[PATCH] D49962: AMDGPU/R600: Convert kernel param loads to use PARAM_I_ADDRESS

[Jan Vesely via Phabricator via llvm-commits]

jvesely created this revision.
jvesely added a reviewer: arsenm.
Herald added subscribers: t-tye, tpr, dstuttard, yaxunl, nhaehnle, wdng, kzhuravl.

Non ext aligned i32 loads are still optimized to use CONSTANT_BUFFER_0.
Fixes remaining regressions since r337535


Repository:
  rL LLVM

https://reviews.llvm.org/D49962

Files:
  lib/Target/AMDGPU/R600ISelLowering.cpp
  lib/Target/AMDGPU/R600ISelLowering.h
  test/CodeGen/AMDGPU/kernel-args.ll
  test/CodeGen/AMDGPU/r600.extract-lowbits.ll
  test/CodeGen/AMDGPU/r600.work-item-intrinsics.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D49962.157875.patch
Type: text/x-patch
Size: 16550 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180728/8f7854f1/attachment.bin>