[all-commits] [llvm/llvm-project] b0aa19: [AMDGPU] Promote recursive loads from kernel argum...

[Stanislav Mekhanoshin via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: b0aa1946dfe1d204e49b8238c4960f64a68f31d5
      https://github.com/llvm/llvm-project/commit/b0aa1946dfe1d204e49b8238c4960f64a68f31d5
  Author: Stanislav Mekhanoshin <Stanislav.Mekhanoshin at amd.com>
  Date:   2022-02-17 (Thu, 17 Feb 2022)

  Changed paths:
    M clang/test/CodeGenCUDA/amdgpu-kernel-arg-pointer-type.cu
    M llvm/lib/Target/AMDGPU/AMDGPUPromoteKernelArguments.cpp
    M llvm/test/CodeGen/AMDGPU/promote-kernel-arguments.ll

  Log Message:
  -----------
  [AMDGPU] Promote recursive loads from kernel argument to constant

Not clobbered pointer load chains are promoted to global now. That
is possible to promote these loads itself into constant address
space. Loaded pointers still need to point to global because we
need to be able to store into that pointer and because an actual
load from it may occur after a clobber.

Differential Revision: https://reviews.llvm.org/D119886