]> granicus.if.org Git - llvm/commit
[AMDGPU] SDWA: add support for GFX9 in peephole pass
authorSam Kolton <Sam.Kolton@amd.com>
Thu, 22 Jun 2017 06:26:41 +0000 (06:26 +0000)
committerSam Kolton <Sam.Kolton@amd.com>
Thu, 22 Jun 2017 06:26:41 +0000 (06:26 +0000)
commite88fc4046f70131e67411cee503201e3e9665891
tree11cc099de3b8fcc1bd9b79371e38188c7b0213b6
parente696366e67bc9f94db16e20821115d51a3f1d41b
[AMDGPU] SDWA: add support for GFX9 in peephole pass

Summary:
Added support based on merged SDWA pseudo instructions. Now peephole allow one scalar operand, omod and clamp modifiers.
Added several subtarget features for GFX9 SDWA.
This diff also contains changes from D34026.
Depends D34026

Reviewers: vpykhtin, rampitec, arsenm

Subscribers: kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye

Differential Revision: https://reviews.llvm.org/D34241

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@305986 91177308-0d34-0410-b5e6-96231b3b80d8
12 files changed:
lib/Target/AMDGPU/AMDGPU.td
lib/Target/AMDGPU/AMDGPUSubtarget.cpp
lib/Target/AMDGPU/AMDGPUSubtarget.h
lib/Target/AMDGPU/SIInstrInfo.cpp
lib/Target/AMDGPU/SIPeepholeSDWA.cpp
lib/Target/AMDGPU/VOPInstructions.td
test/CodeGen/AMDGPU/fneg.f16.ll
test/CodeGen/AMDGPU/fpext.f16.ll
test/CodeGen/AMDGPU/llvm.rint.f16.ll
test/CodeGen/AMDGPU/sdwa-gfx9.mir [new file with mode: 0644]
test/CodeGen/AMDGPU/sdwa-peephole.ll
test/CodeGen/AMDGPU/sdwa-scalar-ops.mir