[PATCH] D50200: AMDGPU: Handle "uniform-work-group-size" attribute
Aakanksha Patil via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Thu Nov 29 11:55:13 PST 2018
aakanksha555 added inline comments.
================
Comment at: lib/Target/AMDGPU/AMDGPUAnnotateKernelFeatures.cpp:261-263
+ // If the attribute is absent, set it as false
+ Caller.addFnAttr("uniform-work-group-size", "false");
+ Callee.addFnAttr("uniform-work-group-size", "false");
----------------
arsenm wrote:
> Can you avoid adding the attribute if not originally present?
If I don't add the attribute if not originally present, it can create a discrepancy in certain scenarios.
For eg. A function is called by two kernels, one without the attribute and the other with uniform-work-group-attribute = true. The function will be set as uniform-work-group-attribute = true, which may not be the correct approach.
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D50200/new/
https://reviews.llvm.org/D50200
More information about the llvm-commits
mailing list