[PATCH] D149740: [InlineCost][TargetTransformInfo][AMDGPU] Consider cost of alloca instructions in the caller (1/2)

[Juan Manuel Martinez Caamaño via Phabricator via llvm-commits]

jmmartinez created this revision.
Herald added subscribers: kosarev, ChuanqiXu, haicheng, hiraditya, tpr, dstuttard, yaxunl, kzhuravl.
Herald added a project: All.
jmmartinez requested review of this revision.
Herald added subscribers: llvm-commits, wdng.
Herald added a project: LLVM.

On AMDGPU, alloca instructions have penalty that can
be avoided when SROA is applied after inlining.

This patch introduces the default implementation of
TargetTransformInfo::getCallerAllocaCost.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D149740

Files:
  llvm/include/llvm/Analysis/TargetTransformInfo.h
  llvm/include/llvm/Analysis/TargetTransformInfoImpl.h
  llvm/include/llvm/CodeGen/BasicTTIImpl.h
  llvm/lib/Analysis/InlineCost.cpp
  llvm/lib/Analysis/TargetTransformInfo.cpp

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D149740.519037.patch
Type: text/x-patch
Size: 4141 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230503/33b10224/attachment.bin>