[all-commits] [llvm/llvm-project] a397c1: [AMDGPU] Tune perfhint analysis to account access ...

[Stanislav Mekhanoshin via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: a397c1c82f1c49106d1459c8f755e4a52743b882
      https://github.com/llvm/llvm-project/commit/a397c1c82f1c49106d1459c8f755e4a52743b882
  Author: Stanislav Mekhanoshin <Stanislav.Mekhanoshin at amd.com>
  Date:   2021-07-21 (Wed, 21 Jul 2021)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUPerfHintAnalysis.cpp
    M llvm/lib/Target/AMDGPU/AMDGPUPerfHintAnalysis.h
    M llvm/test/CodeGen/AMDGPU/perfhint.ll

  Log Message:
  -----------
  [AMDGPU] Tune perfhint analysis to account access width

A function with less memory instructions but wider access
is the same as a function with more but narrower accesses
in terms of memory boundness. In fact the pass would give
different answers before and after vectorization without
this change.

Differential Revision: https://reviews.llvm.org/D105651