[PATCH] D17875: AMDGPU: Cost model for basic integer operations

[Matt Arsenault via llvm-commits]

arsenm created this revision.
arsenm added a reviewer: tstellarAMD.
arsenm added a subscriber: llvm-commits.
Herald added a subscriber: arsenm.

This resolves bug 21148 by preventing promotion to i64 induction variables.

http://reviews.llvm.org/D17875

Files:
  lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
  test/Analysis/CostModel/AMDGPU/add-sub.ll
  test/Analysis/CostModel/AMDGPU/add.ll
  test/Analysis/CostModel/AMDGPU/bit-ops.ll
  test/Analysis/CostModel/AMDGPU/mul.ll
  test/Analysis/CostModel/AMDGPU/shifts.ll
  test/Transforms/IndVarSimplify/AMDGPU/lit.local.cfg
  test/Transforms/IndVarSimplify/AMDGPU/no-widen-to-i64.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D17875.49792.patch
Type: text/x-patch
Size: 20221 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20160304/a268d66d/attachment.bin>