The PE build of vkd3d uses Windows critical sections for synchronisation, and
these slow down on the very high lock/unlock rate during multithreaded
descriptor copying in Shadow of the Tomb Raider. This patch speeds up the
demo by about 8%. By comparison, using SRW locks in the allocators and
locking them for read only where applicable is about 4% faster.
--
https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/67
Today, the Rich Edit control handles the WM_IME_STARTCOMPOSITION message
by the computing the composition start position (imeStartIndex) _before_
calling ME_DeleteSelection(), which shifts the character positions after
the range of deletion. If the selection were not empty, imeStartIndex
immediately becomes stale, since it does not take into account the
number of deleted characters before it.
Fix this by computing imeStartIndex after the ME_DeleteSelection() call.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54318
cc/ @aricstewart (has no Developer access to wine/wine; cannot be assigned as a reviewer yet)
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/2001
## Summary
Tests for UrlFixupW and semi stub function implementaion.
## Test results
- Tests are 100% ok on my local WIN 10 pc. Wine test bot seems to be okay too.
- Wine tests on debian machine
| function type | result |
| ------ | ------ |
| stub | 0020:url: 69802 tests executed (0 marked as todo, 0 as flaky, **200** failures), 0 skipped. |
| semi stub| 0020:url: 69802 tests executed (0 marked as todo, 0 as flaky, **100** failures), 0 skipped. |
## Todos
For follow up merge request(s).
- [ ] Add more tests for priority investigation, who url scheme fix up is working on windows.
- [ ] Implement missing parts of fix up.
## Conclusion
Semi stub implementation is an improvement to the existing solution. A grep showed that wine internals do not call this function, therefore it should not introduce any new bugs. Windows applications using/relying on the function should now work better, if not they may reveal already existing bugs.
--
v5: shlwapi: fixup leading and trailing typos in UrlFixupW
shlwapi: check for colon in UrlFixupW
kernelbase: Restructure UrlFixupW()
shlwapi/tests: Add tests for UrlFixupW.
https://gitlab.winehq.org/wine/wine/-/merge_requests/1825
First relevant commit:
windows.media.speech: Add a worker thread to the recognition session.
--
v7: windows.media.speech: Store recorded audio in a temporary ringbuffer.
windows.media.speech: Partially implement the speech recognizer state.
windows.media.speech: Add an audio capturing system.
https://gitlab.winehq.org/wine/wine/-/merge_requests/1948
Call MoveFileWithProgress from MoveFileTransacted such that the stub functions.
This enables saving for Affinity Photo.
--
v5: kernel32: Unstub MoveFileTransactedW, Use MoveFileWithProgressW
https://gitlab.winehq.org/wine/wine/-/merge_requests/1444
If a hlsl_ir_load loads a variable whose components are stored from different
instructions, copy propagation doesn't replace it.
But if all these instructions are constants (which currently is the case
for value constructors), the load can be replaced with a constant value, which
is what the first patch of this series does.
For instance, this shader:
```
sampler s;
Texture2D t;
float4 main() : sv_target
{
return t.Gather(s, float2(0.6, 0.6), int2(0, 0));
}
```
results in the following IR before applying the patch:
```
float | 6.00000024e-01
float | 6.00000024e-01
uint | 0
| = (<constructor-2>[@4].x @2)
uint | 1
| = (<constructor-2>[@6].x @3)
float2 | <constructor-2>
int | 0
int | 0
uint | 0
| = (<constructor-5>[@11].x @9)
uint | 1
| = (<constructor-5>[@13].x @10)
int2 | <constructor-5>
float4 | gather_red(resource = t, sampler = s, coords = @8, offset = @15)
| return
| = (<output-sv_target0> @16)
```
and this IR afterwards:
```
float2 | {6.00000024e-01 6.00000024e-01 }
int2 | {0 0 }
float4 | gather_red(resource = t, sampler = s, coords = @2, offset = @3)
| return
| = (<output-sv_target0> @4)
```
This is required to write texel_offsets as aoffimmi modifiers in the sm4 backend, since it expects the texel_offset arguments to be hlsl_ir_constant.
This series also:
* Allows Gather() methods to use aoffimmi modifiers instead of an additional source register (which is the only way allowed for shader model 4.1), when possible.
* Adds support to texel_offsets in the Load() method via aoffimmi modifiers (the only allowed method).
--
v10: vkd3d-shader/hlsl: Fold swizzle chains.
vkd3d-shader/hlsl: Apply copy propagation to swizzled loads.
vkd3d-shader/hlsl: Use aoffimmis when writing gather resource loads.
vkd3d-shader/hlsl: Replace loads with constants in copy prop.
vkd3d-shader/hlsl: Synthesize the swizzle and replace the instruction inside of copy_propagation_compute_replacement().
vkd3d-shader/hlsl: Call copy_propagation_get_value() directly in copy_propagation_transform_object_load().
vkd3d-shader/hlsl: Add some swizzle manipulation definitions.
tests: Test constant propagation through swizzles.
vkd3d-shader/hlsl: Support offset argument for the texture Load() method.
tests: Test offset argument for the texture Load() method.
https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/51