[PATCH] D58512: AMDGPU/GlobalISel: Insert waterfall loop for vector indexing

Matt Arsenault via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Thu Feb 21 08:10:54 PST 2019


arsenm created this revision.
arsenm added reviewers: tstellar, scott.linder, rampitec, qcolombet.
Herald added subscribers: jdoerfert, Petar.Avramovic, arphaman, t-tye, tpr, dstuttard, kristof.beyls, rovka, yaxunl, nhaehnle, wdng, jvesely, kzhuravl.

The register index can only really be an SGPR. Lie that a VGPR index
is legal, and then rewrite the instruction in a waterfall loop to
handle the index.


https://reviews.llvm.org/D58512

Files:
  lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
  lib/Target/AMDGPU/AMDGPURegisterBankInfo.h
  test/CodeGen/AMDGPU/GlobalISel/regbankselect-extract-vector-elt.mir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D58512.187801.patch
Type: text/x-patch
Size: 16877 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190221/c90e016c/attachment.bin>


More information about the llvm-commits mailing list