[all-commits] [llvm/llvm-project] 1dfd1b: [AMDGPU] Tune threshold for cmp/select vector lowe...

[Stanislav Mekhanoshin via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 1dfd1b3e4b2b4ba73416bf9fb5cc8fa837116ce3
      https://github.com/llvm/llvm-project/commit/1dfd1b3e4b2b4ba73416bf9fb5cc8fa837116ce3
  Author: Stanislav Mekhanoshin <Stanislav.Mekhanoshin at amd.com>
  Date:   2020-05-21 (Thu, 21 May 2020)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    M llvm/test/CodeGen/AMDGPU/extract_vector_dynelt.ll
    M llvm/test/CodeGen/AMDGPU/insert_vector_dynelt.ll

  Log Message:
  -----------
  [AMDGPU] Tune threshold for cmp/select vector lowering

It was set in total vector size while the idea was to limit
a number of instructions. Now it started to work with doubles
and thresholds needs to be updated.

Differential Revision: https://reviews.llvm.org/D80322