[PATCH] D141544: [AMDGPU] Mark wmma intrinsics as source of divergence
Stanislav Mekhanoshin via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Wed Jan 11 13:30:41 PST 2023
rampitec created this revision.
rampitec added reviewers: arsenm, foad.
Herald added subscribers: kosarev, StephenFan, kerbowa, hiraditya, tpr, dstuttard, yaxunl, jvesely, kzhuravl.
Herald added a project: All.
rampitec requested review of this revision.
Herald added a subscriber: wdng.
Herald added a project: LLVM.
I do not believe any code can hit this, but these do not give
a uniform answer with all unifirm sources.
https://reviews.llvm.org/D141544
Files:
llvm/lib/Target/AMDGPU/AMDGPUSearchableTables.td
Index: llvm/lib/Target/AMDGPU/AMDGPUSearchableTables.td
===================================================================
--- llvm/lib/Target/AMDGPU/AMDGPUSearchableTables.td
+++ llvm/lib/Target/AMDGPU/AMDGPUSearchableTables.td
@@ -362,6 +362,12 @@
def : SourceOfDivergence<int_amdgcn_smfmac_f32_32x32x32_bf8_fp8>;
def : SourceOfDivergence<int_amdgcn_smfmac_f32_32x32x32_fp8_bf8>;
def : SourceOfDivergence<int_amdgcn_smfmac_f32_32x32x32_fp8_fp8>;
+def : SourceOfDivergence<int_amdgcn_wmma_f32_16x16x16_f16>;
+def : SourceOfDivergence<int_amdgcn_wmma_f32_16x16x16_bf16>;
+def : SourceOfDivergence<int_amdgcn_wmma_f16_16x16x16_f16>;
+def : SourceOfDivergence<int_amdgcn_wmma_bf16_16x16x16_bf16>;
+def : SourceOfDivergence<int_amdgcn_wmma_i32_16x16x16_iu8>;
+def : SourceOfDivergence<int_amdgcn_wmma_i32_16x16x16_iu4>;
// The dummy boolean output is divergent from the IR's perspective,
// but the mask results are uniform. These produce a divergent and
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D141544.488362.patch
Type: text/x-patch
Size: 960 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230111/47b54b37/attachment.bin>
More information about the llvm-commits
mailing list