[PATCH] D35839: [X86] SET0 to use XMM registers where possible PR26018 PR32862

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon added a comment.

A couple of minors, but otherwise seems almost ready.



================
Comment at: test/CodeGen/X86/vec_uint_to_fp-fastmath.ll:1
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -mtriple=x86_64 -enable-unsafe-fp-math \
----------------
Don't regenerate this file (we want the symbols to still be tested) - just manually tweak the xor cases.


================
Comment at: test/CodeGen/X86/vector-shuffle-combining-avx.ll:7
 ; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx2 | FileCheck %s --check-prefix=X64
-; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx512f | FileCheck %s --check-prefix=X64
+; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx512f | FileCheck %s --check-prefix=X64AVX512F
 ;
----------------
RKSimon wrote:
> Please can improve the test prefixes to avoid so much duplication:
> ```
> ; RUN: llc < %s -mtriple=i686-unknown -mattr=+avx | FileCheck %s --check-prefix=X32 --check-prefix=X32-AVX
> ; RUN: llc < %s -mtriple=i686-unknown -mattr=+avx2 | FileCheck %s --check-prefix=X32 --check-prefix=X32-AVX
> ; RUN: llc < %s -mtriple=i686-unknown -mattr=+avx512f | FileCheck %s --check-prefix=X32 --check-prefix=X32-AVX512
> ; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx | FileCheck %s --check-prefix=X64 --check-prefix=X64-AVX
> ; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx2 | FileCheck %s --check-prefix=X64 --check-prefix=X64-AVX
> ; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx512f | FileCheck %s --check-prefix=X64 --check-prefix=X64-AVX512
> ```
This still stands


https://reviews.llvm.org/D35839