[PATCH] D37348: Implement custom lowering for ISD::CTTZ_ZERO_UNDEF and ISD::CTTZ.
Wei Ding via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Wed Oct 11 16:27:40 PDT 2017
wdng added inline comments.
================
Comment at: test/CodeGen/AMDGPU/cttz_zero_undef.ll:175-176
+; SI-NOSDWA-DAG: v_or_b32_e32
+; SI-SDWA-DAG: v_or_b32_sdwa
+; SI-SDWA-DAG: v_or_b32_sdwa
+; SI: v_or_b32_e32 [[VAL1:v[0-9]+]], v{{[0-9]+}}, v{{[0-9]+}}
----------------
arsenm wrote:
> Using 2 -DAGs with identical lines doesn't do anything. It will pass with only one
No, it won't work if I remove the -DAG. As the generated instructions get interleaved with each other.
Repository:
rL LLVM
https://reviews.llvm.org/D37348
More information about the llvm-commits
mailing list