7594187538
softgpu: Skip sample lookup if masked.
...
Was hoping making other things faster would make this unnecessary or
worse, but it hasn't seemed to. This gives a pretty decent improvement in
most places (~4%.)
2022-01-02 13:47:14 -08:00
a0fe4d06bf
softgpu: Stop specializing on miplevels.
...
Now that samplerjit is processing mips, it no longer helps. Just
complexity now.
2022-01-02 13:47:14 -08:00
e4673a5fa4
softgpu: Separately profile verts and lighting.
2022-01-02 13:46:11 -08:00
d3f0af7458
Merge pull request #15273 from unknownbrackets/softjit-bloom
...
Optimize software renderer handling of common bloom operations
2022-01-02 18:11:07 +01:00
c07ca2d89d
Merge pull request #15272 from unknownbrackets/softgpu-meminfo
...
softgpu: Add code for tracking GPU writes
2022-01-02 18:09:16 +01:00
c7062d7063
Merge pull request #15271 from unknownbrackets/samplerjit-color16
...
samplerjit: Decode colors in parallel
2022-01-02 17:55:46 +01:00
a259761262
samplerjit: Use nearest func in fast path too.
...
This uses the more optimal tex funcs.
2022-01-02 08:48:16 -08:00
ba17f538d6
softjit: Avoid const temp registers.
...
Was trying to make sure register allocation was okay in the worst case.
2022-01-02 08:47:04 -08:00
e93c709f5c
sofjit: Correctly poison memory.
...
Noticed this wasn't breakpoints when reviewing some assembly output.
2022-01-02 08:47:04 -08:00
745c35f320
softjit: Small bloom optimization.
...
Another common case, src*dst + dst*0. Can skip the add.
2022-01-02 08:47:04 -08:00
355bad666c
softjit: Optimize common case bloom blending.
...
Bloom often uses fixed ONE + ONE, which is a lot less work for us. And
bloom often runs over and over again on pixels, so saving work is good.
2022-01-02 08:47:04 -08:00
6fb5d82fe0
Merge pull request #15264 from unknownbrackets/samplerjit-vec
...
A couple more smaller samplerjit optimizations
2022-01-02 17:32:54 +01:00
496545e55c
softgpu: Add code for tracking GPU writes.
...
Unfortunately, it has a pretty noticeable speed impact, even at the basic
"assume everything's written" level. Compiled off by default, but at
least it's there.
Doesn't account for tests (i.e. alpha test skipping write) so still not
perfectly accurate.
2022-01-02 08:28:30 -08:00
0eec4e7e4d
samplerjit: Decode colors in parallel.
...
Not used in a ton of games, but a decent improvement where it is used.
2022-01-02 08:27:55 -08:00
cb1f26122d
Merge pull request #15269 from unknownbrackets/softgpu-opt
...
softgpu: Reduce interpolation if not needed
2022-01-02 09:47:19 +01:00
da38c027b5
Merge pull request #15268 from unknownbrackets/samplerjit-nearest
...
Implement nearest in samplerjit, like linear
2022-01-02 09:46:29 +01:00
025ac99f2f
softgpu: Reduce interpolation if not needed.
...
About 3% gain in some areas.
2022-01-01 18:34:04 -08:00
7060035303
samplerjit: Implement nearest in jit.
...
This uses the tex func and similar within jit.
2022-01-01 16:58:05 -08:00
91c9343e87
samplerjit: Refactor and reuse constant pool.
...
It's just here to be rip accessible, the fixed values can be output just
once.
2022-01-01 16:58:05 -08:00
40240be91c
samplerjit: Update nearest args, temp disable jit.
...
This temporarily disables jit for nearest, but refactors to use the new
arg structure. It now matches linear.
2022-01-01 16:58:05 -08:00
5f84de7de7
softjit: Small optimizations.
2022-01-01 16:58:04 -08:00
06e954fe2a
samplerjit: Create a separate fetch func.
...
This allows nearest to become more similar to linear, where it applies the
texture function.
2022-01-01 16:58:04 -08:00
3bc6009158
samplerjit: Refactor sampler ID calculation.
...
Make it the same as pixel func IDs.
2022-01-01 16:58:04 -08:00
d41e42d247
softgpu: Correct off-by-one scissor mask.
...
Fixes Brave Story in the software renderer. Was overwriting display list
data in the stride gap.
2022-01-01 16:42:36 -08:00
b35ca3d472
softgpu: Cleanup min/max tri range handling.
...
The previous looked like it had off by one errors. This is simpler.
2022-01-01 16:42:36 -08:00
e82fd3bd33
GPU: Avoid spline crashes on bad data.
...
If we get 0 prims, we can generate confusing index bounds and go out of
bounds. Similarly, if we get a crazy number of control points and fail to
allocate, we can crash.
2022-01-01 16:40:59 -08:00
12405709f0
softgpu: Skip processing scissored triangles.
...
If only one side was scissored (common), we might even put it on a thread,
which ended up as a lot of overhead. Gives 3-4% improvement in some
places.
2022-01-01 16:40:34 -08:00
6aec68aa5c
samplerjit: Correct wrong bufw at mip levels.
...
Oops, was always using the base bufw.
2022-01-01 16:40:02 -08:00
dbb015f427
samplerjit: Oops, fix Linux mipmap handling.
2022-01-01 16:40:02 -08:00
8c31f1bb38
softjit: Fix regcache error when clearing.
...
Happens for non-through clears.
2022-01-01 16:40:01 -08:00
8ea67b571b
samplerjit: Tiny dependency optimizations.
...
This had a small but measureable impact (~0.3%.)
2021-12-31 08:11:57 -08:00
fc3688d273
samplerjit: Small AVX optimization to modulate.
...
Only gives about 0.5% but it's still something.
2021-12-31 08:10:04 -08:00
244b0a86f6
Merge pull request #15262 from unknownbrackets/samplerjit-vec
...
samplerjit: Use SSSE3/SSE4 in linear filtering
2021-12-31 09:29:59 +01:00
33e9841a4a
softgpu: Skip zero size triangles.
...
These were drawing before, incorrectly, which caused artifacts.
Noticeable in Blade Dancer.
2021-12-31 00:20:12 -08:00
1addf84e90
samplerjit: Use SSSE3/SSE4 in linear filtering.
2021-12-30 23:22:56 -08:00
147b81d6f7
x64jit: Add AVX/AVX2 encodings.
...
Also fix the FMA double ones, which were passing W wrongly.
2021-12-29 19:46:26 -08:00
4bd94a4e5e
samplerjit: Pass funcs as an argument.
...
Seeing computing the ID in some profiles, so want to avoid computing per
thread/invocation.
2021-12-29 07:11:53 -08:00
28cfbe0e5a
samplerjit: Add an alternate profiling method.
...
This is more useful to group common operations together for profiling.
2021-12-29 07:11:39 -08:00
3aedea89eb
samplerjit: Correct level lookup offset.
2021-12-29 07:09:36 -08:00
bf06342f9d
samplerjit: Minor SSE4 optimizations.
...
These seem to be a bit faster.
2021-12-29 07:07:35 -08:00
631706a8ba
samplerjit: Set stackArgPos_ early.
...
Unfortunately, this has to match the value set lower...
2021-12-28 20:21:21 -08:00
74eb450e76
samplerjit: Move texture function into jit.
...
Could do this also for nearest, might end up with a third set of functions
there for a direct sample lookup (for debug funcs.)
2021-12-28 17:52:17 -08:00
940e6bb1d7
samplerjit: Lookup both mip tex values.
2021-12-28 16:22:54 -08:00
6b55d328e5
samplerjit: Use regcache for linear filtering.
...
This makes it easier to reuse for mipmap filtering.
2021-12-28 15:37:25 -08:00
cdf14c8579
samplerjit: Calculate mip level U/V/offsets.
...
Not actually doing the sampling for the second mip level in the single jit
pass yet, but close.
2021-12-28 14:12:58 -08:00
a4558a5736
samplerjit: Take texptr/bufw as arrays.
...
Prep for moving mip map sampling into linear.
2021-12-28 12:04:16 -08:00
4864850b3b
samplerjit: Handle mipmap width/height in S/T calc.
2021-12-28 11:29:29 -08:00
a84accf713
samplerjit: Move S/T calculation into jit.
...
Gives a pretty decent 5-10% improvement in many places.
2021-12-28 09:58:23 -08:00
476dfdf731
samplerjit: Add more bits for S/T, skip multiply.
...
For now, we're not using those other bits yet.
2021-12-27 18:24:37 -08:00
9cc0883d53
softgpu: Correct non-SSE T clamp.
2021-12-27 15:31:37 -08:00
Unknown W. Brackets