined(VK_KHR_external_semaphore_fd) */ +#if defined(VK_KHR_external_semaphore_win32) + PFN_vkGetSemaphoreWin32HandleKHR vkGetSemaphoreWin32HandleKHR; + PFN_vkImportSemaphoreWin32HandleKHR vkImportSemaphoreWin32HandleKHR; +#endif /* defined(VK_KHR_external_semaphore_win32) */ +#if defined(VK_KHR_get_memory_requirements2) + PFN_vkGetBufferMemoryRequirements2KHR vkGetBufferMemoryRequirements2KHR; + PFN_vkGetImageMemoryRequirements2KHR vkGetImageMemoryRequirements2KHR; + PFN_vkGetImageSparseMemoryRequirements2KHR vkGetImageSparseMemoryRequirements2KHR; +#endif /* defined(VK_KHR_get_memory_requirements2) */ +#if defined(VK_KHR_maintenance1) + PFN_vkTrimCommandPoolKHR vkTrimCommandPoolKHR; +#endif /* defined(VK_KHR_maintenance1) */ +#if defined(VK_KHR_maintenance3) + PFN_vkGetDescriptorSetLayoutSupportKHR vkGetDescriptorSetLayoutSupportKHR; +#endif /* defined(VK_KHR_maintenance3) */ +#if defined(VK_KHR_performance_query) + PFN_vkAcquireProfilingLockKHR vkAcquireProfilingLockKHR; + PFN_vkReleaseProfilingLockKHR vkReleaseProfilingLockKHR; +#endif /* defined(VK_KHR_performance_query) */ +#if defined(VK_KHR_pipeline_executable_properties) + PFN_vkGetPipelineExecutableInternalRepresentationsKHR vkGetPipelineExecutableInternalRepresentationsKHR; + PFN_vkGetPipelineExecutablePropertiesKHR vkGetPipelineExecutablePropertiesKHR; + PFN_vkGetPipelineExecutableStatisticsKHR vkGetPipelineExecutableStatisticsKHR; +#endif /* defined(VK_KHR_pipeline_executable_properties) */ +#if defined(VK_KHR_push_descriptor) + PFN_vkCmdPushDescriptorSetKHR vkCmdPushDescriptorSetKHR; +#endif /* defined(VK_KHR_push_descriptor) */ +#if defined(VK_KHR_ray_tracing) + PFN_vkBindAccelerationStructureMemoryKHR vkBindAccelerationStructureMemoryKHR; + PFN_vkBuildAccelerationStructureKHR vkBuildAccelerationStructureKHR; + PFN_vkCmdBuildAccelerationStructureIndirectKHR vkCmdBuildAccelerationStructureIndirectKHR; + PFN_vkCmdBuildAccelerationStructureKHR vkCmdBuildAccelerationStructureKHR; + PFN_vkCmdCopyAccelerationStructureKHR vkCmdCopyAccelerationStructureKHR; + PFN_vkCmdCopyAccelerationStructureToMemoryKHR vkCmdCopyAccelerationStructureToMemoryKHR; + PFN_vkCmdCopyMemoryToAccelerationStructureKHR vkCmdCopyMemoryToAccelerationStructureKHR; + PFN_vkCmdTraceRaysIndirectKHR vkCmdTraceRaysIndirectKHR; + PFN_vkCmdTraceRaysKHR vkCmdTraceRaysKHR; + PFN_vkCmdWriteAccelerationStructuresPropertiesKHR vkCmdWriteAccelerationStructuresPropertiesKHR; + PFN_vkCopyAccelerationStructureKHR vkCopyAccelerationStructureKHR; + PFN_vkCopyAccelerationStructureToMemoryKHR vkCopyAccelerationStructureToMemoryKHR; + PFN_vkCopyMemoryToAccelerationStructureKHR vkCopyMemoryToAccelerationStructureKHR; + PFN_vkCreateAccelerationStructureKHR vkCreateAccelerationStructureKHR; + PFN_vkCreateRayTracingPipelinesKHR vkCreateRayTracingPipelinesKHR; + PFN_vkDestroyAccelerationStructureKHR vkDestroyAccelerationStructureKHR; + PFN_vkGetAccelerationStructureDeviceAddressKHR vkGetAccelerationStructureDeviceAddressKHR; + PFN_vkGetAccelerationStructureMemoryRequirementsKHR vkGetAccelerationStructureMemoryRequirementsKHR; + PFN_vkGetDeviceAccelerationStructureCompatibilityKHR vkGetDeviceAccelerationStructureCompatibilityKHR; + PFN_vkGetRayTracingCaptureReplayShaderGroupHandlesKHR vkGetRayTracingCaptureReplayShaderGroupHandlesKHR; + PFN_vkGetRayTracingShaderGroupHandlesKHR vkGetRayTracingShaderGroupHandlesKHR; + PFN_vkWriteAccelerationStructuresPropertiesKHR vkWriteAccelerationStructuresPropertiesKHR; +#endif /* defined(VK_KHR_ray_tracing) */ +#if defined(VK_KHR_sampler_ycbcr_conversion) + PFN_vkCreateSamplerYcbcrConversionKHR vkCreateSamplerYcbcrConversionKHR; + PFN_vkDestroySamplerYcbcrConversionKHR vkDestroySamplerYcbcrConversionKHR; +#endif /* defined(VK_KHR_sampler_ycbcr_conversion) */ +#if defined(VK_KHR_shared_presentable_image) + PFN_vkGetSwapchainStatusKHR vkGetSwapchainStatusKHR; +#endif /* defined(VK_KHR_shared_presentable_image) */ +#if defined(VK_KHR_swapchain) + PFN_vkAcquireNextImageKHR vkAcquireNextImageKHR; + PFN_vkCreateSwapchainKHR vkCreateSwapchainKHR; + PFN_vkDestroySwapchainKHR vkDestroySwapchainKHR; + PFN_vkGetSwapchainImagesKHR vkGetSwapchainImagesKHR; + PFN_vkQueuePresentKHR vkQueuePresentKHR; +#endif /* defined(VK_KHR_swapchain) */ +#if defined(VK_KHR_timeline_semaphore) + PFN_vkGetSemaphoreCounterValueKHR vkGetSemaphoreCounterValueKHR; + PFN_vkSignalSemaphoreKHR vkSignalSemaphoreKHR; + PFN_vkWaitSemaphoresKHR vkWaitSemaphoresKHR; +#endif /* defined(VK_KHR_timeline_semaphore) */ +#if defined(VK_NVX_image_view_handle) + PFN_vkGetImageViewAddressNVX vkGetImageViewAddressNVX; + PFN_vkGetImageViewHandleNVX vkGetImageViewHandleNVX; +#endif /* defined(VK_NVX_image_view_handle) */ +#if defined(VK_NV_clip_space_w_scaling) + PFN_vkCmdSetViewportWScalingNV vkCmdSetViewportWScalingNV; +#endif /* defined(VK_NV_clip_space_w_scaling) */ +#if defined(VK_NV_device_diagnostic_checkpoints) + PFN_vkCmdSetCheckpointNV vkCmdSetCheckpointNV; + PFN_vkGetQueueCheckpointDataNV vkGetQueueCheckpointDataNV; +#endif /* defined(VK_NV_device_diagnostic_checkpoints) */ +#if defined(VK_NV_device_generated_commands) + PFN_vkCmdBindPipelineShaderGroupNV vkCmdBindPipelineShaderGroupNV; + PFN_vkCmdExecuteGeneratedCommandsNV vkCmdExecuteGeneratedCommandsNV; + PFN_vkCmdPreprocessGeneratedCommandsNV vkCmdPreprocessGeneratedCommandsNV; + PFN_vkCreateIndirectCommandsLayoutNV vkCreateIndirectCommandsLayoutNV; + PFN_vkDestroyIndirectCommandsLayoutNV vkDestroyIndirectCommandsLayoutNV; + PFN_vkGetGeneratedCommandsMemoryRequirementsNV vkGetGeneratedCommandsMemoryRequirementsNV; +#endif /* defined(VK_NV_device_generated_commands) */ +#if defined(VK_NV_external_memory_win32) + PFN_vkGetMemoryWin32HandleNV vkGetMemoryWin32HandleNV; +#endif /* defined(VK_NV_external_memory_win32) */ +#if defined(VK_NV_mesh_shader) + PFN_vkCmdDrawMeshTasksIndirectCountNV vkCmdDrawMeshTasksIndirectCountNV; + PFN_vkCmdDrawMeshTasksIndirectNV vkCmdDrawMeshTasksIndirectNV; + PFN_vkCmdDrawMeshTasksNV vkCmdDrawMeshTasksNV; +#endif /* defined(VK_NV_mesh_shader) */ +#if defined(VK_NV_ray_tracing) + PFN_vkBindAccelerationStructureMemoryNV vkBindAccelerationStructureMemoryNV; + PFN_vkCmdBuildAccelerationStructureNV vkCmdBuildAccelerationStructureNV; + PFN_vkCmdCopyAccelerationStructureNV vkCmdCopyAccelerationStructureNV; + PFN_vkCmdTraceRaysNV vkCmdTraceRaysNV; + PFN_vkCmdWriteAccelerationStructuresPropertiesNV vkCmdWriteAccelerationStructuresPropertiesNV; + PFN_vkCompileDeferredNV vkCompileDeferredNV; + PFN_vkCreateAccelerationStructureNV vkCreateAccelerationStructureNV; + PFN_vkCreateRayTracingPipelinesNV vkCreateRayTracingPipelinesNV; + PFN_vkDestroyAccelerationStructureNV vkDestroyAccelerationStructureNV; + PFN_vkGetAccelerationStructureHandleNV vkGetAccelerationStructureHandleNV; + PFN_vkGetAccelerationStructureMemoryRequirementsNV vkGetAccelerationStructureMemoryRequirementsNV; + PFN_vkGetRayTracingShaderGroupHandlesNV vkGetRayTracingShaderGroupHandlesNV; +#endif /* defined(VK_NV_ray_tracing) */ +#if defined(VK_NV_scissor_exclusive) + PFN_vkCmdSetExclusiveScissorNV vkCmdSetExclusiveScissorNV; +#endif /* defined(VK_NV_scissor_exclusive) */ +#if defined(VK_NV_shading_rate_image) + PFN_vkCmdBindShadingRateImageNV vkCmdBindShadingRateImageNV; + PFN_vkCmdSetCoarseSampleOrderNV vkCmdSetCoarseSampleOrderNV; + PFN_vkCmdSetViewportShadingRatePaletteNV vkCmdSetViewportShadingRatePaletteNV; +#endif /* defined(VK_NV_shading_rate_image) */ +#if (defined(VK_EXT_full_screen_exclusive) && defined(VK_KHR_device_group)) || (defined(VK_EXT_full_screen_exclusive) && defined(VK_VERSION_1_1)) + PFN_vkGetDeviceGroupSurfacePresentModes2EXT vkGetDeviceGroupSurfacePresentModes2EXT; +#endif /* (defined(VK_EXT_full_screen_exclusive) && defined(VK_KHR_device_group)) || (defined(VK_EXT_full_screen_exclusive) && defined(VK_VERSION_1_1)) */ +#if (defined(VK_KHR_descriptor_update_template) && defined(VK_KHR_push_descriptor)) || (defined(VK_KHR_push_descriptor) && defined(VK_VERSION_1_1)) || (defined(VK_KHR_push_descriptor) && defined(VK_KHR_descriptor_update_template)) + PFN_vkCmdPushDescriptorSetWithTemplateKHR vkCmdPushDescriptorSetWithTemplateKHR; +#endif /* (defined(VK_KHR_descriptor_update_template) && defined(VK_KHR_push_descriptor)) || (defined(VK_KHR_push_descriptor) && defined(VK_VERSION_1_1)) || (defined(VK_KHR_push_descriptor) && defined(VK_KHR_descriptor_update_template)) */ +#if (defined(VK_KHR_device_group) && defined(VK_KHR_surface)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) + PFN_vkGetDeviceGroupPresentCapabilitiesKHR vkGetDeviceGroupPresentCapabilitiesKHR; + PFN_vkGetDeviceGroupSurfacePresentModesKHR vkGetDeviceGroupSurfacePresentModesKHR; +#endif /* (defined(VK_KHR_device_group) && defined(VK_KHR_surface)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) */ +#if (defined(VK_KHR_device_group) && defined(VK_KHR_swapchain)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) + PFN_vkAcquireNextImage2KHR vkAcquireNextImage2KHR; +#endif /* (defined(VK_KHR_device_group) && defined(VK_KHR_swapchain)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) */ + /* VOLK_GENERATE_DEVICE_TABLE */ +}; + +/* VOLK_GENERATE_PROTOTYPES_H */ +#if defined(VK_VERSION_1_0) +extern PFN_vkAllocateCommandBuffers vkAllocateCommandBuffers; +extern PFN_vkAllocateDescriptorSets vkAllocateDescriptorSets; +extern PFN_vkAllocateMemory vkAllocateMemory; +extern PFN_vkBeginCommandBuffer vkBeginCommandBuffer; +extern PFN_vkBindBufferMemory vkBindBufferMemory; +extern PFN_vkBindImageMemory vkBindImageMemory; +extern PFN_vkCmdBeginQuery vkCmdBeginQuery; +extern PFN_vkCmdBeginRenderPass vkCmdBeginRenderPass; +extern PFN_vkCmdBindDescriptorSets vkCmdBindDescriptorSets; +extern PFN_vkCmdBindIndexBuffer vkCmdBindIndexBuffer; +extern PFN_vkCmdBindPipeline vkCmdBindPipeline; +extern PFN_vkCmdBindVertexBuffers vkCmdBindVertexBuffers; +extern PFN_vkCmdBlitImage vkCmdBlitImage; +extern PFN_vkCmdClearAttachments vkCmdClearAttachments; +extern PFN_vkCmdClearColorImage vkCmdClearColorImage; +extern PFN_vkCmdClearDepthStencilImage vkCmdClearDepthStencilImage; +extern PFN_vkCmdCopyBuffer vkCmdCopyBuffer; +extern PFN_vkCmdCopyBufferToImage vkCmdCopyBufferToImage; +extern PFN_vkCmdCopyImage vkCmdCopyImage; +extern PFN_vkCmdCopyImageToBuffer vkCmdCopyImageToBuffer; +extern PFN_vkCmdCopyQueryPoolResults vkCmdCopyQueryPoolResults; +extern PFN_vkCmdDispatch vkCmdDispatch; +extern PFN_vkCmdDispatchIndirect vkCmdDispatchIndirect; +extern PFN_vkCmdDraw vkCmdDraw; +extern PFN_vkCmdDrawIndexed vkCmdDrawIndexed; +extern PFN_vkCmdDrawIndexedIndirect vkCmdDrawIndexedIndirect; +extern PFN_vkCmdDrawIndirect vkCmdDrawIndirect; +extern PFN_vkCmdEndQuery vkCmdEndQuery; +extern PFN_vkCmdEndRenderPass vkCmdEndRenderPass; +extern PFN_vkCmdExecuteCommands vkCmdExecuteCommands; +extern PFN_vkCmdFillBuffer vkCmdFillBuffer; +extern PFN_vkCmdNextSubpass vkCmdNextSubpass; +extern PFN_vkCmdPipelineBarrier vkCmdPipelineBarrier; +extern PFN_vkCmdPushConstants vkCmdPushConstants; +extern PFN_vkCmdResetEvent vkCmdResetEvent; +extern PFN_vkCmdResetQueryPool vkCmdResetQueryPool; +extern PFN_vkCmdResolveImage vkCmdResolveImage; +extern PFN_vkCmdSetBlendConstants vkCmdSetBlendConstants; +extern PFN_vkCmdSetDepthBias vkCmdSetDepthBias; +extern PFN_vkCmdSetDepthBounds vkCmdSetDepthBounds; +extern PFN_vkCmdSetEvent vkCmdSetEvent; +extern PFN_vkCmdSetLineWidth vkCmdSetLineWidth; +extern PFN_vkCmdSetScissor vkCmdSetScissor; +extern PFN_vkCmdSetStencilCompareMask vkCmdSetStencilCompareMask; +extern PFN_vkCmdSetStencilReference vkCmdSetStencilReference; +extern PFN_vkCmdSetStencilWriteMask vkCmdSetStencilWriteMask; +extern PFN_vkCmdSetViewport vkCmdSetViewport; +extern PFN_vkCmdUpdateBuffer vkCmdUpdateBuffer; +extern PFN_vkCmdWaitEvents vkCmdWaitEvents; +extern PFN_vkCmdWriteTimestamp vkCmdWriteTimestamp; +extern PFN_vkCreateBuffer vkCreateBuffer; +extern PFN_vkCreateBufferView vkCreateBufferView; +extern PFN_vkCreateCommandPool vkCreateCommandPool; +extern PFN_vkCreateComputePipelines vkCreateComputePipelines; +extern PFN_vkCreateDescriptorPool vkCreateDescriptorPool; +extern PFN_vkCreateDescriptorSetLayout vkCreateDescriptorSetLayout; +extern PFN_vkCreateDevice vkCreateDevice; +extern PFN_vkCreateEvent vkCreateEvent; +extern PFN_vkCreateFence vkCreateFence; +extern PFN_vkCreateFramebuffer vkCreateFramebuffer; +extern PFN_vkCreateGraphicsPipelines vkCreateGraphicsPipelines; +extern PFN_vkCreateImage vkCreateImage; +extern PFN_vkCreateImageView vkCreateImageView; +extern PFN_vkCreateInstance vkCreateInstance; +extern PFN_vkCreatePipelineCache vkCreatePipelineCache; +extern PFN_vkCreatePipelineLayout vkCreatePipelineLayout; +extern PFN_vkCreateQueryPool vkCreateQueryPool; +extern PFN_vkCreateRenderPass vkCreateRenderPass; +extern PFN_vkCreateSampler vkCreateSampler; +extern PFN_vkCreateSemaphore vkCreateSemaphore; +extern PFN_vkCreateShaderModule vkCreateShaderModule; +extern PFN_vkDestroyBuffer vkDestroyBuffer; +extern PFN_vkDestroyBufferView vkDestroyBufferView; +extern PFN_vkDestroyCommandPool vkDestroyCommandPool; +extern PFN_vkDestroyDescriptorPool vkDestroyDescriptorPool; +extern PFN_vkDestroyDescriptorSetLayout vkDestroyDescriptorSetLayout; +extern PFN_vkDestroyDevice vkDestroyDevice; +extern PFN_vkDestroyEvent vkDestroyEvent; +extern PFN_vkDestroyFence vkDestroyFence; +extern PFN_vkDestroyFramebuffer vkDestroyFramebuffer; +extern PFN_vkDestroyImage vkDestroyImage; +extern PFN_vkDestroyImageView vkDestroyImageView; +extern PFN_vkDestroyInstance vkDestroyInstance; +extern PFN_vkDestroyPipeline vkDestroyPipeline; +extern PFN_vkDestroyPipelineCache vkDestroyPipelineCache; +extern PFN_vkDestroyPipelineLayout vkDestroyPipelineLayout; +extern PFN_vkDestroyQueryPool vkDestroyQueryPool; +extern PFN_vkDestroyRenderPass vkDestroyRenderPass; +extern PFN_vkDestroySampler vkDestroySampler; +extern PFN_vkDestroySemaphore vkDestroySemaphore; +extern PFN_vkDestroyShaderModule vkDestroyShaderModule; +extern PFN_vkDeviceWaitIdle vkDeviceWaitIdle; +extern PFN_vkEndCommandBuffer vkEndCommandBuffer; +extern PFN_vkEnumerateDeviceExtensionProperties vkEnumerateDeviceExtensionProperties; +extern PFN_vkEnumerateDeviceLayerProperties vkEnumerateDeviceLayerProperties; +extern PFN_vkEnumerateInstanceExtensionProperties vkEnumerateInstanceExtensionProperties; +extern PFN_vkEnumerateInstanceLayerProperties vkEnumerateInstanceLayerProperties; +extern PFN_vkEnumeratePhysicalDevices vkEnumeratePhysicalDevices; +extern PFN_vkFlushMappedMemoryRanges vkFlushMappedMemoryRanges; +extern PFN_vkFreeCommandBuffers vkFreeCommandBuffers; +extern PFN_vkFreeDescriptorSets vkFreeDescriptorSets; +extern PFN_vkFreeMemory vkFreeMemory; +extern PFN_vkGetBufferMemoryRequirements vkGetBufferMemoryRequirements; +extern PFN_vkGetDeviceMemoryCommitment vkGetDeviceMemoryCommitment; +extern PFN_vkGetDeviceProcAddr vkGetDeviceProcAddr; +extern PFN_vkGetDeviceQueue vkGetDeviceQueue; +extern PFN_vkGetEventStatus vkGetEventStatus; +extern PFN_vkGetFenceStatus vkGetFenceStatus; +extern PFN_vkGetImageMemoryRequirements vkGetImageMemoryRequirements; +extern PFN_vkGetImageSparseMemoryRequirements vkGetImageSparseMemoryRequirements; +extern PFN_vkGetImageSubresourceLayout vkGetImageSubresourceLayout; +extern PFN_vkGetInstanceProcAddr vkGetInstanceProcAddr; +extern PFN_vkGetPhysicalDeviceFeatures vkGetPhysicalDeviceFeatures; +extern PFN_vkGetPhysicalDeviceFormatProperties vkGetPhysicalDeviceFormatProperties; +extern PFN_vkGetPhysicalDeviceImageFormatProperties vkGetPhysicalDeviceImageFormatProperties; +extern PFN_vkGetPhysicalDeviceMemoryProperties vkGetPhysicalDeviceMemoryProperties; +extern PFN_vkGetPhysicalDeviceProperties vkGetPhysicalDeviceProperties; +extern PFN_vkGetPhysicalDeviceQueueFamilyProperties vkGetPhysicalDeviceQueueFamilyProperties; +extern PFN_vkGetPhysicalDeviceSparseImageFormatProperties vkGetPhysicalDeviceSparseImageFormatProperties; +extern PFN_vkGetPipelineCacheData vkGetPipelineCacheData; +extern PFN_vkGetQueryPoolResults vkGetQueryPoolResults; +extern PFN_vkGetRenderAreaGranularity vkGetRenderAreaGranularity; +extern PFN_vkInvalidateMappedMemoryRanges vkInvalidateMappedMemoryRanges; +extern PFN_vkMapMemory vkMapMemory; +extern PFN_vkMergePipelineCaches vkMergePipelineCaches; +extern PFN_vkQueueBindSparse vkQueueBindSparse; +extern PFN_vkQueueSubmit vkQueueSubmit; +extern PFN_vkQueueWaitIdle vkQueueWaitIdle; +extern PFN_vkResetCommandBuffer vkResetCommandBuffer; +extern PFN_vkResetCommandPool vkResetCommandPool; +extern PFN_vkResetDescriptorPool vkResetDescriptorPool; +extern PFN_vkResetEvent vkResetEvent; +extern PFN_vkResetFences vkResetFences; +extern PFN_vkSetEvent vkSetEvent; +extern PFN_vkUnmapMemory vkUnmapMemory; +extern PFN_vkUpdateDescriptorSets vkUpdateDescriptorSets; +extern PFN_vkWaitForFences vkWaitForFences; +#endif /* defined(VK_VERSION_1_0) */ +#if defined(VK_VERSION_1_1) +extern PFN_vkBindBufferMemory2 vkBindBufferMemory2; +extern PFN_vkBindImageMemory2 vkBindImageMemory2; +extern PFN_vkCmdDispatchBase vkCmdDispatchBase; +extern PFN_vkCmdSetDeviceMask vkCmdSetDeviceMask; +extern PFN_vkCreateDescriptorUpdateTemplate vkCreateDescriptorUpdateTemplate; +extern PFN_vkCreateSamplerYcbcrConversion vkCreateSamplerYcbcrConversion; +extern PFN_vkDestroyDescriptorUpdateTemplate vkDestroyDescriptorUpdateTemplate; +extern PFN_vkDestroySamplerYcbcrConversion vkDestroySamplerYcbcrConversion; +extern PFN_vkEnumerateInstanceVersion vkEnumerateInstanceVersion; +extern PFN_vkEnumeratePhysicalDeviceGroups vkEnumeratePhysicalDeviceGroups; +extern PFN_vkGetBufferMemoryRequirements2 vkGetBufferMemoryRequirements2; +extern PFN_vkGetDescriptorSetLayoutSupport vkGetDescriptorSetLayoutSupport; +extern PFN_vkGetDeviceGroupPeerMemoryFeatures vkGetDeviceGroupPeerMemoryFeatures; +extern PFN_vkGetDeviceQueue2 vkGetDeviceQueue2; +extern PFN_vkGetImageMemoryRequirements2 vkGetImageMemoryRequirements2; +extern PFN_vkGetImageSparseMemoryRequirements2 vkGetImageSparseMemoryRequirements2; +extern PFN_vkGetPhysicalDeviceExternalBufferProperties vkGetPhysicalDeviceExternalBufferProperties; +extern PFN_vkGetPhysicalDeviceExternalFenceProperties vkGetPhysicalDeviceExternalFenceProperties; +extern PFN_vkGetPhysicalDeviceExternalSemaphoreProperties vkGetPhysicalDeviceExternalSemaphoreProperties; +extern PFN_vkGetPhysicalDeviceFeatures2 vkGetPhysicalDeviceFeatures2; +extern PFN_vkGetPhysicalDeviceFormatProperties2 vkGetPhysicalDeviceFormatProperties2; +extern PFN_vkGetPhysicalDeviceImageFormatProperties2 vkGetPhysicalDeviceImageFormatProperties2; +extern PFN_vkGetPhysicalDeviceMemoryProperties2 vkGetPhysicalDeviceMemoryProperties2; +extern PFN_vkGetPhysicalDeviceProperties2 vkGetPhysicalDeviceProperties2; +extern PFN_vkGetPhysicalDeviceQueueFamilyProperties2 vkGetPhysicalDeviceQueueFamilyProperties2; +extern PFN_vkGetPhysicalDeviceSparseImageFormatProperties2 vkGetPhysicalDeviceSparseImageFormatProperties2; +extern PFN_vkTrimCommandPool vkTrimCommandPool; +extern PFN_vkUpdateDescriptorSetWithTemplate vkUpdateDescriptorSetWithTemplate; +#endif /* defined(VK_VERSION_1_1) */ +#if defined(VK_VERSION_1_2) +extern PFN_vkCmdBeginRenderPass2 vkCmdBeginRenderPass2; +extern PFN_vkCmdDrawIndexedIndirectCount vkCmdDrawIndexedIndirectCount; +extern PFN_vkCmdDrawIndirectCount vkCmdDrawIndirectCount; +extern PFN_vkCmdEndRenderPass2 vkCmdEndRenderPass2; +extern PFN_vkCmdNextSubpass2 vkCmdNextSubpass2; +extern PFN_vkCreateRenderPass2 vkCreateRenderPass2; +extern PFN_vkGetBufferDeviceAddress vkGetBufferDeviceAddress; +extern PFN_vkGetBufferOpaqueCaptureAddress vkGetBufferOpaqueCaptureAddress; +extern PFN_vkGetDeviceMemoryOpaqueCaptureAddress vkGetDeviceMemoryOpaqueCaptureAddress; +extern PFN_vkGetSemaphoreCounterValue vkGetSemaphoreCounterValue; +extern PFN_vkResetQueryPool vkResetQueryPool; +extern PFN_vkSignalSemaphore vkSignalSemaphore; +extern PFN_vkWaitSemaphores vkWaitSemaphores; +#endif /* defined(VK_VERSION_1_2) */ +#if defined(VK_AMD_buffer_marker) +extern PFN_vkCmdWriteBufferMarkerAMD vkCmdWriteBufferMarkerAMD; +#endif /* defined(VK_AMD_buffer_marker) */ +#if defined(VK_AMD_display_native_hdr) +extern PFN_vkSetLocalDimmingAMD vkSetLocalDimmingAMD; +#endif /* defined(VK_AMD_display_native_hdr) */ +#if defined(VK_AMD_draw_indirect_count) +extern PFN_vkCmdDrawIndexedIndirectCountAMD vkCmdDrawIndexedIndirectCountAMD; +extern PFN_vkCmdDrawIndirectCountAMD vkCmdDrawIndirectCountAMD; +#endif /* defined(VK_AMD_draw_indirect_count) */ +#if defined(VK_AMD_shader_info) +extern PFN_vkGetShaderInfoAMD vkGetShaderInfoAMD; +#endif /* defined(VK_AMD_shader_info) */ +#if defined(VK_ANDROID_external_memory_android_hardware_buffer) +extern PFN_vkGetAndroidHardwareBufferPropertiesANDROID vkGetAndroidHardwareBufferPropertiesANDROID; +extern PFN_vkGetMemoryAndroidHardwareBufferANDROID vkGetMemoryAndroidHardwareBufferANDROID; +#endif /* defined(VK_ANDROID_external_memory_android_hardware_buffer) */ +#if defined(VK_EXT_acquire_xlib_display) +extern PFN_vkAcquireXlibDisplayEXT vkAcquireXlibDisplayEXT; +extern PFN_vkGetRandROutputDisplayEXT vkGetRandROutputDisplayEXT; +#endif /* defined(VK_EXT_acquire_xlib_display) */ +#if defined(VK_EXT_buffer_device_address) +extern PFN_vkGetBufferDeviceAddressEXT vkGetBufferDeviceAddressEXT; +#endif /* defined(VK_EXT_buffer_device_address) */ +#if defined(VK_EXT_calibrated_timestamps) +extern PFN_vkGetCalibratedTimestampsEXT vkGetCalibratedTimestampsEXT; +extern PFN_vkGetPhysicalDeviceCalibrateableTimeDomainsEXT vkGetPhysicalDeviceCalibrateableTimeDomainsEXT; +#endif /* defined(VK_EXT_calibrated_timestamps) */ +#if defined(VK_EXT_conditional_rendering) +extern PFN_vkCmdBeginConditionalRenderingEXT vkCmdBeginConditionalRenderingEXT; +extern PFN_vkCmdEndConditionalRenderingEXT vkCmdEndConditionalRenderingEXT; +#endif /* defined(VK_EXT_conditional_rendering) */ +#if defined(VK_EXT_debug_marker) +extern PFN_vkCmdDebugMarkerBeginEXT vkCmdDebugMarkerBeginEXT; +extern PFN_vkCmdDebugMarkerEndEXT vkCmdDebugMarkerEndEXT; +extern PFN_vkCmdDebugMarkerInsertEXT vkCmdDebugMarkerInsertEXT; +extern PFN_vkDebugMarkerSetObjectNameEXT vkDebugMarkerSetObjectNameEXT; +extern PFN_vkDebugMarkerSetObjectTagEXT vkDebugMarkerSetObjectTagEXT; +#endif /* defined(VK_EXT_debug_marker) */ +#if defined(VK_EXT_debug_report) +extern PFN_vkCreateDebugReportCallbackEXT vkCreateDebugReportCallbackEXT; +extern PFN_vkDebugReportMessageEXT vkDebugReportMessageEXT; +extern PFN_vkDestroyDebugReportCallbackEXT vkDestroyDebugReportCallbackEXT; +#endif /* defined(VK_EXT_debug_report) */ +#if defined(VK_EXT_debug_utils) +extern PFN_vkCmdBeginDebugUtilsLabelEXT vkCmdBeginDebugUtilsLabelEXT; +extern PFN_vkCmdEndDebugUtilsLabelEXT vkCmdEndDebugUtilsLabelEXT; +extern PFN_vkCmdInsertDebugUtilsLabelEXT vkCmdInsertDebugUtilsLabelEXT; +extern PFN_vkCreateDebugUtilsMessengerEXT vkCreateDebugUtilsMessengerEXT; +extern PFN_vkDestroyDebugUtilsMessengerEXT vkDestroyDebugUtilsMessengerEXT; +extern PFN_vkQueueBeginDebugUtilsLabelEXT vkQueueBeginDebugUtilsLabelEXT; +extern PFN_vkQueueEndDebugUtilsLabelEXT vkQueueEndDebugUtilsLabelEXT; +extern PFN_vkQueueInsertDebugUtilsLabelEXT vkQueueInsertDebugUtilsLabelEXT; +extern PFN_vkSetDebugUtilsObjectNameEXT vkSetDebugUtilsObjectNameEXT; +extern PFN_vkSetDebugUtilsObjectTagEXT vkSetDebugUtilsObjectTagEXT; +extern PFN_vkSubmitDebugUtilsMessageEXT vkSubmitDebugUtilsMessageEXT; +#endif /* defined(VK_EXT_debug_utils) */ +#if defined(VK_EXT_direct_mode_display) +extern PFN_vkReleaseDisplayEXT vkReleaseDisplayEXT; +#endif /* defined(VK_EXT_direct_mode_display) */ +#if defined(VK_EXT_directfb_surface) +extern PFN_vkCreateDirectFBSurfaceEXT vkCreateDirectFBSurfaceEXT; +extern PFN_vkGetPhysicalDeviceDirectFBPresentationSupportEXT vkGetPhysicalDeviceDirectFBPresentationSupportEXT; +#endif /* defined(VK_EXT_directfb_surface) */ +#if defined(VK_EXT_discard_rectangles) +extern PFN_vkCmdSetDiscardRectangleEXT vkCmdSetDiscardRectangleEXT; +#endif /* defined(VK_EXT_discard_rectangles) */ +#if defined(VK_EXT_display_control) +extern PFN_vkDisplayPowerControlEXT vkDisplayPowerControlEXT; +extern PFN_vkGetSwapchainCounterEXT vkGetSwapchainCounterEXT; +extern PFN_vkRegisterDeviceEventEXT vkRegisterDeviceEventEXT; +extern PFN_vkRegisterDisplayEventEXT vkRegisterDisplayEventEXT; +#endif /* defined(VK_EXT_display_control) */ +#if defined(VK_EXT_display_surface_counter) +extern PFN_vkGetPhysicalDeviceSurfaceCapabilities2EXT vkGetPhysicalDeviceSurfaceCapabilities2EXT; +#endif /* defined(VK_EXT_display_surface_counter) */ +#if defined(VK_EXT_extended_dynamic_state) +extern PFN_vkCmdBindVertexBuffers2EXT vkCmdBindVertexBuffers2EXT; +extern PFN_vkCmdSetCullModeEXT vkCmdSetCullModeEXT; +extern PFN_vkCmdSetDepthBoundsTestEnableEXT vkCmdSetDepthBoundsTestEnableEXT; +extern PFN_vkCmdSetDepthCompareOpEXT vkCmdSetDepthCompareOpEXT; +extern PFN_vkCmdSetDepthTestEnableEXT vkCmdSetDepthTestEnableEXT; +extern PFN_vkCmdSetDepthWriteEnableEXT vkCmdSetDepthWriteEnableEXT; +extern PFN_vkCmdSetFrontFaceEXT vkCmdSetFrontFaceEXT; +extern PFN_vkCmdSetPrimitiveTopologyEXT vkCmdSetPrimitiveTopologyEXT; +extern PFN_vkCmdSetScissorWithCountEXT vkCmdSetScissorWithCountEXT; +extern PFN_vkCmdSetStencilOpEXT vkCmdSetStencilOpEXT; +extern PFN_vkCmdSetStencilTestEnableEXT vkCmdSetStencilTestEnableEXT; +extern PFN_vkCmdSetViewportWithCountEXT vkCmdSetViewportWithCountEXT; +#endif /* defined(VK_EXT_extended_dynamic_state) */ +#if defined(VK_EXT_external_memory_host) +extern PFN_vkGetMemoryHostPointerPropertiesEXT vkGetMemoryHostPointerPropertiesEXT; +#endif /* defined(VK_EXT_external_memory_host) */ +#if defined(VK_EXT_full_screen_exclusive) +extern PFN_vkAcquireFullScreenExclusiveModeEXT vkAcquireFullScreenExclusiveModeEXT; +extern PFN_vkGetPhysicalDeviceSurfacePresentModes2EXT vkGetPhysicalDeviceSurfacePresentModes2EXT; +extern PFN_vkReleaseFullScreenExclusiveModeEXT vkReleaseFullScreenExclusiveModeEXT; +#endif /* defined(VK_EXT_full_screen_exclusive) */ +#if defined(VK_EXT_hdr_metadata) +extern PFN_vkSetHdrMetadataEXT vkSetHdrMetadataEXT; +#endif /* defined(VK_EXT_hdr_metadata) */ +#if defined(VK_EXT_headless_surface) +extern PFN_vkCreateHeadlessSurfaceEXT vkCreateHeadlessSurfaceEXT; +#endif /* defined(VK_EXT_headless_surface) */ +#if defined(VK_EXT_host_query_reset) +extern PFN_vkResetQueryPoolEXT vkResetQueryPoolEXT; +#endif /* defined(VK_EXT_host_query_reset) */ +#if defined(VK_EXT_image_drm_format_modifier) +extern PFN_vkGetImageDrmFormatModifierPropertiesEXT vkGetImageDrmFormatModifierPropertiesEXT; +#endif /* defined(VK_EXT_image_drm_format_modifier) */ +#if defined(VK_EXT_line_rasterization) +extern PFN_vkCmdSetLineStippleEXT vkCmdSetLineStippleEXT; +#endif /* defined(VK_EXT_line_rasterization) */ +#if defined(VK_EXT_metal_surface) +extern PFN_vkCreateMetalSurfaceEXT vkCreateMetalSurfaceEXT; +#endif /* defined(VK_EXT_metal_surface) */ +#if defined(VK_EXT_private_data) +extern PFN_vkCreatePrivateDataSlotEXT vkCreatePrivateDataSlotEXT; +extern PFN_vkDestroyPrivateDataSlotEXT vkDestroyPrivateDataSlotEXT; +extern PFN_vkGetPrivateDataEXT vkGetPrivateDataEXT; +extern PFN_vkSetPrivateDataEXT vkSetPrivateDataEXT; +#endif /* defined(VK_EXT_private_data) */ +#if defined(VK_EXT_sample_locations) +extern PFN_vkCmdSetSampleLocationsEXT vkCmdSetSampleLocationsEXT; +extern PFN_vkGetPhysicalDeviceMultisamplePropertiesEXT vkGetPhysicalDeviceMultisamplePropertiesEXT; +#endif /* defined(VK_EXT_sample_locations) */ +#if defined(VK_EXT_tooling_info) +extern PFN_vkGetPhysicalDeviceToolPropertiesEXT vkGetPhysicalDeviceToolPropertiesEXT; +#endif /* defined(VK_EXT_tooling_info) */ +#if defined(VK_EXT_transform_feedback) +extern PFN_vkCmdBeginQueryIndexedEXT vkCmdBeginQueryIndexedEXT; +extern PFN_vkCmdBeginTransformFeedbackEXT vkCmdBeginTransformFeedbackEXT; +extern PFN_vkCmdBindTransformFeedbackBuffersEXT vkCmdBindTransformFeedbackBuffersEXT; +extern PFN_vkCmdDrawIndirectByteCountEXT vkCmdDrawIndirectByteCountEXT; +extern PFN_vkCmdEndQueryIndexedEXT vkCmdEndQueryIndexedEXT; +extern PFN_vkCmdEndTransformFeedbackEXT vkCmdEndTransformFeedbackEXT; +#endif /* defined(VK_EXT_transform_feedback) */ +#if defined(VK_EXT_validation_cache) +extern PFN_vkCreateValidationCacheEXT vkCreateValidationCacheEXT; +extern PFN_vkDestroyValidationCacheEXT vkDestroyValidationCacheEXT; +extern PFN_vkGetValidationCacheDataEXT vkGetValidationCacheDataEXT; +extern PFN_vkMergeValidationCachesEXT vkMergeValidationCachesEXT; +#endif /* defined(VK_EXT_validation_cache) */ +#if defined(VK_FUCHSIA_imagepipe_surface) +extern PFN_vkCreateImagePipeSurfaceFUCHSIA vkCreateImagePipeSurfaceFUCHSIA; +#endif /* defined(VK_FUCHSIA_imagepipe_surface) */ +#if defined(VK_GGP_stream_descriptor_surface) +extern PFN_vkCreateStreamDescriptorSurfaceGGP vkCreateStreamDescriptorSurfaceGGP; +#endif /* defined(VK_GGP_stream_descriptor_surface) */ +#if defined(VK_GOOGLE_display_timing) +extern PFN_vkGetPastPresentationTimingGOOGLE vkGetPastPresentationTimingGOOGLE; +extern PFN_vkGetRefreshCycleDurationGOOGLE vkGetRefreshCycleDurationGOOGLE; +#endif /* defined(VK_GOOGLE_display_timing) */ +#if defined(VK_INTEL_performance_query) +extern PFN_vkAcquirePerformanceConfigurationINTEL vkAcquirePerformanceConfigurationINTEL; +extern PFN_vkCmdSetPerformanceMarkerINTEL vkCmdSetPerformanceMarkerINTEL; +extern PFN_vkCmdSetPerformanceOverrideINTEL vkCmdSetPerformanceOverrideINTEL; +extern PFN_vkCmdSetPerformanceStreamMarkerINTEL vkCmdSetPerformanceStreamMarkerINTEL; +extern PFN_vkGetPerformanceParameterINTEL vkGetPerformanceParameterINTEL; +extern PFN_vkInitializePerformanceApiINTEL vkInitializePerformanceApiINTEL; +extern PFN_vkQueueSetPerformanceConfigurationINTEL vkQueueSetPerformanceConfigurationINTEL; +extern PFN_vkReleasePerformanceConfigurationINTEL vkReleasePerformanceConfigurationINTEL; +extern PFN_vkUninitializePerformanceApiINTEL vkUninitializePerformanceApiINTEL; +#endif /* defined(VK_INTEL_performance_query) */ +#if defined(VK_KHR_android_surface) +extern PFN_vkCreateAndroidSurfaceKHR vkCreateAndroidSurfaceKHR; +#endif /* defined(VK_KHR_android_surface) */ +#if defined(VK_KHR_bind_memory2) +extern PFN_vkBindBufferMemory2KHR vkBindBufferMemory2KHR; +extern PFN_vkBindImageMemory2KHR vkBindImageMemory2KHR; +#endif /* defined(VK_KHR_bind_memory2) */ +#if defined(VK_KHR_buffer_device_address) +extern PFN_vkGetBufferDeviceAddressKHR vkGetBufferDeviceAddressKHR; +extern PFN_vkGetBufferOpaqueCaptureAddressKHR vkGetBufferOpaqueCaptureAddressKHR; +extern PFN_vkGetDeviceMemoryOpaqueCaptureAddressKHR vkGetDeviceMemoryOpaqueCaptureAddressKHR; +#endif /* defined(VK_KHR_buffer_device_address) */ +#if defined(VK_KHR_create_renderpass2) +extern PFN_vkCmdBeginRenderPass2KHR vkCmdBeginRenderPass2KHR; +extern PFN_vkCmdEndRenderPass2KHR vkCmdEndRenderPass2KHR; +extern PFN_vkCmdNextSubpass2KHR vkCmdNextSubpass2KHR; +extern PFN_vkCreateRenderPass2KHR vkCreateRenderPass2KHR; +#endif /* defined(VK_KHR_create_renderpass2) */ +#if defined(VK_KHR_deferred_host_operations) +extern PFN_vkCreateDeferredOperationKHR vkCreateDeferredOperationKHR; +extern PFN_vkDeferredOperationJoinKHR vkDeferredOperationJoinKHR; +extern PFN_vkDestroyDeferredOperationKHR vkDestroyDeferredOperationKHR; +extern PFN_vkGetDeferredOperationMaxConcurrencyKHR vkGetDeferredOperationMaxConcurrencyKHR; +extern PFN_vkGetDeferredOperationResultKHR vkGetDeferredOperationResultKHR; +#endif /* defined(VK_KHR_deferred_host_operations) */ +#if defined(VK_KHR_descriptor_update_template) +extern PFN_vkCreateDescriptorUpdateTemplateKHR vkCreateDescriptorUpdateTemplateKHR; +extern PFN_vkDestroyDescriptorUpdateTemplateKHR vkDestroyDescriptorUpdateTemplateKHR; +extern PFN_vkUpdateDescriptorSetWithTemplateKHR vkUpdateDescriptorSetWithTemplateKHR; +#endif /* defined(VK_KHR_descriptor_update_template) */ +#if defined(VK_KHR_device_group) +extern PFN_vkCmdDispatchBaseKHR vkCmdDispatchBaseKHR; +extern PFN_vkCmdSetDeviceMaskKHR vkCmdSetDeviceMaskKHR; +extern PFN_vkGetDeviceGroupPeerMemoryFeaturesKHR vkGetDeviceGroupPeerMemoryFeaturesKHR; +#endif /* defined(VK_KHR_device_group) */ +#if defined(VK_KHR_device_group_creation) +extern PFN_vkEnumeratePhysicalDeviceGroupsKHR vkEnumeratePhysicalDeviceGroupsKHR; +#endif /* defined(VK_KHR_device_group_creation) */ +#if defined(VK_KHR_display) +extern PFN_vkCreateDisplayModeKHR vkCreateDisplayModeKHR; +extern PFN_vkCreateDisplayPlaneSurfaceKHR vkCreateDisplayPlaneSurfaceKHR; +extern PFN_vkGetDisplayModePropertiesKHR vkGetDisplayModePropertiesKHR; +extern PFN_vkGetDisplayPlaneCapabilitiesKHR vkGetDisplayPlaneCapabilitiesKHR; +extern PFN_vkGetDisplayPlaneSupportedDisplaysKHR vkGetDisplayPlaneSupportedDisplaysKHR; +extern PFN_vkGetPhysicalDeviceDisplayPlanePropertiesKHR vkGetPhysicalDeviceDisplayPlanePropertiesKHR; +extern PFN_vkGetPhysicalDeviceDisplayPropertiesKHR vkGetPhysicalDeviceDisplayPropertiesKHR; +#endif /* defined(VK_KHR_display) */ +#if defined(VK_KHR_display_swapchain) +extern PFN_vkCreateSharedSwapchainsKHR vkCreateSharedSwapchainsKHR; +#endif /* defined(VK_KHR_display_swapchain) */ +#if defined(VK_KHR_draw_indirect_count) +extern PFN_vkCmdDrawIndexedIndirectCountKHR vkCmdDrawIndexedIndirectCountKHR; +extern PFN_vkCmdDrawIndirectCountKHR vkCmdDrawIndirectCountKHR; +#endif /* defined(VK_KHR_draw_indirect_count) */ +#if defined(VK_KHR_external_fence_capabilities) +extern PFN_vkGetPhysicalDeviceExternalFencePropertiesKHR vkGetPhysicalDeviceExternalFencePropertiesKHR; +#endif /* defined(VK_KHR_external_fence_capabilities) */ +#if defined(VK_KHR_external_fence_fd) +extern PFN_vkGetFenceFdKHR vkGetFenceFdKHR; +extern PFN_vkImportFenceFdKHR vkImportFenceFdKHR; +#endif /* defined(VK_KHR_external_fence_fd) */ +#if defined(VK_KHR_external_fence_win32) +extern PFN_vkGetFenceWin32HandleKHR vkGetFenceWin32HandleKHR; +extern PFN_vkImportFenceWin32HandleKHR vkImportFenceWin32HandleKHR; +#endif /* defined(VK_KHR_external_fence_win32) */ +#if defined(VK_KHR_external_memory_capabilities) +extern PFN_vkGetPhysicalDeviceExternalBufferPropertiesKHR vkGetPhysicalDeviceExternalBufferPropertiesKHR; +#endif /* defined(VK_KHR_external_memory_capabilities) */ +#if defined(VK_KHR_external_memory_fd) +extern PFN_vkGetMemoryFdKHR vkGetMemoryFdKHR; +extern PFN_vkGetMemoryFdPropertiesKHR vkGetMemoryFdPropertiesKHR; +#endif /* defined(VK_KHR_external_memory_fd) */ +#if defined(VK_KHR_external_memory_win32) +extern PFN_vkGetMemoryWin32HandleKHR vkGetMemoryWin32HandleKHR; +extern PFN_vkGetMemoryWin32HandlePropertiesKHR vkGetMemoryWin32HandlePropertiesKHR; +#endif /* defined(VK_KHR_external_memory_win32) */ +#if defined(VK_KHR_external_semaphore_capabilities) +extern PFN_vkGetPhysicalDeviceExternalSemaphorePropertiesKHR vkGetPhysicalDeviceExternalSemaphorePropertiesKHR; +#endif /* defined(VK_KHR_external_semaphore_capabilities) */ +#if defined(VK_KHR_external_semaphore_fd) +extern PFN_vkGetSemaphoreFdKHR vkGetSemaphoreFdKHR; +extern PFN_vkImportSemaphoreFdKHR vkImportSemaphoreFdKHR; +#endif /* defined(VK_KHR_external_semaphore_fd) */ +#if defined(VK_KHR_external_semaphore_win32) +extern PFN_vkGetSemaphoreWin32HandleKHR vkGetSemaphoreWin32HandleKHR; +extern PFN_vkImportSemaphoreWin32HandleKHR vkImportSemaphoreWin32HandleKHR; +#endif /* defined(VK_KHR_external_semaphore_win32) */ +#if defined(VK_KHR_get_display_properties2) +extern PFN_vkGetDisplayModeProperties2KHR vkGetDisplayModeProperties2KHR; +extern PFN_vkGetDisplayPlaneCapabilities2KHR vkGetDisplayPlaneCapabilities2KHR; +extern PFN_vkGetPhysicalDeviceDisplayPlaneProperties2KHR vkGetPhysicalDeviceDisplayPlaneProperties2KHR; +extern PFN_vkGetPhysicalDeviceDisplayProperties2KHR vkGetPhysicalDeviceDisplayProperties2KHR; +#endif /* defined(VK_KHR_get_display_properties2) */ +#if defined(VK_KHR_get_memory_requirements2) +extern PFN_vkGetBufferMemoryRequirements2KHR vkGetBufferMemoryRequirements2KHR; +extern PFN_vkGetImageMemoryRequirements2KHR vkGetImageMemoryRequirements2KHR; +extern PFN_vkGetImageSparseMemoryRequirements2KHR vkGetImageSparseMemoryRequirements2KHR; +#endif /* defined(VK_KHR_get_memory_requirements2) */ +#if defined(VK_KHR_get_physical_device_properties2) +extern PFN_vkGetPhysicalDeviceFeatures2KHR vkGetPhysicalDeviceFeatures2KHR; +extern PFN_vkGetPhysicalDeviceFormatProperties2KHR vkGetPhysicalDeviceFormatProperties2KHR; +extern PFN_vkGetPhysicalDeviceImageFormatProperties2KHR vkGetPhysicalDeviceImageFormatProperties2KHR; +extern PFN_vkGetPhysicalDeviceMemoryProperties2KHR vkGetPhysicalDeviceMemoryProperties2KHR; +extern PFN_vkGetPhysicalDeviceProperties2KHR vkGetPhysicalDeviceProperties2KHR; +extern PFN_vkGetPhysicalDeviceQueueFamilyProperties2KHR vkGetPhysicalDeviceQueueFamilyProperties2KHR; +extern PFN_vkGetPhysicalDeviceSparseImageFormatProperties2KHR vkGetPhysicalDeviceSparseImageFormatProperties2KHR; +#endif /* defined(VK_KHR_get_physical_device_properties2) */ +#if defined(VK_KHR_get_surface_capabilities2) +extern PFN_vkGetPhysicalDeviceSurfaceCapabilities2KHR vkGetPhysicalDeviceSurfaceCapabilities2KHR; +extern PFN_vkGetPhysicalDeviceSurfaceFormats2KHR vkGetPhysicalDeviceSurfaceFormats2KHR; +#endif /* defined(VK_KHR_get_surface_capabilities2) */ +#if defined(VK_KHR_maintenance1) +extern PFN_vkTrimCommandPoolKHR vkTrimCommandPoolKHR; +#endif /* defined(VK_KHR_maintenance1) */ +#if defined(VK_KHR_maintenance3) +extern PFN_vkGetDescriptorSetLayoutSupportKHR vkGetDescriptorSetLayoutSupportKHR; +#endif /* defined(VK_KHR_maintenance3) */ +#if defined(VK_KHR_performance_query) +extern PFN_vkAcquireProfilingLockKHR vkAcquireProfilingLockKHR; +extern PFN_vkEnumeratePhysicalDeviceQueueFamilyPerformanceQueryCountersKHR vkEnumeratePhysicalDeviceQueueFamilyPerformanceQueryCountersKHR; +extern PFN_vkGetPhysicalDeviceQueueFamilyPerformanceQueryPassesKHR vkGetPhysicalDeviceQueueFamilyPerformanceQueryPassesKHR; +extern PFN_vkReleaseProfilingLockKHR vkReleaseProfilingLockKHR; +#endif /* defined(VK_KHR_performance_query) */ +#if defined(VK_KHR_pipeline_executable_properties) +extern PFN_vkGetPipelineExecutableInternalRepresentationsKHR vkGetPipelineExecutableInternalRepresentationsKHR; +extern PFN_vkGetPipelineExecutablePropertiesKHR vkGetPipelineExecutablePropertiesKHR; +extern PFN_vkGetPipelineExecutableStatisticsKHR vkGetPipelineExecutableStatisticsKHR; +#endif /* defined(VK_KHR_pipeline_executable_properties) */ +#if defined(VK_KHR_push_descriptor) +extern PFN_vkCmdPushDescriptorSetKHR vkCmdPushDescriptorSetKHR; +#endif /* defined(VK_KHR_push_descriptor) */ +#if defined(VK_KHR_ray_tracing) +extern PFN_vkBindAccelerationStructureMemoryKHR vkBindAccelerationStructureMemoryKHR; +extern PFN_vkBuildAccelerationStructureKHR vkBuildAccelerationStructureKHR; +extern PFN_vkCmdBuildAccelerationStructureIndirectKHR vkCmdBuildAccelerationStructureIndirectKHR; +extern PFN_vkCmdBuildAccelerationStructureKHR vkCmdBuildAccelerationStructureKHR; +extern PFN_vkCmdCopyAccelerationStructureKHR vkCmdCopyAccelerationStructureKHR; +extern PFN_vkCmdCopyAccelerationStructureToMemoryKHR vkCmdCopyAccelerationStructureToMemoryKHR; +extern PFN_vkCmdCopyMemoryToAccelerationStructureKHR vkCmdCopyMemoryToAccelerationStructureKHR; +extern PFN_vkCmdTraceRaysIndirectKHR vkCmdTraceRaysIndirectKHR; +extern PFN_vkCmdTraceRaysKHR vkCmdTraceRaysKHR; +extern PFN_vkCmdWriteAccelerationStructuresPropertiesKHR vkCmdWriteAccelerationStructuresPropertiesKHR; +extern PFN_vkCopyAccelerationStructureKHR vkCopyAccelerationStructureKHR; +extern PFN_vkCopyAccelerationStructureToMemoryKHR vkCopyAccelerationStructureToMemoryKHR; +extern PFN_vkCopyMemoryToAccelerationStructureKHR vkCopyMemoryToAccelerationStructureKHR; +extern PFN_vkCreateAccelerationStructureKHR vkCreateAccelerationStructureKHR; +extern PFN_vkCreateRayTracingPipelinesKHR vkCreateRayTracingPipelinesKHR; +extern PFN_vkDestroyAccelerationStructureKHR vkDestroyAccelerationStructureKHR; +extern PFN_vkGetAccelerationStructureDeviceAddressKHR vkGetAccelerationStructureDeviceAddressKHR; +extern PFN_vkGetAccelerationStructureMemoryRequirementsKHR vkGetAccelerationStructureMemoryRequirementsKHR; +extern PFN_vkGetDeviceAccelerationStructureCompatibilityKHR vkGetDeviceAccelerationStructureCompatibilityKHR; +extern PFN_vkGetRayTracingCaptureReplayShaderGroupHandlesKHR vkGetRayTracingCaptureReplayShaderGroupHandlesKHR; +extern PFN_vkGetRayTracingShaderGroupHandlesKHR vkGetRayTracingShaderGroupHandlesKHR; +extern PFN_vkWriteAccelerationStructuresPropertiesKHR vkWriteAccelerationStructuresPropertiesKHR; +#endif /* defined(VK_KHR_ray_tracing) */ +#if defined(VK_KHR_sampler_ycbcr_conversion) +extern PFN_vkCreateSamplerYcbcrConversionKHR vkCreateSamplerYcbcrConversionKHR; +extern PFN_vkDestroySamplerYcbcrConversionKHR vkDestroySamplerYcbcrConversionKHR; +#endif /* defined(VK_KHR_sampler_ycbcr_conversion) */ +#if defined(VK_KHR_shared_presentable_image) +extern PFN_vkGetSwapchainStatusKHR vkGetSwapchainStatusKHR; +#endif /* defined(VK_KHR_shared_presentable_image) */ +#if defined(VK_KHR_surface) +extern PFN_vkDestroySurfaceKHR vkDestroySurfaceKHR; +extern PFN_vkGetPhysicalDeviceSurfaceCapabilitiesKHR vkGetPhysicalDeviceSurfaceCapabilitiesKHR; +extern PFN_vkGetPhysicalDeviceSurfaceFormatsKHR vkGetPhysicalDeviceSurfaceFormatsKHR; +extern PFN_vkGetPhysicalDeviceSurfacePresentModesKHR vkGetPhysicalDeviceSurfacePresentModesKHR; +extern PFN_vkGetPhysicalDeviceSurfaceSupportKHR vkGetPhysicalDeviceSurfaceSupportKHR; +#endif /* defined(VK_KHR_surface) */ +#if defined(VK_KHR_swapchain) +extern PFN_vkAcquireNextImageKHR vkAcquireNextImageKHR; +extern PFN_vkCreateSwapchainKHR vkCreateSwapchainKHR; +extern PFN_vkDestroySwapchainKHR vkDestroySwapchainKHR; +extern PFN_vkGetSwapchainImagesKHR vkGetSwapchainImagesKHR; +extern PFN_vkQueuePresentKHR vkQueuePresentKHR; +#endif /* defined(VK_KHR_swapchain) */ +#if defined(VK_KHR_timeline_semaphore) +extern PFN_vkGetSemaphoreCounterValueKHR vkGetSemaphoreCounterValueKHR; +extern PFN_vkSignalSemaphoreKHR vkSignalSemaphoreKHR; +extern PFN_vkWaitSemaphoresKHR vkWaitSemaphoresKHR; +#endif /* defined(VK_KHR_timeline_semaphore) */ +#if defined(VK_KHR_wayland_surface) +extern PFN_vkCreateWaylandSurfaceKHR vkCreateWaylandSurfaceKHR; +extern PFN_vkGetPhysicalDeviceWaylandPresentationSupportKHR vkGetPhysicalDeviceWaylandPresentationSupportKHR; +#endif /* defined(VK_KHR_wayland_surface) */ +#if defined(VK_KHR_win32_surface) +extern PFN_vkCreateWin32SurfaceKHR vkCreateWin32SurfaceKHR; +extern PFN_vkGetPhysicalDeviceWin32PresentationSupportKHR vkGetPhysicalDeviceWin32PresentationSupportKHR; +#endif /* defined(VK_KHR_win32_surface) */ +#if defined(VK_KHR_xcb_surface) +extern PFN_vkCreateXcbSurfaceKHR vkCreateXcbSurfaceKHR; +extern PFN_vkGetPhysicalDeviceXcbPresentationSupportKHR vkGetPhysicalDeviceXcbPresentationSupportKHR; +#endif /* defined(VK_KHR_xcb_surface) */ +#if defined(VK_KHR_xlib_surface) +extern PFN_vkCreateXlibSurfaceKHR vkCreateXlibSurfaceKHR; +extern PFN_vkGetPhysicalDeviceXlibPresentationSupportKHR vkGetPhysicalDeviceXlibPresentationSupportKHR; +#endif /* defined(VK_KHR_xlib_surface) */ +#if defined(VK_MVK_ios_surface) +extern PFN_vkCreateIOSSurfaceMVK vkCreateIOSSurfaceMVK; +#endif /* defined(VK_MVK_ios_surface) */ +#if defined(VK_MVK_macos_surface) +extern PFN_vkCreateMacOSSurfaceMVK vkCreateMacOSSurfaceMVK; +#endif /* defined(VK_MVK_macos_surface) */ +#if defined(VK_NN_vi_surface) +extern PFN_vkCreateViSurfaceNN vkCreateViSurfaceNN; +#endif /* defined(VK_NN_vi_surface) */ +#if defined(VK_NVX_image_view_handle) +extern PFN_vkGetImageViewAddressNVX vkGetImageViewAddressNVX; +extern PFN_vkGetImageViewHandleNVX vkGetImageViewHandleNVX; +#endif /* defined(VK_NVX_image_view_handle) */ +#if defined(VK_NV_clip_space_w_scaling) +extern PFN_vkCmdSetViewportWScalingNV vkCmdSetViewportWScalingNV; +#endif /* defined(VK_NV_clip_space_w_scaling) */ +#if defined(VK_NV_cooperative_matrix) +extern PFN_vkGetPhysicalDeviceCooperativeMatrixPropertiesNV vkGetPhysicalDeviceCooperativeMatrixPropertiesNV; +#endif /* defined(VK_NV_cooperative_matrix) */ +#if defined(VK_NV_coverage_reduction_mode) +extern PFN_vkGetPhysicalDeviceSupportedFramebufferMixedSamplesCombinationsNV vkGetPhysicalDeviceSupportedFramebufferMixedSamplesCombinationsNV; +#endif /* defined(VK_NV_coverage_reduction_mode) */ +#if defined(VK_NV_device_diagnostic_checkpoints) +extern PFN_vkCmdSetCheckpointNV vkCmdSetCheckpointNV; +extern PFN_vkGetQueueCheckpointDataNV vkGetQueueCheckpointDataNV; +#endif /* defined(VK_NV_device_diagnostic_checkpoints) */ +#if defined(VK_NV_device_generated_commands) +extern PFN_vkCmdBindPipelineShaderGroupNV vkCmdBindPipelineShaderGroupNV; +extern PFN_vkCmdExecuteGeneratedCommandsNV vkCmdExecuteGeneratedCommandsNV; +extern PFN_vkCmdPreprocessGeneratedCommandsNV vkCmdPreprocessGeneratedCommandsNV; +extern PFN_vkCreateIndirectCommandsLayoutNV vkCreateIndirectCommandsLayoutNV; +extern PFN_vkDestroyIndirectCommandsLayoutNV vkDestroyIndirectCommandsLayoutNV; +extern PFN_vkGetGeneratedCommandsMemoryRequirementsNV vkGetGeneratedCommandsMemoryRequirementsNV; +#endif /* defined(VK_NV_device_generated_commands) */ +#if defined(VK_NV_external_memory_capabilities) +extern PFN_vkGetPhysicalDeviceExternalImageFormatPropertiesNV vkGetPhysicalDeviceExternalImageFormatPropertiesNV; +#endif /* defined(VK_NV_external_memory_capabilities) */ +#if defined(VK_NV_external_memory_win32) +extern PFN_vkGetMemoryWin32HandleNV vkGetMemoryWin32HandleNV; +#endif /* defined(VK_NV_external_memory_win32) */ +#if defined(VK_NV_mesh_shader) +extern PFN_vkCmdDrawMeshTasksIndirectCountNV vkCmdDrawMeshTasksIndirectCountNV; +extern PFN_vkCmdDrawMeshTasksIndirectNV vkCmdDrawMeshTasksIndirectNV; +extern PFN_vkCmdDrawMeshTasksNV vkCmdDrawMeshTasksNV; +#endif /* defined(VK_NV_mesh_shader) */ +#if defined(VK_NV_ray_tracing) +extern PFN_vkBindAccelerationStructureMemoryNV vkBindAccelerationStructureMemoryNV; +extern PFN_vkCmdBuildAccelerationStructureNV vkCmdBuildAccelerationStructureNV; +extern PFN_vkCmdCopyAccelerationStructureNV vkCmdCopyAccelerationStructureNV; +extern PFN_vkCmdTraceRaysNV vkCmdTraceRaysNV; +extern PFN_vkCmdWriteAccelerationStructuresPropertiesNV vkCmdWriteAccelerationStructuresPropertiesNV; +extern PFN_vkCompileDeferredNV vkCompileDeferredNV; +extern PFN_vkCreateAccelerationStructureNV vkCreateAccelerationStructureNV; +extern PFN_vkCreateRayTracingPipelinesNV vkCreateRayTracingPipelinesNV; +extern PFN_vkDestroyAccelerationStructureNV vkDestroyAccelerationStructureNV; +extern PFN_vkGetAccelerationStructureHandleNV vkGetAccelerationStructureHandleNV; +extern PFN_vkGetAccelerationStructureMemoryRequirementsNV vkGetAccelerationStructureMemoryRequirementsNV; +extern PFN_vkGetRayTracingShaderGroupHandlesNV vkGetRayTracingShaderGroupHandlesNV; +#endif /* defined(VK_NV_ray_tracing) */ +#if defined(VK_NV_scissor_exclusive) +extern PFN_vkCmdSetExclusiveScissorNV vkCmdSetExclusiveScissorNV; +#endif /* defined(VK_NV_scissor_exclusive) */ +#if defined(VK_NV_shading_rate_image) +extern PFN_vkCmdBindShadingRateImageNV vkCmdBindShadingRateImageNV; +extern PFN_vkCmdSetCoarseSampleOrderNV vkCmdSetCoarseSampleOrderNV; +extern PFN_vkCmdSetViewportShadingRatePaletteNV vkCmdSetViewportShadingRatePaletteNV; +#endif /* defined(VK_NV_shading_rate_image) */ +#if (defined(VK_EXT_full_screen_exclusive) && defined(VK_KHR_device_group)) || (defined(VK_EXT_full_screen_exclusive) && defined(VK_VERSION_1_1)) +extern PFN_vkGetDeviceGroupSurfacePresentModes2EXT vkGetDeviceGroupSurfacePresentModes2EXT; +#endif /* (defined(VK_EXT_full_screen_exclusive) && defined(VK_KHR_device_group)) || (defined(VK_EXT_full_screen_exclusive) && defined(VK_VERSION_1_1)) */ +#if (defined(VK_KHR_descriptor_update_template) && defined(VK_KHR_push_descriptor)) || (defined(VK_KHR_push_descriptor) && defined(VK_VERSION_1_1)) || (defined(VK_KHR_push_descriptor) && defined(VK_KHR_descriptor_update_template)) +extern PFN_vkCmdPushDescriptorSetWithTemplateKHR vkCmdPushDescriptorSetWithTemplateKHR; +#endif /* (defined(VK_KHR_descriptor_update_template) && defined(VK_KHR_push_descriptor)) || (defined(VK_KHR_push_descriptor) && defined(VK_VERSION_1_1)) || (defined(VK_KHR_push_descriptor) && defined(VK_KHR_descriptor_update_template)) */ +#if (defined(VK_KHR_device_group) && defined(VK_KHR_surface)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) +extern PFN_vkGetDeviceGroupPresentCapabilitiesKHR vkGetDeviceGroupPresentCapabilitiesKHR; +extern PFN_vkGetDeviceGroupSurfacePresentModesKHR vkGetDeviceGroupSurfacePresentModesKHR; +extern PFN_vkGetPhysicalDevicePresentRectanglesKHR vkGetPhysicalDevicePresentRectanglesKHR; +#endif /* (defined(VK_KHR_device_group) && defined(VK_KHR_surface)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) */ +#if (defined(VK_KHR_device_group) && defined(VK_KHR_swapchain)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) +extern PFN_vkAcquireNextImage2KHR vkAcquireNextImage2KHR; +#endif /* (defined(VK_KHR_device_group) && defined(VK_KHR_swapchain)) || (defined(VK_KHR_swapchain) && defined(VK_VERSION_1_1)) */ +/* VOLK_GENERATE_PROTOTYPES_H */ + +#ifdef __cplusplus +} +#endif + +#endif + +#ifdef VOLK_IMPLEMENTATION +#undef VOLK_IMPLEMENTATION +// Prevent tools like dependency checkers that don't evaluate +// macros from detecting a cyclic dependency. +#define VOLK_SOURCE "volk.c" +#include VOLK_SOURCE +#endif + +/** + * Copyright (c) 2018-2019 Arseny Kapoulkine + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. +*/ +/* clang-format on */