[PATCH] D69661: [AMDGPU] Fix vccz after v_readlane/v_readfirstlane to vcc_lo/hi

[Jay Foad via Phabricator via llvm-commits]

foad created this revision.
Herald added subscribers: hiraditya, t-tye, tpr, dstuttard, yaxunl, nhaehnle, wdng, jvesely, kzhuravl, arsenm.
Herald added a project: LLVM.
foad added reviewers: tpr, dstuttard, critson, nhaehnle, arsenm.

Up to gfx9, writes to vcc_lo and vcc_hi by instructions like
v_readlane and v_readfirstlane do not update vccz to reflect the new
value of vcc. Fix it by reusing part of the existing vccz bug handling
code, which inserts an "s_mov_b64 vcc, vcc" instruction to restore vccz
just before an instruction that needs the correct value.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D69661

Files:
  llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
  llvm/test/CodeGen/AMDGPU/reload-vcc-vccz.mir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D69661.227277.patch
Type: text/x-patch
Size: 4570 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20191031/d83a7cd2/attachment.bin>