[PATCH] D26824: AMDGPU: Make m0 unallocatable

[Matt Arsenault via llvm-commits]

arsenm created this revision.
arsenm added a subscriber: llvm-commits.
Herald added subscribers: tony-tye, yaxunl, nhaehnle, wdng, kzhuravl, qcolombet.
Herald added a reviewer: tstellarAMD.

m0 may need to be written for spill code, so
we don't want general code uses relying on the
value stored in it.

      

This introduces a few code quality regressions where copies
from m0 are not coalesced into copies of a copy of m0.


https://reviews.llvm.org/D26824

Files:
  lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp
  lib/Target/AMDGPU/SIISelLowering.cpp
  lib/Target/AMDGPU/SIInstrInfo.cpp
  lib/Target/AMDGPU/SIRegisterInfo.cpp
  lib/Target/AMDGPU/SIRegisterInfo.td
  lib/Target/AMDGPU/SIWholeQuadMode.cpp
  test/CodeGen/AMDGPU/detect-dead-lanes.mir
  test/CodeGen/AMDGPU/inline-constraints.ll
  test/CodeGen/AMDGPU/llvm.amdgcn.readfirstlane.ll
  test/CodeGen/AMDGPU/llvm.amdgcn.readlane.ll
  test/CodeGen/AMDGPU/read_register.ll
  test/CodeGen/AMDGPU/spill-m0.ll
  test/CodeGen/MIR/AMDGPU/si-fix-sgpr-copies.mir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D26824.78446.patch
Type: text/x-patch
Size: 14838 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20161118/c1358738/attachment.bin>