AMDGPU: Custom lower v4i16/v4f16 vector operations
authorMatt Arsenault <Matthew.Arsenault@amd.com>
Wed, 16 May 2018 11:47:30 +0000 (11:47 +0000)
committerMatt Arsenault <Matthew.Arsenault@amd.com>
Wed, 16 May 2018 11:47:30 +0000 (11:47 +0000)
commit67a9815a5c9d0355020e17c0a494f7eaa1df21fe
tree3422ea805c0a69f22bcdec55e8c684b282498b59
parentca22d427b94de946d4ef32b8acbdfb7e62e7cfa4
AMDGPU: Custom lower v4i16/v4f16 vector operations

Avoids stack access.

Also handle extract hi elt pattern from truncate + shift
to avoid a couple test regressions.

llvm-svn: 332453
llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
llvm/lib/Target/AMDGPU/AMDGPUISelLowering.h
llvm/lib/Target/AMDGPU/SIISelLowering.cpp
llvm/lib/Target/AMDGPU/SIISelLowering.h
llvm/test/CodeGen/AMDGPU/extload-align.ll
llvm/test/CodeGen/AMDGPU/extract_vector_elt-f16.ll
llvm/test/CodeGen/AMDGPU/extract_vector_elt-i16.ll
llvm/test/CodeGen/AMDGPU/insert_vector_elt.ll
llvm/test/CodeGen/AMDGPU/insert_vector_elt.v2i16.ll
llvm/test/CodeGen/AMDGPU/min.ll