voices. Because this wasn't done, note_on() could try to access
the sample for a voice that had never been played yet and access
random memory. There may be other places where it's a problem, but
this is where I noticed it, by chance.
SVN r1182 (trunk)
- Fixed: (SBARINFO patch) DrawInventoryBar has a missing argument in
one of its drawgraphic calls.
- Added Gez's patch for Heretic's GIMME cheat.
- Externalized some cheat strings.
SVN r1178 (trunk)
- Fixed: The recent changes in the DECORATE parser require the special parameter
to A_CallSpecial to be an expression, not a constant.
- Removed game filters from old style decorations. No WAD in existence ever
used them and removing them allows to make the parser more robust.
SVN r1170 (trunk)
to reference its arguments directly, except 'self'. This may be helpful if it
becomes necessary to restructure this code once DoomScript becomes real.
- Removed the last remaining global variable related to action functions:
pStateCall. Everything is passed properly as function arguments now.
- fixed: PlayerChunk was missing the NOSKIN flag.
SVN r1169 (trunk)
the texture manager even if it is from Raven's IWADs.
- Removed code related to internal ActorInfo definitions from dobjtype.cpp.
- removed unneeded file dehackedactions.h
SVN r1162 (trunk)
After finding out that gas does have directives to describe the .eh_frame
metadata, I figured that would be significantly easier and quicker than
trying to locate all the scattered docs I would need to read to figure out
how to construct it by hand. Unfortunately, this now means I have to
maintain two versions of exactly the same code. :( (But unless I add 32-bit
color rendering in the future, the chances that I will have to touch it
again are quite slim.)
SVN r1159 (trunk)
internal actor classes which no longer exist.
- Changed the state structure so that the Tics value doesn't need to be hacked
into misc1 with SF_BIGTIC anymore.
- Changed sprite processing so that sprite names are converted to indices
during parsing so that an additional postprocessing step is no longer needed.
- Fixed: Sprite names in DECORATE were case sensitive.
- Exported AActor's defaults to DECORATE and removed all code for the
internal property parser which is no longer needed.
SVN r1146 (trunk)
- Made the speed a parameter to A_RaiseMobj and A_SinkMobj and deleted
GetRaiseSpeed and GetSinkSpeed.
- Added some remaining DECORATE conversions for Hexen by Karate Chris.
SVN r1144 (trunk)
- Removed AT_GAME_SET because it's no longer used anywhere.
- Converted the last remaining global classes to DECORATE.
- Fixed: Inventory.PickupFlash requires an class name as parameter not an
integer. Some Hexen definitions got it wrong.
- Converted Hexen's Pig to DECORATE.
- Replaced the ActorInfo definitions of all internal inventory classes with
DECORATE definitions.
- Added option to specify a powerup's duration in second by using a negative
number.
SVN r1137 (trunk)
* Added: The ability to have drawkeybar auto detect spacing.
* Added: Offset parameter to drawkeybar to allow two key bars with
different keys.
* Added: Multi-row/column keybar parameters. Spacing can also be auto.
These defualt to left to right/top to bottom but can be switched.
* Added: Drawshadow flag to drawnumber. This will draw a solid color and
translucent number under the normal number.
* Added: hexenarmor to drawimage. This takes a parameter for a hexen
armor type and will fade the image like the hexen status bar.
* Added: centerbottom offset to draw(switchable)image.
* Added: translucent flag to drawinventorybar.
* Fixed: Accidentally removed flag from DrawTexture that allowed negative
coordinates to work with fullscreenoffsets. Hopefully this is the last
major bug in the fullscreenoffsets system.
SVN r1135 (trunk)
registers AMD64 provides, this routine still needs to be written as self-
modifying code for maximum performance. The additional registers do allow
for further optimization over the x86 version by allowing all four pixels
to be in flight at the same time. The end result is that AMD64 ASM is about
2.18 times faster than AMD64 C and about 1.06 times faster than x86 ASM.
(For further comparison, AMD64 C and x86 C are practically the same for
this function.) Should I port any more assembly to AMD64, mvlineasm4 is the
most likely candidate, but it's not used enough at this point to bother.
Also, this may or may not work with Linux at the moment, since it doesn't
have the eh_handler metadata. Win64 is easier, since I just need to
structure the function prologue and epilogue properly and use some
assembler directives/macros to automatically generate the metadata. And
that brings up another point: You need YASM to assemble the AMD64 code,
because NASM doesn't support the Win64 metadata directives.
- Added an SSE version of DoBlending. This is strictly C intrinsics.
VC++ still throws around unneccessary register moves. GCC seems to be
pretty close to optimal, requiring only about 2 cycles/color. They're
both faster than my hand-written MMX routine, so I don't need to feel
bad about not hand-optimizing this for x64 builds.
- Removed an extra instruction from DoBlending_MMX, transposed two
instructions, and unrolled it once, shaving off about 80 cycles from the
time required to blend 256 palette entries. Why? Because I tried writing
a C version of the routine using compiler intrinsics and was appalled by
all the extra movq's VC++ added to the code. GCC was better, but still
generated extra instructions. I only wanted a C version because I can't
use inline assembly with VC++'s x64 compiler, and x64 assembly is a bit
of a pain. (It's a pain because Linux and Windows have different calling
conventions, and you need to maintain extra metadata for functions.) So,
the assembly version stays and the C version stays out.
- Removed all the pixel doubling r_detail modes, since the one platform they
were intended to assist (486) actually sees very little benefit from them.
- Rewrote CheckMMX in C and renamed it to CheckCPU.
- Fixed: CPUID function 0x80000005 is specified to return detailed L1 cache
only for AMD processors, so we must not use it on other architectures, or
we end up overwriting the L1 cache line size with 0 or some other number
we don't actually understand.
SVN r1134 (trunk)
