caio/clang-p2996
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
9f5ef42cbd355fcdbda5ced9ccd21cb65dd67d0f
clang-p2996/llvm/test/CodeGen/X86/fminimum-fmaximum.ll

1959 lines
68 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefixes=SSE2
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefixes=AVX,AVX1
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f | FileCheck %s --check-prefixes=AVX,AVX512,AVX512F
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512dq | FileCheck %s --check-prefixes=AVX,AVX512,AVX512DQ
; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+avx | FileCheck %s --check-prefixes=X86
declare float @llvm.maximum.f32(float, float)
declare double @llvm.maximum.f64(double, double)
declare float @llvm.minimum.f32(float, float)
declare double @llvm.minimum.f64(double, double)
declare <2 x double> @llvm.minimum.v2f64(<2 x double>, <2 x double>)
declare <4 x float> @llvm.maximum.v4f32(<4 x float>, <4 x float>)
;
; fmaximum
;
define float @test_fmaximum(float %x, float %y) nounwind {
; SSE2-LABEL: test_fmaximum:
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: movd %xmm0, %eax
; SSE2-NEXT: testl %eax, %eax
; SSE2-NEXT: movdqa %xmm0, %xmm3
; SSE2-NEXT: js .LBB0_2
; SSE2-NEXT: # %bb.1:
; SSE2-NEXT: movdqa %xmm1, %xmm3
; SSE2-NEXT: .LBB0_2:
; SSE2-NEXT: movdqa %xmm3, %xmm0
; SSE2-NEXT: cmpunordss %xmm3, %xmm0
; SSE2-NEXT: movaps %xmm0, %xmm4
; SSE2-NEXT: andps %xmm3, %xmm4
; SSE2-NEXT: js .LBB0_4
; SSE2-NEXT: # %bb.3:
; SSE2-NEXT: movdqa %xmm2, %xmm1
; SSE2-NEXT: .LBB0_4:
; SSE2-NEXT: maxss %xmm1, %xmm3
; SSE2-NEXT: andnps %xmm3, %xmm0
; SSE2-NEXT: orps %xmm4, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fmaximum:
; AVX1: # %bb.0:
; AVX1-NEXT: vmovd %xmm0, %eax
; AVX1-NEXT: testl %eax, %eax
; AVX1-NEXT: js .LBB0_1
; AVX1-NEXT: # %bb.2:
; AVX1-NEXT: vmovdqa %xmm0, %xmm2
; AVX1-NEXT: jmp .LBB0_3
; AVX1-NEXT: .LBB0_1:
; AVX1-NEXT: vmovdqa %xmm1, %xmm2
; AVX1-NEXT: vmovdqa %xmm0, %xmm1
; AVX1-NEXT: .LBB0_3:
; AVX1-NEXT: vmaxss %xmm2, %xmm1, %xmm0
; AVX1-NEXT: vcmpunordss %xmm1, %xmm1, %xmm2
; AVX1-NEXT: vblendvps %xmm2, %xmm1, %xmm0, %xmm0
; AVX1-NEXT: retq
;
; AVX512-LABEL: test_fmaximum:
; AVX512: # %bb.0:
; AVX512-NEXT: vmovd %xmm0, %eax
; AVX512-NEXT: testl %eax, %eax
; AVX512-NEXT: sets %al
; AVX512-NEXT: kmovw %eax, %k1
; AVX512-NEXT: vmovdqa %xmm0, %xmm2
; AVX512-NEXT: vmovss %xmm1, %xmm2, %xmm2 {%k1}
; AVX512-NEXT: vmovss %xmm0, %xmm1, %xmm1 {%k1}
; AVX512-NEXT: vmaxss %xmm2, %xmm1, %xmm0
; AVX512-NEXT: vcmpunordss %xmm1, %xmm1, %k1
; AVX512-NEXT: vmovss %xmm1, %xmm0, %xmm0 {%k1}
; AVX512-NEXT: retq
;
; X86-LABEL: test_fmaximum:
; X86: # %bb.0:
; X86-NEXT: pushl %eax
; X86-NEXT: vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-NEXT: vmovd {{.*#+}} xmm2 = mem[0],zero,zero,zero
; X86-NEXT: vmovd %xmm2, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: js .LBB0_1
; X86-NEXT: # %bb.2:
; X86-NEXT: vmovdqa %xmm2, %xmm1
; X86-NEXT: jmp .LBB0_3
; X86-NEXT: .LBB0_1:
; X86-NEXT: vmovdqa %xmm0, %xmm1
; X86-NEXT: vmovdqa %xmm2, %xmm0
; X86-NEXT: .LBB0_3:
; X86-NEXT: vmaxss %xmm1, %xmm0, %xmm1
; X86-NEXT: vcmpunordss %xmm0, %xmm0, %xmm2
; X86-NEXT: vblendvps %xmm2, %xmm0, %xmm1, %xmm0
; X86-NEXT: vmovss %xmm0, (%esp)
; X86-NEXT: flds (%esp)
; X86-NEXT: popl %eax
; X86-NEXT: retl
%1 = tail call float @llvm.maximum.f32(float %x, float %y)
ret float %1
}
define <4 x float> @test_fmaximum_scalarize(<4 x float> %x, <4 x float> %y) "no-nans-fp-math"="true" "no-signed-zeros-fp-math"="true" {
; SSE2-LABEL: test_fmaximum_scalarize:
; SSE2: # %bb.0:
; SSE2-NEXT: movaps %xmm1, %xmm2
; SSE2-NEXT: shufps {{.*#+}} xmm2 = xmm2[3,3],xmm1[3,3]
; SSE2-NEXT: movaps %xmm0, %xmm3
; SSE2-NEXT: shufps {{.*#+}} xmm3 = xmm3[3,3],xmm0[3,3]
; SSE2-NEXT: maxss %xmm2, %xmm3
; SSE2-NEXT: movaps %xmm1, %xmm2
; SSE2-NEXT: unpckhpd {{.*#+}} xmm2 = xmm2[1],xmm1[1]
; SSE2-NEXT: movaps %xmm0, %xmm4
; SSE2-NEXT: unpckhpd {{.*#+}} xmm4 = xmm4[1],xmm0[1]
; SSE2-NEXT: maxss %xmm2, %xmm4
; SSE2-NEXT: unpcklps {{.*#+}} xmm4 = xmm4[0],xmm3[0],xmm4[1],xmm3[1]
; SSE2-NEXT: movaps %xmm0, %xmm2
; SSE2-NEXT: maxss %xmm1, %xmm2
; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[1,1,1,1]
; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[1,1,1,1]
; SSE2-NEXT: maxss %xmm1, %xmm0
; SSE2-NEXT: unpcklps {{.*#+}} xmm2 = xmm2[0],xmm0[0],xmm2[1],xmm0[1]
; SSE2-NEXT: movlhps {{.*#+}} xmm2 = xmm2[0],xmm4[0]
; SSE2-NEXT: movaps %xmm2, %xmm0
; SSE2-NEXT: retq
;
; AVX-LABEL: test_fmaximum_scalarize:
; AVX: # %bb.0:
; AVX-NEXT: vmaxss %xmm1, %xmm0, %xmm2
; AVX-NEXT: vmovshdup {{.*#+}} xmm3 = xmm1[1,1,3,3]
; AVX-NEXT: vmovshdup {{.*#+}} xmm4 = xmm0[1,1,3,3]
; AVX-NEXT: vmaxss %xmm3, %xmm4, %xmm3
; AVX-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0],xmm3[0],xmm2[2,3]
; AVX-NEXT: vshufpd {{.*#+}} xmm3 = xmm1[1,0]
; AVX-NEXT: vshufpd {{.*#+}} xmm4 = xmm0[1,0]
; AVX-NEXT: vmaxss %xmm3, %xmm4, %xmm3
; AVX-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1],xmm3[0],xmm2[3]
; AVX-NEXT: vshufps {{.*#+}} xmm1 = xmm1[3,3,3,3]
; AVX-NEXT: vshufps {{.*#+}} xmm0 = xmm0[3,3,3,3]
; AVX-NEXT: vmaxss %xmm1, %xmm0, %xmm0
; AVX-NEXT: vinsertps {{.*#+}} xmm0 = xmm2[0,1,2],xmm0[0]
; AVX-NEXT: retq
;
; X86-LABEL: test_fmaximum_scalarize:
; X86: # %bb.0:
; X86-NEXT: vmaxss %xmm1, %xmm0, %xmm2
; X86-NEXT: vmovshdup {{.*#+}} xmm3 = xmm1[1,1,3,3]
; X86-NEXT: vmovshdup {{.*#+}} xmm4 = xmm0[1,1,3,3]
; X86-NEXT: vmaxss %xmm3, %xmm4, %xmm3
; X86-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0],xmm3[0],xmm2[2,3]
; X86-NEXT: vshufpd {{.*#+}} xmm3 = xmm1[1,0]
; X86-NEXT: vshufpd {{.*#+}} xmm4 = xmm0[1,0]
; X86-NEXT: vmaxss %xmm3, %xmm4, %xmm3
; X86-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1],xmm3[0],xmm2[3]
; X86-NEXT: vshufps {{.*#+}} xmm1 = xmm1[3,3,3,3]
; X86-NEXT: vshufps {{.*#+}} xmm0 = xmm0[3,3,3,3]
; X86-NEXT: vmaxss %xmm1, %xmm0, %xmm0
; X86-NEXT: vinsertps {{.*#+}} xmm0 = xmm2[0,1,2],xmm0[0]
; X86-NEXT: retl
%r = call <4 x float> @llvm.maximum.v4f32(<4 x float> %x, <4 x float> %y)
ret <4 x float> %r
}
define float @test_fmaximum_nan0(float %x, float %y) {
; SSE2-LABEL: test_fmaximum_nan0:
; SSE2: # %bb.0:
; SSE2-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; SSE2-NEXT: retq
;
; AVX-LABEL: test_fmaximum_nan0:
; AVX: # %bb.0:
; AVX-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; AVX-NEXT: retq
;
; X86-LABEL: test_fmaximum_nan0:
; X86: # %bb.0:
; X86-NEXT: flds {{\.?LCPI[0-9]+_[0-9]+}}
; X86-NEXT: retl
%1 = tail call float @llvm.maximum.f32(float 0x7fff000000000000, float %y)
ret float %1
}
define float @test_fmaximum_nan1(float %x, float %y) {
; SSE2-LABEL: test_fmaximum_nan1:
; SSE2: # %bb.0:
; SSE2-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; SSE2-NEXT: retq
;
; AVX-LABEL: test_fmaximum_nan1:
; AVX: # %bb.0:
; AVX-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; AVX-NEXT: retq
;
; X86-LABEL: test_fmaximum_nan1:
; X86: # %bb.0:
; X86-NEXT: flds {{\.?LCPI[0-9]+_[0-9]+}}
; X86-NEXT: retl
%1 = tail call float @llvm.maximum.f32(float %x, float 0x7fff000000000000)
ret float %1
}
define float @test_fmaximum_nnan(float %x, float %y) nounwind {
; SSE2-LABEL: test_fmaximum_nnan:
; SSE2: # %bb.0:
; SSE2-NEXT: movaps %xmm0, %xmm2
; SSE2-NEXT: addss %xmm1, %xmm2
; SSE2-NEXT: subss %xmm1, %xmm0
; SSE2-NEXT: movd %xmm2, %eax
; SSE2-NEXT: testl %eax, %eax
; SSE2-NEXT: js .LBB4_1
; SSE2-NEXT: # %bb.2:
; SSE2-NEXT: maxss %xmm2, %xmm0
; SSE2-NEXT: retq
; SSE2-NEXT: .LBB4_1:
; SSE2-NEXT: movaps %xmm0, %xmm1
; SSE2-NEXT: movaps %xmm2, %xmm0
; SSE2-NEXT: maxss %xmm1, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fmaximum_nnan:
; AVX1: # %bb.0:
; AVX1-NEXT: vaddss %xmm1, %xmm0, %xmm2
; AVX1-NEXT: vsubss %xmm1, %xmm0, %xmm0
; AVX1-NEXT: vmovd %xmm2, %eax
; AVX1-NEXT: testl %eax, %eax
; AVX1-NEXT: js .LBB4_1
; AVX1-NEXT: # %bb.2:
; AVX1-NEXT: vmaxss %xmm2, %xmm0, %xmm0
; AVX1-NEXT: retq
; AVX1-NEXT: .LBB4_1:
; AVX1-NEXT: vmovaps %xmm0, %xmm1
; AVX1-NEXT: vmaxss %xmm1, %xmm2, %xmm0
; AVX1-NEXT: retq
;
; AVX512F-LABEL: test_fmaximum_nnan:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vaddss %xmm1, %xmm0, %xmm2
; AVX512F-NEXT: vsubss %xmm1, %xmm0, %xmm0
; AVX512F-NEXT: vmovd %xmm2, %eax
; AVX512F-NEXT: testl %eax, %eax
; AVX512F-NEXT: sets %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovaps %xmm2, %xmm1
; AVX512F-NEXT: vmovss %xmm0, %xmm1, %xmm1 {%k1}
; AVX512F-NEXT: vmovss %xmm2, %xmm0, %xmm0 {%k1}
; AVX512F-NEXT: vmaxss %xmm1, %xmm0, %xmm0
; AVX512F-NEXT: retq
;
; AVX512DQ-LABEL: test_fmaximum_nnan:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vaddss %xmm1, %xmm0, %xmm2
; AVX512DQ-NEXT: vsubss %xmm1, %xmm0, %xmm0
; AVX512DQ-NEXT: vfpclassss $3, %xmm0, %k0
; AVX512DQ-NEXT: kmovw %k0, %k1
; AVX512DQ-NEXT: vmovaps %xmm2, %xmm1
; AVX512DQ-NEXT: vmovss %xmm0, %xmm1, %xmm1 {%k1}
; AVX512DQ-NEXT: vmovss %xmm2, %xmm0, %xmm0 {%k1}
; AVX512DQ-NEXT: vmaxss %xmm1, %xmm0, %xmm0
; AVX512DQ-NEXT: retq
;
; X86-LABEL: test_fmaximum_nnan:
; X86: # %bb.0:
; X86-NEXT: pushl %eax
; X86-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-NEXT: vmovss {{.*#+}} xmm2 = mem[0],zero,zero,zero
; X86-NEXT: vaddss %xmm0, %xmm2, %xmm1
; X86-NEXT: vsubss %xmm0, %xmm2, %xmm0
; X86-NEXT: vmovd %xmm1, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: js .LBB4_1
; X86-NEXT: # %bb.2:
; X86-NEXT: vmovaps %xmm1, %xmm2
; X86-NEXT: jmp .LBB4_3
; X86-NEXT: .LBB4_1:
; X86-NEXT: vmovaps %xmm0, %xmm2
; X86-NEXT: vmovaps %xmm1, %xmm0
; X86-NEXT: .LBB4_3:
; X86-NEXT: vmaxss %xmm2, %xmm0, %xmm0
; X86-NEXT: vmovss %xmm0, (%esp)
; X86-NEXT: flds (%esp)
; X86-NEXT: popl %eax
; X86-NEXT: retl
%1 = fadd nnan float %x, %y
%2 = fsub nnan float %x, %y
%3 = tail call float @llvm.maximum.f32(float %1, float %2)
ret float %3
}
define double @test_fmaximum_zero0(double %x, double %y) nounwind {
; SSE2-LABEL: test_fmaximum_zero0:
; SSE2: # %bb.0:
; SSE2-NEXT: movapd %xmm1, %xmm0
; SSE2-NEXT: cmpunordsd %xmm1, %xmm0
; SSE2-NEXT: movapd %xmm0, %xmm2
; SSE2-NEXT: andpd %xmm1, %xmm2
; SSE2-NEXT: xorpd %xmm3, %xmm3
; SSE2-NEXT: maxsd %xmm3, %xmm1
; SSE2-NEXT: andnpd %xmm1, %xmm0
; SSE2-NEXT: orpd %xmm2, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fmaximum_zero0:
; AVX1: # %bb.0:
; AVX1-NEXT: vxorpd %xmm0, %xmm0, %xmm0
; AVX1-NEXT: vmaxsd %xmm0, %xmm1, %xmm0
; AVX1-NEXT: vcmpunordsd %xmm1, %xmm1, %xmm2
; AVX1-NEXT: vblendvpd %xmm2, %xmm1, %xmm0, %xmm0
; AVX1-NEXT: retq
;
; AVX512-LABEL: test_fmaximum_zero0:
; AVX512: # %bb.0:
; AVX512-NEXT: vxorpd %xmm0, %xmm0, %xmm0
; AVX512-NEXT: vmaxsd %xmm0, %xmm1, %xmm0
; AVX512-NEXT: vcmpunordsd %xmm1, %xmm1, %k1
; AVX512-NEXT: vmovsd %xmm1, %xmm0, %xmm0 {%k1}
; AVX512-NEXT: retq
;
; X86-LABEL: test_fmaximum_zero0:
; X86: # %bb.0:
; X86-NEXT: pushl %ebp
; X86-NEXT: movl %esp, %ebp
; X86-NEXT: andl $-8, %esp
; X86-NEXT: subl $8, %esp
; X86-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; X86-NEXT: vxorpd %xmm1, %xmm1, %xmm1
; X86-NEXT: vmaxsd %xmm1, %xmm0, %xmm1
; X86-NEXT: vcmpunordsd %xmm0, %xmm0, %xmm2
; X86-NEXT: vblendvpd %xmm2, %xmm0, %xmm1, %xmm0
; X86-NEXT: vmovlpd %xmm0, (%esp)
; X86-NEXT: fldl (%esp)
; X86-NEXT: movl %ebp, %esp
; X86-NEXT: popl %ebp
; X86-NEXT: retl
%1 = tail call double @llvm.maximum.f64(double 0.0, double %y)
ret double %1
}
define double @test_fmaximum_zero1(double %x, double %y) nounwind {
; SSE2-LABEL: test_fmaximum_zero1:
; SSE2: # %bb.0:
; SSE2-NEXT: movapd %xmm0, %xmm1
; SSE2-NEXT: cmpunordsd %xmm0, %xmm1
; SSE2-NEXT: movapd %xmm1, %xmm2
; SSE2-NEXT: andpd %xmm0, %xmm2
; SSE2-NEXT: xorpd %xmm3, %xmm3
; SSE2-NEXT: maxsd %xmm3, %xmm0
; SSE2-NEXT: andnpd %xmm0, %xmm1
; SSE2-NEXT: orpd %xmm2, %xmm1
; SSE2-NEXT: movapd %xmm1, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fmaximum_zero1:
; AVX1: # %bb.0:
; AVX1-NEXT: vxorpd %xmm1, %xmm1, %xmm1
; AVX1-NEXT: vmaxsd %xmm1, %xmm0, %xmm1
; AVX1-NEXT: vcmpunordsd %xmm0, %xmm0, %xmm2
; AVX1-NEXT: vblendvpd %xmm2, %xmm0, %xmm1, %xmm0
; AVX1-NEXT: retq
;
; AVX512-LABEL: test_fmaximum_zero1:
; AVX512: # %bb.0:
; AVX512-NEXT: vxorpd %xmm1, %xmm1, %xmm1
; AVX512-NEXT: vmaxsd %xmm1, %xmm0, %xmm1
; AVX512-NEXT: vcmpunordsd %xmm0, %xmm0, %k1
; AVX512-NEXT: vmovsd %xmm0, %xmm1, %xmm1 {%k1}
; AVX512-NEXT: vmovapd %xmm1, %xmm0
; AVX512-NEXT: retq
;
; X86-LABEL: test_fmaximum_zero1:
; X86: # %bb.0:
; X86-NEXT: pushl %ebp
; X86-NEXT: movl %esp, %ebp
; X86-NEXT: andl $-8, %esp
; X86-NEXT: subl $8, %esp
; X86-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; X86-NEXT: vxorpd %xmm1, %xmm1, %xmm1
; X86-NEXT: vmaxsd %xmm1, %xmm0, %xmm1
; X86-NEXT: vcmpunordsd %xmm0, %xmm0, %xmm2
; X86-NEXT: vblendvpd %xmm2, %xmm0, %xmm1, %xmm0
; X86-NEXT: vmovlpd %xmm0, (%esp)
; X86-NEXT: fldl (%esp)
; X86-NEXT: movl %ebp, %esp
; X86-NEXT: popl %ebp
; X86-NEXT: retl
%1 = tail call double @llvm.maximum.f64(double %x, double 0.0)
ret double %1
}
define double @test_fmaximum_zero2(double %x, double %y) {
; SSE2-LABEL: test_fmaximum_zero2:
; SSE2: # %bb.0:
; SSE2-NEXT: xorps %xmm0, %xmm0
; SSE2-NEXT: retq
;
; AVX-LABEL: test_fmaximum_zero2:
; AVX: # %bb.0:
; AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
; AVX-NEXT: retq
;
; X86-LABEL: test_fmaximum_zero2:
; X86: # %bb.0:
; X86-NEXT: fldz
; X86-NEXT: retl
%1 = tail call double @llvm.maximum.f64(double 0.0, double -0.0)
ret double %1
}
define float @test_fmaximum_nsz(float %x, float %y) "no-signed-zeros-fp-math"="true" nounwind {
; SSE2-LABEL: test_fmaximum_nsz:
; SSE2: # %bb.0:
; SSE2-NEXT: movaps %xmm0, %xmm2
; SSE2-NEXT: cmpunordss %xmm0, %xmm2
; SSE2-NEXT: movaps %xmm2, %xmm3
; SSE2-NEXT: andps %xmm0, %xmm3
; SSE2-NEXT: maxss %xmm1, %xmm0
; SSE2-NEXT: andnps %xmm0, %xmm2
; SSE2-NEXT: orps %xmm3, %xmm2
; SSE2-NEXT: movaps %xmm2, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fmaximum_nsz:
; AVX1: # %bb.0:
; AVX1-NEXT: vmaxss %xmm1, %xmm0, %xmm1
; AVX1-NEXT: vcmpunordss %xmm0, %xmm0, %xmm2
; AVX1-NEXT: vblendvps %xmm2, %xmm0, %xmm1, %xmm0
; AVX1-NEXT: retq
;
; AVX512-LABEL: test_fmaximum_nsz:
; AVX512: # %bb.0:
; AVX512-NEXT: vmaxss %xmm1, %xmm0, %xmm1
; AVX512-NEXT: vcmpunordss %xmm0, %xmm0, %k1
; AVX512-NEXT: vmovss %xmm0, %xmm1, %xmm1 {%k1}
; AVX512-NEXT: vmovaps %xmm1, %xmm0
; AVX512-NEXT: retq
;
; X86-LABEL: test_fmaximum_nsz:
; X86: # %bb.0:
; X86-NEXT: pushl %eax
; X86-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-NEXT: vcmpunordss %xmm0, %xmm0, %xmm1
; X86-NEXT: vmaxss {{[0-9]+}}(%esp), %xmm0, %xmm2
; X86-NEXT: vblendvps %xmm1, %xmm0, %xmm2, %xmm0
; X86-NEXT: vmovss %xmm0, (%esp)
; X86-NEXT: flds (%esp)
; X86-NEXT: popl %eax
; X86-NEXT: retl
%1 = tail call float @llvm.maximum.f32(float %x, float %y)
ret float %1
}
define float @test_fmaximum_combine_cmps(float %x, float %y) nounwind {
; SSE2-LABEL: test_fmaximum_combine_cmps:
; SSE2: # %bb.0:
; SSE2-NEXT: divss %xmm0, %xmm1
; SSE2-NEXT: movd %xmm0, %eax
; SSE2-NEXT: testl %eax, %eax
; SSE2-NEXT: movaps %xmm0, %xmm3
; SSE2-NEXT: js .LBB9_2
; SSE2-NEXT: # %bb.1:
; SSE2-NEXT: movaps %xmm1, %xmm3
; SSE2-NEXT: .LBB9_2:
; SSE2-NEXT: movaps %xmm3, %xmm2
; SSE2-NEXT: cmpunordss %xmm3, %xmm2
; SSE2-NEXT: movaps %xmm2, %xmm4
; SSE2-NEXT: andps %xmm3, %xmm4
; SSE2-NEXT: js .LBB9_4
; SSE2-NEXT: # %bb.3:
; SSE2-NEXT: movaps %xmm0, %xmm1
; SSE2-NEXT: .LBB9_4:
; SSE2-NEXT: maxss %xmm1, %xmm3
; SSE2-NEXT: andnps %xmm3, %xmm2
; SSE2-NEXT: orps %xmm4, %xmm2
; SSE2-NEXT: movaps %xmm2, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fmaximum_combine_cmps:
; AVX1: # %bb.0:
; AVX1-NEXT: vdivss %xmm0, %xmm1, %xmm1
; AVX1-NEXT: vmovd %xmm0, %eax
; AVX1-NEXT: testl %eax, %eax
; AVX1-NEXT: js .LBB9_1
; AVX1-NEXT: # %bb.2:
; AVX1-NEXT: vmovaps %xmm0, %xmm2
; AVX1-NEXT: jmp .LBB9_3
; AVX1-NEXT: .LBB9_1:
; AVX1-NEXT: vmovaps %xmm1, %xmm2
; AVX1-NEXT: vmovaps %xmm0, %xmm1
; AVX1-NEXT: .LBB9_3:
; AVX1-NEXT: vmaxss %xmm2, %xmm1, %xmm0
; AVX1-NEXT: vcmpunordss %xmm1, %xmm1, %xmm2
; AVX1-NEXT: vblendvps %xmm2, %xmm1, %xmm0, %xmm0
; AVX1-NEXT: retq
;
; AVX512F-LABEL: test_fmaximum_combine_cmps:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vdivss %xmm0, %xmm1, %xmm1
; AVX512F-NEXT: vmovd %xmm0, %eax
; AVX512F-NEXT: testl %eax, %eax
; AVX512F-NEXT: sets %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovaps %xmm0, %xmm2
; AVX512F-NEXT: vmovss %xmm1, %xmm2, %xmm2 {%k1}
; AVX512F-NEXT: vmovss %xmm0, %xmm1, %xmm1 {%k1}
; AVX512F-NEXT: vmaxss %xmm2, %xmm1, %xmm0
; AVX512F-NEXT: vcmpunordss %xmm1, %xmm1, %k1
; AVX512F-NEXT: vmovss %xmm1, %xmm0, %xmm0 {%k1}
; AVX512F-NEXT: retq
;
; AVX512DQ-LABEL: test_fmaximum_combine_cmps:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vdivss %xmm0, %xmm1, %xmm1
; AVX512DQ-NEXT: vfpclassss $3, %xmm0, %k0
; AVX512DQ-NEXT: kmovw %k0, %k1
; AVX512DQ-NEXT: vmovaps %xmm1, %xmm2
; AVX512DQ-NEXT: vmovss %xmm0, %xmm2, %xmm2 {%k1}
; AVX512DQ-NEXT: vmovss %xmm1, %xmm0, %xmm0 {%k1}
; AVX512DQ-NEXT: vmaxss %xmm2, %xmm0, %xmm0
; AVX512DQ-NEXT: retq
;
; X86-LABEL: test_fmaximum_combine_cmps:
; X86: # %bb.0:
; X86-NEXT: pushl %eax
; X86-NEXT: vmovss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; X86-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-NEXT: vdivss %xmm1, %xmm0, %xmm0
; X86-NEXT: vmovd %xmm1, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: js .LBB9_1
; X86-NEXT: # %bb.2:
; X86-NEXT: vmovaps %xmm1, %xmm2
; X86-NEXT: jmp .LBB9_3
; X86-NEXT: .LBB9_1:
; X86-NEXT: vmovaps %xmm0, %xmm2
; X86-NEXT: vmovaps %xmm1, %xmm0
; X86-NEXT: .LBB9_3:
; X86-NEXT: vmaxss %xmm2, %xmm0, %xmm1
; X86-NEXT: vcmpunordss %xmm0, %xmm0, %xmm2
; X86-NEXT: vblendvps %xmm2, %xmm0, %xmm1, %xmm0
; X86-NEXT: vmovss %xmm0, (%esp)
; X86-NEXT: flds (%esp)
; X86-NEXT: popl %eax
; X86-NEXT: retl
%1 = fdiv nnan float %y, %x
%2 = tail call float @llvm.maximum.f32(float %x, float %1)
ret float %2
}
;
; fminimum
;
define float @test_fminimum(float %x, float %y) nounwind {
; SSE2-LABEL: test_fminimum:
; SSE2: # %bb.0:
; SSE2-NEXT: movd %xmm0, %eax
; SSE2-NEXT: testl %eax, %eax
; SSE2-NEXT: movdqa %xmm1, %xmm3
; SSE2-NEXT: js .LBB10_2
; SSE2-NEXT: # %bb.1:
; SSE2-NEXT: movdqa %xmm0, %xmm3
; SSE2-NEXT: .LBB10_2:
; SSE2-NEXT: movdqa %xmm3, %xmm2
; SSE2-NEXT: cmpunordss %xmm3, %xmm2
; SSE2-NEXT: movaps %xmm2, %xmm4
; SSE2-NEXT: andps %xmm3, %xmm4
; SSE2-NEXT: js .LBB10_4
; SSE2-NEXT: # %bb.3:
; SSE2-NEXT: movdqa %xmm1, %xmm0
; SSE2-NEXT: .LBB10_4:
; SSE2-NEXT: minss %xmm0, %xmm3
; SSE2-NEXT: andnps %xmm3, %xmm2
; SSE2-NEXT: orps %xmm4, %xmm2
; SSE2-NEXT: movaps %xmm2, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fminimum:
; AVX1: # %bb.0:
; AVX1-NEXT: vmovd %xmm0, %eax
; AVX1-NEXT: testl %eax, %eax
; AVX1-NEXT: js .LBB10_1
; AVX1-NEXT: # %bb.2:
; AVX1-NEXT: vmovdqa %xmm1, %xmm2
; AVX1-NEXT: jmp .LBB10_3
; AVX1-NEXT: .LBB10_1:
; AVX1-NEXT: vmovdqa %xmm0, %xmm2
; AVX1-NEXT: vmovdqa %xmm1, %xmm0
; AVX1-NEXT: .LBB10_3:
; AVX1-NEXT: vminss %xmm2, %xmm0, %xmm1
; AVX1-NEXT: vcmpunordss %xmm0, %xmm0, %xmm2
; AVX1-NEXT: vblendvps %xmm2, %xmm0, %xmm1, %xmm0
; AVX1-NEXT: retq
;
; AVX512-LABEL: test_fminimum:
; AVX512: # %bb.0:
; AVX512-NEXT: vmovd %xmm0, %eax
; AVX512-NEXT: testl %eax, %eax
; AVX512-NEXT: sets %al
; AVX512-NEXT: kmovw %eax, %k1
; AVX512-NEXT: vmovaps %xmm1, %xmm2
; AVX512-NEXT: vmovss %xmm0, %xmm2, %xmm2 {%k1}
; AVX512-NEXT: vmovss %xmm1, %xmm0, %xmm0 {%k1}
; AVX512-NEXT: vminss %xmm2, %xmm0, %xmm1
; AVX512-NEXT: vcmpunordss %xmm0, %xmm0, %k1
; AVX512-NEXT: vmovss %xmm0, %xmm1, %xmm1 {%k1}
; AVX512-NEXT: vmovaps %xmm1, %xmm0
; AVX512-NEXT: retq
;
; X86-LABEL: test_fminimum:
; X86: # %bb.0:
; X86-NEXT: pushl %eax
; X86-NEXT: vmovd {{.*#+}} xmm1 = mem[0],zero,zero,zero
; X86-NEXT: vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-NEXT: vmovd %xmm0, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: js .LBB10_1
; X86-NEXT: # %bb.2:
; X86-NEXT: vmovdqa %xmm1, %xmm2
; X86-NEXT: jmp .LBB10_3
; X86-NEXT: .LBB10_1:
; X86-NEXT: vmovdqa %xmm0, %xmm2
; X86-NEXT: vmovdqa %xmm1, %xmm0
; X86-NEXT: .LBB10_3:
; X86-NEXT: vminss %xmm2, %xmm0, %xmm1
; X86-NEXT: vcmpunordss %xmm0, %xmm0, %xmm2
; X86-NEXT: vblendvps %xmm2, %xmm0, %xmm1, %xmm0
; X86-NEXT: vmovss %xmm0, (%esp)
; X86-NEXT: flds (%esp)
; X86-NEXT: popl %eax
; X86-NEXT: retl
%1 = tail call float @llvm.minimum.f32(float %x, float %y)
ret float %1
}
define <2 x double> @test_fminimum_scalarize(<2 x double> %x, <2 x double> %y) "no-nans-fp-math"="true" "no-signed-zeros-fp-math"="true" {
; SSE2-LABEL: test_fminimum_scalarize:
; SSE2: # %bb.0:
; SSE2-NEXT: movapd %xmm0, %xmm2
; SSE2-NEXT: minsd %xmm1, %xmm2
; SSE2-NEXT: unpckhpd {{.*#+}} xmm1 = xmm1[1,1]
; SSE2-NEXT: unpckhpd {{.*#+}} xmm0 = xmm0[1,1]
; SSE2-NEXT: minsd %xmm1, %xmm0
; SSE2-NEXT: unpcklpd {{.*#+}} xmm2 = xmm2[0],xmm0[0]
; SSE2-NEXT: movapd %xmm2, %xmm0
; SSE2-NEXT: retq
;
; AVX-LABEL: test_fminimum_scalarize:
; AVX: # %bb.0:
; AVX-NEXT: vminsd %xmm1, %xmm0, %xmm2
; AVX-NEXT: vshufpd {{.*#+}} xmm1 = xmm1[1,0]
; AVX-NEXT: vshufpd {{.*#+}} xmm0 = xmm0[1,0]
; AVX-NEXT: vminsd %xmm1, %xmm0, %xmm0
; AVX-NEXT: vunpcklpd {{.*#+}} xmm0 = xmm2[0],xmm0[0]
; AVX-NEXT: retq
;
; X86-LABEL: test_fminimum_scalarize:
; X86: # %bb.0:
; X86-NEXT: vminsd %xmm1, %xmm0, %xmm2
; X86-NEXT: vshufpd {{.*#+}} xmm1 = xmm1[1,0]
; X86-NEXT: vshufpd {{.*#+}} xmm0 = xmm0[1,0]
; X86-NEXT: vminsd %xmm1, %xmm0, %xmm0
; X86-NEXT: vunpcklpd {{.*#+}} xmm0 = xmm2[0],xmm0[0]
; X86-NEXT: retl
%r = call <2 x double> @llvm.minimum.v2f64(<2 x double> %x, <2 x double> %y)
ret <2 x double> %r
}
define float @test_fminimum_nan0(float %x, float %y) {
; SSE2-LABEL: test_fminimum_nan0:
; SSE2: # %bb.0:
; SSE2-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; SSE2-NEXT: retq
;
; AVX-LABEL: test_fminimum_nan0:
; AVX: # %bb.0:
; AVX-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; AVX-NEXT: retq
;
; X86-LABEL: test_fminimum_nan0:
; X86: # %bb.0:
; X86-NEXT: flds {{\.?LCPI[0-9]+_[0-9]+}}
; X86-NEXT: retl
%1 = tail call float @llvm.minimum.f32(float 0x7fff000000000000, float %y)
ret float %1
}
define float @test_fminimum_nan1(float %x, float %y) {
; SSE2-LABEL: test_fminimum_nan1:
; SSE2: # %bb.0:
; SSE2-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; SSE2-NEXT: retq
;
; AVX-LABEL: test_fminimum_nan1:
; AVX: # %bb.0:
; AVX-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; AVX-NEXT: retq
;
; X86-LABEL: test_fminimum_nan1:
; X86: # %bb.0:
; X86-NEXT: flds {{\.?LCPI[0-9]+_[0-9]+}}
; X86-NEXT: retl
%1 = tail call float @llvm.minimum.f32(float %x, float 0x7fff000000000000)
ret float %1
}
define double @test_fminimum_nnan(double %x, double %y) "no-nans-fp-math"="true" nounwind {
; SSE2-LABEL: test_fminimum_nnan:
; SSE2: # %bb.0:
; SSE2-NEXT: movq %xmm0, %rax
; SSE2-NEXT: testq %rax, %rax
; SSE2-NEXT: js .LBB14_1
; SSE2-NEXT: # %bb.2:
; SSE2-NEXT: minsd %xmm1, %xmm0
; SSE2-NEXT: retq
; SSE2-NEXT: .LBB14_1:
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: movapd %xmm1, %xmm0
; SSE2-NEXT: minsd %xmm2, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fminimum_nnan:
; AVX1: # %bb.0:
; AVX1-NEXT: vmovq %xmm0, %rax
; AVX1-NEXT: testq %rax, %rax
; AVX1-NEXT: js .LBB14_1
; AVX1-NEXT: # %bb.2:
; AVX1-NEXT: vminsd %xmm1, %xmm0, %xmm0
; AVX1-NEXT: retq
; AVX1-NEXT: .LBB14_1:
; AVX1-NEXT: vmovdqa %xmm0, %xmm2
; AVX1-NEXT: vminsd %xmm2, %xmm1, %xmm0
; AVX1-NEXT: retq
;
; AVX512F-LABEL: test_fminimum_nnan:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovq %xmm0, %rax
; AVX512F-NEXT: testq %rax, %rax
; AVX512F-NEXT: sets %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovapd %xmm1, %xmm2
; AVX512F-NEXT: vmovsd %xmm0, %xmm2, %xmm2 {%k1}
; AVX512F-NEXT: vmovsd %xmm1, %xmm0, %xmm0 {%k1}
; AVX512F-NEXT: vminsd %xmm2, %xmm0, %xmm0
; AVX512F-NEXT: retq
;
; AVX512DQ-LABEL: test_fminimum_nnan:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vfpclasssd $5, %xmm1, %k0
; AVX512DQ-NEXT: kmovw %k0, %k1
; AVX512DQ-NEXT: vmovapd %xmm0, %xmm2
; AVX512DQ-NEXT: vmovsd %xmm1, %xmm2, %xmm2 {%k1}
; AVX512DQ-NEXT: vmovsd %xmm0, %xmm1, %xmm1 {%k1}
; AVX512DQ-NEXT: vminsd %xmm2, %xmm1, %xmm0
; AVX512DQ-NEXT: retq
;
; X86-LABEL: test_fminimum_nnan:
; X86: # %bb.0:
; X86-NEXT: pushl %ebp
; X86-NEXT: movl %esp, %ebp
; X86-NEXT: andl $-8, %esp
; X86-NEXT: subl $8, %esp
; X86-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; X86-NEXT: vmovsd {{.*#+}} xmm1 = mem[0],zero
; X86-NEXT: vmovsd {{.*#+}} xmm2 = mem[0],zero
; X86-NEXT: vextractps $1, %xmm2, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: js .LBB14_1
; X86-NEXT: # %bb.2:
; X86-NEXT: vmovapd %xmm1, %xmm2
; X86-NEXT: jmp .LBB14_3
; X86-NEXT: .LBB14_1:
; X86-NEXT: vmovapd %xmm0, %xmm2
; X86-NEXT: vmovapd %xmm1, %xmm0
; X86-NEXT: .LBB14_3:
; X86-NEXT: vminsd %xmm2, %xmm0, %xmm0
; X86-NEXT: vmovsd %xmm0, (%esp)
; X86-NEXT: fldl (%esp)
; X86-NEXT: movl %ebp, %esp
; X86-NEXT: popl %ebp
; X86-NEXT: retl
%1 = tail call double @llvm.minimum.f64(double %x, double %y)
ret double %1
}
define double @test_fminimum_zero0(double %x, double %y) nounwind {
; SSE2-LABEL: test_fminimum_zero0:
; SSE2: # %bb.0:
; SSE2-NEXT: movapd %xmm1, %xmm0
; SSE2-NEXT: cmpunordsd %xmm1, %xmm0
; SSE2-NEXT: movapd %xmm0, %xmm2
; SSE2-NEXT: andpd %xmm1, %xmm2
; SSE2-NEXT: minsd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
; SSE2-NEXT: andnpd %xmm1, %xmm0
; SSE2-NEXT: orpd %xmm2, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fminimum_zero0:
; AVX1: # %bb.0:
; AVX1-NEXT: vcmpunordsd %xmm1, %xmm1, %xmm0
; AVX1-NEXT: vminsd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm2
; AVX1-NEXT: vblendvpd %xmm0, %xmm1, %xmm2, %xmm0
; AVX1-NEXT: retq
;
; AVX512-LABEL: test_fminimum_zero0:
; AVX512: # %bb.0:
; AVX512-NEXT: vcmpunordsd %xmm1, %xmm1, %k1
; AVX512-NEXT: vminsd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm0
; AVX512-NEXT: vmovsd %xmm1, %xmm0, %xmm0 {%k1}
; AVX512-NEXT: retq
;
; X86-LABEL: test_fminimum_zero0:
; X86: # %bb.0:
; X86-NEXT: pushl %ebp
; X86-NEXT: movl %esp, %ebp
; X86-NEXT: andl $-8, %esp
; X86-NEXT: subl $8, %esp
; X86-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; X86-NEXT: vcmpunordsd %xmm0, %xmm0, %xmm1
; X86-NEXT: vminsd {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0, %xmm2
; X86-NEXT: vblendvpd %xmm1, %xmm0, %xmm2, %xmm0
; X86-NEXT: vmovlpd %xmm0, (%esp)
; X86-NEXT: fldl (%esp)
; X86-NEXT: movl %ebp, %esp
; X86-NEXT: popl %ebp
; X86-NEXT: retl
%1 = tail call double @llvm.minimum.f64(double -0.0, double %y)
ret double %1
}
define double @test_fminimum_zero1(double %x, double %y) nounwind {
; SSE2-LABEL: test_fminimum_zero1:
; SSE2: # %bb.0:
; SSE2-NEXT: movapd %xmm0, %xmm1
; SSE2-NEXT: cmpunordsd %xmm0, %xmm1
; SSE2-NEXT: movapd %xmm1, %xmm2
; SSE2-NEXT: andpd %xmm0, %xmm2
; SSE2-NEXT: minsd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE2-NEXT: andnpd %xmm0, %xmm1
; SSE2-NEXT: orpd %xmm2, %xmm1
; SSE2-NEXT: movapd %xmm1, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fminimum_zero1:
; AVX1: # %bb.0:
; AVX1-NEXT: vcmpunordsd %xmm0, %xmm0, %xmm1
; AVX1-NEXT: vminsd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm2
; AVX1-NEXT: vblendvpd %xmm1, %xmm0, %xmm2, %xmm0
; AVX1-NEXT: retq
;
; AVX512-LABEL: test_fminimum_zero1:
; AVX512: # %bb.0:
; AVX512-NEXT: vcmpunordsd %xmm0, %xmm0, %k1
; AVX512-NEXT: vminsd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm1
; AVX512-NEXT: vmovsd %xmm0, %xmm1, %xmm1 {%k1}
; AVX512-NEXT: vmovapd %xmm1, %xmm0
; AVX512-NEXT: retq
;
; X86-LABEL: test_fminimum_zero1:
; X86: # %bb.0:
; X86-NEXT: pushl %ebp
; X86-NEXT: movl %esp, %ebp
; X86-NEXT: andl $-8, %esp
; X86-NEXT: subl $8, %esp
; X86-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; X86-NEXT: vcmpunordsd %xmm0, %xmm0, %xmm1
; X86-NEXT: vminsd {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0, %xmm2
; X86-NEXT: vblendvpd %xmm1, %xmm0, %xmm2, %xmm0
; X86-NEXT: vmovlpd %xmm0, (%esp)
; X86-NEXT: fldl (%esp)
; X86-NEXT: movl %ebp, %esp
; X86-NEXT: popl %ebp
; X86-NEXT: retl
%1 = tail call double @llvm.minimum.f64(double %x, double -0.0)
ret double %1
}
define double @test_fminimum_zero2(double %x, double %y) {
; SSE2-LABEL: test_fminimum_zero2:
; SSE2: # %bb.0:
; SSE2-NEXT: movsd {{.*#+}} xmm0 = mem[0],zero
; SSE2-NEXT: retq
;
; AVX-LABEL: test_fminimum_zero2:
; AVX: # %bb.0:
; AVX-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; AVX-NEXT: retq
;
; X86-LABEL: test_fminimum_zero2:
; X86: # %bb.0:
; X86-NEXT: fldz
; X86-NEXT: fchs
; X86-NEXT: retl
%1 = tail call double @llvm.minimum.f64(double -0.0, double 0.0)
ret double %1
}
define float @test_fminimum_nsz(float %x, float %y) nounwind {
; SSE2-LABEL: test_fminimum_nsz:
; SSE2: # %bb.0:
; SSE2-NEXT: movaps %xmm0, %xmm2
; SSE2-NEXT: cmpunordss %xmm0, %xmm2
; SSE2-NEXT: movaps %xmm2, %xmm3
; SSE2-NEXT: andps %xmm0, %xmm3
; SSE2-NEXT: minss %xmm1, %xmm0
; SSE2-NEXT: andnps %xmm0, %xmm2
; SSE2-NEXT: orps %xmm3, %xmm2
; SSE2-NEXT: movaps %xmm2, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fminimum_nsz:
; AVX1: # %bb.0:
; AVX1-NEXT: vminss %xmm1, %xmm0, %xmm1
; AVX1-NEXT: vcmpunordss %xmm0, %xmm0, %xmm2
; AVX1-NEXT: vblendvps %xmm2, %xmm0, %xmm1, %xmm0
; AVX1-NEXT: retq
;
; AVX512-LABEL: test_fminimum_nsz:
; AVX512: # %bb.0:
; AVX512-NEXT: vminss %xmm1, %xmm0, %xmm1
; AVX512-NEXT: vcmpunordss %xmm0, %xmm0, %k1
; AVX512-NEXT: vmovss %xmm0, %xmm1, %xmm1 {%k1}
; AVX512-NEXT: vmovaps %xmm1, %xmm0
; AVX512-NEXT: retq
;
; X86-LABEL: test_fminimum_nsz:
; X86: # %bb.0:
; X86-NEXT: pushl %eax
; X86-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-NEXT: vcmpunordss %xmm0, %xmm0, %xmm1
; X86-NEXT: vminss {{[0-9]+}}(%esp), %xmm0, %xmm2
; X86-NEXT: vblendvps %xmm1, %xmm0, %xmm2, %xmm0
; X86-NEXT: vmovss %xmm0, (%esp)
; X86-NEXT: flds (%esp)
; X86-NEXT: popl %eax
; X86-NEXT: retl
%1 = tail call nsz float @llvm.minimum.f32(float %x, float %y)
ret float %1
}
define float @test_fminimum_combine_cmps(float %x, float %y) nounwind {
; SSE2-LABEL: test_fminimum_combine_cmps:
; SSE2: # %bb.0:
; SSE2-NEXT: divss %xmm0, %xmm1
; SSE2-NEXT: movd %xmm0, %eax
; SSE2-NEXT: testl %eax, %eax
; SSE2-NEXT: movaps %xmm1, %xmm3
; SSE2-NEXT: js .LBB19_2
; SSE2-NEXT: # %bb.1:
; SSE2-NEXT: movaps %xmm0, %xmm3
; SSE2-NEXT: .LBB19_2:
; SSE2-NEXT: movaps %xmm3, %xmm2
; SSE2-NEXT: cmpunordss %xmm3, %xmm2
; SSE2-NEXT: movaps %xmm2, %xmm4
; SSE2-NEXT: andps %xmm3, %xmm4
; SSE2-NEXT: js .LBB19_4
; SSE2-NEXT: # %bb.3:
; SSE2-NEXT: movaps %xmm1, %xmm0
; SSE2-NEXT: .LBB19_4:
; SSE2-NEXT: minss %xmm0, %xmm3
; SSE2-NEXT: andnps %xmm3, %xmm2
; SSE2-NEXT: orps %xmm4, %xmm2
; SSE2-NEXT: movaps %xmm2, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fminimum_combine_cmps:
; AVX1: # %bb.0:
; AVX1-NEXT: vdivss %xmm0, %xmm1, %xmm2
; AVX1-NEXT: vmovd %xmm0, %eax
; AVX1-NEXT: testl %eax, %eax
; AVX1-NEXT: js .LBB19_1
; AVX1-NEXT: # %bb.2:
; AVX1-NEXT: vmovaps %xmm2, %xmm1
; AVX1-NEXT: jmp .LBB19_3
; AVX1-NEXT: .LBB19_1:
; AVX1-NEXT: vmovaps %xmm0, %xmm1
; AVX1-NEXT: vmovaps %xmm2, %xmm0
; AVX1-NEXT: .LBB19_3:
; AVX1-NEXT: vminss %xmm1, %xmm0, %xmm1
; AVX1-NEXT: vcmpunordss %xmm0, %xmm0, %xmm2
; AVX1-NEXT: vblendvps %xmm2, %xmm0, %xmm1, %xmm0
; AVX1-NEXT: retq
;
; AVX512F-LABEL: test_fminimum_combine_cmps:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vdivss %xmm0, %xmm1, %xmm1
; AVX512F-NEXT: vmovd %xmm0, %eax
; AVX512F-NEXT: testl %eax, %eax
; AVX512F-NEXT: sets %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovaps %xmm1, %xmm2
; AVX512F-NEXT: vmovss %xmm0, %xmm2, %xmm2 {%k1}
; AVX512F-NEXT: vmovss %xmm1, %xmm0, %xmm0 {%k1}
; AVX512F-NEXT: vminss %xmm2, %xmm0, %xmm1
; AVX512F-NEXT: vcmpunordss %xmm0, %xmm0, %k1
; AVX512F-NEXT: vmovss %xmm0, %xmm1, %xmm1 {%k1}
; AVX512F-NEXT: vmovaps %xmm1, %xmm0
; AVX512F-NEXT: retq
;
; AVX512DQ-LABEL: test_fminimum_combine_cmps:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vdivss %xmm0, %xmm1, %xmm1
; AVX512DQ-NEXT: vfpclassss $5, %xmm0, %k0
; AVX512DQ-NEXT: kmovw %k0, %k1
; AVX512DQ-NEXT: vmovaps %xmm1, %xmm2
; AVX512DQ-NEXT: vmovss %xmm0, %xmm2, %xmm2 {%k1}
; AVX512DQ-NEXT: vmovss %xmm1, %xmm0, %xmm0 {%k1}
; AVX512DQ-NEXT: vminss %xmm2, %xmm0, %xmm0
; AVX512DQ-NEXT: retq
;
; X86-LABEL: test_fminimum_combine_cmps:
; X86: # %bb.0:
; X86-NEXT: pushl %eax
; X86-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-NEXT: vmovss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; X86-NEXT: vdivss %xmm0, %xmm1, %xmm2
; X86-NEXT: vmovd %xmm0, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: js .LBB19_1
; X86-NEXT: # %bb.2:
; X86-NEXT: vmovaps %xmm2, %xmm1
; X86-NEXT: jmp .LBB19_3
; X86-NEXT: .LBB19_1:
; X86-NEXT: vmovaps %xmm0, %xmm1
; X86-NEXT: vmovaps %xmm2, %xmm0
; X86-NEXT: .LBB19_3:
; X86-NEXT: vminss %xmm1, %xmm0, %xmm1
; X86-NEXT: vcmpunordss %xmm0, %xmm0, %xmm2
; X86-NEXT: vblendvps %xmm2, %xmm0, %xmm1, %xmm0
; X86-NEXT: vmovss %xmm0, (%esp)
; X86-NEXT: flds (%esp)
; X86-NEXT: popl %eax
; X86-NEXT: retl
%1 = fdiv nnan float %y, %x
%2 = tail call float @llvm.minimum.f32(float %x, float %1)
ret float %2
}
define <2 x double> @test_fminimum_vector(<2 x double> %x, <2 x double> %y) {
; SSE2-LABEL: test_fminimum_vector:
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa %xmm0, %xmm4
; SSE2-NEXT: movq %xmm0, %rax
; SSE2-NEXT: testq %rax, %rax
; SSE2-NEXT: movdqa %xmm1, %xmm5
; SSE2-NEXT: js .LBB20_2
; SSE2-NEXT: # %bb.1:
; SSE2-NEXT: movdqa %xmm4, %xmm5
; SSE2-NEXT: .LBB20_2:
; SSE2-NEXT: movdqa %xmm5, %xmm2
; SSE2-NEXT: cmpunordsd %xmm5, %xmm2
; SSE2-NEXT: js .LBB20_4
; SSE2-NEXT: # %bb.3:
; SSE2-NEXT: movdqa %xmm1, %xmm4
; SSE2-NEXT: .LBB20_4:
; SSE2-NEXT: movapd %xmm2, %xmm3
; SSE2-NEXT: andpd %xmm5, %xmm3
; SSE2-NEXT: minsd %xmm4, %xmm5
; SSE2-NEXT: andnpd %xmm5, %xmm2
; SSE2-NEXT: punpckhqdq {{.*#+}} xmm1 = xmm1[1,1]
; SSE2-NEXT: punpckhqdq {{.*#+}} xmm0 = xmm0[1,1]
; SSE2-NEXT: movq %xmm0, %rax
; SSE2-NEXT: testq %rax, %rax
; SSE2-NEXT: movdqa %xmm1, %xmm4
; SSE2-NEXT: js .LBB20_6
; SSE2-NEXT: # %bb.5:
; SSE2-NEXT: movdqa %xmm0, %xmm4
; SSE2-NEXT: .LBB20_6:
; SSE2-NEXT: orpd %xmm3, %xmm2
; SSE2-NEXT: movdqa %xmm4, %xmm3
; SSE2-NEXT: cmpunordsd %xmm4, %xmm3
; SSE2-NEXT: movapd %xmm3, %xmm5
; SSE2-NEXT: andpd %xmm4, %xmm5
; SSE2-NEXT: js .LBB20_8
; SSE2-NEXT: # %bb.7:
; SSE2-NEXT: movdqa %xmm1, %xmm0
; SSE2-NEXT: .LBB20_8:
; SSE2-NEXT: minsd %xmm0, %xmm4
; SSE2-NEXT: andnpd %xmm4, %xmm3
; SSE2-NEXT: orpd %xmm5, %xmm3
; SSE2-NEXT: unpcklpd {{.*#+}} xmm2 = xmm2[0],xmm3[0]
; SSE2-NEXT: movapd %xmm2, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fminimum_vector:
; AVX1: # %bb.0:
; AVX1-NEXT: vmovq %xmm0, %rax
; AVX1-NEXT: testq %rax, %rax
; AVX1-NEXT: js .LBB20_1
; AVX1-NEXT: # %bb.2:
; AVX1-NEXT: vmovdqa %xmm1, %xmm2
; AVX1-NEXT: vmovdqa %xmm0, %xmm3
; AVX1-NEXT: jmp .LBB20_3
; AVX1-NEXT: .LBB20_1:
; AVX1-NEXT: vmovdqa %xmm0, %xmm2
; AVX1-NEXT: vmovdqa %xmm1, %xmm3
; AVX1-NEXT: .LBB20_3:
; AVX1-NEXT: vminsd %xmm2, %xmm3, %xmm2
; AVX1-NEXT: vcmpunordsd %xmm3, %xmm3, %xmm4
; AVX1-NEXT: vblendvpd %xmm4, %xmm3, %xmm2, %xmm2
; AVX1-NEXT: vshufpd {{.*#+}} xmm1 = xmm1[1,0]
; AVX1-NEXT: vshufpd {{.*#+}} xmm0 = xmm0[1,0]
; AVX1-NEXT: vmovq %xmm0, %rax
; AVX1-NEXT: testq %rax, %rax
; AVX1-NEXT: js .LBB20_4
; AVX1-NEXT: # %bb.5:
; AVX1-NEXT: vmovapd %xmm1, %xmm3
; AVX1-NEXT: jmp .LBB20_6
; AVX1-NEXT: .LBB20_4:
; AVX1-NEXT: vmovapd %xmm0, %xmm3
; AVX1-NEXT: vmovapd %xmm1, %xmm0
; AVX1-NEXT: .LBB20_6:
; AVX1-NEXT: vminsd %xmm3, %xmm0, %xmm1
; AVX1-NEXT: vcmpunordsd %xmm0, %xmm0, %xmm3
; AVX1-NEXT: vblendvpd %xmm3, %xmm0, %xmm1, %xmm0
; AVX1-NEXT: vunpcklpd {{.*#+}} xmm0 = xmm2[0],xmm0[0]
; AVX1-NEXT: retq
;
; AVX512-LABEL: test_fminimum_vector:
; AVX512: # %bb.0:
; AVX512-NEXT: vmovq %xmm0, %rax
; AVX512-NEXT: testq %rax, %rax
; AVX512-NEXT: sets %al
; AVX512-NEXT: kmovw %eax, %k1
; AVX512-NEXT: vmovdqa %xmm0, %xmm2
; AVX512-NEXT: vmovsd %xmm1, %xmm2, %xmm2 {%k1}
; AVX512-NEXT: vshufpd {{.*#+}} xmm3 = xmm1[1,0]
; AVX512-NEXT: vmovsd %xmm0, %xmm1, %xmm1 {%k1}
; AVX512-NEXT: vminsd %xmm1, %xmm2, %xmm1
; AVX512-NEXT: vcmpunordsd %xmm2, %xmm2, %k1
; AVX512-NEXT: vmovsd %xmm2, %xmm1, %xmm1 {%k1}
; AVX512-NEXT: vshufpd {{.*#+}} xmm0 = xmm0[1,0]
; AVX512-NEXT: vmovq %xmm0, %rax
; AVX512-NEXT: testq %rax, %rax
; AVX512-NEXT: sets %al
; AVX512-NEXT: kmovw %eax, %k1
; AVX512-NEXT: vmovapd %xmm3, %xmm2
; AVX512-NEXT: vmovsd %xmm0, %xmm2, %xmm2 {%k1}
; AVX512-NEXT: vmovsd %xmm3, %xmm0, %xmm0 {%k1}
; AVX512-NEXT: vminsd %xmm2, %xmm0, %xmm2
; AVX512-NEXT: vcmpunordsd %xmm0, %xmm0, %k1
; AVX512-NEXT: vmovsd %xmm0, %xmm2, %xmm2 {%k1}
; AVX512-NEXT: vunpcklpd {{.*#+}} xmm0 = xmm1[0],xmm2[0]
; AVX512-NEXT: retq
;
; X86-LABEL: test_fminimum_vector:
; X86: # %bb.0:
; X86-NEXT: vshufpd {{.*#+}} xmm2 = xmm0[1,0]
; X86-NEXT: vshufpd {{.*#+}} xmm3 = xmm1[1,0]
; X86-NEXT: vextractps $3, %xmm0, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: js .LBB20_1
; X86-NEXT: # %bb.2:
; X86-NEXT: vmovapd %xmm3, %xmm4
; X86-NEXT: jmp .LBB20_3
; X86-NEXT: .LBB20_1:
; X86-NEXT: vmovapd %xmm2, %xmm4
; X86-NEXT: vmovapd %xmm3, %xmm2
; X86-NEXT: .LBB20_3:
; X86-NEXT: vminsd %xmm4, %xmm2, %xmm3
; X86-NEXT: vcmpunordsd %xmm2, %xmm2, %xmm4
; X86-NEXT: vblendvpd %xmm4, %xmm2, %xmm3, %xmm2
; X86-NEXT: vextractps $1, %xmm0, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: js .LBB20_4
; X86-NEXT: # %bb.5:
; X86-NEXT: vmovapd %xmm1, %xmm3
; X86-NEXT: jmp .LBB20_6
; X86-NEXT: .LBB20_4:
; X86-NEXT: vmovapd %xmm0, %xmm3
; X86-NEXT: vmovapd %xmm1, %xmm0
; X86-NEXT: .LBB20_6:
; X86-NEXT: vminsd %xmm3, %xmm0, %xmm1
; X86-NEXT: vcmpunordsd %xmm0, %xmm0, %xmm3
; X86-NEXT: vblendvpd %xmm3, %xmm0, %xmm1, %xmm0
; X86-NEXT: vunpcklpd {{.*#+}} xmm0 = xmm0[0],xmm2[0]
; X86-NEXT: retl
%r = call <2 x double> @llvm.minimum.v2f64(<2 x double> %x, <2 x double> %y)
ret <2 x double> %r
}
define <4 x float> @test_fmaximum_vector(<4 x float> %x, <4 x float> %y) "no-nans-fp-math"="true" "no-signed-zeros-fp-math"="true" {
; SSE2-LABEL: test_fmaximum_vector:
; SSE2: # %bb.0:
; SSE2-NEXT: movaps %xmm1, %xmm2
; SSE2-NEXT: shufps {{.*#+}} xmm2 = xmm2[3,3],xmm1[3,3]
; SSE2-NEXT: movaps %xmm0, %xmm3
; SSE2-NEXT: shufps {{.*#+}} xmm3 = xmm3[3,3],xmm0[3,3]
; SSE2-NEXT: maxss %xmm2, %xmm3
; SSE2-NEXT: movaps %xmm1, %xmm2
; SSE2-NEXT: unpckhpd {{.*#+}} xmm2 = xmm2[1],xmm1[1]
; SSE2-NEXT: movaps %xmm0, %xmm4
; SSE2-NEXT: unpckhpd {{.*#+}} xmm4 = xmm4[1],xmm0[1]
; SSE2-NEXT: maxss %xmm2, %xmm4
; SSE2-NEXT: unpcklps {{.*#+}} xmm4 = xmm4[0],xmm3[0],xmm4[1],xmm3[1]
; SSE2-NEXT: movaps %xmm0, %xmm2
; SSE2-NEXT: maxss %xmm1, %xmm2
; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[1,1,1,1]
; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[1,1,1,1]
; SSE2-NEXT: maxss %xmm1, %xmm0
; SSE2-NEXT: unpcklps {{.*#+}} xmm2 = xmm2[0],xmm0[0],xmm2[1],xmm0[1]
; SSE2-NEXT: movlhps {{.*#+}} xmm2 = xmm2[0],xmm4[0]
; SSE2-NEXT: movaps %xmm2, %xmm0
; SSE2-NEXT: retq
;
; AVX-LABEL: test_fmaximum_vector:
; AVX: # %bb.0:
; AVX-NEXT: vmaxss %xmm1, %xmm0, %xmm2
; AVX-NEXT: vmovshdup {{.*#+}} xmm3 = xmm1[1,1,3,3]
; AVX-NEXT: vmovshdup {{.*#+}} xmm4 = xmm0[1,1,3,3]
; AVX-NEXT: vmaxss %xmm3, %xmm4, %xmm3
; AVX-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0],xmm3[0],xmm2[2,3]
; AVX-NEXT: vshufpd {{.*#+}} xmm3 = xmm1[1,0]
; AVX-NEXT: vshufpd {{.*#+}} xmm4 = xmm0[1,0]
; AVX-NEXT: vmaxss %xmm3, %xmm4, %xmm3
; AVX-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1],xmm3[0],xmm2[3]
; AVX-NEXT: vshufps {{.*#+}} xmm1 = xmm1[3,3,3,3]
; AVX-NEXT: vshufps {{.*#+}} xmm0 = xmm0[3,3,3,3]
; AVX-NEXT: vmaxss %xmm1, %xmm0, %xmm0
; AVX-NEXT: vinsertps {{.*#+}} xmm0 = xmm2[0,1,2],xmm0[0]
; AVX-NEXT: retq
;
; X86-LABEL: test_fmaximum_vector:
; X86: # %bb.0:
; X86-NEXT: vmaxss %xmm1, %xmm0, %xmm2
; X86-NEXT: vmovshdup {{.*#+}} xmm3 = xmm1[1,1,3,3]
; X86-NEXT: vmovshdup {{.*#+}} xmm4 = xmm0[1,1,3,3]
; X86-NEXT: vmaxss %xmm3, %xmm4, %xmm3
; X86-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0],xmm3[0],xmm2[2,3]
; X86-NEXT: vshufpd {{.*#+}} xmm3 = xmm1[1,0]
; X86-NEXT: vshufpd {{.*#+}} xmm4 = xmm0[1,0]
; X86-NEXT: vmaxss %xmm3, %xmm4, %xmm3
; X86-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1],xmm3[0],xmm2[3]
; X86-NEXT: vshufps {{.*#+}} xmm1 = xmm1[3,3,3,3]
; X86-NEXT: vshufps {{.*#+}} xmm0 = xmm0[3,3,3,3]
; X86-NEXT: vmaxss %xmm1, %xmm0, %xmm0
; X86-NEXT: vinsertps {{.*#+}} xmm0 = xmm2[0,1,2],xmm0[0]
; X86-NEXT: retl
%r = call <4 x float> @llvm.maximum.v4f32(<4 x float> %x, <4 x float> %y)
ret <4 x float> %r
}
define <2 x double> @test_fminimum_vector_zero(<2 x double> %x) {
; SSE2-LABEL: test_fminimum_vector_zero:
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: movq %xmm0, %rax
; SSE2-NEXT: testq %rax, %rax
; SSE2-NEXT: pxor %xmm4, %xmm4
; SSE2-NEXT: js .LBB22_2
; SSE2-NEXT: # %bb.1:
; SSE2-NEXT: movdqa %xmm2, %xmm4
; SSE2-NEXT: .LBB22_2:
; SSE2-NEXT: movdqa %xmm4, %xmm1
; SSE2-NEXT: cmpunordsd %xmm4, %xmm1
; SSE2-NEXT: js .LBB22_4
; SSE2-NEXT: # %bb.3:
; SSE2-NEXT: pxor %xmm2, %xmm2
; SSE2-NEXT: .LBB22_4:
; SSE2-NEXT: movapd %xmm1, %xmm3
; SSE2-NEXT: andpd %xmm4, %xmm3
; SSE2-NEXT: minsd %xmm2, %xmm4
; SSE2-NEXT: andnpd %xmm4, %xmm1
; SSE2-NEXT: punpckhqdq {{.*#+}} xmm0 = xmm0[1,1]
; SSE2-NEXT: movq %xmm0, %rax
; SSE2-NEXT: testq %rax, %rax
; SSE2-NEXT: pxor %xmm2, %xmm2
; SSE2-NEXT: js .LBB22_6
; SSE2-NEXT: # %bb.5:
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: .LBB22_6:
; SSE2-NEXT: orpd %xmm3, %xmm1
; SSE2-NEXT: movdqa %xmm2, %xmm3
; SSE2-NEXT: cmpunordsd %xmm2, %xmm3
; SSE2-NEXT: movapd %xmm3, %xmm4
; SSE2-NEXT: andpd %xmm2, %xmm4
; SSE2-NEXT: js .LBB22_8
; SSE2-NEXT: # %bb.7:
; SSE2-NEXT: pxor %xmm0, %xmm0
; SSE2-NEXT: .LBB22_8:
; SSE2-NEXT: minsd %xmm0, %xmm2
; SSE2-NEXT: andnpd %xmm2, %xmm3
; SSE2-NEXT: orpd %xmm4, %xmm3
; SSE2-NEXT: unpcklpd {{.*#+}} xmm1 = xmm1[0],xmm3[0]
; SSE2-NEXT: movapd %xmm1, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fminimum_vector_zero:
; AVX1: # %bb.0:
; AVX1-NEXT: vmovq %xmm0, %rax
; AVX1-NEXT: testq %rax, %rax
; AVX1-NEXT: js .LBB22_1
; AVX1-NEXT: # %bb.2:
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX1-NEXT: vmovdqa %xmm0, %xmm2
; AVX1-NEXT: jmp .LBB22_3
; AVX1-NEXT: .LBB22_1:
; AVX1-NEXT: vmovdqa %xmm0, %xmm1
; AVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2
; AVX1-NEXT: .LBB22_3:
; AVX1-NEXT: vminsd %xmm1, %xmm2, %xmm1
; AVX1-NEXT: vcmpunordsd %xmm2, %xmm2, %xmm3
; AVX1-NEXT: vblendvpd %xmm3, %xmm2, %xmm1, %xmm1
; AVX1-NEXT: vshufpd {{.*#+}} xmm0 = xmm0[1,0]
; AVX1-NEXT: vmovq %xmm0, %rax
; AVX1-NEXT: testq %rax, %rax
; AVX1-NEXT: js .LBB22_4
; AVX1-NEXT: # %bb.5:
; AVX1-NEXT: vxorpd %xmm2, %xmm2, %xmm2
; AVX1-NEXT: jmp .LBB22_6
; AVX1-NEXT: .LBB22_4:
; AVX1-NEXT: vmovapd %xmm0, %xmm2
; AVX1-NEXT: vxorpd %xmm0, %xmm0, %xmm0
; AVX1-NEXT: .LBB22_6:
; AVX1-NEXT: vminsd %xmm2, %xmm0, %xmm2
; AVX1-NEXT: vcmpunordsd %xmm0, %xmm0, %xmm3
; AVX1-NEXT: vblendvpd %xmm3, %xmm0, %xmm2, %xmm0
; AVX1-NEXT: vunpcklpd {{.*#+}} xmm0 = xmm1[0],xmm0[0]
; AVX1-NEXT: retq
;
; AVX512F-LABEL: test_fminimum_vector_zero:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovq %xmm0, %rax
; AVX512F-NEXT: testq %rax, %rax
; AVX512F-NEXT: sets %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovsd %xmm0, %xmm0, %xmm1 {%k1} {z}
; AVX512F-NEXT: vxorpd %xmm2, %xmm2, %xmm2
; AVX512F-NEXT: vshufpd {{.*#+}} xmm3 = xmm0[1,0]
; AVX512F-NEXT: vmovsd %xmm2, %xmm0, %xmm0 {%k1}
; AVX512F-NEXT: vminsd %xmm1, %xmm0, %xmm1
; AVX512F-NEXT: vcmpunordsd %xmm0, %xmm0, %k1
; AVX512F-NEXT: vmovsd %xmm0, %xmm1, %xmm1 {%k1}
; AVX512F-NEXT: vmovq %xmm3, %rax
; AVX512F-NEXT: testq %rax, %rax
; AVX512F-NEXT: sets %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovsd %xmm3, %xmm3, %xmm0 {%k1} {z}
; AVX512F-NEXT: vmovsd %xmm2, %xmm3, %xmm3 {%k1}
; AVX512F-NEXT: vminsd %xmm0, %xmm3, %xmm0
; AVX512F-NEXT: vcmpunordsd %xmm3, %xmm3, %k1
; AVX512F-NEXT: vmovsd %xmm3, %xmm0, %xmm0 {%k1}
; AVX512F-NEXT: vunpcklpd {{.*#+}} xmm0 = xmm1[0],xmm0[0]
; AVX512F-NEXT: retq
;
; AVX512DQ-LABEL: test_fminimum_vector_zero:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vfpclasssd $5, %xmm0, %k0
; AVX512DQ-NEXT: kmovw %k0, %k1
; AVX512DQ-NEXT: vmovsd %xmm0, %xmm0, %xmm1 {%k1} {z}
; AVX512DQ-NEXT: vxorpd %xmm2, %xmm2, %xmm2
; AVX512DQ-NEXT: vshufpd {{.*#+}} xmm3 = xmm0[1,0]
; AVX512DQ-NEXT: vmovsd %xmm2, %xmm0, %xmm0 {%k1}
; AVX512DQ-NEXT: vminsd %xmm1, %xmm0, %xmm0
; AVX512DQ-NEXT: vfpclasssd $5, %xmm3, %k0
; AVX512DQ-NEXT: kmovw %k0, %k1
; AVX512DQ-NEXT: vmovsd %xmm3, %xmm3, %xmm1 {%k1} {z}
; AVX512DQ-NEXT: vmovsd %xmm2, %xmm3, %xmm3 {%k1}
; AVX512DQ-NEXT: vminsd %xmm1, %xmm3, %xmm1
; AVX512DQ-NEXT: vunpcklpd {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; AVX512DQ-NEXT: retq
;
; X86-LABEL: test_fminimum_vector_zero:
; X86: # %bb.0:
; X86-NEXT: vshufpd {{.*#+}} xmm1 = xmm0[1,0]
; X86-NEXT: vextractps $3, %xmm0, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: js .LBB22_1
; X86-NEXT: # %bb.2:
; X86-NEXT: vxorpd %xmm2, %xmm2, %xmm2
; X86-NEXT: jmp .LBB22_3
; X86-NEXT: .LBB22_1:
; X86-NEXT: vmovapd %xmm1, %xmm2
; X86-NEXT: vxorpd %xmm1, %xmm1, %xmm1
; X86-NEXT: .LBB22_3:
; X86-NEXT: vminsd %xmm2, %xmm1, %xmm2
; X86-NEXT: vcmpunordsd %xmm1, %xmm1, %xmm3
; X86-NEXT: vblendvpd %xmm3, %xmm1, %xmm2, %xmm1
; X86-NEXT: vextractps $1, %xmm0, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: js .LBB22_4
; X86-NEXT: # %bb.5:
; X86-NEXT: vxorpd %xmm2, %xmm2, %xmm2
; X86-NEXT: jmp .LBB22_6
; X86-NEXT: .LBB22_4:
; X86-NEXT: vmovapd %xmm0, %xmm2
; X86-NEXT: vxorpd %xmm0, %xmm0, %xmm0
; X86-NEXT: .LBB22_6:
; X86-NEXT: vminsd %xmm2, %xmm0, %xmm2
; X86-NEXT: vcmpunordsd %xmm0, %xmm0, %xmm3
; X86-NEXT: vblendvpd %xmm3, %xmm0, %xmm2, %xmm0
; X86-NEXT: vunpcklpd {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; X86-NEXT: retl
%r = call <2 x double> @llvm.minimum.v2f64(<2 x double> %x, <2 x double> <double 0., double 0.>)
ret <2 x double> %r
}
define <4 x float> @test_fmaximum_vector_signed_zero(<4 x float> %x) {
; SSE2-LABEL: test_fmaximum_vector_signed_zero:
; SSE2: # %bb.0:
; SSE2-NEXT: movaps %xmm0, %xmm3
; SSE2-NEXT: shufps {{.*#+}} xmm3 = xmm3[3,3],xmm0[3,3]
; SSE2-NEXT: movd %xmm3, %eax
; SSE2-NEXT: testl %eax, %eax
; SSE2-NEXT: movss {{.*#+}} xmm2 = mem[0],zero,zero,zero
; SSE2-NEXT: movaps %xmm3, %xmm4
; SSE2-NEXT: js .LBB23_2
; SSE2-NEXT: # %bb.1:
; SSE2-NEXT: movaps %xmm2, %xmm4
; SSE2-NEXT: .LBB23_2:
; SSE2-NEXT: movaps %xmm2, %xmm1
; SSE2-NEXT: js .LBB23_4
; SSE2-NEXT: # %bb.3:
; SSE2-NEXT: movaps %xmm3, %xmm1
; SSE2-NEXT: .LBB23_4:
; SSE2-NEXT: movaps %xmm0, %xmm3
; SSE2-NEXT: unpckhpd {{.*#+}} xmm3 = xmm3[1],xmm0[1]
; SSE2-NEXT: movd %xmm3, %eax
; SSE2-NEXT: testl %eax, %eax
; SSE2-NEXT: movaps %xmm3, %xmm6
; SSE2-NEXT: js .LBB23_6
; SSE2-NEXT: # %bb.5:
; SSE2-NEXT: movaps %xmm2, %xmm6
; SSE2-NEXT: .LBB23_6:
; SSE2-NEXT: movaps %xmm2, %xmm7
; SSE2-NEXT: js .LBB23_8
; SSE2-NEXT: # %bb.7:
; SSE2-NEXT: movaps %xmm3, %xmm7
; SSE2-NEXT: .LBB23_8:
; SSE2-NEXT: movaps %xmm4, %xmm5
; SSE2-NEXT: cmpunordss %xmm4, %xmm5
; SSE2-NEXT: movaps %xmm4, %xmm8
; SSE2-NEXT: maxss %xmm1, %xmm8
; SSE2-NEXT: movaps %xmm6, %xmm3
; SSE2-NEXT: cmpunordss %xmm6, %xmm3
; SSE2-NEXT: movaps %xmm6, %xmm1
; SSE2-NEXT: maxss %xmm7, %xmm1
; SSE2-NEXT: movd %xmm0, %eax
; SSE2-NEXT: testl %eax, %eax
; SSE2-NEXT: movaps %xmm0, %xmm7
; SSE2-NEXT: js .LBB23_10
; SSE2-NEXT: # %bb.9:
; SSE2-NEXT: movaps %xmm2, %xmm7
; SSE2-NEXT: .LBB23_10:
; SSE2-NEXT: andps %xmm5, %xmm4
; SSE2-NEXT: andnps %xmm8, %xmm5
; SSE2-NEXT: andps %xmm3, %xmm6
; SSE2-NEXT: andnps %xmm1, %xmm3
; SSE2-NEXT: movaps %xmm7, %xmm1
; SSE2-NEXT: cmpunordss %xmm7, %xmm1
; SSE2-NEXT: movaps %xmm2, %xmm8
; SSE2-NEXT: js .LBB23_12
; SSE2-NEXT: # %bb.11:
; SSE2-NEXT: movaps %xmm0, %xmm8
; SSE2-NEXT: .LBB23_12:
; SSE2-NEXT: orps %xmm4, %xmm5
; SSE2-NEXT: orps %xmm6, %xmm3
; SSE2-NEXT: movaps %xmm1, %xmm6
; SSE2-NEXT: andps %xmm7, %xmm6
; SSE2-NEXT: maxss %xmm8, %xmm7
; SSE2-NEXT: andnps %xmm7, %xmm1
; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[1,1,1,1]
; SSE2-NEXT: movd %xmm0, %eax
; SSE2-NEXT: testl %eax, %eax
; SSE2-NEXT: movaps %xmm0, %xmm4
; SSE2-NEXT: js .LBB23_14
; SSE2-NEXT: # %bb.13:
; SSE2-NEXT: movaps %xmm2, %xmm4
; SSE2-NEXT: .LBB23_14:
; SSE2-NEXT: unpcklps {{.*#+}} xmm3 = xmm3[0],xmm5[0],xmm3[1],xmm5[1]
; SSE2-NEXT: orps %xmm6, %xmm1
; SSE2-NEXT: movaps %xmm4, %xmm5
; SSE2-NEXT: cmpunordss %xmm4, %xmm5
; SSE2-NEXT: movaps %xmm5, %xmm6
; SSE2-NEXT: andps %xmm4, %xmm6
; SSE2-NEXT: js .LBB23_16
; SSE2-NEXT: # %bb.15:
; SSE2-NEXT: movaps %xmm0, %xmm2
; SSE2-NEXT: .LBB23_16:
; SSE2-NEXT: maxss %xmm2, %xmm4
; SSE2-NEXT: andnps %xmm4, %xmm5
; SSE2-NEXT: orps %xmm6, %xmm5
; SSE2-NEXT: unpcklps {{.*#+}} xmm1 = xmm1[0],xmm5[0],xmm1[1],xmm5[1]
; SSE2-NEXT: movlhps {{.*#+}} xmm1 = xmm1[0],xmm3[0]
; SSE2-NEXT: movaps %xmm1, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fmaximum_vector_signed_zero:
; AVX1: # %bb.0:
; AVX1-NEXT: vmovd %xmm0, %eax
; AVX1-NEXT: testl %eax, %eax
; AVX1-NEXT: vmovd {{.*#+}} xmm1 = mem[0],zero,zero,zero
; AVX1-NEXT: js .LBB23_1
; AVX1-NEXT: # %bb.2:
; AVX1-NEXT: vmovdqa %xmm0, %xmm2
; AVX1-NEXT: vmovdqa %xmm1, %xmm3
; AVX1-NEXT: jmp .LBB23_3
; AVX1-NEXT: .LBB23_1:
; AVX1-NEXT: vmovdqa %xmm1, %xmm2
; AVX1-NEXT: vmovdqa %xmm0, %xmm3
; AVX1-NEXT: .LBB23_3:
; AVX1-NEXT: vmaxss %xmm2, %xmm3, %xmm2
; AVX1-NEXT: vcmpunordss %xmm3, %xmm3, %xmm4
; AVX1-NEXT: vblendvps %xmm4, %xmm3, %xmm2, %xmm2
; AVX1-NEXT: vmovshdup {{.*#+}} xmm3 = xmm0[1,1,3,3]
; AVX1-NEXT: vmovd %xmm3, %eax
; AVX1-NEXT: testl %eax, %eax
; AVX1-NEXT: js .LBB23_4
; AVX1-NEXT: # %bb.5:
; AVX1-NEXT: vmovdqa %xmm3, %xmm4
; AVX1-NEXT: vmovdqa %xmm1, %xmm3
; AVX1-NEXT: jmp .LBB23_6
; AVX1-NEXT: .LBB23_4:
; AVX1-NEXT: vmovdqa %xmm1, %xmm4
; AVX1-NEXT: .LBB23_6:
; AVX1-NEXT: vmaxss %xmm4, %xmm3, %xmm4
; AVX1-NEXT: vcmpunordss %xmm3, %xmm3, %xmm5
; AVX1-NEXT: vblendvps %xmm5, %xmm3, %xmm4, %xmm3
; AVX1-NEXT: vunpcklps {{.*#+}} xmm2 = xmm2[0],xmm3[0],xmm2[1],xmm3[1]
; AVX1-NEXT: vshufpd {{.*#+}} xmm3 = xmm0[1,0]
; AVX1-NEXT: vmovd %xmm3, %eax
; AVX1-NEXT: testl %eax, %eax
; AVX1-NEXT: js .LBB23_7
; AVX1-NEXT: # %bb.8:
; AVX1-NEXT: vmovdqa %xmm3, %xmm4
; AVX1-NEXT: vmovdqa %xmm1, %xmm3
; AVX1-NEXT: jmp .LBB23_9
; AVX1-NEXT: .LBB23_7:
; AVX1-NEXT: vmovdqa %xmm1, %xmm4
; AVX1-NEXT: .LBB23_9:
; AVX1-NEXT: vmaxss %xmm4, %xmm3, %xmm4
; AVX1-NEXT: vcmpunordss %xmm3, %xmm3, %xmm5
; AVX1-NEXT: vblendvps %xmm5, %xmm3, %xmm4, %xmm3
; AVX1-NEXT: vmovlhps {{.*#+}} xmm2 = xmm2[0],xmm3[0]
; AVX1-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[3,3,3,3]
; AVX1-NEXT: vmovd %xmm0, %eax
; AVX1-NEXT: testl %eax, %eax
; AVX1-NEXT: js .LBB23_10
; AVX1-NEXT: # %bb.11:
; AVX1-NEXT: vmovdqa %xmm0, %xmm3
; AVX1-NEXT: jmp .LBB23_12
; AVX1-NEXT: .LBB23_10:
; AVX1-NEXT: vmovdqa %xmm1, %xmm3
; AVX1-NEXT: vmovdqa %xmm0, %xmm1
; AVX1-NEXT: .LBB23_12:
; AVX1-NEXT: vmaxss %xmm3, %xmm1, %xmm0
; AVX1-NEXT: vcmpunordss %xmm1, %xmm1, %xmm3
; AVX1-NEXT: vblendvps %xmm3, %xmm1, %xmm0, %xmm0
; AVX1-NEXT: vinsertps {{.*#+}} xmm0 = xmm2[0,1,2],xmm0[0]
; AVX1-NEXT: retq
;
; AVX512F-LABEL: test_fmaximum_vector_signed_zero:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovd %xmm0, %eax
; AVX512F-NEXT: testl %eax, %eax
; AVX512F-NEXT: sets %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; AVX512F-NEXT: vmovaps %xmm1, %xmm2
; AVX512F-NEXT: vmovss %xmm0, %xmm2, %xmm2 {%k1}
; AVX512F-NEXT: vmovshdup {{.*#+}} xmm3 = xmm0[1,1,3,3]
; AVX512F-NEXT: vshufpd {{.*#+}} xmm4 = xmm0[1,0]
; AVX512F-NEXT: vpshufd {{.*#+}} xmm5 = xmm0[3,3,3,3]
; AVX512F-NEXT: vmovss %xmm1, %xmm0, %xmm0 {%k1}
; AVX512F-NEXT: vmaxss %xmm0, %xmm2, %xmm0
; AVX512F-NEXT: vcmpunordss %xmm2, %xmm2, %k1
; AVX512F-NEXT: vmovss %xmm2, %xmm0, %xmm0 {%k1}
; AVX512F-NEXT: vmovd %xmm3, %eax
; AVX512F-NEXT: testl %eax, %eax
; AVX512F-NEXT: sets %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovaps %xmm1, %xmm2
; AVX512F-NEXT: vmovss %xmm3, %xmm2, %xmm2 {%k1}
; AVX512F-NEXT: vmovss %xmm1, %xmm3, %xmm3 {%k1}
; AVX512F-NEXT: vmaxss %xmm3, %xmm2, %xmm3
; AVX512F-NEXT: vcmpunordss %xmm2, %xmm2, %k1
; AVX512F-NEXT: vmovss %xmm2, %xmm3, %xmm3 {%k1}
; AVX512F-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0],xmm3[0],xmm0[2,3]
; AVX512F-NEXT: vmovd %xmm4, %eax
; AVX512F-NEXT: testl %eax, %eax
; AVX512F-NEXT: sets %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovaps %xmm1, %xmm2
; AVX512F-NEXT: vmovss %xmm4, %xmm2, %xmm2 {%k1}
; AVX512F-NEXT: vmovss %xmm1, %xmm4, %xmm4 {%k1}
; AVX512F-NEXT: vmaxss %xmm4, %xmm2, %xmm3
; AVX512F-NEXT: vcmpunordss %xmm2, %xmm2, %k1
; AVX512F-NEXT: vmovss %xmm2, %xmm3, %xmm3 {%k1}
; AVX512F-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0,1],xmm3[0],xmm0[3]
; AVX512F-NEXT: vmovd %xmm5, %eax
; AVX512F-NEXT: testl %eax, %eax
; AVX512F-NEXT: sets %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovdqa %xmm5, %xmm2
; AVX512F-NEXT: vmovss %xmm1, %xmm2, %xmm2 {%k1}
; AVX512F-NEXT: vmovss %xmm5, %xmm1, %xmm1 {%k1}
; AVX512F-NEXT: vmaxss %xmm2, %xmm1, %xmm2
; AVX512F-NEXT: vcmpunordss %xmm1, %xmm1, %k1
; AVX512F-NEXT: vmovss %xmm1, %xmm2, %xmm2 {%k1}
; AVX512F-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0,1,2],xmm2[0]
; AVX512F-NEXT: retq
;
; AVX512DQ-LABEL: test_fmaximum_vector_signed_zero:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vfpclassss $3, %xmm0, %k0
; AVX512DQ-NEXT: kmovw %k0, %eax
; AVX512DQ-NEXT: vmovss {{.*#+}} xmm3 = mem[0],zero,zero,zero
; AVX512DQ-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
; AVX512DQ-NEXT: vfpclassss $3, %xmm1, %k0
; AVX512DQ-NEXT: kmovw %k0, %k1
; AVX512DQ-NEXT: vmovaps %xmm3, %xmm2
; AVX512DQ-NEXT: vmovss %xmm1, %xmm2, %xmm2 {%k1}
; AVX512DQ-NEXT: vmovss %xmm3, %xmm1, %xmm1 {%k1}
; AVX512DQ-NEXT: vshufpd {{.*#+}} xmm4 = xmm0[1,0]
; AVX512DQ-NEXT: vfpclassss $3, %xmm4, %k0
; AVX512DQ-NEXT: kmovw %k0, %k1
; AVX512DQ-NEXT: vmovaps %xmm3, %xmm5
; AVX512DQ-NEXT: vmovss %xmm4, %xmm5, %xmm5 {%k1}
; AVX512DQ-NEXT: vmovss %xmm3, %xmm4, %xmm4 {%k1}
; AVX512DQ-NEXT: vshufps {{.*#+}} xmm6 = xmm0[3,3,3,3]
; AVX512DQ-NEXT: vfpclassss $3, %xmm6, %k0
; AVX512DQ-NEXT: kmovw %k0, %k1
; AVX512DQ-NEXT: vmovaps %xmm3, %xmm7
; AVX512DQ-NEXT: vmovss %xmm6, %xmm7, %xmm7 {%k1}
; AVX512DQ-NEXT: vmovss %xmm3, %xmm6, %xmm6 {%k1}
; AVX512DQ-NEXT: kmovw %eax, %k1
; AVX512DQ-NEXT: vmovaps %xmm0, %xmm8
; AVX512DQ-NEXT: vmovss %xmm3, %xmm8, %xmm8 {%k1}
; AVX512DQ-NEXT: vmovss %xmm0, %xmm3, %xmm3 {%k1}
; AVX512DQ-NEXT: vmaxss %xmm3, %xmm8, %xmm0
; AVX512DQ-NEXT: vmaxss %xmm2, %xmm1, %xmm1
; AVX512DQ-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[2,3]
; AVX512DQ-NEXT: vmaxss %xmm5, %xmm4, %xmm1
; AVX512DQ-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0],xmm0[3]
; AVX512DQ-NEXT: vmaxss %xmm7, %xmm6, %xmm1
; AVX512DQ-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0,1,2],xmm1[0]
; AVX512DQ-NEXT: retq
;
; X86-LABEL: test_fmaximum_vector_signed_zero:
; X86: # %bb.0:
; X86-NEXT: vmovd %xmm0, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: vmovd {{.*#+}} xmm1 = mem[0],zero,zero,zero
; X86-NEXT: js .LBB23_1
; X86-NEXT: # %bb.2:
; X86-NEXT: vmovdqa %xmm0, %xmm2
; X86-NEXT: vmovdqa %xmm1, %xmm3
; X86-NEXT: jmp .LBB23_3
; X86-NEXT: .LBB23_1:
; X86-NEXT: vmovdqa %xmm1, %xmm2
; X86-NEXT: vmovdqa %xmm0, %xmm3
; X86-NEXT: .LBB23_3:
; X86-NEXT: vmaxss %xmm2, %xmm3, %xmm2
; X86-NEXT: vcmpunordss %xmm3, %xmm3, %xmm4
; X86-NEXT: vblendvps %xmm4, %xmm3, %xmm2, %xmm2
; X86-NEXT: vmovshdup {{.*#+}} xmm3 = xmm0[1,1,3,3]
; X86-NEXT: vmovd %xmm3, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: js .LBB23_4
; X86-NEXT: # %bb.5:
; X86-NEXT: vmovdqa %xmm3, %xmm4
; X86-NEXT: vmovdqa %xmm1, %xmm3
; X86-NEXT: jmp .LBB23_6
; X86-NEXT: .LBB23_4:
; X86-NEXT: vmovdqa %xmm1, %xmm4
; X86-NEXT: .LBB23_6:
; X86-NEXT: vmaxss %xmm4, %xmm3, %xmm4
; X86-NEXT: vcmpunordss %xmm3, %xmm3, %xmm5
; X86-NEXT: vblendvps %xmm5, %xmm3, %xmm4, %xmm3
; X86-NEXT: vunpcklps {{.*#+}} xmm2 = xmm2[0],xmm3[0],xmm2[1],xmm3[1]
; X86-NEXT: vshufpd {{.*#+}} xmm3 = xmm0[1,0]
; X86-NEXT: vmovd %xmm3, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: js .LBB23_7
; X86-NEXT: # %bb.8:
; X86-NEXT: vmovdqa %xmm3, %xmm4
; X86-NEXT: vmovdqa %xmm1, %xmm3
; X86-NEXT: jmp .LBB23_9
; X86-NEXT: .LBB23_7:
; X86-NEXT: vmovdqa %xmm1, %xmm4
; X86-NEXT: .LBB23_9:
; X86-NEXT: vmaxss %xmm4, %xmm3, %xmm4
; X86-NEXT: vcmpunordss %xmm3, %xmm3, %xmm5
; X86-NEXT: vblendvps %xmm5, %xmm3, %xmm4, %xmm3
; X86-NEXT: vmovlhps {{.*#+}} xmm2 = xmm2[0],xmm3[0]
; X86-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[3,3,3,3]
; X86-NEXT: vmovd %xmm0, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: js .LBB23_10
; X86-NEXT: # %bb.11:
; X86-NEXT: vmovdqa %xmm0, %xmm3
; X86-NEXT: jmp .LBB23_12
; X86-NEXT: .LBB23_10:
; X86-NEXT: vmovdqa %xmm1, %xmm3
; X86-NEXT: vmovdqa %xmm0, %xmm1
; X86-NEXT: .LBB23_12:
; X86-NEXT: vmaxss %xmm3, %xmm1, %xmm0
; X86-NEXT: vcmpunordss %xmm1, %xmm1, %xmm3
; X86-NEXT: vblendvps %xmm3, %xmm1, %xmm0, %xmm0
; X86-NEXT: vinsertps {{.*#+}} xmm0 = xmm2[0,1,2],xmm0[0]
; X86-NEXT: retl
%r = call <4 x float> @llvm.maximum.v4f32(<4 x float> %x, <4 x float> <float -0., float -0., float -0., float -0.>)
ret <4 x float> %r
}
define <2 x double> @test_fminimum_vector_partially_zero(<2 x double> %x) {
; SSE2-LABEL: test_fminimum_vector_partially_zero:
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: movq %xmm0, %rax
; SSE2-NEXT: testq %rax, %rax
; SSE2-NEXT: pxor %xmm3, %xmm3
; SSE2-NEXT: js .LBB24_2
; SSE2-NEXT: # %bb.1:
; SSE2-NEXT: movdqa %xmm2, %xmm3
; SSE2-NEXT: .LBB24_2:
; SSE2-NEXT: movdqa %xmm3, %xmm1
; SSE2-NEXT: cmpunordsd %xmm3, %xmm1
; SSE2-NEXT: movapd %xmm1, %xmm4
; SSE2-NEXT: andpd %xmm3, %xmm4
; SSE2-NEXT: js .LBB24_4
; SSE2-NEXT: # %bb.3:
; SSE2-NEXT: pxor %xmm2, %xmm2
; SSE2-NEXT: .LBB24_4:
; SSE2-NEXT: minsd %xmm2, %xmm3
; SSE2-NEXT: andnpd %xmm3, %xmm1
; SSE2-NEXT: orpd %xmm4, %xmm1
; SSE2-NEXT: punpckhqdq {{.*#+}} xmm0 = xmm0[1,1]
; SSE2-NEXT: movsd {{.*#+}} xmm2 = mem[0],zero
; SSE2-NEXT: minsd %xmm0, %xmm2
; SSE2-NEXT: unpcklpd {{.*#+}} xmm1 = xmm1[0],xmm2[0]
; SSE2-NEXT: movapd %xmm1, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fminimum_vector_partially_zero:
; AVX1: # %bb.0:
; AVX1-NEXT: vmovq %xmm0, %rax
; AVX1-NEXT: testq %rax, %rax
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX1-NEXT: js .LBB24_1
; AVX1-NEXT: # %bb.2:
; AVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2
; AVX1-NEXT: vmovdqa %xmm0, %xmm1
; AVX1-NEXT: jmp .LBB24_3
; AVX1-NEXT: .LBB24_1:
; AVX1-NEXT: vmovdqa %xmm0, %xmm2
; AVX1-NEXT: .LBB24_3:
; AVX1-NEXT: vminsd %xmm2, %xmm1, %xmm2
; AVX1-NEXT: vcmpunordsd %xmm1, %xmm1, %xmm3
; AVX1-NEXT: vblendvpd %xmm3, %xmm1, %xmm2, %xmm1
; AVX1-NEXT: vshufpd {{.*#+}} xmm0 = xmm0[1,0]
; AVX1-NEXT: vmovsd {{.*#+}} xmm2 = mem[0],zero
; AVX1-NEXT: vminsd %xmm0, %xmm2, %xmm0
; AVX1-NEXT: vunpcklpd {{.*#+}} xmm0 = xmm1[0],xmm0[0]
; AVX1-NEXT: retq
;
; AVX512F-LABEL: test_fminimum_vector_partially_zero:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovq %xmm0, %rax
; AVX512F-NEXT: testq %rax, %rax
; AVX512F-NEXT: sets %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovsd %xmm0, %xmm0, %xmm1 {%k1} {z}
; AVX512F-NEXT: vxorpd %xmm2, %xmm2, %xmm2
; AVX512F-NEXT: vshufpd {{.*#+}} xmm3 = xmm0[1,0]
; AVX512F-NEXT: vmovsd %xmm2, %xmm0, %xmm0 {%k1}
; AVX512F-NEXT: vminsd %xmm1, %xmm0, %xmm1
; AVX512F-NEXT: vcmpunordsd %xmm0, %xmm0, %k1
; AVX512F-NEXT: vmovsd %xmm0, %xmm1, %xmm1 {%k1}
; AVX512F-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; AVX512F-NEXT: vminsd %xmm3, %xmm0, %xmm0
; AVX512F-NEXT: vunpcklpd {{.*#+}} xmm0 = xmm1[0],xmm0[0]
; AVX512F-NEXT: retq
;
; AVX512DQ-LABEL: test_fminimum_vector_partially_zero:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vfpclasssd $5, %xmm0, %k0
; AVX512DQ-NEXT: kmovw %k0, %k1
; AVX512DQ-NEXT: vmovsd %xmm0, %xmm0, %xmm1 {%k1} {z}
; AVX512DQ-NEXT: vxorpd %xmm2, %xmm2, %xmm2
; AVX512DQ-NEXT: vshufpd {{.*#+}} xmm3 = xmm0[1,0]
; AVX512DQ-NEXT: vmovsd %xmm2, %xmm0, %xmm0 {%k1}
; AVX512DQ-NEXT: vminsd %xmm1, %xmm0, %xmm0
; AVX512DQ-NEXT: vmovsd {{.*#+}} xmm1 = mem[0],zero
; AVX512DQ-NEXT: vminsd %xmm3, %xmm1, %xmm1
; AVX512DQ-NEXT: vunpcklpd {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; AVX512DQ-NEXT: retq
;
; X86-LABEL: test_fminimum_vector_partially_zero:
; X86: # %bb.0:
; X86-NEXT: vextractps $1, %xmm0, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: vxorpd %xmm1, %xmm1, %xmm1
; X86-NEXT: js .LBB24_1
; X86-NEXT: # %bb.2:
; X86-NEXT: vxorpd %xmm2, %xmm2, %xmm2
; X86-NEXT: vmovapd %xmm0, %xmm1
; X86-NEXT: jmp .LBB24_3
; X86-NEXT: .LBB24_1:
; X86-NEXT: vmovapd %xmm0, %xmm2
; X86-NEXT: .LBB24_3:
; X86-NEXT: vminsd %xmm2, %xmm1, %xmm2
; X86-NEXT: vcmpunordsd %xmm1, %xmm1, %xmm3
; X86-NEXT: vblendvpd %xmm3, %xmm1, %xmm2, %xmm1
; X86-NEXT: vshufpd {{.*#+}} xmm0 = xmm0[1,0]
; X86-NEXT: vmovsd {{.*#+}} xmm2 = mem[0],zero
; X86-NEXT: vminsd %xmm0, %xmm2, %xmm0
; X86-NEXT: vunpcklpd {{.*#+}} xmm0 = xmm1[0],xmm0[0]
; X86-NEXT: retl
%r = call <2 x double> @llvm.minimum.v2f64(<2 x double> %x, <2 x double> <double 0., double 5.>)
ret <2 x double> %r
}
define <4 x float> @test_fmaximum_vector_non_zero(<4 x float> %x) {
; SSE2-LABEL: test_fmaximum_vector_non_zero:
; SSE2: # %bb.0:
; SSE2-NEXT: movaps %xmm0, %xmm1
; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[3,3],xmm0[3,3]
; SSE2-NEXT: movss {{.*#+}} xmm2 = mem[0],zero,zero,zero
; SSE2-NEXT: maxss %xmm1, %xmm2
; SSE2-NEXT: movaps %xmm0, %xmm1
; SSE2-NEXT: unpckhpd {{.*#+}} xmm1 = xmm1[1],xmm0[1]
; SSE2-NEXT: movss {{.*#+}} xmm3 = mem[0],zero,zero,zero
; SSE2-NEXT: maxss %xmm1, %xmm3
; SSE2-NEXT: unpcklps {{.*#+}} xmm3 = xmm3[0],xmm2[0],xmm3[1],xmm2[1]
; SSE2-NEXT: movss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; SSE2-NEXT: maxss %xmm0, %xmm1
; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[1,1,1,1]
; SSE2-NEXT: movss {{.*#+}} xmm2 = mem[0],zero,zero,zero
; SSE2-NEXT: maxss %xmm0, %xmm2
; SSE2-NEXT: unpcklps {{.*#+}} xmm1 = xmm1[0],xmm2[0],xmm1[1],xmm2[1]
; SSE2-NEXT: movlhps {{.*#+}} xmm1 = xmm1[0],xmm3[0]
; SSE2-NEXT: movaps %xmm1, %xmm0
; SSE2-NEXT: retq
;
; AVX-LABEL: test_fmaximum_vector_non_zero:
; AVX: # %bb.0:
; AVX-NEXT: vmovss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; AVX-NEXT: vmaxss %xmm0, %xmm1, %xmm1
; AVX-NEXT: vmovshdup {{.*#+}} xmm2 = xmm0[1,1,3,3]
; AVX-NEXT: vmovss {{.*#+}} xmm3 = mem[0],zero,zero,zero
; AVX-NEXT: vmaxss %xmm2, %xmm3, %xmm2
; AVX-NEXT: vinsertps {{.*#+}} xmm1 = xmm1[0],xmm2[0],xmm1[2,3]
; AVX-NEXT: vshufpd {{.*#+}} xmm2 = xmm0[1,0]
; AVX-NEXT: vmovss {{.*#+}} xmm3 = mem[0],zero,zero,zero
; AVX-NEXT: vmaxss %xmm2, %xmm3, %xmm2
; AVX-NEXT: vinsertps {{.*#+}} xmm1 = xmm1[0,1],xmm2[0],xmm1[3]
; AVX-NEXT: vshufps {{.*#+}} xmm0 = xmm0[3,3,3,3]
; AVX-NEXT: vmovss {{.*#+}} xmm2 = mem[0],zero,zero,zero
; AVX-NEXT: vmaxss %xmm0, %xmm2, %xmm0
; AVX-NEXT: vinsertps {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[0]
; AVX-NEXT: retq
;
; X86-LABEL: test_fmaximum_vector_non_zero:
; X86: # %bb.0:
; X86-NEXT: vmovss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; X86-NEXT: vmaxss %xmm0, %xmm1, %xmm1
; X86-NEXT: vmovshdup {{.*#+}} xmm2 = xmm0[1,1,3,3]
; X86-NEXT: vmovss {{.*#+}} xmm3 = mem[0],zero,zero,zero
; X86-NEXT: vmaxss %xmm2, %xmm3, %xmm2
; X86-NEXT: vinsertps {{.*#+}} xmm1 = xmm1[0],xmm2[0],xmm1[2,3]
; X86-NEXT: vshufpd {{.*#+}} xmm2 = xmm0[1,0]
; X86-NEXT: vmovss {{.*#+}} xmm3 = mem[0],zero,zero,zero
; X86-NEXT: vmaxss %xmm2, %xmm3, %xmm2
; X86-NEXT: vinsertps {{.*#+}} xmm1 = xmm1[0,1],xmm2[0],xmm1[3]
; X86-NEXT: vshufps {{.*#+}} xmm0 = xmm0[3,3,3,3]
; X86-NEXT: vmovss {{.*#+}} xmm2 = mem[0],zero,zero,zero
; X86-NEXT: vmaxss %xmm0, %xmm2, %xmm0
; X86-NEXT: vinsertps {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[0]
; X86-NEXT: retl
%r = call <4 x float> @llvm.maximum.v4f32(<4 x float> %x, <4 x float> <float 5., float 4., float 3., float 2.>)
ret <4 x float> %r
}
define <2 x double> @test_fminimum_vector_nan(<2 x double> %x) {
; SSE2-LABEL: test_fminimum_vector_nan:
; SSE2: # %bb.0:
; SSE2-NEXT: movq %xmm0, %rax
; SSE2-NEXT: testq %rax, %rax
; SSE2-NEXT: pxor %xmm2, %xmm2
; SSE2-NEXT: js .LBB26_2
; SSE2-NEXT: # %bb.1:
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: .LBB26_2:
; SSE2-NEXT: movdqa %xmm2, %xmm1
; SSE2-NEXT: cmpunordsd %xmm2, %xmm1
; SSE2-NEXT: movapd %xmm1, %xmm3
; SSE2-NEXT: andpd %xmm2, %xmm3
; SSE2-NEXT: js .LBB26_4
; SSE2-NEXT: # %bb.3:
; SSE2-NEXT: pxor %xmm0, %xmm0
; SSE2-NEXT: .LBB26_4:
; SSE2-NEXT: minsd %xmm0, %xmm2
; SSE2-NEXT: andnpd %xmm2, %xmm1
; SSE2-NEXT: orpd %xmm3, %xmm1
; SSE2-NEXT: shufpd {{.*#+}} xmm1 = xmm1[0],mem[1]
; SSE2-NEXT: movapd %xmm1, %xmm0
; SSE2-NEXT: retq
;
; AVX1-LABEL: test_fminimum_vector_nan:
; AVX1: # %bb.0:
; AVX1-NEXT: vmovq %xmm0, %rax
; AVX1-NEXT: testq %rax, %rax
; AVX1-NEXT: js .LBB26_1
; AVX1-NEXT: # %bb.2:
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX1-NEXT: jmp .LBB26_3
; AVX1-NEXT: .LBB26_1:
; AVX1-NEXT: vmovdqa %xmm0, %xmm1
; AVX1-NEXT: vpxor %xmm0, %xmm0, %xmm0
; AVX1-NEXT: .LBB26_3:
; AVX1-NEXT: vminsd %xmm1, %xmm0, %xmm1
; AVX1-NEXT: vcmpunordsd %xmm0, %xmm0, %xmm2
; AVX1-NEXT: vblendvpd %xmm2, %xmm0, %xmm1, %xmm0
; AVX1-NEXT: vblendpd {{.*#+}} xmm0 = xmm0[0],mem[1]
; AVX1-NEXT: retq
;
; AVX512F-LABEL: test_fminimum_vector_nan:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovq %xmm0, %rax
; AVX512F-NEXT: testq %rax, %rax
; AVX512F-NEXT: sets %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovsd %xmm0, %xmm0, %xmm1 {%k1} {z}
; AVX512F-NEXT: vxorpd %xmm2, %xmm2, %xmm2
; AVX512F-NEXT: vmovsd %xmm2, %xmm0, %xmm0 {%k1}
; AVX512F-NEXT: vminsd %xmm1, %xmm0, %xmm1
; AVX512F-NEXT: vcmpunordsd %xmm0, %xmm0, %k1
; AVX512F-NEXT: vmovsd %xmm0, %xmm1, %xmm1 {%k1}
; AVX512F-NEXT: vblendpd {{.*#+}} xmm0 = xmm1[0],mem[1]
; AVX512F-NEXT: retq
;
; AVX512DQ-LABEL: test_fminimum_vector_nan:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vfpclasssd $5, %xmm0, %k0
; AVX512DQ-NEXT: kmovw %k0, %k1
; AVX512DQ-NEXT: vmovsd %xmm0, %xmm0, %xmm1 {%k1} {z}
; AVX512DQ-NEXT: vxorpd %xmm2, %xmm2, %xmm2
; AVX512DQ-NEXT: vmovsd %xmm2, %xmm0, %xmm0 {%k1}
; AVX512DQ-NEXT: vminsd %xmm1, %xmm0, %xmm0
; AVX512DQ-NEXT: vblendpd {{.*#+}} xmm0 = xmm0[0],mem[1]
; AVX512DQ-NEXT: retq
;
; X86-LABEL: test_fminimum_vector_nan:
; X86: # %bb.0:
; X86-NEXT: vextractps $1, %xmm0, %eax
; X86-NEXT: testl %eax, %eax
; X86-NEXT: js .LBB26_1
; X86-NEXT: # %bb.2:
; X86-NEXT: vxorpd %xmm1, %xmm1, %xmm1
; X86-NEXT: jmp .LBB26_3
; X86-NEXT: .LBB26_1:
; X86-NEXT: vmovapd %xmm0, %xmm1
; X86-NEXT: vxorpd %xmm0, %xmm0, %xmm0
; X86-NEXT: .LBB26_3:
; X86-NEXT: vminsd %xmm1, %xmm0, %xmm1
; X86-NEXT: vcmpunordsd %xmm0, %xmm0, %xmm2
; X86-NEXT: vblendvpd %xmm2, %xmm0, %xmm1, %xmm0
; X86-NEXT: vblendpd {{.*#+}} xmm0 = xmm0[0],mem[1]
; X86-NEXT: retl
%r = call <2 x double> @llvm.minimum.v2f64(<2 x double> %x, <2 x double> <double 0., double 0x7fff000000000000>)
ret <2 x double> %r
}
Reference in New Issue View Git Blame Copy Permalink