[PATCH] D54073: [x86] allow vector load narrowing with multi-use values

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon accepted this revision.
RKSimon added a comment.
This revision is now accepted and ready to land.

LGTM - thanks



================
Comment at: test/CodeGen/X86/pr34653.ll:2
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc %s -O0 -mtriple=x86_64-unknown-unknown -mattr=+avx512f -o - | FileCheck %s
 
----------------
spatel wrote:
> This is a -O0 test, so doesn't matter? With optimization, everything disappears:
> 	
> ```
> pushq	%rbp
> movq	%rsp, %rbp
> andq	$-512, %rsp             ## imm = 0xFE00
> subq	$1024, %rsp             ## imm = 0x400
> movq	%rsp, %rdi
> callq	_test
> movq	%rbp, %rsp
> popq	%rbp
> 
> ```
Agreed - it's annoying but doesn't matter.


https://reviews.llvm.org/D54073