a82ca1cd1b
We try to only use X32 for gnux32 triple tests. Use no_x86_scrub_mem_shuffle so the test shows updated shuffle intermediate and the +4 offset into the constant pool vector entry
22 lines
1.0 KiB
LLVM
22 lines
1.0 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --no_x86_scrub_mem_shuffle --version 4
|
|
; RUN: llc -mtriple=i686-unknown-unknown -mattr=+sse4.1 < %s | FileCheck %s --check-prefix=X86
|
|
; RUN: llc -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 < %s | FileCheck %s --check-prefix=X64
|
|
|
|
; Test for case where insertps folds the load of an insertion element from a constant pool.
|
|
|
|
define <4 x float> @fold_from_constantpool(<4 x float> %a) {
|
|
; X86-LABEL: fold_from_constantpool:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: insertps $0, {{\.?LCPI[0-9]+_[0-9]+}}+4, %xmm0 # xmm0 = mem[0],xmm0[1,2,3]
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: fold_from_constantpool:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: insertps $0, {{\.?LCPI[0-9]+_[0-9]+}}+4(%rip), %xmm0 # xmm0 = mem[0],xmm0[1,2,3]
|
|
; X64-NEXT: retq
|
|
%1 = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %a, <4 x float> <float 0.0, float 1.0, float 0.0, float 0.0>, i8 64)
|
|
ret <4 x float> %1
|
|
}
|
|
|
|
declare <4 x float> @llvm.x86.sse41.insertps(<4 x float>, <4 x float>, i8) nounwind readnone
|