projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: b50327e) | patch
[AMDGPU] Cast sub-dword elements to i32 in concat_vectors
authorStanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com>
Mon, 9 Jan 2023 22:19:46 +0000 (14:19 -0800)
committerStanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com>
Mon, 9 Jan 2023 23:35:49 +0000 (15:35 -0800)
commitc8ed36281a929c34477ffc391a8a3e84d1a07a92
tree1ec03a8019ed6eb5f873822ca4784735e7030c57tree | snapshot
parentb50327eea603358846743ff784ddf2099fc35587commit | diff
[AMDGPU] Cast sub-dword elements to i32 in concat_vectors

This produces better code by avoiding repacking in some cases.

Fixes: SWDEV-373436

Differential Revision: https://reviews.llvm.org/D141329
llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/vector_shuffle.packed.ll diff | blob | history
Domain: System / Toolchain;