[PATCH] D100543: [AMDGPU] Fix large return values with amdgpu_gfx

[Sebastian Neubauer via Phabricator via llvm-commits]

sebastian-ne created this revision.
sebastian-ne added reviewers: arsenm, madhur13490.
Herald added subscribers: kerbowa, hiraditya, t-tye, tpr, dstuttard, yaxunl, nhaehnle, jvesely, kzhuravl.
sebastian-ne requested review of this revision.
Herald added subscribers: llvm-commits, wdng.
Herald added a project: LLVM.

Returning in memory is not supported, so fall back to sret.
Also, extend i1 and i16 to i32. Otherwise, they would be passed through
memory.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D100543

Files:
  llvm/lib/Target/AMDGPU/AMDGPUCallingConv.td
  llvm/test/CodeGen/AMDGPU/GlobalISel/irtranslator-call-return-values.ll
  llvm/test/CodeGen/AMDGPU/gfx-callable-return-types.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D100543.337694.patch
Type: text/x-patch
Size: 86105 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210415/371af3b2/attachment-0001.bin>