[PATCH] D39616: [AMDGPU] Remove hardcoded address space value from AMDGPULibFunc

[Yaxun Liu via Phabricator via llvm-commits]

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

AMDGPULibFunc hardcodes address space values of the old address space mapping,
which causes invalid addrspacecast instructions and undefined functions in
APPSDK sample MonteCarloAsianDP.

This patch fixes that.


https://reviews.llvm.org/D39616

Files:
  lib/Target/AMDGPU/AMDGPULibCalls.cpp
  lib/Target/AMDGPU/AMDGPULibFunc.cpp
  lib/Target/AMDGPU/AMDGPULibFunc.h
  test/CodeGen/AMDGPU/simplify-libcalls.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D39616.121534.patch
Type: text/x-patch
Size: 21946 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20171103/b2ce5293/attachment.bin>