Commit graph

1181 commits

Author SHA1 Message Date
Bill Currie
75f19f7243 Completely rework the vulkan related api
Things don't work yet, but this feels much cleaner.
2019-07-12 13:34:31 +09:00
Bill Currie
c0bc5cfad6 Implement swapchain creation 2019-07-11 00:58:14 +09:00
Bill Currie
2771e9c573 Correct extension handling
I had missed a critical bit from the cookbook.
2019-07-10 13:16:46 +09:00
Bill Currie
31ead15e96 Add comment about developer flags 2019-07-09 22:04:48 +09:00
Bill Currie
8a3cd224a3 Add vulkan developer flag 2019-07-09 20:14:30 +09:00
Bill Currie
68449d0f6f Create a window and a surface for vulkan
Yay, segfaults in R_Init :)
2019-07-09 16:33:44 +09:00
Bill Currie
b3d982bfc3 Remove global vulkan_ctx
While I can't say that I'm happy with the details of vulkan_ctx_t, I am
pretty sure I don't want to be limited to having only one.
2019-07-09 11:54:23 +09:00
Bill Currie
8ee06d75a9 Separate render and presentation initialization
This paves the way for clean initialization of the Vulkan renderer, and
very much cleans up the older renderer initialization code as gl and sw
are no longer intertwined.
2019-07-09 01:00:47 +09:00
Bill Currie
cb10175824 Pass vid_internal to R_LoadModule
This fixes the segfault and pushes things very much in the desired
direction of proper system independence for rendering and presentation
separation (though things were headed in the right direction before).
2019-07-08 14:04:03 +09:00
Bill Currie
6ee2df8445 Move vid callbacks into vid_internal
Currently segfaults because viddef is an alias for *r_data->vid, which
has not yet been initialized (chicken and egg).
2019-07-08 13:40:29 +09:00
Bill Currie
3e1520c246 Move vid callback access into vid
The plan is to move the callbacks into a "private" struct.
2019-07-08 12:46:22 +09:00
Bill Currie
940ef833ae Shutdown Vulkan properly
Well, as properly as can be considering how little is started up :P
2019-07-07 14:34:02 +09:00
Bill Currie
ab08e4f207 Create a logical device with a single queue
A single graphics-capable queue should be enough for now. However, I'm
not sure I'm happy with a lot of the code: it's a bit difficult to write
flexibly configured code for Vulkan (or seems to be at this stage),
especially in C.
2019-07-07 01:28:05 +09:00
Bill Currie
2bc78e7f0a Start work on a Vulkan-based renderer
Doesn't do much other than create an instance and enumerate some stuff,
but the build system is working.
2019-07-06 14:56:15 +09:00
Bill Currie
ab71311174 Correct some comments 2019-07-06 14:44:22 +09:00
Bill Currie
2ee36e8b99 Rename qfot_type_t.ty to meta
When I did up a ruamoko implementation of a type encoding dumper, I
named the field meta, which feels much better than ty.
2019-06-09 18:10:07 +09:00
Bill Currie
f093516962 Move qfo type encoding structs
This makes them available to the gamecode VM, and thus the debugger.
2019-06-09 18:07:58 +09:00
Bill Currie
f58c2fef5a Switch QF's quaterions from wxyz to xyzw.
After messing with SIMD stuff for a little, I think I now understand why
the industry went with xyzw instead of the mathematical wxyz. Anyway, this
will make for less pain in the future (assuming I got everything).
2018-10-12 14:33:55 +09:00
Bill Currie
364c2d2c2b Add a stack and push and pop instructions.
The stack is for data, not return addresses.
2018-10-11 13:24:03 +09:00
Bill Currie
9501d30a8f Fix up a missing const attribute.
The problem with having such a complex build system :P
2018-10-11 11:29:12 +09:00
Bill Currie
82809eb856 Simplify the call to PR_LoadProgsFile.
I've decided that setting pr.max_edicts and pr.zone_size as part of the
local progs initialization rather than in PR_LoadProgsFile makes more
sense. For one, it is unlikely for the limits to change every time progs is
reloaded. Also, they seem to be a property of the VM rather than the progs.
However, there is nothing stopping the caller from updating max_edicts and
zone_size every call.
2018-10-11 10:06:48 +09:00
Bill Currie
4872c56e2d Mark the end of the v6 opcodes. 2018-10-11 10:05:36 +09:00
Bill Currie
34bcf7faab Do a pure/const/noreturn/format attribute pass.
I always wanted these, but as gcc now provides warnings for functions that
could do with such attributes, finding all the functions is much easier.
2018-10-09 12:42:21 +09:00
Bill Currie
89daf15215 Fix a c&p error in a doc comment. 2018-08-26 15:05:13 +09:00
Bill Currie
4f58429137 Fix an unhealthy pile of gcc 8 warnings.
While some of the warnings were merely annoying, some where actual bugs or
unearthed bugs in related code.
2018-08-20 00:05:00 +09:00
Bill Currie
0de0eb2fc7 Fix a bunch of bit-rot. 2018-06-08 13:21:30 +09:00
Bill Currie
e1f9cb2028 Create an input subsection for the docs. 2016-01-04 17:24:13 +09:00
Bill Currie
fb3396a21d Fix a typo in the key docs. 2016-01-04 16:20:57 +09:00
Bill Currie
aecea5c350 Document most of they key binding system. 2016-01-04 16:11:15 +09:00
Bill Currie
7c16acf4aa Make Key_GetBinding take an imt instead of a name. 2016-01-04 16:06:48 +09:00
Bill Currie
546e333a3c Allow Sys_Error to be hooked.
This makes debugging builtins that wrap normal functions a little easier by
giving a progs dump when such an error occurs.
2016-01-03 23:04:00 +09:00
Bill Currie
f0dc995c67 Handle c99 inline vs gnu89 inline.
The meaning of "extern inline" and "inline" swaps between the two, and
gcc-5 uses c99 inline.
2015-08-18 00:11:03 +09:00
Bill Currie
d1f1375df5 Increase number of supported joystick buttons.
Also fix a bug where despite supporting 32 buttons, only 18 were actually
supported, and a similar issue for the number of axes.

My saitek x52 has 34 buttons and 10 axes. Whee.
2014-08-17 11:35:40 +09:00
Bill Currie
3fb9fb9373 Attempt to autoconfiscate gcc_struct.
It seems some tool-chains get rather picky about such things.
2014-02-03 16:47:51 +09:00
Bill Currie
9d71d9bb7a Remove maxwarpwidth and maxwarpheight.
Only the sw32 renderer even tried to use them, and they didn't get set
properly. This fixes JohnnyonFlame's waterwarp issue.
2014-02-02 14:41:58 +09:00
Bill Currie
8bea6a66ca Create _QFS_VOpenFile and QFS_VOpenFile.
_QFS_VOpenFile is actually _QFS_FOpenFile reimplemented to take vpath start
and end parameters so the search can be limited. QFS_VOpenFile,
_QFS_FOpenFile, and QFS_FOpenFile are all wrappers for _QFS_VOpenFile.
2014-01-28 16:22:05 +09:00
Bill Currie
3efb0c538f Separate file search from loading.
QFS_LoadFile (and its wrappers) now  take a file handle rather than a
path. This will make vpath usage a little cleaner to implement.
2014-01-28 16:22:05 +09:00
Bill Currie
328b997843 Save the vpath in the model.
This will be used for restricting searches for auxiliary model files.
2014-01-28 16:22:05 +09:00
Bill Currie
a561477274 Nuke QFS_LoadStackFile().
It was used in only one place and is really not necessary (these days,
anyway).
2014-01-28 16:22:05 +09:00
Bill Currie
25a060f369 Remove the foundname param from _QFS_FOpenFile.
It's now obsolete with qfs_foundfile.
2014-01-28 16:22:05 +09:00
Bill Currie
576c73cf7c Replace qfs_file_from_pak with qfs_foundfile.
This exposes the found name, vpath and in_pak status.
2014-01-28 16:22:05 +09:00
Bill Currie
e5148493db Prepend qfs_ to file_from_pak.
A little better name-spacing.
2014-01-28 16:22:05 +09:00
Bill Currie
53d267c030 Correct some doc comments. 2014-01-28 16:22:05 +09:00
Bill Currie
6b87ddbb7c Add QFS_FindFile.
QFS_FindFile's search can be constrained to begin or end at a certain
vpath.
2014-01-28 16:22:04 +09:00
Bill Currie
a5db0b1a65 Remove the now unnecessary GLSL_CompileShaderS().
Now that the transition to using shader effects is comlete, there's no
need for the wrapper.
2014-01-28 16:22:04 +09:00
Bill Currie
4bd82eb2b8 Add some documentation.
Nowhere near adequate.
2014-01-28 16:22:04 +09:00
Bill Currie
96c40cb9a1 Add basic support for shader "effects".
Again, based on The OpenGL Shader Wrangler. The wrangling part is not used
yet, but the shader compiler has been modified to take the built up shader.
Just to keep things compiling, a wrapper has been temporarily created.
2014-01-28 16:22:03 +09:00
Bill Currie
d912970cd3 Add support for segmented text files.
The idea comes from The OpenGL Shader Wrangler
(http://prideout.net/blog/?p=11). Text files are broken up into chunks via
lines beginning with -- (^-- in regex). The chunks are optionally named
with tags of the form: [0-9A-Za-z._]+. Unnamed chunks cannot be found.
Searching for chunks looks for the longest tag that matches the beginning
of the search tag (eg, a chunk named "Vertex" will be found with a search
tag of "Vertex.foo"). Note that '.' forms the units for the searc
("Vertex.foo" will not find "Vertex.f").

Unlike glsw, this implementation does not have the concept of effects keys
as that will be separate. Also, this implementation takes strings rather
than file names (thus is more generally useful).
2014-01-28 16:22:03 +09:00
Bill Currie
247f3be0c0 Create reentrant versions of the set functions that need it.
Only the functions that allocate or delete sets or iterators.
2013-03-18 13:20:39 +09:00
Bill Currie
27bb337a60 Use set_bits_t for holding set elements.
set_bits_t is now 64 bits for x86_64 machines (in linux, anyway). This gave
qfvis a huge speed boost: from ~815s to ~720s.

Also, expose some of the set internals so custom set operators can be
created.
2013-03-16 21:26:49 +09:00