Ryujinx/moltenvk
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge pull request #574 from mbarriault/master

Browse Source 
Fix zero local threadgroup size in indirect rendering.
This commit is contained in:
Bill Hollings 2019-04-15 17:33:35 -04:00 committed by GitHub
commit 6bbd0eca75
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
MoltenVK/MoltenVK/Commands/MVKCmdDraw.mm
View File

@ -601,7 +601,7 @@ void MVKCmdDrawIndirect::encode(MVKCommandEncoder* cmdEncoder) {
sizeof(_drawCount),
5);
[mtlTessCtlEncoder dispatchThreadgroups: MTLSizeMake(mvkCeilingDivide(_drawCount, mtlConvertState.threadExecutionWidth), 1, 1)
threadsPerThreadgroup: MTLSizeMake(mtlConvertState.threadExecutionWidth, 0, 0)];
threadsPerThreadgroup: MTLSizeMake(mtlConvertState.threadExecutionWidth, 1, 1)];
}
cmdEncoder->finalizeDrawState(stage); // Ensure all updated state has been submitted to Metal