[PATCH] D76772: [AMDGPU] Add __builtin_amdgcn_workgroup_size_x/y/z

[Yaxun Liu via Phabricator via cfe-commits]

yaxunl marked 2 inline comments as done.
yaxunl added inline comments.


================
Comment at: clang/test/CodeGenCUDA/amdgpu-workgroup-size.cu:2
+// RUN: %clang_cc1 -triple amdgcn-amd-amdhsa \
+// RUN:     -fcuda-is-device -emit-llvm -o - -x hip %s \
+// RUN:     | FileCheck %s
----------------
arsenm wrote:
> I assume the addrspacecast got optimized out? Should this disable llvm passes?
We did not emit addrspacecast here since we only need return the loaded value.

HIP by default uses -O0, therefore no need to disable llvm passes.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D76772/new/

https://reviews.llvm.org/D76772