HomePhabricator

[AMDGPU] Support idot2 pattern.

Authored by FarhanaAleen on Aug 21 2018, 9:21 AM.

Description

[AMDGPU] Support idot2 pattern.

Summary: Transform add (mul ((i32)S0.x, (i32)S1.x),

add( mul ((i32)S0.y, (i32)S1.y), (i32)S3) => i/udot2((v2i16)S0, (v2i16)S1, (i32)S3)

Author: FarhanaAleen

Reviewed By: arsenm

Subscribers: llvm-commits, AMDGPU

Differential Revision: https://reviews.llvm.org/D50024

llvm-svn: 340295

Details

Committed
FarhanaAleenAug 21 2018, 9:21 AM
Reviewer
arsenm
Differential Revision
D50024: [AMDGPU] Support idot2 pattern.
Parents
rG95f21584a9b3: lldbtest.py: Unconditionally set the clang module cache path.
Branches
Unknown
Tags
Unknown