Skip to content

Workaround for GetKernelThreadGroupSizes returning wrong data #760

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Jun 9, 2020
Merged

Workaround for GetKernelThreadGroupSizes returning wrong data #760

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
2026b3f
Hard code group size
FrancescoC-unity Jun 8, 2020
6eec2bc
changelog
FrancescoC-unity Jun 8, 2020
1028761
Add comment
FrancescoC-unity Jun 9, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions com.unity.render-pipelines.high-definition/CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -649,6 +649,7 @@ and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.
- Fix an issue with the half resolution Mode (performance)
- Fix an issue with the color intensity of emissive for performance rtgi
- Fixed issue with rendering being mostly broken when target platform disables VR.
- Workaround an issue caused by GetKernelThreadGroupSizes failing to retrieve correct group size.

### Changed
- Improve MIP selection for decals on Transparents
Expand Down
6 changes: 5 additions & 1 deletion com.unity.render-pipelines.high-definition/Runtime/PostProcessing/PostProcessSystem.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2725,7 +2725,11 @@ static void DoColorGrading( in ColorGradingParameters parameters,

// Generate the lut
// See the note about Metal & Intel in LutBuilder3D.compute
builderCS.GetKernelThreadGroupSizes(builderKernel, out uint threadX, out uint threadY, out uint threadZ);
// GetKernelThreadGroupSizes is currently broken on some binary versions.
//builderCS.GetKernelThreadGroupSizes(builderKernel, out uint threadX, out uint threadY, out uint threadZ);
uint threadX = 4;
uint threadY = 4;
uint threadZ = 4;
cmd.DispatchCompute(builderCS, builderKernel,
(int)((parameters.lutSize + threadX - 1u) / threadX),
(int)((parameters.lutSize + threadY - 1u) / threadY),
Expand Down
1 change: 1 addition & 0 deletions ...nity.render-pipelines.high-definition/Runtime/PostProcessing/Shaders/LutBuilder3D.compute
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -231,6 +231,7 @@ float3 Tonemap(float3 colorLinear)
// allow anything higher than 256 threads. We'll use 4x4x4 then.
// Ultimately it would nice to expose `maxTotalThreadsPerThreadgroup` for Metal...
// Source: https://developer.apple.com/metal/Metal-Feature-Set-Tables.pdf
// It is important to keep this in sync with the group-size declared in PostProcessSystem.cs
[numthreads(4,4,4)]
void KBuild(uint3 dispatchThreadId : SV_DispatchThreadID)
{
Expand Down