[PATCH] D80129: AMDGPU: Handle rewriting ptrmask for more address spaces

[Matt Arsenault via Phabricator via llvm-commits]

arsenm created this revision.
arsenm added reviewers: reshabh, rampitec, b-sumner.
Herald added subscribers: kerbowa, hiraditya, t-tye, tpr, dstuttard, yaxunl, nhaehnle, wdng, jvesely, kzhuravl.
Herald added a project: LLVM.
arsenm added a parent revision: D80038: InferAddressSpaces: Handle ptrmask intrinsic.
arsenm updated this revision to Diff 264642.
arsenm added a comment.

Adjust mask type to canonical pointer size


If this mask only clears bits in the low 32-bit half of a flat
pointer, these bits are always preserved in the result address
space. If the high bits are modified, they may need to be preserved
for some kind of user pointer tagging.


https://reviews.llvm.org/D80129

Files:
  llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
  llvm/test/Transforms/InferAddressSpaces/AMDGPU/ptrmask.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D80129.264642.patch
Type: text/x-patch
Size: 12147 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200518/c9bf4515/attachment.bin>