[llvm] r322385 - [X86][SSE] Force blend domains on stack folding tests
Simon Pilgrim via llvm-commits
llvm-commits at lists.llvm.org
Fri Jan 12 10:05:29 PST 2018
Author: rksimon
Date: Fri Jan 12 10:05:29 2018
New Revision: 322385
URL: http://llvm.org/viewvc/llvm-project?rev=322385&view=rev
Log:
[X86][SSE] Force blend domains on stack folding tests
Modified:
llvm/trunk/test/CodeGen/X86/stack-folding-fp-avx1.ll
llvm/trunk/test/CodeGen/X86/stack-folding-fp-sse42.ll
Modified: llvm/trunk/test/CodeGen/X86/stack-folding-fp-avx1.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/stack-folding-fp-avx1.ll?rev=322385&r1=322384&r2=322385&view=diff
==============================================================================
--- llvm/trunk/test/CodeGen/X86/stack-folding-fp-avx1.ll (original)
+++ llvm/trunk/test/CodeGen/X86/stack-folding-fp-avx1.ll Fri Jan 12 10:05:29 2018
@@ -223,7 +223,9 @@ define <2 x double> @stack_fold_blendpd(
;CHECK: vblendpd $2, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
%1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
%2 = select <2 x i1> <i1 1, i1 0>, <2 x double> %a0, <2 x double> %a1
- ret <2 x double> %2
+ ; fadd forces execution domain
+ %3 = fadd <2 x double> %2, <double 0x0, double 0x0>
+ ret <2 x double> %3
}
define <4 x double> @stack_fold_blendpd_ymm(<4 x double> %a0, <4 x double> %a1) {
@@ -231,15 +233,18 @@ define <4 x double> @stack_fold_blendpd_
;CHECK: vblendpd $6, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
%1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
%2 = select <4 x i1> <i1 1, i1 0, i1 0, i1 1>, <4 x double> %a0, <4 x double> %a1
- ret <4 x double> %2
-}
+ ; fadd forces execution domain
+ %3 = fadd <4 x double> %2, <double 0x0, double 0x0, double 0x0, double 0x0>
+ ret <4 x double> %3}
define <4 x float> @stack_fold_blendps(<4 x float> %a0, <4 x float> %a1) {
;CHECK-LABEL: stack_fold_blendps
;CHECK: vblendps $6, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
%1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
%2 = select <4 x i1> <i1 1, i1 0, i1 0, i1 1>, <4 x float> %a0, <4 x float> %a1
- ret <4 x float> %2
+ ; fadd forces execution domain
+ %3 = fadd <4 x float> %2, <float 0x0, float 0x0, float 0x0, float 0x0>
+ ret <4 x float> %3
}
define <8 x float> @stack_fold_blendps_ymm(<8 x float> %a0, <8 x float> %a1) {
@@ -247,7 +252,9 @@ define <8 x float> @stack_fold_blendps_y
;CHECK: vblendps $102, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
%1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
%2 = select <8 x i1> <i1 1, i1 0, i1 0, i1 1, i1 1, i1 0, i1 0, i1 1>, <8 x float> %a0, <8 x float> %a1
- ret <8 x float> %2
+ ; fadd forces execution domain
+ %3 = fadd <8 x float> %2, <float 0x0, float 0x0, float 0x0, float 0x0, float 0x0, float 0x0, float 0x0, float 0x0>
+ ret <8 x float> %3
}
define <2 x double> @stack_fold_blendvpd(<2 x double> %a0, <2 x double> %a1, <2 x double> %c) {
Modified: llvm/trunk/test/CodeGen/X86/stack-folding-fp-sse42.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/stack-folding-fp-sse42.ll?rev=322385&r1=322384&r2=322385&view=diff
==============================================================================
--- llvm/trunk/test/CodeGen/X86/stack-folding-fp-sse42.ll (original)
+++ llvm/trunk/test/CodeGen/X86/stack-folding-fp-sse42.ll Fri Jan 12 10:05:29 2018
@@ -135,7 +135,9 @@ define <2 x double> @stack_fold_blendpd(
;CHECK: blendpd $2, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
%1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
%2 = select <2 x i1> <i1 1, i1 0>, <2 x double> %a0, <2 x double> %a1
- ret <2 x double> %2
+ ; fadd forces execution domain
+ %3 = fadd <2 x double> %2, <double 0x0, double 0x0>
+ ret <2 x double> %3
}
define <4 x float> @stack_fold_blendps(<4 x float> %a0, <4 x float> %a1) {
@@ -143,7 +145,9 @@ define <4 x float> @stack_fold_blendps(<
;CHECK: blendps $6, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
%1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
%2 = select <4 x i1> <i1 1, i1 0, i1 0, i1 1>, <4 x float> %a0, <4 x float> %a1
- ret <4 x float> %2
+ ; fadd forces execution domain
+ %3 = fadd <4 x float> %2, <float 0x0, float 0x0, float 0x0, float 0x0>
+ ret <4 x float> %3
}
define <2 x double> @stack_fold_blendvpd(<2 x double> %a0, <2 x double> %a1, <2 x double> %c) {
More information about the llvm-commits
mailing list