[PATCH] D61986: AMDGPU/GlobalISel: Use subreg index instead of extra unmerge

[Matt Arsenault via Phabricator via llvm-commits]

arsenm created this revision.
arsenm added reviewers: rampitec, tstellar, scott.linder.
Herald added subscribers: Petar.Avramovic, arphaman, t-tye, tpr, dstuttard, kristof.beyls, rovka, yaxunl, nhaehnle, wdng, jvesely, kzhuravl.
arsenm added a parent revision: D61985: AMDGPU/GlobalISel: Use waterfall loop for buffer_load.

This saves instructions and extra steps, but I'm not sure about
introducing subregister indexes at this point.


https://reviews.llvm.org/D61986

Files:
  lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
  test/CodeGen/AMDGPU/GlobalISel/regbankselect-amdgcn-buffer-load.mir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D61986.199728.patch
Type: text/x-patch
Size: 9762 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190516/1f4d52d1/attachment.bin>