[X86] SET0 to use XMM registers where possible PR26018 PR32862
Differential Revision: https://reviews.llvm.org/D35839 llvm-svn: 309298
This commit is contained in:
Dinar Temirbulatov
parent
a67dfe3b04
commit
aead31a36f
@ -7705,9 +7705,14 @@ bool X86InstrInfo::expandPostRAPseudo(MachineInstr &MI) const {
|
||||
case X86::FsFLD0SS:
|
||||
case X86::FsFLD0SD:
|
||||
return Expand2AddrUndef(MIB, get(HasAVX ? X86::VXORPSrr : X86::XORPSrr));
|
||||
case X86::AVX_SET0:
|
||||
case X86::AVX_SET0: {
|
||||
assert(HasAVX && "AVX not supported");
|
||||
return Expand2AddrUndef(MIB, get(X86::VXORPSYrr));
|
||||
const TargetRegisterInfo *TRI = &getRegisterInfo();
|
||||
unsigned SrcReg = MIB->getOperand(0).getReg();
|
||||
unsigned XReg = TRI->getSubReg(SrcReg, X86::sub_xmm);
|
||||
MIB->getOperand(0).setReg(XReg);
|
||||
return Expand2AddrUndef(MIB, get(X86::VXORPSrr));
|
||||
}
|
||||
case X86::AVX512_128_SET0:
|
||||
case X86::AVX512_FsFLD0SS:
|
||||
case X86::AVX512_FsFLD0SD: {
|
||||
@ -7726,9 +7731,13 @@ bool X86InstrInfo::expandPostRAPseudo(MachineInstr &MI) const {
|
||||
bool HasVLX = Subtarget.hasVLX();
|
||||
unsigned SrcReg = MIB->getOperand(0).getReg();
|
||||
const TargetRegisterInfo *TRI = &getRegisterInfo();
|
||||
if (HasVLX || TRI->getEncodingValue(SrcReg) < 16)
|
||||
return Expand2AddrUndef(MIB,
|
||||
get(HasVLX ? X86::VPXORDZ256rr : X86::VXORPSYrr));
|
||||
if (HasVLX)
|
||||
return Expand2AddrUndef(MIB, get(X86::VPXORDZ256rr));
|
||||
if (TRI->getEncodingValue(SrcReg) < 16) {
|
||||
unsigned XReg = TRI->getSubReg(SrcReg, X86::sub_xmm);
|
||||
MIB->getOperand(0).setReg(XReg);
|
||||
return Expand2AddrUndef(MIB, get(X86::VXORPSrr));
|
||||
}
|
||||
// Extended register without VLX. Use a larger XOR.
|
||||
SrcReg = TRI->getMatchingSuperReg(SrcReg, X86::sub_ymm, &X86::VR512RegClass);
|
||||
MIB->getOperand(0).setReg(SrcReg);
|
||||
|
||||
@ -8,7 +8,7 @@ define void @endless_loop() {
|
||||
; CHECK-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,0,1,1]
|
||||
; CHECK-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[0,1,0,1]
|
||||
; CHECK-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm1
|
||||
; CHECK-NEXT: vxorps %ymm2, %ymm2, %ymm2
|
||||
; CHECK-NEXT: vxorps %xmm2, %xmm2, %xmm2
|
||||
; CHECK-NEXT: vblendps {{.*#+}} ymm1 = ymm2[0,1,2,3,4,5,6],ymm1[7]
|
||||
; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm2[1,2,3,4,5,6,7]
|
||||
; CHECK-NEXT: vmovaps %ymm0, (%eax)
|
||||
|
||||
@ -8,7 +8,7 @@ define void @func() nounwind ssp {
|
||||
; CHECK-LABEL: func:
|
||||
; CHECK: ## BB#0:
|
||||
; CHECK-NEXT: vmovups 0, %xmm0
|
||||
; CHECK-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; CHECK-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; CHECK-NEXT: vblendps {{.*#+}} ymm2 = ymm0[0,1,2,3],ymm1[4,5,6,7]
|
||||
; CHECK-NEXT: vpermilps {{.*#+}} xmm0 = xmm0[1,2,3,3]
|
||||
; CHECK-NEXT: vbroadcastss 32, %xmm3
|
||||
@ -26,6 +26,7 @@ define void @func() nounwind ssp {
|
||||
; CHECK-NEXT: vmovaps %ymm0, (%rax)
|
||||
; CHECK-NEXT: vzeroupper
|
||||
; CHECK-NEXT: retq
|
||||
; CHECK-NEXT: ## -- End function
|
||||
%tmp = load <4 x float>, <4 x float>* null, align 1
|
||||
%tmp14 = getelementptr <4 x float>, <4 x float>* null, i32 2
|
||||
%tmp15 = load <4 x float>, <4 x float>* %tmp14, align 1
|
||||
|
||||
@ -4,7 +4,7 @@
|
||||
define void @bad_cast() {
|
||||
; CHECK-LABEL: bad_cast:
|
||||
; CHECK: # BB#0:
|
||||
; CHECK-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; CHECK-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; CHECK-NEXT: vmovaps %xmm0, (%eax)
|
||||
; CHECK-NEXT: movl $0, (%eax)
|
||||
; CHECK-NEXT: vzeroupper
|
||||
|
||||
@ -157,7 +157,7 @@ define <32 x i8> @allones_v32i8() nounwind {
|
||||
;
|
||||
; X32-AVX1-LABEL: allones_v32i8:
|
||||
; X32-AVX1: # BB#0:
|
||||
; X32-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: retl
|
||||
;
|
||||
@ -174,7 +174,7 @@ define <32 x i8> @allones_v32i8() nounwind {
|
||||
;
|
||||
; X64-AVX1-LABEL: allones_v32i8:
|
||||
; X64-AVX1: # BB#0:
|
||||
; X64-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: retq
|
||||
;
|
||||
@ -194,7 +194,7 @@ define <16 x i16> @allones_v16i16() nounwind {
|
||||
;
|
||||
; X32-AVX1-LABEL: allones_v16i16:
|
||||
; X32-AVX1: # BB#0:
|
||||
; X32-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: retl
|
||||
;
|
||||
@ -211,7 +211,7 @@ define <16 x i16> @allones_v16i16() nounwind {
|
||||
;
|
||||
; X64-AVX1-LABEL: allones_v16i16:
|
||||
; X64-AVX1: # BB#0:
|
||||
; X64-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: retq
|
||||
;
|
||||
@ -231,7 +231,7 @@ define <8 x i32> @allones_v8i32() nounwind {
|
||||
;
|
||||
; X32-AVX1-LABEL: allones_v8i32:
|
||||
; X32-AVX1: # BB#0:
|
||||
; X32-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: retl
|
||||
;
|
||||
@ -248,7 +248,7 @@ define <8 x i32> @allones_v8i32() nounwind {
|
||||
;
|
||||
; X64-AVX1-LABEL: allones_v8i32:
|
||||
; X64-AVX1: # BB#0:
|
||||
; X64-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: retq
|
||||
;
|
||||
@ -268,7 +268,7 @@ define <4 x i64> @allones_v4i64() nounwind {
|
||||
;
|
||||
; X32-AVX1-LABEL: allones_v4i64:
|
||||
; X32-AVX1: # BB#0:
|
||||
; X32-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: retl
|
||||
;
|
||||
@ -285,7 +285,7 @@ define <4 x i64> @allones_v4i64() nounwind {
|
||||
;
|
||||
; X64-AVX1-LABEL: allones_v4i64:
|
||||
; X64-AVX1: # BB#0:
|
||||
; X64-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: retq
|
||||
;
|
||||
@ -305,7 +305,7 @@ define <4 x double> @allones_v4f64() nounwind {
|
||||
;
|
||||
; X32-AVX1-LABEL: allones_v4f64:
|
||||
; X32-AVX1: # BB#0:
|
||||
; X32-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: retl
|
||||
;
|
||||
@ -322,7 +322,7 @@ define <4 x double> @allones_v4f64() nounwind {
|
||||
;
|
||||
; X64-AVX1-LABEL: allones_v4f64:
|
||||
; X64-AVX1: # BB#0:
|
||||
; X64-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: retq
|
||||
;
|
||||
@ -342,7 +342,7 @@ define <4 x double> @allones_v4f64_optsize() nounwind optsize {
|
||||
;
|
||||
; X32-AVX1-LABEL: allones_v4f64_optsize:
|
||||
; X32-AVX1: # BB#0:
|
||||
; X32-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: retl
|
||||
;
|
||||
@ -359,7 +359,7 @@ define <4 x double> @allones_v4f64_optsize() nounwind optsize {
|
||||
;
|
||||
; X64-AVX1-LABEL: allones_v4f64_optsize:
|
||||
; X64-AVX1: # BB#0:
|
||||
; X64-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: retq
|
||||
;
|
||||
@ -379,7 +379,7 @@ define <8 x float> @allones_v8f32() nounwind {
|
||||
;
|
||||
; X32-AVX1-LABEL: allones_v8f32:
|
||||
; X32-AVX1: # BB#0:
|
||||
; X32-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: retl
|
||||
;
|
||||
@ -396,7 +396,7 @@ define <8 x float> @allones_v8f32() nounwind {
|
||||
;
|
||||
; X64-AVX1-LABEL: allones_v8f32:
|
||||
; X64-AVX1: # BB#0:
|
||||
; X64-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: retq
|
||||
;
|
||||
@ -416,7 +416,7 @@ define <8 x float> @allones_v8f32_optsize() nounwind optsize {
|
||||
;
|
||||
; X32-AVX1-LABEL: allones_v8f32_optsize:
|
||||
; X32-AVX1: # BB#0:
|
||||
; X32-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: retl
|
||||
;
|
||||
@ -433,7 +433,7 @@ define <8 x float> @allones_v8f32_optsize() nounwind optsize {
|
||||
;
|
||||
; X64-AVX1-LABEL: allones_v8f32_optsize:
|
||||
; X64-AVX1: # BB#0:
|
||||
; X64-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: retq
|
||||
;
|
||||
@ -455,7 +455,7 @@ define <64 x i8> @allones_v64i8() nounwind {
|
||||
;
|
||||
; X32-AVX1-LABEL: allones_v64i8:
|
||||
; X32-AVX1: # BB#0:
|
||||
; X32-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vmovaps %ymm0, %ymm1
|
||||
; X32-AVX1-NEXT: retl
|
||||
@ -487,7 +487,7 @@ define <64 x i8> @allones_v64i8() nounwind {
|
||||
;
|
||||
; X64-AVX1-LABEL: allones_v64i8:
|
||||
; X64-AVX1: # BB#0:
|
||||
; X64-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vmovaps %ymm0, %ymm1
|
||||
; X64-AVX1-NEXT: retq
|
||||
@ -522,7 +522,7 @@ define <32 x i16> @allones_v32i16() nounwind {
|
||||
;
|
||||
; X32-AVX1-LABEL: allones_v32i16:
|
||||
; X32-AVX1: # BB#0:
|
||||
; X32-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vmovaps %ymm0, %ymm1
|
||||
; X32-AVX1-NEXT: retl
|
||||
@ -554,7 +554,7 @@ define <32 x i16> @allones_v32i16() nounwind {
|
||||
;
|
||||
; X64-AVX1-LABEL: allones_v32i16:
|
||||
; X64-AVX1: # BB#0:
|
||||
; X64-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vmovaps %ymm0, %ymm1
|
||||
; X64-AVX1-NEXT: retq
|
||||
@ -589,7 +589,7 @@ define <16 x i32> @allones_v16i32() nounwind {
|
||||
;
|
||||
; X32-AVX1-LABEL: allones_v16i32:
|
||||
; X32-AVX1: # BB#0:
|
||||
; X32-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vmovaps %ymm0, %ymm1
|
||||
; X32-AVX1-NEXT: retl
|
||||
@ -615,7 +615,7 @@ define <16 x i32> @allones_v16i32() nounwind {
|
||||
;
|
||||
; X64-AVX1-LABEL: allones_v16i32:
|
||||
; X64-AVX1: # BB#0:
|
||||
; X64-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vmovaps %ymm0, %ymm1
|
||||
; X64-AVX1-NEXT: retq
|
||||
@ -644,7 +644,7 @@ define <8 x i64> @allones_v8i64() nounwind {
|
||||
;
|
||||
; X32-AVX1-LABEL: allones_v8i64:
|
||||
; X32-AVX1: # BB#0:
|
||||
; X32-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vmovaps %ymm0, %ymm1
|
||||
; X32-AVX1-NEXT: retl
|
||||
@ -670,7 +670,7 @@ define <8 x i64> @allones_v8i64() nounwind {
|
||||
;
|
||||
; X64-AVX1-LABEL: allones_v8i64:
|
||||
; X64-AVX1: # BB#0:
|
||||
; X64-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vmovaps %ymm0, %ymm1
|
||||
; X64-AVX1-NEXT: retq
|
||||
@ -699,7 +699,7 @@ define <8 x double> @allones_v8f64() nounwind {
|
||||
;
|
||||
; X32-AVX1-LABEL: allones_v8f64:
|
||||
; X32-AVX1: # BB#0:
|
||||
; X32-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vmovaps %ymm0, %ymm1
|
||||
; X32-AVX1-NEXT: retl
|
||||
@ -725,7 +725,7 @@ define <8 x double> @allones_v8f64() nounwind {
|
||||
;
|
||||
; X64-AVX1-LABEL: allones_v8f64:
|
||||
; X64-AVX1: # BB#0:
|
||||
; X64-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vmovaps %ymm0, %ymm1
|
||||
; X64-AVX1-NEXT: retq
|
||||
@ -754,7 +754,7 @@ define <16 x float> @allones_v16f32() nounwind {
|
||||
;
|
||||
; X32-AVX1-LABEL: allones_v16f32:
|
||||
; X32-AVX1: # BB#0:
|
||||
; X32-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX1-NEXT: vmovaps %ymm0, %ymm1
|
||||
; X32-AVX1-NEXT: retl
|
||||
@ -780,7 +780,7 @@ define <16 x float> @allones_v16f32() nounwind {
|
||||
;
|
||||
; X64-AVX1-LABEL: allones_v16f32:
|
||||
; X64-AVX1: # BB#0:
|
||||
; X64-AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX1-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX1-NEXT: vmovaps %ymm0, %ymm1
|
||||
; X64-AVX1-NEXT: retq
|
||||
|
||||
@ -21,7 +21,7 @@ define void @zero256() nounwind ssp {
|
||||
; CHECK-LABEL: zero256:
|
||||
; CHECK: ## BB#0:
|
||||
; CHECK-NEXT: movq _x@{{.*}}(%rip), %rax
|
||||
; CHECK-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; CHECK-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; CHECK-NEXT: vmovaps %ymm0, (%rax)
|
||||
; CHECK-NEXT: movq _y@{{.*}}(%rip), %rax
|
||||
; CHECK-NEXT: vmovaps %ymm0, (%rax)
|
||||
@ -36,7 +36,7 @@ define void @zero256() nounwind ssp {
|
||||
define void @ones([0 x float]* nocapture %RET, [0 x float]* nocapture %aFOO) nounwind {
|
||||
; CHECK-LABEL: ones:
|
||||
; CHECK: ## BB#0: ## %allocas
|
||||
; CHECK-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; CHECK-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; CHECK-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; CHECK-NEXT: vmovaps %ymm0, (%rdi)
|
||||
; CHECK-NEXT: vzeroupper
|
||||
@ -54,7 +54,7 @@ float>* %ptr2vec615, align 32
|
||||
define void @ones2([0 x i32]* nocapture %RET, [0 x i32]* nocapture %aFOO) nounwind {
|
||||
; CHECK-LABEL: ones2:
|
||||
; CHECK: ## BB#0: ## %allocas
|
||||
; CHECK-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; CHECK-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; CHECK-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
|
||||
; CHECK-NEXT: vmovaps %ymm0, (%rdi)
|
||||
; CHECK-NEXT: vzeroupper
|
||||
|
||||
@ -10,7 +10,7 @@ define <8 x float> @castA(<4 x float> %m) nounwind uwtable readnone ssp {
|
||||
; AVX-LABEL: castA:
|
||||
; AVX: ## BB#0:
|
||||
; AVX-NEXT: ## kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; AVX-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
|
||||
; AVX-NEXT: retq
|
||||
%shuffle.i = shufflevector <4 x float> %m, <4 x float> zeroinitializer, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 4, i32 4, i32 4>
|
||||
@ -21,7 +21,7 @@ define <4 x double> @castB(<2 x double> %m) nounwind uwtable readnone ssp {
|
||||
; AVX-LABEL: castB:
|
||||
; AVX: ## BB#0:
|
||||
; AVX-NEXT: ## kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; AVX-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
|
||||
; AVX-NEXT: retq
|
||||
%shuffle.i = shufflevector <2 x double> %m, <2 x double> zeroinitializer, <4 x i32> <i32 0, i32 1, i32 2, i32 2>
|
||||
@ -34,14 +34,14 @@ define <4 x i64> @castC(<2 x i64> %m) nounwind uwtable readnone ssp {
|
||||
; AVX1-LABEL: castC:
|
||||
; AVX1: ## BB#0:
|
||||
; AVX1-NEXT: ## kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; AVX1-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
; AVX2-LABEL: castC:
|
||||
; AVX2: ## BB#0:
|
||||
; AVX2-NEXT: ## kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
|
||||
; AVX2-NEXT: retq
|
||||
%shuffle.i = shufflevector <2 x i64> %m, <2 x i64> zeroinitializer, <4 x i32> <i32 0, i32 1, i32 2, i32 2>
|
||||
|
||||
@ -7,14 +7,14 @@
|
||||
define <8 x float> @sitofp_insert_zero_v8i32(<8 x i32> %a0) {
|
||||
; X86-LABEL: sitofp_insert_zero_v8i32:
|
||||
; X86: # BB#0:
|
||||
; X86-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X86-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4,5],ymm0[6,7]
|
||||
; X86-NEXT: vcvtdq2ps %ymm0, %ymm0
|
||||
; X86-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: sitofp_insert_zero_v8i32:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4,5],ymm0[6,7]
|
||||
; X64-NEXT: vcvtdq2ps %ymm0, %ymm0
|
||||
; X64-NEXT: retq
|
||||
@ -29,14 +29,14 @@ define <8 x float> @sitofp_insert_zero_v8i32(<8 x i32> %a0) {
|
||||
define <8 x float> @sitofp_shuffle_zero_v8i32(<8 x i32> %a0) {
|
||||
; X86-LABEL: sitofp_shuffle_zero_v8i32:
|
||||
; X86: # BB#0:
|
||||
; X86-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X86-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4],ymm0[5],ymm1[6],ymm0[7]
|
||||
; X86-NEXT: vcvtdq2ps %ymm0, %ymm0
|
||||
; X86-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: sitofp_shuffle_zero_v8i32:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4],ymm0[5],ymm1[6],ymm0[7]
|
||||
; X64-NEXT: vcvtdq2ps %ymm0, %ymm0
|
||||
; X64-NEXT: retq
|
||||
@ -48,7 +48,7 @@ define <8 x float> @sitofp_shuffle_zero_v8i32(<8 x i32> %a0) {
|
||||
define <8 x float> @sitofp_insert_allbits_v8i32(<8 x i32> %a0) {
|
||||
; X86-LABEL: sitofp_insert_allbits_v8i32:
|
||||
; X86: # BB#0:
|
||||
; X86-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X86-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4,5],ymm0[6,7]
|
||||
; X86-NEXT: vcvtdq2ps %ymm0, %ymm0
|
||||
@ -56,7 +56,7 @@ define <8 x float> @sitofp_insert_allbits_v8i32(<8 x i32> %a0) {
|
||||
;
|
||||
; X64-LABEL: sitofp_insert_allbits_v8i32:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4,5],ymm0[6,7]
|
||||
; X64-NEXT: vcvtdq2ps %ymm0, %ymm0
|
||||
@ -72,7 +72,7 @@ define <8 x float> @sitofp_insert_allbits_v8i32(<8 x i32> %a0) {
|
||||
define <8 x float> @sitofp_shuffle_allbits_v8i32(<8 x i32> %a0) {
|
||||
; X86-LABEL: sitofp_shuffle_allbits_v8i32:
|
||||
; X86: # BB#0:
|
||||
; X86-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X86-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4],ymm0[5],ymm1[6],ymm0[7]
|
||||
; X86-NEXT: vcvtdq2ps %ymm0, %ymm0
|
||||
@ -80,7 +80,7 @@ define <8 x float> @sitofp_shuffle_allbits_v8i32(<8 x i32> %a0) {
|
||||
;
|
||||
; X64-LABEL: sitofp_shuffle_allbits_v8i32:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4],ymm0[5],ymm1[6],ymm0[7]
|
||||
; X64-NEXT: vcvtdq2ps %ymm0, %ymm0
|
||||
@ -93,7 +93,7 @@ define <8 x float> @sitofp_shuffle_allbits_v8i32(<8 x i32> %a0) {
|
||||
define <8 x float> @sitofp_insert_constants_v8i32(<8 x i32> %a0) {
|
||||
; X86-LABEL: sitofp_insert_constants_v8i32:
|
||||
; X86: # BB#0:
|
||||
; X86-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X86-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1,2,3,4,5,6,7]
|
||||
; X86-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1],ymm1[2],ymm0[3,4,5,6,7]
|
||||
@ -108,7 +108,7 @@ define <8 x float> @sitofp_insert_constants_v8i32(<8 x i32> %a0) {
|
||||
;
|
||||
; X64-LABEL: sitofp_insert_constants_v8i32:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1,2,3,4,5,6,7]
|
||||
; X64-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1],ymm1[2],ymm0[3,4,5,6,7]
|
||||
|
||||
@ -99,7 +99,7 @@ define <8 x float> @test_mm256_and_ps(<8 x float> %a0, <8 x float> %a1) nounwind
|
||||
define <4 x double> @test_mm256_andnot_pd(<4 x double> %a0, <4 x double> %a1) nounwind {
|
||||
; X32-LABEL: test_mm256_andnot_pd:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vxorps %ymm2, %ymm2, %ymm2
|
||||
; X32-NEXT: vxorps %xmm2, %xmm2, %xmm2
|
||||
; X32-NEXT: vcmptrueps %ymm2, %ymm2, %ymm2
|
||||
; X32-NEXT: vxorps %ymm2, %ymm0, %ymm0
|
||||
; X32-NEXT: vandps %ymm1, %ymm0, %ymm0
|
||||
@ -107,7 +107,7 @@ define <4 x double> @test_mm256_andnot_pd(<4 x double> %a0, <4 x double> %a1) no
|
||||
;
|
||||
; X64-LABEL: test_mm256_andnot_pd:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm2, %ymm2, %ymm2
|
||||
; X64-NEXT: vxorps %xmm2, %xmm2, %xmm2
|
||||
; X64-NEXT: vcmptrueps %ymm2, %ymm2, %ymm2
|
||||
; X64-NEXT: vxorps %ymm2, %ymm0, %ymm0
|
||||
; X64-NEXT: vandps %ymm1, %ymm0, %ymm0
|
||||
@ -2945,12 +2945,12 @@ define <8 x float> @test_mm256_setr_ps(float %a0, float %a1, float %a2, float %a
|
||||
define <4 x double> @test_mm256_setzero_pd() nounwind {
|
||||
; X32-LABEL: test_mm256_setzero_pd:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: test_mm256_setzero_pd:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-NEXT: retq
|
||||
ret <4 x double> zeroinitializer
|
||||
}
|
||||
@ -2958,12 +2958,12 @@ define <4 x double> @test_mm256_setzero_pd() nounwind {
|
||||
define <8 x float> @test_mm256_setzero_ps() nounwind {
|
||||
; X32-LABEL: test_mm256_setzero_ps:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: test_mm256_setzero_ps:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-NEXT: retq
|
||||
ret <8 x float> zeroinitializer
|
||||
}
|
||||
@ -2971,12 +2971,12 @@ define <8 x float> @test_mm256_setzero_ps() nounwind {
|
||||
define <4 x i64> @test_mm256_setzero_si256() nounwind {
|
||||
; X32-LABEL: test_mm256_setzero_si256:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: test_mm256_setzero_si256:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-NEXT: retq
|
||||
ret <4 x i64> zeroinitializer
|
||||
}
|
||||
|
||||
@ -661,7 +661,7 @@ define void @test_x86_avx_storeu_pd_256(i8* %a0, <4 x double> %a1) {
|
||||
; X86-LABEL: test_x86_avx_storeu_pd_256:
|
||||
; X86: # BB#0:
|
||||
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X86-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X86-NEXT: vaddpd %ymm1, %ymm0, %ymm0
|
||||
; X86-NEXT: vmovupd %ymm0, (%eax)
|
||||
; X86-NEXT: vzeroupper
|
||||
@ -669,7 +669,7 @@ define void @test_x86_avx_storeu_pd_256(i8* %a0, <4 x double> %a1) {
|
||||
;
|
||||
; X64-LABEL: test_x86_avx_storeu_pd_256:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vaddpd %ymm1, %ymm0, %ymm0
|
||||
; X64-NEXT: vmovupd %ymm0, (%rdi)
|
||||
; X64-NEXT: vzeroupper
|
||||
|
||||
@ -975,7 +975,7 @@ define void @movnt_pd(i8* %p, <4 x double> %a1) nounwind {
|
||||
; AVX-LABEL: movnt_pd:
|
||||
; AVX: # BB#0:
|
||||
; AVX-NEXT: movl {{[0-9]+}}(%esp), %eax # encoding: [0x8b,0x44,0x24,0x04]
|
||||
; AVX-NEXT: vxorpd %ymm1, %ymm1, %ymm1 # encoding: [0xc5,0xf5,0x57,0xc9]
|
||||
; AVX-NEXT: vxorpd %xmm1, %xmm1, %xmm1 # encoding: [0xc5,0xf1,0x57,0xc9]
|
||||
; AVX-NEXT: vaddpd %ymm1, %ymm0, %ymm0 # encoding: [0xc5,0xfd,0x58,0xc1]
|
||||
; AVX-NEXT: vmovntpd %ymm0, (%eax) # encoding: [0xc5,0xfd,0x2b,0x00]
|
||||
; AVX-NEXT: vzeroupper # encoding: [0xc5,0xf8,0x77]
|
||||
|
||||
@ -87,7 +87,7 @@ define <8 x float> @mov00(<8 x float> %v, float * %ptr) nounwind {
|
||||
; CHECK_O0-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
|
||||
; CHECK_O0-NEXT: # implicit-def: %YMM1
|
||||
; CHECK_O0-NEXT: vmovaps %xmm0, %xmm1
|
||||
; CHECK_O0-NEXT: vxorps %ymm2, %ymm2, %ymm2
|
||||
; CHECK_O0-NEXT: vxorps %xmm2, %xmm2, %xmm2
|
||||
; CHECK_O0-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm2[1,2,3,4,5,6,7]
|
||||
; CHECK_O0-NEXT: retq
|
||||
%val = load float, float* %ptr
|
||||
@ -106,7 +106,7 @@ define <4 x double> @mov01(<4 x double> %v, double * %ptr) nounwind {
|
||||
; CHECK_O0-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
|
||||
; CHECK_O0-NEXT: # implicit-def: %YMM1
|
||||
; CHECK_O0-NEXT: vmovaps %xmm0, %xmm1
|
||||
; CHECK_O0-NEXT: vxorps %ymm2, %ymm2, %ymm2
|
||||
; CHECK_O0-NEXT: vxorps %xmm2, %xmm2, %xmm2
|
||||
; CHECK_O0-NEXT: vblendpd {{.*#+}} ymm0 = ymm1[0],ymm2[1,2,3]
|
||||
; CHECK_O0-NEXT: retq
|
||||
%val = load double, double* %ptr
|
||||
|
||||
@ -6,7 +6,7 @@ define <4 x double> @andpd256(<4 x double> %y, <4 x double> %x) nounwind uwtable
|
||||
; CHECK-LABEL: andpd256:
|
||||
; CHECK: # BB#0: # %entry
|
||||
; CHECK-NEXT: vandpd %ymm0, %ymm1, %ymm0
|
||||
; CHECK-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; CHECK-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; CHECK-NEXT: vaddpd %ymm1, %ymm0, %ymm0
|
||||
; CHECK-NEXT: retq
|
||||
entry:
|
||||
@ -23,7 +23,7 @@ define <4 x double> @andpd256fold(<4 x double> %y) nounwind uwtable readnone ssp
|
||||
; CHECK-LABEL: andpd256fold:
|
||||
; CHECK: # BB#0: # %entry
|
||||
; CHECK-NEXT: vandpd {{.*}}(%rip), %ymm0, %ymm0
|
||||
; CHECK-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; CHECK-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; CHECK-NEXT: vaddpd %ymm1, %ymm0, %ymm0
|
||||
; CHECK-NEXT: retq
|
||||
entry:
|
||||
@ -64,7 +64,7 @@ define <4 x double> @xorpd256(<4 x double> %y, <4 x double> %x) nounwind uwtable
|
||||
; CHECK-LABEL: xorpd256:
|
||||
; CHECK: # BB#0: # %entry
|
||||
; CHECK-NEXT: vxorpd %ymm0, %ymm1, %ymm0
|
||||
; CHECK-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; CHECK-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; CHECK-NEXT: vaddpd %ymm1, %ymm0, %ymm0
|
||||
; CHECK-NEXT: retq
|
||||
entry:
|
||||
@ -81,7 +81,7 @@ define <4 x double> @xorpd256fold(<4 x double> %y) nounwind uwtable readnone ssp
|
||||
; CHECK-LABEL: xorpd256fold:
|
||||
; CHECK: # BB#0: # %entry
|
||||
; CHECK-NEXT: vxorpd {{.*}}(%rip), %ymm0, %ymm0
|
||||
; CHECK-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; CHECK-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; CHECK-NEXT: vaddpd %ymm1, %ymm0, %ymm0
|
||||
; CHECK-NEXT: retq
|
||||
entry:
|
||||
@ -122,7 +122,7 @@ define <4 x double> @orpd256(<4 x double> %y, <4 x double> %x) nounwind uwtable
|
||||
; CHECK-LABEL: orpd256:
|
||||
; CHECK: # BB#0: # %entry
|
||||
; CHECK-NEXT: vorpd %ymm0, %ymm1, %ymm0
|
||||
; CHECK-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; CHECK-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; CHECK-NEXT: vaddpd %ymm1, %ymm0, %ymm0
|
||||
; CHECK-NEXT: retq
|
||||
entry:
|
||||
@ -139,7 +139,7 @@ define <4 x double> @orpd256fold(<4 x double> %y) nounwind uwtable readnone ssp
|
||||
; CHECK-LABEL: orpd256fold:
|
||||
; CHECK: # BB#0: # %entry
|
||||
; CHECK-NEXT: vorpd {{.*}}(%rip), %ymm0, %ymm0
|
||||
; CHECK-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; CHECK-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; CHECK-NEXT: vaddpd %ymm1, %ymm0, %ymm0
|
||||
; CHECK-NEXT: retq
|
||||
entry:
|
||||
@ -180,7 +180,7 @@ define <4 x double> @andnotpd256(<4 x double> %y, <4 x double> %x) nounwind uwta
|
||||
; CHECK-LABEL: andnotpd256:
|
||||
; CHECK: # BB#0: # %entry
|
||||
; CHECK-NEXT: vandnpd %ymm0, %ymm1, %ymm0
|
||||
; CHECK-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; CHECK-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; CHECK-NEXT: vaddpd %ymm1, %ymm0, %ymm0
|
||||
; CHECK-NEXT: retq
|
||||
entry:
|
||||
@ -198,7 +198,7 @@ define <4 x double> @andnotpd256fold(<4 x double> %y, <4 x double>* nocapture %x
|
||||
; CHECK-LABEL: andnotpd256fold:
|
||||
; CHECK: # BB#0: # %entry
|
||||
; CHECK-NEXT: vandnpd (%rdi), %ymm0, %ymm0
|
||||
; CHECK-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; CHECK-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; CHECK-NEXT: vaddpd %ymm1, %ymm0, %ymm0
|
||||
; CHECK-NEXT: retq
|
||||
entry:
|
||||
|
||||
@ -6,7 +6,7 @@ define <8 x i32> @select00(i32 %a, <8 x i32> %b) nounwind {
|
||||
; X86-LABEL: select00:
|
||||
; X86: # BB#0:
|
||||
; X86-NEXT: cmpl $255, {{[0-9]+}}(%esp)
|
||||
; X86-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X86-NEXT: je .LBB0_2
|
||||
; X86-NEXT: # BB#1:
|
||||
; X86-NEXT: vmovaps %ymm0, %ymm1
|
||||
@ -16,7 +16,7 @@ define <8 x i32> @select00(i32 %a, <8 x i32> %b) nounwind {
|
||||
;
|
||||
; X64-LABEL: select00:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: cmpl $255, %edi
|
||||
; X64-NEXT: je .LBB0_2
|
||||
; X64-NEXT: # BB#1:
|
||||
@ -34,7 +34,7 @@ define <4 x i64> @select01(i32 %a, <4 x i64> %b) nounwind {
|
||||
; X86-LABEL: select01:
|
||||
; X86: # BB#0:
|
||||
; X86-NEXT: cmpl $255, {{[0-9]+}}(%esp)
|
||||
; X86-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X86-NEXT: je .LBB1_2
|
||||
; X86-NEXT: # BB#1:
|
||||
; X86-NEXT: vmovaps %ymm0, %ymm1
|
||||
@ -44,7 +44,7 @@ define <4 x i64> @select01(i32 %a, <4 x i64> %b) nounwind {
|
||||
;
|
||||
; X64-LABEL: select01:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: cmpl $255, %edi
|
||||
; X64-NEXT: je .LBB1_2
|
||||
; X64-NEXT: # BB#1:
|
||||
|
||||
@ -236,7 +236,7 @@ define <8 x i32> @PR29088(<4 x i32>* %p0, <8 x float>* %p1) {
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
||||
; X32-NEXT: vmovaps (%ecx), %xmm0
|
||||
; X32-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vmovaps %ymm1, (%eax)
|
||||
; X32-NEXT: vinsertf128 $1, %xmm0, %ymm0, %ymm0
|
||||
; X32-NEXT: retl
|
||||
@ -244,7 +244,7 @@ define <8 x i32> @PR29088(<4 x i32>* %p0, <8 x float>* %p1) {
|
||||
; X64-LABEL: PR29088:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vmovaps (%rdi), %xmm0
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vmovaps %ymm1, (%rsi)
|
||||
; X64-NEXT: vinsertf128 $1, %xmm0, %ymm0, %ymm0
|
||||
; X64-NEXT: retq
|
||||
|
||||
@ -119,7 +119,7 @@ entry:
|
||||
define void @t9(i64* %p) {
|
||||
; CHECK-LABEL: t9:
|
||||
; CHECK: ## BB#0:
|
||||
; CHECK-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; CHECK-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; CHECK-NEXT: vmovups %ymm0, (%rdi)
|
||||
; CHECK-NEXT: vzeroupper
|
||||
; CHECK-NEXT: retq
|
||||
|
||||
@ -335,7 +335,7 @@ define <4 x double> @shuffle_v4f64_zz01_optsize(<4 x double> %a) optsize {
|
||||
define <4 x double> @shuffle_v4f64_zz23(<4 x double> %a) {
|
||||
; ALL-LABEL: shuffle_v4f64_zz23:
|
||||
; ALL: # BB#0:
|
||||
; ALL-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; ALL-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; ALL-NEXT: vblendpd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,3]
|
||||
; ALL-NEXT: retq
|
||||
%s = shufflevector <4 x double> %a, <4 x double> <double 0.0, double 0.0, double undef, double undef>, <4 x i32> <i32 4, i32 5, i32 2, i32 3>
|
||||
@ -344,7 +344,7 @@ define <4 x double> @shuffle_v4f64_zz23(<4 x double> %a) {
|
||||
define <4 x double> @shuffle_v4f64_zz23_optsize(<4 x double> %a) optsize {
|
||||
; ALL-LABEL: shuffle_v4f64_zz23_optsize:
|
||||
; ALL: # BB#0:
|
||||
; ALL-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; ALL-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; ALL-NEXT: vblendpd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,3]
|
||||
; ALL-NEXT: retq
|
||||
%s = shufflevector <4 x double> %a, <4 x double> <double 0.0, double 0.0, double undef, double undef>, <4 x i32> <i32 4, i32 5, i32 2, i32 3>
|
||||
@ -371,7 +371,7 @@ define <4 x double> @shuffle_v4f64_zz45_optsize(<4 x double> %a) optsize {
|
||||
define <4 x double> @shuffle_v4f64_zz67(<4 x double> %a) {
|
||||
; ALL-LABEL: shuffle_v4f64_zz67:
|
||||
; ALL: # BB#0:
|
||||
; ALL-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; ALL-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; ALL-NEXT: vblendpd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,3]
|
||||
; ALL-NEXT: retq
|
||||
%s = shufflevector <4 x double> <double 0.0, double 0.0, double undef, double undef>, <4 x double> %a, <4 x i32> <i32 0, i32 1, i32 6, i32 7>
|
||||
@ -380,7 +380,7 @@ define <4 x double> @shuffle_v4f64_zz67(<4 x double> %a) {
|
||||
define <4 x double> @shuffle_v4f64_zz67_optsize(<4 x double> %a) optsize {
|
||||
; ALL-LABEL: shuffle_v4f64_zz67_optsize:
|
||||
; ALL: # BB#0:
|
||||
; ALL-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; ALL-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; ALL-NEXT: vblendpd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,3]
|
||||
; ALL-NEXT: retq
|
||||
%s = shufflevector <4 x double> <double 0.0, double 0.0, double undef, double undef>, <4 x double> %a, <4 x i32> <i32 0, i32 1, i32 6, i32 7>
|
||||
@ -390,7 +390,7 @@ define <4 x double> @shuffle_v4f64_zz67_optsize(<4 x double> %a) optsize {
|
||||
define <4 x double> @shuffle_v4f64_01zz(<4 x double> %a) {
|
||||
; ALL-LABEL: shuffle_v4f64_01zz:
|
||||
; ALL: # BB#0:
|
||||
; ALL-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; ALL-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; ALL-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
|
||||
; ALL-NEXT: retq
|
||||
%s = shufflevector <4 x double> %a, <4 x double> <double 0.0, double 0.0, double undef, double undef>, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
|
||||
@ -399,7 +399,7 @@ define <4 x double> @shuffle_v4f64_01zz(<4 x double> %a) {
|
||||
define <4 x double> @shuffle_v4f64_01zz_optsize(<4 x double> %a) optsize {
|
||||
; ALL-LABEL: shuffle_v4f64_01zz_optsize:
|
||||
; ALL: # BB#0:
|
||||
; ALL-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; ALL-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; ALL-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
|
||||
; ALL-NEXT: retq
|
||||
%s = shufflevector <4 x double> %a, <4 x double> <double 0.0, double 0.0, double undef, double undef>, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
|
||||
@ -426,7 +426,7 @@ define <4 x double> @shuffle_v4f64_23zz_optsize(<4 x double> %a) optsize {
|
||||
define <4 x double> @shuffle_v4f64_45zz(<4 x double> %a) {
|
||||
; ALL-LABEL: shuffle_v4f64_45zz:
|
||||
; ALL: # BB#0:
|
||||
; ALL-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; ALL-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; ALL-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
|
||||
; ALL-NEXT: retq
|
||||
%s = shufflevector <4 x double> <double 0.0, double 0.0, double undef, double undef>, <4 x double> %a, <4 x i32> <i32 4, i32 5, i32 0, i32 1>
|
||||
@ -435,7 +435,7 @@ define <4 x double> @shuffle_v4f64_45zz(<4 x double> %a) {
|
||||
define <4 x double> @shuffle_v4f64_45zz_optsize(<4 x double> %a) optsize {
|
||||
; ALL-LABEL: shuffle_v4f64_45zz_optsize:
|
||||
; ALL: # BB#0:
|
||||
; ALL-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; ALL-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; ALL-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
|
||||
; ALL-NEXT: retq
|
||||
%s = shufflevector <4 x double> <double 0.0, double 0.0, double undef, double undef>, <4 x double> %a, <4 x i32> <i32 4, i32 5, i32 0, i32 1>
|
||||
|
||||
@ -293,13 +293,13 @@ define <16 x i16> @mul_const3(<16 x i16> %x) {
|
||||
define <4 x i64> @mul_const4(<4 x i64> %x) {
|
||||
; X32-LABEL: mul_const4:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vpsubq %ymm0, %ymm1, %ymm0
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: mul_const4:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vpsubq %ymm0, %ymm1, %ymm0
|
||||
; X64-NEXT: retq
|
||||
%y = mul <4 x i64> %x, <i64 -1, i64 -1, i64 -1, i64 -1>
|
||||
@ -309,12 +309,12 @@ define <4 x i64> @mul_const4(<4 x i64> %x) {
|
||||
define <8 x i32> @mul_const5(<8 x i32> %x) {
|
||||
; X32-LABEL: mul_const5:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: mul_const5:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-NEXT: retq
|
||||
%y = mul <8 x i32> %x, <i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
|
||||
ret <8 x i32> %y
|
||||
|
||||
@ -55,14 +55,14 @@ define <8 x float> @test_x86_avx2_gather_d_ps_256(i8* %a1, <8 x i32> %idx, <8 x
|
||||
; X32-LABEL: test_x86_avx2_gather_d_ps_256:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-NEXT: vxorps %ymm2, %ymm2, %ymm2
|
||||
; X32-NEXT: vxorps %xmm2, %xmm2, %xmm2
|
||||
; X32-NEXT: vgatherdps %ymm1, (%eax,%ymm0,4), %ymm2
|
||||
; X32-NEXT: vmovaps %ymm2, %ymm0
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: test_x86_avx2_gather_d_ps_256:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm2, %ymm2, %ymm2
|
||||
; X64-NEXT: vxorps %xmm2, %xmm2, %xmm2
|
||||
; X64-NEXT: vgatherdps %ymm1, (%rdi,%ymm0,4), %ymm2
|
||||
; X64-NEXT: vmovaps %ymm2, %ymm0
|
||||
; X64-NEXT: retq
|
||||
@ -78,14 +78,14 @@ define <4 x double> @test_x86_avx2_gather_d_pd_256(i8* %a1, <4 x i32> %idx, <4 x
|
||||
; X32-LABEL: test_x86_avx2_gather_d_pd_256:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-NEXT: vxorpd %ymm2, %ymm2, %ymm2
|
||||
; X32-NEXT: vxorpd %xmm2, %xmm2, %xmm2
|
||||
; X32-NEXT: vgatherdpd %ymm1, (%eax,%xmm0,8), %ymm2
|
||||
; X32-NEXT: vmovapd %ymm2, %ymm0
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: test_x86_avx2_gather_d_pd_256:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorpd %ymm2, %ymm2, %ymm2
|
||||
; X64-NEXT: vxorpd %xmm2, %xmm2, %xmm2
|
||||
; X64-NEXT: vgatherdpd %ymm1, (%rdi,%xmm0,8), %ymm2
|
||||
; X64-NEXT: vmovapd %ymm2, %ymm0
|
||||
; X64-NEXT: retq
|
||||
|
||||
@ -1114,7 +1114,7 @@ define <4 x i64> @test_mm256_i32gather_epi32(i32 *%a0, <4 x i64> %a1) {
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-NEXT: vpcmpeqd %ymm2, %ymm2, %ymm2
|
||||
; X32-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vpgatherdd %ymm2, (%eax,%ymm0,2), %ymm1
|
||||
; X32-NEXT: vmovdqa %ymm1, %ymm0
|
||||
; X32-NEXT: retl
|
||||
@ -1122,7 +1122,7 @@ define <4 x i64> @test_mm256_i32gather_epi32(i32 *%a0, <4 x i64> %a1) {
|
||||
; X64-LABEL: test_mm256_i32gather_epi32:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vpcmpeqd %ymm2, %ymm2, %ymm2
|
||||
; X64-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vpgatherdd %ymm2, (%rdi,%ymm0,2), %ymm1
|
||||
; X64-NEXT: vmovdqa %ymm1, %ymm0
|
||||
; X64-NEXT: retq
|
||||
@ -1201,7 +1201,7 @@ define <4 x i64> @test_mm256_i32gather_epi64(i64 *%a0, <2 x i64> %a1) {
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-NEXT: vpcmpeqd %ymm2, %ymm2, %ymm2
|
||||
; X32-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vpgatherdq %ymm2, (%eax,%xmm0,2), %ymm1
|
||||
; X32-NEXT: vmovdqa %ymm1, %ymm0
|
||||
; X32-NEXT: retl
|
||||
@ -1209,7 +1209,7 @@ define <4 x i64> @test_mm256_i32gather_epi64(i64 *%a0, <2 x i64> %a1) {
|
||||
; X64-LABEL: test_mm256_i32gather_epi64:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vpcmpeqd %ymm2, %ymm2, %ymm2
|
||||
; X64-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vpgatherdq %ymm2, (%rdi,%xmm0,2), %ymm1
|
||||
; X64-NEXT: vmovdqa %ymm1, %ymm0
|
||||
; X64-NEXT: retq
|
||||
@ -1285,7 +1285,7 @@ define <4 x double> @test_mm256_i32gather_pd(double *%a0, <2 x i64> %a1) {
|
||||
; X32-LABEL: test_mm256_i32gather_pd:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vcmpeqpd %ymm1, %ymm1, %ymm2
|
||||
; X32-NEXT: vgatherdpd %ymm2, (%eax,%xmm0,2), %ymm1
|
||||
; X32-NEXT: vmovapd %ymm1, %ymm0
|
||||
@ -1293,7 +1293,7 @@ define <4 x double> @test_mm256_i32gather_pd(double *%a0, <2 x i64> %a1) {
|
||||
;
|
||||
; X64-LABEL: test_mm256_i32gather_pd:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vcmpeqpd %ymm1, %ymm1, %ymm2
|
||||
; X64-NEXT: vgatherdpd %ymm2, (%rdi,%xmm0,2), %ymm1
|
||||
; X64-NEXT: vmovapd %ymm1, %ymm0
|
||||
@ -1371,7 +1371,7 @@ define <8 x float> @test_mm256_i32gather_ps(float *%a0, <4 x i64> %a1) {
|
||||
; X32-LABEL: test_mm256_i32gather_ps:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vcmpeqps %ymm1, %ymm1, %ymm2
|
||||
; X32-NEXT: vgatherdps %ymm2, (%eax,%ymm0,2), %ymm1
|
||||
; X32-NEXT: vmovaps %ymm1, %ymm0
|
||||
@ -1379,7 +1379,7 @@ define <8 x float> @test_mm256_i32gather_ps(float *%a0, <4 x i64> %a1) {
|
||||
;
|
||||
; X64-LABEL: test_mm256_i32gather_ps:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vcmpeqps %ymm1, %ymm1, %ymm2
|
||||
; X64-NEXT: vgatherdps %ymm2, (%rdi,%ymm0,2), %ymm1
|
||||
; X64-NEXT: vmovaps %ymm1, %ymm0
|
||||
@ -1545,7 +1545,7 @@ define <4 x i64> @test_mm256_i64gather_epi64(i64 *%a0, <4 x i64> %a1) {
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-NEXT: vpcmpeqd %ymm2, %ymm2, %ymm2
|
||||
; X32-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vpgatherqq %ymm2, (%eax,%ymm0,2), %ymm1
|
||||
; X32-NEXT: vmovdqa %ymm1, %ymm0
|
||||
; X32-NEXT: retl
|
||||
@ -1553,7 +1553,7 @@ define <4 x i64> @test_mm256_i64gather_epi64(i64 *%a0, <4 x i64> %a1) {
|
||||
; X64-LABEL: test_mm256_i64gather_epi64:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vpcmpeqd %ymm2, %ymm2, %ymm2
|
||||
; X64-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vpgatherqq %ymm2, (%rdi,%ymm0,2), %ymm1
|
||||
; X64-NEXT: vmovdqa %ymm1, %ymm0
|
||||
; X64-NEXT: retq
|
||||
@ -1625,7 +1625,7 @@ define <4 x double> @test_mm256_i64gather_pd(double *%a0, <4 x i64> %a1) {
|
||||
; X32-LABEL: test_mm256_i64gather_pd:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vcmpeqpd %ymm1, %ymm1, %ymm2
|
||||
; X32-NEXT: vgatherqpd %ymm2, (%eax,%ymm0,2), %ymm1
|
||||
; X32-NEXT: vmovapd %ymm1, %ymm0
|
||||
@ -1633,7 +1633,7 @@ define <4 x double> @test_mm256_i64gather_pd(double *%a0, <4 x i64> %a1) {
|
||||
;
|
||||
; X64-LABEL: test_mm256_i64gather_pd:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vcmpeqpd %ymm1, %ymm1, %ymm2
|
||||
; X64-NEXT: vgatherqpd %ymm2, (%rdi,%ymm0,2), %ymm1
|
||||
; X64-NEXT: vmovapd %ymm1, %ymm0
|
||||
|
||||
@ -455,13 +455,13 @@ define <32 x i8> @shr9(<32 x i8> %A) nounwind {
|
||||
define <32 x i8> @sra_v32i8_7(<32 x i8> %A) nounwind {
|
||||
; X32-LABEL: sra_v32i8_7:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vpcmpgtb %ymm0, %ymm1, %ymm0
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: sra_v32i8_7:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vpcmpgtb %ymm0, %ymm1, %ymm0
|
||||
; X64-NEXT: retq
|
||||
%B = ashr <32 x i8> %A, <i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7, i8 7>
|
||||
|
||||
@ -1135,52 +1135,99 @@ eintry:
|
||||
}
|
||||
|
||||
define void @isel_crash_32b(i8* %cV_R.addr) {
|
||||
; X32-LABEL: isel_crash_32b:
|
||||
; X32: ## BB#0: ## %eintry
|
||||
; X32-NEXT: pushl %ebp
|
||||
; X32-NEXT: Lcfi1:
|
||||
; X32-NEXT: .cfi_def_cfa_offset 8
|
||||
; X32-NEXT: Lcfi2:
|
||||
; X32-NEXT: .cfi_offset %ebp, -8
|
||||
; X32-NEXT: movl %esp, %ebp
|
||||
; X32-NEXT: Lcfi3:
|
||||
; X32-NEXT: .cfi_def_cfa_register %ebp
|
||||
; X32-NEXT: andl $-32, %esp
|
||||
; X32-NEXT: subl $128, %esp
|
||||
; X32-NEXT: movl 8(%ebp), %eax
|
||||
; X32-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-NEXT: vmovaps %ymm0, (%esp)
|
||||
; X32-NEXT: vpbroadcastb (%eax), %ymm1
|
||||
; X32-NEXT: vmovaps %ymm0, {{[0-9]+}}(%esp)
|
||||
; X32-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%esp)
|
||||
; X32-NEXT: movl %ebp, %esp
|
||||
; X32-NEXT: popl %ebp
|
||||
; X32-NEXT: vzeroupper
|
||||
; X32-NEXT: retl
|
||||
; X32-AVX2-LABEL: isel_crash_32b:
|
||||
; X32-AVX2: ## BB#0: ## %eintry
|
||||
; X32-AVX2-NEXT: pushl %ebp
|
||||
; X32-AVX2-NEXT: Lcfi1:
|
||||
; X32-AVX2-NEXT: .cfi_def_cfa_offset 8
|
||||
; X32-AVX2-NEXT: Lcfi2:
|
||||
; X32-AVX2-NEXT: .cfi_offset %ebp, -8
|
||||
; X32-AVX2-NEXT: movl %esp, %ebp
|
||||
; X32-AVX2-NEXT: Lcfi3:
|
||||
; X32-AVX2-NEXT: .cfi_def_cfa_register %ebp
|
||||
; X32-AVX2-NEXT: andl $-32, %esp
|
||||
; X32-AVX2-NEXT: subl $128, %esp
|
||||
; X32-AVX2-NEXT: movl 8(%ebp), %eax
|
||||
; X32-AVX2-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX2-NEXT: vmovaps %ymm0, (%esp)
|
||||
; X32-AVX2-NEXT: vpbroadcastb (%eax), %ymm1
|
||||
; X32-AVX2-NEXT: vmovaps %ymm0, {{[0-9]+}}(%esp)
|
||||
; X32-AVX2-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%esp)
|
||||
; X32-AVX2-NEXT: movl %ebp, %esp
|
||||
; X32-AVX2-NEXT: popl %ebp
|
||||
; X32-AVX2-NEXT: vzeroupper
|
||||
; X32-AVX2-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: isel_crash_32b:
|
||||
; X64: ## BB#0: ## %eintry
|
||||
; X64-NEXT: pushq %rbp
|
||||
; X64-NEXT: Lcfi0:
|
||||
; X64-NEXT: .cfi_def_cfa_offset 16
|
||||
; X64-NEXT: Lcfi1:
|
||||
; X64-NEXT: .cfi_offset %rbp, -16
|
||||
; X64-NEXT: movq %rsp, %rbp
|
||||
; X64-NEXT: Lcfi2:
|
||||
; X64-NEXT: .cfi_def_cfa_register %rbp
|
||||
; X64-NEXT: andq $-32, %rsp
|
||||
; X64-NEXT: subq $128, %rsp
|
||||
; X64-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-NEXT: vmovaps %ymm0, (%rsp)
|
||||
; X64-NEXT: movb (%rdi), %al
|
||||
; X64-NEXT: vmovd %eax, %xmm1
|
||||
; X64-NEXT: vpbroadcastb %xmm1, %ymm1
|
||||
; X64-NEXT: vmovaps %ymm0, {{[0-9]+}}(%rsp)
|
||||
; X64-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%rsp)
|
||||
; X64-NEXT: movq %rbp, %rsp
|
||||
; X64-NEXT: popq %rbp
|
||||
; X64-NEXT: vzeroupper
|
||||
; X64-NEXT: retq
|
||||
; X64-AVX2-LABEL: isel_crash_32b:
|
||||
; X64-AVX2: ## BB#0: ## %eintry
|
||||
; X64-AVX2-NEXT: pushq %rbp
|
||||
; X64-AVX2-NEXT: Lcfi0:
|
||||
; X64-AVX2-NEXT: .cfi_def_cfa_offset 16
|
||||
; X64-AVX2-NEXT: Lcfi1:
|
||||
; X64-AVX2-NEXT: .cfi_offset %rbp, -16
|
||||
; X64-AVX2-NEXT: movq %rsp, %rbp
|
||||
; X64-AVX2-NEXT: Lcfi2:
|
||||
; X64-AVX2-NEXT: .cfi_def_cfa_register %rbp
|
||||
; X64-AVX2-NEXT: andq $-32, %rsp
|
||||
; X64-AVX2-NEXT: subq $128, %rsp
|
||||
; X64-AVX2-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX2-NEXT: vmovaps %ymm0, (%rsp)
|
||||
; X64-AVX2-NEXT: movb (%rdi), %al
|
||||
; X64-AVX2-NEXT: vmovd %eax, %xmm1
|
||||
; X64-AVX2-NEXT: vpbroadcastb %xmm1, %ymm1
|
||||
; X64-AVX2-NEXT: vmovaps %ymm0, {{[0-9]+}}(%rsp)
|
||||
; X64-AVX2-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%rsp)
|
||||
; X64-AVX2-NEXT: movq %rbp, %rsp
|
||||
; X64-AVX2-NEXT: popq %rbp
|
||||
; X64-AVX2-NEXT: vzeroupper
|
||||
; X64-AVX2-NEXT: retq
|
||||
;
|
||||
; X32-AVX512VL-LABEL: isel_crash_32b:
|
||||
; X32-AVX512VL: ## BB#0: ## %eintry
|
||||
; X32-AVX512VL-NEXT: pushl %ebp
|
||||
; X32-AVX512VL-NEXT: Lcfi1:
|
||||
; X32-AVX512VL-NEXT: .cfi_def_cfa_offset 8
|
||||
; X32-AVX512VL-NEXT: Lcfi2:
|
||||
; X32-AVX512VL-NEXT: .cfi_offset %ebp, -8
|
||||
; X32-AVX512VL-NEXT: movl %esp, %ebp
|
||||
; X32-AVX512VL-NEXT: Lcfi3:
|
||||
; X32-AVX512VL-NEXT: .cfi_def_cfa_register %ebp
|
||||
; X32-AVX512VL-NEXT: andl $-32, %esp
|
||||
; X32-AVX512VL-NEXT: subl $128, %esp
|
||||
; X32-AVX512VL-NEXT: movl 8(%ebp), %eax
|
||||
; X32-AVX512VL-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX512VL-NEXT: vmovaps %ymm0, (%esp)
|
||||
; X32-AVX512VL-NEXT: vpbroadcastb (%eax), %ymm1
|
||||
; X32-AVX512VL-NEXT: vmovaps %ymm0, {{[0-9]+}}(%esp)
|
||||
; X32-AVX512VL-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%esp)
|
||||
; X32-AVX512VL-NEXT: movl %ebp, %esp
|
||||
; X32-AVX512VL-NEXT: popl %ebp
|
||||
; X32-AVX512VL-NEXT: vzeroupper
|
||||
; X32-AVX512VL-NEXT: retl
|
||||
;
|
||||
; X64-AVX512VL-LABEL: isel_crash_32b:
|
||||
; X64-AVX512VL: ## BB#0: ## %eintry
|
||||
; X64-AVX512VL-NEXT: pushq %rbp
|
||||
; X64-AVX512VL-NEXT: Lcfi0:
|
||||
; X64-AVX512VL-NEXT: .cfi_def_cfa_offset 16
|
||||
; X64-AVX512VL-NEXT: Lcfi1:
|
||||
; X64-AVX512VL-NEXT: .cfi_offset %rbp, -16
|
||||
; X64-AVX512VL-NEXT: movq %rsp, %rbp
|
||||
; X64-AVX512VL-NEXT: Lcfi2:
|
||||
; X64-AVX512VL-NEXT: .cfi_def_cfa_register %rbp
|
||||
; X64-AVX512VL-NEXT: andq $-32, %rsp
|
||||
; X64-AVX512VL-NEXT: subq $128, %rsp
|
||||
; X64-AVX512VL-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX512VL-NEXT: vmovaps %ymm0, (%rsp)
|
||||
; X64-AVX512VL-NEXT: movb (%rdi), %al
|
||||
; X64-AVX512VL-NEXT: vmovd %eax, %xmm1
|
||||
; X64-AVX512VL-NEXT: vpbroadcastb %xmm1, %ymm1
|
||||
; X64-AVX512VL-NEXT: vmovaps %ymm0, {{[0-9]+}}(%rsp)
|
||||
; X64-AVX512VL-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%rsp)
|
||||
; X64-AVX512VL-NEXT: movq %rbp, %rsp
|
||||
; X64-AVX512VL-NEXT: popq %rbp
|
||||
; X64-AVX512VL-NEXT: vzeroupper
|
||||
; X64-AVX512VL-NEXT: retq
|
||||
eintry:
|
||||
%__a.addr.i = alloca <4 x i64>, align 16
|
||||
%__b.addr.i = alloca <4 x i64>, align 16
|
||||
@ -1237,52 +1284,99 @@ entry:
|
||||
}
|
||||
|
||||
define void @isel_crash_16w(i16* %cV_R.addr) {
|
||||
; X32-LABEL: isel_crash_16w:
|
||||
; X32: ## BB#0: ## %eintry
|
||||
; X32-NEXT: pushl %ebp
|
||||
; X32-NEXT: Lcfi5:
|
||||
; X32-NEXT: .cfi_def_cfa_offset 8
|
||||
; X32-NEXT: Lcfi6:
|
||||
; X32-NEXT: .cfi_offset %ebp, -8
|
||||
; X32-NEXT: movl %esp, %ebp
|
||||
; X32-NEXT: Lcfi7:
|
||||
; X32-NEXT: .cfi_def_cfa_register %ebp
|
||||
; X32-NEXT: andl $-32, %esp
|
||||
; X32-NEXT: subl $128, %esp
|
||||
; X32-NEXT: movl 8(%ebp), %eax
|
||||
; X32-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-NEXT: vmovaps %ymm0, (%esp)
|
||||
; X32-NEXT: vpbroadcastw (%eax), %ymm1
|
||||
; X32-NEXT: vmovaps %ymm0, {{[0-9]+}}(%esp)
|
||||
; X32-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%esp)
|
||||
; X32-NEXT: movl %ebp, %esp
|
||||
; X32-NEXT: popl %ebp
|
||||
; X32-NEXT: vzeroupper
|
||||
; X32-NEXT: retl
|
||||
; X32-AVX2-LABEL: isel_crash_16w:
|
||||
; X32-AVX2: ## BB#0: ## %eintry
|
||||
; X32-AVX2-NEXT: pushl %ebp
|
||||
; X32-AVX2-NEXT: Lcfi5:
|
||||
; X32-AVX2-NEXT: .cfi_def_cfa_offset 8
|
||||
; X32-AVX2-NEXT: Lcfi6:
|
||||
; X32-AVX2-NEXT: .cfi_offset %ebp, -8
|
||||
; X32-AVX2-NEXT: movl %esp, %ebp
|
||||
; X32-AVX2-NEXT: Lcfi7:
|
||||
; X32-AVX2-NEXT: .cfi_def_cfa_register %ebp
|
||||
; X32-AVX2-NEXT: andl $-32, %esp
|
||||
; X32-AVX2-NEXT: subl $128, %esp
|
||||
; X32-AVX2-NEXT: movl 8(%ebp), %eax
|
||||
; X32-AVX2-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX2-NEXT: vmovaps %ymm0, (%esp)
|
||||
; X32-AVX2-NEXT: vpbroadcastw (%eax), %ymm1
|
||||
; X32-AVX2-NEXT: vmovaps %ymm0, {{[0-9]+}}(%esp)
|
||||
; X32-AVX2-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%esp)
|
||||
; X32-AVX2-NEXT: movl %ebp, %esp
|
||||
; X32-AVX2-NEXT: popl %ebp
|
||||
; X32-AVX2-NEXT: vzeroupper
|
||||
; X32-AVX2-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: isel_crash_16w:
|
||||
; X64: ## BB#0: ## %eintry
|
||||
; X64-NEXT: pushq %rbp
|
||||
; X64-NEXT: Lcfi3:
|
||||
; X64-NEXT: .cfi_def_cfa_offset 16
|
||||
; X64-NEXT: Lcfi4:
|
||||
; X64-NEXT: .cfi_offset %rbp, -16
|
||||
; X64-NEXT: movq %rsp, %rbp
|
||||
; X64-NEXT: Lcfi5:
|
||||
; X64-NEXT: .cfi_def_cfa_register %rbp
|
||||
; X64-NEXT: andq $-32, %rsp
|
||||
; X64-NEXT: subq $128, %rsp
|
||||
; X64-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-NEXT: vmovaps %ymm0, (%rsp)
|
||||
; X64-NEXT: movw (%rdi), %ax
|
||||
; X64-NEXT: vmovd %eax, %xmm1
|
||||
; X64-NEXT: vpbroadcastw %xmm1, %ymm1
|
||||
; X64-NEXT: vmovaps %ymm0, {{[0-9]+}}(%rsp)
|
||||
; X64-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%rsp)
|
||||
; X64-NEXT: movq %rbp, %rsp
|
||||
; X64-NEXT: popq %rbp
|
||||
; X64-NEXT: vzeroupper
|
||||
; X64-NEXT: retq
|
||||
; X64-AVX2-LABEL: isel_crash_16w:
|
||||
; X64-AVX2: ## BB#0: ## %eintry
|
||||
; X64-AVX2-NEXT: pushq %rbp
|
||||
; X64-AVX2-NEXT: Lcfi3:
|
||||
; X64-AVX2-NEXT: .cfi_def_cfa_offset 16
|
||||
; X64-AVX2-NEXT: Lcfi4:
|
||||
; X64-AVX2-NEXT: .cfi_offset %rbp, -16
|
||||
; X64-AVX2-NEXT: movq %rsp, %rbp
|
||||
; X64-AVX2-NEXT: Lcfi5:
|
||||
; X64-AVX2-NEXT: .cfi_def_cfa_register %rbp
|
||||
; X64-AVX2-NEXT: andq $-32, %rsp
|
||||
; X64-AVX2-NEXT: subq $128, %rsp
|
||||
; X64-AVX2-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX2-NEXT: vmovaps %ymm0, (%rsp)
|
||||
; X64-AVX2-NEXT: movw (%rdi), %ax
|
||||
; X64-AVX2-NEXT: vmovd %eax, %xmm1
|
||||
; X64-AVX2-NEXT: vpbroadcastw %xmm1, %ymm1
|
||||
; X64-AVX2-NEXT: vmovaps %ymm0, {{[0-9]+}}(%rsp)
|
||||
; X64-AVX2-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%rsp)
|
||||
; X64-AVX2-NEXT: movq %rbp, %rsp
|
||||
; X64-AVX2-NEXT: popq %rbp
|
||||
; X64-AVX2-NEXT: vzeroupper
|
||||
; X64-AVX2-NEXT: retq
|
||||
;
|
||||
; X32-AVX512VL-LABEL: isel_crash_16w:
|
||||
; X32-AVX512VL: ## BB#0: ## %eintry
|
||||
; X32-AVX512VL-NEXT: pushl %ebp
|
||||
; X32-AVX512VL-NEXT: Lcfi5:
|
||||
; X32-AVX512VL-NEXT: .cfi_def_cfa_offset 8
|
||||
; X32-AVX512VL-NEXT: Lcfi6:
|
||||
; X32-AVX512VL-NEXT: .cfi_offset %ebp, -8
|
||||
; X32-AVX512VL-NEXT: movl %esp, %ebp
|
||||
; X32-AVX512VL-NEXT: Lcfi7:
|
||||
; X32-AVX512VL-NEXT: .cfi_def_cfa_register %ebp
|
||||
; X32-AVX512VL-NEXT: andl $-32, %esp
|
||||
; X32-AVX512VL-NEXT: subl $128, %esp
|
||||
; X32-AVX512VL-NEXT: movl 8(%ebp), %eax
|
||||
; X32-AVX512VL-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX512VL-NEXT: vmovaps %ymm0, (%esp)
|
||||
; X32-AVX512VL-NEXT: vpbroadcastw (%eax), %ymm1
|
||||
; X32-AVX512VL-NEXT: vmovaps %ymm0, {{[0-9]+}}(%esp)
|
||||
; X32-AVX512VL-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%esp)
|
||||
; X32-AVX512VL-NEXT: movl %ebp, %esp
|
||||
; X32-AVX512VL-NEXT: popl %ebp
|
||||
; X32-AVX512VL-NEXT: vzeroupper
|
||||
; X32-AVX512VL-NEXT: retl
|
||||
;
|
||||
; X64-AVX512VL-LABEL: isel_crash_16w:
|
||||
; X64-AVX512VL: ## BB#0: ## %eintry
|
||||
; X64-AVX512VL-NEXT: pushq %rbp
|
||||
; X64-AVX512VL-NEXT: Lcfi3:
|
||||
; X64-AVX512VL-NEXT: .cfi_def_cfa_offset 16
|
||||
; X64-AVX512VL-NEXT: Lcfi4:
|
||||
; X64-AVX512VL-NEXT: .cfi_offset %rbp, -16
|
||||
; X64-AVX512VL-NEXT: movq %rsp, %rbp
|
||||
; X64-AVX512VL-NEXT: Lcfi5:
|
||||
; X64-AVX512VL-NEXT: .cfi_def_cfa_register %rbp
|
||||
; X64-AVX512VL-NEXT: andq $-32, %rsp
|
||||
; X64-AVX512VL-NEXT: subq $128, %rsp
|
||||
; X64-AVX512VL-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX512VL-NEXT: vmovaps %ymm0, (%rsp)
|
||||
; X64-AVX512VL-NEXT: movw (%rdi), %ax
|
||||
; X64-AVX512VL-NEXT: vmovd %eax, %xmm1
|
||||
; X64-AVX512VL-NEXT: vpbroadcastw %xmm1, %ymm1
|
||||
; X64-AVX512VL-NEXT: vmovaps %ymm0, {{[0-9]+}}(%rsp)
|
||||
; X64-AVX512VL-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%rsp)
|
||||
; X64-AVX512VL-NEXT: movq %rbp, %rsp
|
||||
; X64-AVX512VL-NEXT: popq %rbp
|
||||
; X64-AVX512VL-NEXT: vzeroupper
|
||||
; X64-AVX512VL-NEXT: retq
|
||||
eintry:
|
||||
%__a.addr.i = alloca <4 x i64>, align 16
|
||||
%__b.addr.i = alloca <4 x i64>, align 16
|
||||
@ -1349,28 +1443,28 @@ entry:
|
||||
}
|
||||
|
||||
define void @isel_crash_8d(i32* %cV_R.addr) {
|
||||
; X32-LABEL: isel_crash_8d:
|
||||
; X32: ## BB#0: ## %eintry
|
||||
; X32-NEXT: pushl %ebp
|
||||
; X32-NEXT: Lcfi9:
|
||||
; X32-NEXT: .cfi_def_cfa_offset 8
|
||||
; X32-NEXT: Lcfi10:
|
||||
; X32-NEXT: .cfi_offset %ebp, -8
|
||||
; X32-NEXT: movl %esp, %ebp
|
||||
; X32-NEXT: Lcfi11:
|
||||
; X32-NEXT: .cfi_def_cfa_register %ebp
|
||||
; X32-NEXT: andl $-32, %esp
|
||||
; X32-NEXT: subl $128, %esp
|
||||
; X32-NEXT: movl 8(%ebp), %eax
|
||||
; X32-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-NEXT: vmovaps %ymm0, (%esp)
|
||||
; X32-NEXT: vbroadcastss (%eax), %ymm1
|
||||
; X32-NEXT: vmovaps %ymm0, {{[0-9]+}}(%esp)
|
||||
; X32-NEXT: vmovaps %ymm1, {{[0-9]+}}(%esp)
|
||||
; X32-NEXT: movl %ebp, %esp
|
||||
; X32-NEXT: popl %ebp
|
||||
; X32-NEXT: vzeroupper
|
||||
; X32-NEXT: retl
|
||||
; X32-AVX2-LABEL: isel_crash_8d:
|
||||
; X32-AVX2: ## BB#0: ## %eintry
|
||||
; X32-AVX2-NEXT: pushl %ebp
|
||||
; X32-AVX2-NEXT: Lcfi9:
|
||||
; X32-AVX2-NEXT: .cfi_def_cfa_offset 8
|
||||
; X32-AVX2-NEXT: Lcfi10:
|
||||
; X32-AVX2-NEXT: .cfi_offset %ebp, -8
|
||||
; X32-AVX2-NEXT: movl %esp, %ebp
|
||||
; X32-AVX2-NEXT: Lcfi11:
|
||||
; X32-AVX2-NEXT: .cfi_def_cfa_register %ebp
|
||||
; X32-AVX2-NEXT: andl $-32, %esp
|
||||
; X32-AVX2-NEXT: subl $128, %esp
|
||||
; X32-AVX2-NEXT: movl 8(%ebp), %eax
|
||||
; X32-AVX2-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX2-NEXT: vmovaps %ymm0, (%esp)
|
||||
; X32-AVX2-NEXT: vbroadcastss (%eax), %ymm1
|
||||
; X32-AVX2-NEXT: vmovaps %ymm0, {{[0-9]+}}(%esp)
|
||||
; X32-AVX2-NEXT: vmovaps %ymm1, {{[0-9]+}}(%esp)
|
||||
; X32-AVX2-NEXT: movl %ebp, %esp
|
||||
; X32-AVX2-NEXT: popl %ebp
|
||||
; X32-AVX2-NEXT: vzeroupper
|
||||
; X32-AVX2-NEXT: retl
|
||||
;
|
||||
; X64-AVX2-LABEL: isel_crash_8d:
|
||||
; X64-AVX2: ## BB#0: ## %eintry
|
||||
@ -1384,7 +1478,7 @@ define void @isel_crash_8d(i32* %cV_R.addr) {
|
||||
; X64-AVX2-NEXT: .cfi_def_cfa_register %rbp
|
||||
; X64-AVX2-NEXT: andq $-32, %rsp
|
||||
; X64-AVX2-NEXT: subq $128, %rsp
|
||||
; X64-AVX2-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX2-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX2-NEXT: vmovaps %ymm0, (%rsp)
|
||||
; X64-AVX2-NEXT: movl (%rdi), %eax
|
||||
; X64-AVX2-NEXT: vmovd %eax, %xmm1
|
||||
@ -1396,6 +1490,29 @@ define void @isel_crash_8d(i32* %cV_R.addr) {
|
||||
; X64-AVX2-NEXT: vzeroupper
|
||||
; X64-AVX2-NEXT: retq
|
||||
;
|
||||
; X32-AVX512VL-LABEL: isel_crash_8d:
|
||||
; X32-AVX512VL: ## BB#0: ## %eintry
|
||||
; X32-AVX512VL-NEXT: pushl %ebp
|
||||
; X32-AVX512VL-NEXT: Lcfi9:
|
||||
; X32-AVX512VL-NEXT: .cfi_def_cfa_offset 8
|
||||
; X32-AVX512VL-NEXT: Lcfi10:
|
||||
; X32-AVX512VL-NEXT: .cfi_offset %ebp, -8
|
||||
; X32-AVX512VL-NEXT: movl %esp, %ebp
|
||||
; X32-AVX512VL-NEXT: Lcfi11:
|
||||
; X32-AVX512VL-NEXT: .cfi_def_cfa_register %ebp
|
||||
; X32-AVX512VL-NEXT: andl $-32, %esp
|
||||
; X32-AVX512VL-NEXT: subl $128, %esp
|
||||
; X32-AVX512VL-NEXT: movl 8(%ebp), %eax
|
||||
; X32-AVX512VL-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX512VL-NEXT: vmovaps %ymm0, (%esp)
|
||||
; X32-AVX512VL-NEXT: vbroadcastss (%eax), %ymm1
|
||||
; X32-AVX512VL-NEXT: vmovaps %ymm0, {{[0-9]+}}(%esp)
|
||||
; X32-AVX512VL-NEXT: vmovaps %ymm1, {{[0-9]+}}(%esp)
|
||||
; X32-AVX512VL-NEXT: movl %ebp, %esp
|
||||
; X32-AVX512VL-NEXT: popl %ebp
|
||||
; X32-AVX512VL-NEXT: vzeroupper
|
||||
; X32-AVX512VL-NEXT: retl
|
||||
;
|
||||
; X64-AVX512VL-LABEL: isel_crash_8d:
|
||||
; X64-AVX512VL: ## BB#0: ## %eintry
|
||||
; X64-AVX512VL-NEXT: pushq %rbp
|
||||
@ -1488,34 +1605,34 @@ entry:
|
||||
}
|
||||
|
||||
define void @isel_crash_4q(i64* %cV_R.addr) {
|
||||
; X32-LABEL: isel_crash_4q:
|
||||
; X32: ## BB#0: ## %eintry
|
||||
; X32-NEXT: pushl %ebp
|
||||
; X32-NEXT: Lcfi13:
|
||||
; X32-NEXT: .cfi_def_cfa_offset 8
|
||||
; X32-NEXT: Lcfi14:
|
||||
; X32-NEXT: .cfi_offset %ebp, -8
|
||||
; X32-NEXT: movl %esp, %ebp
|
||||
; X32-NEXT: Lcfi15:
|
||||
; X32-NEXT: .cfi_def_cfa_register %ebp
|
||||
; X32-NEXT: andl $-32, %esp
|
||||
; X32-NEXT: subl $128, %esp
|
||||
; X32-NEXT: movl 8(%ebp), %eax
|
||||
; X32-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-NEXT: vmovaps %ymm0, (%esp)
|
||||
; X32-NEXT: movl (%eax), %ecx
|
||||
; X32-NEXT: movl 4(%eax), %eax
|
||||
; X32-NEXT: vmovd %ecx, %xmm1
|
||||
; X32-NEXT: vpinsrd $1, %eax, %xmm1, %xmm1
|
||||
; X32-NEXT: vpinsrd $2, %ecx, %xmm1, %xmm1
|
||||
; X32-NEXT: vpinsrd $3, %eax, %xmm1, %xmm1
|
||||
; X32-NEXT: vinserti128 $1, %xmm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vmovaps %ymm0, {{[0-9]+}}(%esp)
|
||||
; X32-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%esp)
|
||||
; X32-NEXT: movl %ebp, %esp
|
||||
; X32-NEXT: popl %ebp
|
||||
; X32-NEXT: vzeroupper
|
||||
; X32-NEXT: retl
|
||||
; X32-AVX2-LABEL: isel_crash_4q:
|
||||
; X32-AVX2: ## BB#0: ## %eintry
|
||||
; X32-AVX2-NEXT: pushl %ebp
|
||||
; X32-AVX2-NEXT: Lcfi13:
|
||||
; X32-AVX2-NEXT: .cfi_def_cfa_offset 8
|
||||
; X32-AVX2-NEXT: Lcfi14:
|
||||
; X32-AVX2-NEXT: .cfi_offset %ebp, -8
|
||||
; X32-AVX2-NEXT: movl %esp, %ebp
|
||||
; X32-AVX2-NEXT: Lcfi15:
|
||||
; X32-AVX2-NEXT: .cfi_def_cfa_register %ebp
|
||||
; X32-AVX2-NEXT: andl $-32, %esp
|
||||
; X32-AVX2-NEXT: subl $128, %esp
|
||||
; X32-AVX2-NEXT: movl 8(%ebp), %eax
|
||||
; X32-AVX2-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX2-NEXT: vmovaps %ymm0, (%esp)
|
||||
; X32-AVX2-NEXT: movl (%eax), %ecx
|
||||
; X32-AVX2-NEXT: movl 4(%eax), %eax
|
||||
; X32-AVX2-NEXT: vmovd %ecx, %xmm1
|
||||
; X32-AVX2-NEXT: vpinsrd $1, %eax, %xmm1, %xmm1
|
||||
; X32-AVX2-NEXT: vpinsrd $2, %ecx, %xmm1, %xmm1
|
||||
; X32-AVX2-NEXT: vpinsrd $3, %eax, %xmm1, %xmm1
|
||||
; X32-AVX2-NEXT: vinserti128 $1, %xmm1, %ymm1, %ymm1
|
||||
; X32-AVX2-NEXT: vmovaps %ymm0, {{[0-9]+}}(%esp)
|
||||
; X32-AVX2-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%esp)
|
||||
; X32-AVX2-NEXT: movl %ebp, %esp
|
||||
; X32-AVX2-NEXT: popl %ebp
|
||||
; X32-AVX2-NEXT: vzeroupper
|
||||
; X32-AVX2-NEXT: retl
|
||||
;
|
||||
; X64-AVX2-LABEL: isel_crash_4q:
|
||||
; X64-AVX2: ## BB#0: ## %eintry
|
||||
@ -1529,7 +1646,7 @@ define void @isel_crash_4q(i64* %cV_R.addr) {
|
||||
; X64-AVX2-NEXT: .cfi_def_cfa_register %rbp
|
||||
; X64-AVX2-NEXT: andq $-32, %rsp
|
||||
; X64-AVX2-NEXT: subq $128, %rsp
|
||||
; X64-AVX2-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-AVX2-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-AVX2-NEXT: vmovaps %ymm0, (%rsp)
|
||||
; X64-AVX2-NEXT: movq (%rdi), %rax
|
||||
; X64-AVX2-NEXT: vmovq %rax, %xmm1
|
||||
@ -1541,6 +1658,35 @@ define void @isel_crash_4q(i64* %cV_R.addr) {
|
||||
; X64-AVX2-NEXT: vzeroupper
|
||||
; X64-AVX2-NEXT: retq
|
||||
;
|
||||
; X32-AVX512VL-LABEL: isel_crash_4q:
|
||||
; X32-AVX512VL: ## BB#0: ## %eintry
|
||||
; X32-AVX512VL-NEXT: pushl %ebp
|
||||
; X32-AVX512VL-NEXT: Lcfi13:
|
||||
; X32-AVX512VL-NEXT: .cfi_def_cfa_offset 8
|
||||
; X32-AVX512VL-NEXT: Lcfi14:
|
||||
; X32-AVX512VL-NEXT: .cfi_offset %ebp, -8
|
||||
; X32-AVX512VL-NEXT: movl %esp, %ebp
|
||||
; X32-AVX512VL-NEXT: Lcfi15:
|
||||
; X32-AVX512VL-NEXT: .cfi_def_cfa_register %ebp
|
||||
; X32-AVX512VL-NEXT: andl $-32, %esp
|
||||
; X32-AVX512VL-NEXT: subl $128, %esp
|
||||
; X32-AVX512VL-NEXT: movl 8(%ebp), %eax
|
||||
; X32-AVX512VL-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX512VL-NEXT: vmovaps %ymm0, (%esp)
|
||||
; X32-AVX512VL-NEXT: movl (%eax), %ecx
|
||||
; X32-AVX512VL-NEXT: movl 4(%eax), %eax
|
||||
; X32-AVX512VL-NEXT: vmovd %ecx, %xmm1
|
||||
; X32-AVX512VL-NEXT: vpinsrd $1, %eax, %xmm1, %xmm1
|
||||
; X32-AVX512VL-NEXT: vpinsrd $2, %ecx, %xmm1, %xmm1
|
||||
; X32-AVX512VL-NEXT: vpinsrd $3, %eax, %xmm1, %xmm1
|
||||
; X32-AVX512VL-NEXT: vinserti128 $1, %xmm1, %ymm1, %ymm1
|
||||
; X32-AVX512VL-NEXT: vmovaps %ymm0, {{[0-9]+}}(%esp)
|
||||
; X32-AVX512VL-NEXT: vmovdqa %ymm1, {{[0-9]+}}(%esp)
|
||||
; X32-AVX512VL-NEXT: movl %ebp, %esp
|
||||
; X32-AVX512VL-NEXT: popl %ebp
|
||||
; X32-AVX512VL-NEXT: vzeroupper
|
||||
; X32-AVX512VL-NEXT: retl
|
||||
;
|
||||
; X64-AVX512VL-LABEL: isel_crash_4q:
|
||||
; X64-AVX512VL: ## BB#0: ## %eintry
|
||||
; X64-AVX512VL-NEXT: pushq %rbp
|
||||
|
||||
@ -272,7 +272,7 @@ define <8 x i32> @PR29088(<4 x i32>* %p0, <8 x float>* %p1) {
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
||||
; X32-NEXT: vmovaps (%ecx), %xmm0
|
||||
; X32-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vmovaps %ymm1, (%eax)
|
||||
; X32-NEXT: vinsertf128 $1, %xmm0, %ymm0, %ymm0
|
||||
; X32-NEXT: retl
|
||||
@ -280,7 +280,7 @@ define <8 x i32> @PR29088(<4 x i32>* %p0, <8 x float>* %p1) {
|
||||
; X64-LABEL: PR29088:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vmovaps (%rdi), %xmm0
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vmovaps %ymm1, (%rsi)
|
||||
; X64-NEXT: vinsertf128 $1, %xmm0, %ymm0, %ymm0
|
||||
; X64-NEXT: retq
|
||||
|
||||
@ -430,7 +430,7 @@ define <8 x i16> @shl_8i16(<8 x i16> %r, <8 x i16> %a) nounwind {
|
||||
define <16 x i16> @shl_16i16(<16 x i16> %r, <16 x i16> %a) nounwind {
|
||||
; X32-LABEL: shl_16i16:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; X32-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; X32-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm1[4],ymm2[4],ymm1[5],ymm2[5],ymm1[6],ymm2[6],ymm1[7],ymm2[7],ymm1[12],ymm2[12],ymm1[13],ymm2[13],ymm1[14],ymm2[14],ymm1[15],ymm2[15]
|
||||
; X32-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm2[4],ymm0[4],ymm2[5],ymm0[5],ymm2[6],ymm0[6],ymm2[7],ymm0[7],ymm2[12],ymm0[12],ymm2[13],ymm0[13],ymm2[14],ymm0[14],ymm2[15],ymm0[15]
|
||||
; X32-NEXT: vpsllvd %ymm3, %ymm4, %ymm3
|
||||
@ -444,7 +444,7 @@ define <16 x i16> @shl_16i16(<16 x i16> %r, <16 x i16> %a) nounwind {
|
||||
;
|
||||
; X64-LABEL: shl_16i16:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; X64-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; X64-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm1[4],ymm2[4],ymm1[5],ymm2[5],ymm1[6],ymm2[6],ymm1[7],ymm2[7],ymm1[12],ymm2[12],ymm1[13],ymm2[13],ymm1[14],ymm2[14],ymm1[15],ymm2[15]
|
||||
; X64-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm2[4],ymm0[4],ymm2[5],ymm0[5],ymm2[6],ymm0[6],ymm2[7],ymm0[7],ymm2[12],ymm0[12],ymm2[13],ymm0[13],ymm2[14],ymm0[14],ymm2[15],ymm0[15]
|
||||
; X64-NEXT: vpsllvd %ymm3, %ymm4, %ymm3
|
||||
@ -522,7 +522,7 @@ define <8 x i16> @ashr_8i16(<8 x i16> %r, <8 x i16> %a) nounwind {
|
||||
define <16 x i16> @ashr_16i16(<16 x i16> %r, <16 x i16> %a) nounwind {
|
||||
; X32-LABEL: ashr_16i16:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; X32-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; X32-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm1[4],ymm2[4],ymm1[5],ymm2[5],ymm1[6],ymm2[6],ymm1[7],ymm2[7],ymm1[12],ymm2[12],ymm1[13],ymm2[13],ymm1[14],ymm2[14],ymm1[15],ymm2[15]
|
||||
; X32-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm2[4],ymm0[4],ymm2[5],ymm0[5],ymm2[6],ymm0[6],ymm2[7],ymm0[7],ymm2[12],ymm0[12],ymm2[13],ymm0[13],ymm2[14],ymm0[14],ymm2[15],ymm0[15]
|
||||
; X32-NEXT: vpsravd %ymm3, %ymm4, %ymm3
|
||||
@ -536,7 +536,7 @@ define <16 x i16> @ashr_16i16(<16 x i16> %r, <16 x i16> %a) nounwind {
|
||||
;
|
||||
; X64-LABEL: ashr_16i16:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; X64-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; X64-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm1[4],ymm2[4],ymm1[5],ymm2[5],ymm1[6],ymm2[6],ymm1[7],ymm2[7],ymm1[12],ymm2[12],ymm1[13],ymm2[13],ymm1[14],ymm2[14],ymm1[15],ymm2[15]
|
||||
; X64-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm2[4],ymm0[4],ymm2[5],ymm0[5],ymm2[6],ymm0[6],ymm2[7],ymm0[7],ymm2[12],ymm0[12],ymm2[13],ymm0[13],ymm2[14],ymm0[14],ymm2[15],ymm0[15]
|
||||
; X64-NEXT: vpsravd %ymm3, %ymm4, %ymm3
|
||||
@ -640,7 +640,7 @@ define <8 x i16> @lshr_8i16(<8 x i16> %r, <8 x i16> %a) nounwind {
|
||||
define <16 x i16> @lshr_16i16(<16 x i16> %r, <16 x i16> %a) nounwind {
|
||||
; X32-LABEL: lshr_16i16:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; X32-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; X32-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm1[4],ymm2[4],ymm1[5],ymm2[5],ymm1[6],ymm2[6],ymm1[7],ymm2[7],ymm1[12],ymm2[12],ymm1[13],ymm2[13],ymm1[14],ymm2[14],ymm1[15],ymm2[15]
|
||||
; X32-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm2[4],ymm0[4],ymm2[5],ymm0[5],ymm2[6],ymm0[6],ymm2[7],ymm0[7],ymm2[12],ymm0[12],ymm2[13],ymm0[13],ymm2[14],ymm0[14],ymm2[15],ymm0[15]
|
||||
; X32-NEXT: vpsrlvd %ymm3, %ymm4, %ymm3
|
||||
@ -654,7 +654,7 @@ define <16 x i16> @lshr_16i16(<16 x i16> %r, <16 x i16> %a) nounwind {
|
||||
;
|
||||
; X64-LABEL: lshr_16i16:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; X64-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; X64-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm1[4],ymm2[4],ymm1[5],ymm2[5],ymm1[6],ymm2[6],ymm1[7],ymm2[7],ymm1[12],ymm2[12],ymm1[13],ymm2[13],ymm1[14],ymm2[14],ymm1[15],ymm2[15]
|
||||
; X64-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm2[4],ymm0[4],ymm2[5],ymm0[5],ymm2[6],ymm0[6],ymm2[7],ymm0[7],ymm2[12],ymm0[12],ymm2[13],ymm0[13],ymm2[14],ymm0[14],ymm2[15],ymm0[15]
|
||||
; X64-NEXT: vpsrlvd %ymm3, %ymm4, %ymm3
|
||||
|
||||
@ -718,7 +718,7 @@ define <8 x double> @test_mask_vminpd(<8 x double> %dst, <8 x double> %i,
|
||||
; AVX512F-LABEL: test_mask_vminpd:
|
||||
; AVX512F: # BB#0:
|
||||
; AVX512F-NEXT: # kill: %YMM3<def> %YMM3<kill> %ZMM3<def>
|
||||
; AVX512F-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; AVX512F-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; AVX512F-NEXT: vpcmpneqd %zmm4, %zmm3, %k1
|
||||
; AVX512F-NEXT: vminpd %zmm2, %zmm1, %zmm0 {%k1}
|
||||
; AVX512F-NEXT: retq
|
||||
@ -733,7 +733,7 @@ define <8 x double> @test_mask_vminpd(<8 x double> %dst, <8 x double> %i,
|
||||
; AVX512BW-LABEL: test_mask_vminpd:
|
||||
; AVX512BW: # BB#0:
|
||||
; AVX512BW-NEXT: # kill: %YMM3<def> %YMM3<kill> %ZMM3<def>
|
||||
; AVX512BW-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; AVX512BW-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; AVX512BW-NEXT: vpcmpneqd %zmm4, %zmm3, %k1
|
||||
; AVX512BW-NEXT: vminpd %zmm2, %zmm1, %zmm0 {%k1}
|
||||
; AVX512BW-NEXT: retq
|
||||
@ -741,7 +741,7 @@ define <8 x double> @test_mask_vminpd(<8 x double> %dst, <8 x double> %i,
|
||||
; AVX512DQ-LABEL: test_mask_vminpd:
|
||||
; AVX512DQ: # BB#0:
|
||||
; AVX512DQ-NEXT: # kill: %YMM3<def> %YMM3<kill> %ZMM3<def>
|
||||
; AVX512DQ-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; AVX512DQ-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; AVX512DQ-NEXT: vpcmpneqd %zmm4, %zmm3, %k1
|
||||
; AVX512DQ-NEXT: vminpd %zmm2, %zmm1, %zmm0 {%k1}
|
||||
; AVX512DQ-NEXT: retq
|
||||
@ -781,7 +781,7 @@ define <8 x double> @test_mask_vmaxpd(<8 x double> %dst, <8 x double> %i,
|
||||
; AVX512F-LABEL: test_mask_vmaxpd:
|
||||
; AVX512F: # BB#0:
|
||||
; AVX512F-NEXT: # kill: %YMM3<def> %YMM3<kill> %ZMM3<def>
|
||||
; AVX512F-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; AVX512F-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; AVX512F-NEXT: vpcmpneqd %zmm4, %zmm3, %k1
|
||||
; AVX512F-NEXT: vmaxpd %zmm2, %zmm1, %zmm0 {%k1}
|
||||
; AVX512F-NEXT: retq
|
||||
@ -796,7 +796,7 @@ define <8 x double> @test_mask_vmaxpd(<8 x double> %dst, <8 x double> %i,
|
||||
; AVX512BW-LABEL: test_mask_vmaxpd:
|
||||
; AVX512BW: # BB#0:
|
||||
; AVX512BW-NEXT: # kill: %YMM3<def> %YMM3<kill> %ZMM3<def>
|
||||
; AVX512BW-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; AVX512BW-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; AVX512BW-NEXT: vpcmpneqd %zmm4, %zmm3, %k1
|
||||
; AVX512BW-NEXT: vmaxpd %zmm2, %zmm1, %zmm0 {%k1}
|
||||
; AVX512BW-NEXT: retq
|
||||
@ -804,7 +804,7 @@ define <8 x double> @test_mask_vmaxpd(<8 x double> %dst, <8 x double> %i,
|
||||
; AVX512DQ-LABEL: test_mask_vmaxpd:
|
||||
; AVX512DQ: # BB#0:
|
||||
; AVX512DQ-NEXT: # kill: %YMM3<def> %YMM3<kill> %ZMM3<def>
|
||||
; AVX512DQ-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; AVX512DQ-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; AVX512DQ-NEXT: vpcmpneqd %zmm4, %zmm3, %k1
|
||||
; AVX512DQ-NEXT: vmaxpd %zmm2, %zmm1, %zmm0 {%k1}
|
||||
; AVX512DQ-NEXT: retq
|
||||
|
||||
@ -1232,7 +1232,7 @@ define <8 x float> @sitofp_8i1_float(<8 x float> %a) {
|
||||
; NOVLDQ-LABEL: sitofp_8i1_float:
|
||||
; NOVLDQ: # BB#0:
|
||||
; NOVLDQ-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
|
||||
; NOVLDQ-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; NOVLDQ-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; NOVLDQ-NEXT: vcmpltps %zmm0, %zmm1, %k1
|
||||
; NOVLDQ-NEXT: vpternlogq $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
|
||||
; NOVLDQ-NEXT: vpmovqd %zmm0, %ymm0
|
||||
@ -1259,7 +1259,7 @@ define <8 x float> @sitofp_8i1_float(<8 x float> %a) {
|
||||
; AVX512DQ-LABEL: sitofp_8i1_float:
|
||||
; AVX512DQ: # BB#0:
|
||||
; AVX512DQ-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
|
||||
; AVX512DQ-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX512DQ-NEXT: vcmpltps %zmm0, %zmm1, %k0
|
||||
; AVX512DQ-NEXT: vpmovm2d %k0, %zmm0
|
||||
; AVX512DQ-NEXT: vcvtdq2ps %ymm0, %ymm0
|
||||
@ -1301,7 +1301,7 @@ define <4 x float> @sitofp_4i1_float(<4 x float> %a) {
|
||||
define <4 x double> @sitofp_4i1_double(<4 x double> %a) {
|
||||
; NOVL-LABEL: sitofp_4i1_double:
|
||||
; NOVL: # BB#0:
|
||||
; NOVL-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; NOVL-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; NOVL-NEXT: vcmpltpd %ymm0, %ymm1, %ymm0
|
||||
; NOVL-NEXT: vpmovqd %zmm0, %ymm0
|
||||
; NOVL-NEXT: vcvtdq2pd %xmm0, %ymm0
|
||||
@ -1459,7 +1459,7 @@ define <8 x float> @uitofp_8i1_float(<8 x i32> %a) {
|
||||
; NOVL-LABEL: uitofp_8i1_float:
|
||||
; NOVL: # BB#0:
|
||||
; NOVL-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
|
||||
; NOVL-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; NOVL-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; NOVL-NEXT: vpcmpgtd %zmm0, %zmm1, %k1
|
||||
; NOVL-NEXT: vpbroadcastq {{.*}}(%rip), %zmm0 {%k1} {z}
|
||||
; NOVL-NEXT: vpmovqd %zmm0, %ymm0
|
||||
@ -1483,7 +1483,7 @@ define <8 x double> @uitofp_8i1_double(<8 x i32> %a) {
|
||||
; NOVL-LABEL: uitofp_8i1_double:
|
||||
; NOVL: # BB#0:
|
||||
; NOVL-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
|
||||
; NOVL-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; NOVL-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; NOVL-NEXT: vpcmpgtd %zmm0, %zmm1, %k1
|
||||
; NOVL-NEXT: vpbroadcastq {{.*}}(%rip), %zmm0 {%k1} {z}
|
||||
; NOVL-NEXT: vpmovqd %zmm0, %ymm0
|
||||
|
||||
@ -346,7 +346,7 @@ define <8 x i32> @zext_8x8mem_to_8x32(<8 x i8> *%i , <8 x i1> %mask) nounwind re
|
||||
; KNL-NEXT: vpsllq $63, %zmm0, %zmm0
|
||||
; KNL-NEXT: vptestmq %zmm0, %zmm0, %k1
|
||||
; KNL-NEXT: vpmovzxbd {{.*#+}} ymm1 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero
|
||||
; KNL-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; KNL-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; KNL-NEXT: vmovdqa32 %zmm1, %zmm0 {%k1}
|
||||
; KNL-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<kill>
|
||||
; KNL-NEXT: retq
|
||||
@ -370,7 +370,7 @@ define <8 x i32> @sext_8x8mem_to_8x32(<8 x i8> *%i , <8 x i1> %mask) nounwind re
|
||||
; KNL-NEXT: vpsllq $63, %zmm0, %zmm0
|
||||
; KNL-NEXT: vptestmq %zmm0, %zmm0, %k1
|
||||
; KNL-NEXT: vpmovsxbd (%rdi), %ymm1
|
||||
; KNL-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; KNL-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; KNL-NEXT: vmovdqa32 %zmm1, %zmm0 {%k1}
|
||||
; KNL-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<kill>
|
||||
; KNL-NEXT: retq
|
||||
@ -703,7 +703,7 @@ define <8 x i32> @zext_8x16mem_to_8x32(<8 x i16> *%i , <8 x i1> %mask) nounwind
|
||||
; KNL-NEXT: vpsllq $63, %zmm0, %zmm0
|
||||
; KNL-NEXT: vptestmq %zmm0, %zmm0, %k1
|
||||
; KNL-NEXT: vpmovzxwd {{.*#+}} ymm1 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero
|
||||
; KNL-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; KNL-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; KNL-NEXT: vmovdqa32 %zmm1, %zmm0 {%k1}
|
||||
; KNL-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<kill>
|
||||
; KNL-NEXT: retq
|
||||
@ -727,7 +727,7 @@ define <8 x i32> @sext_8x16mem_to_8x32mask(<8 x i16> *%i , <8 x i1> %mask) nounw
|
||||
; KNL-NEXT: vpsllq $63, %zmm0, %zmm0
|
||||
; KNL-NEXT: vptestmq %zmm0, %zmm0, %k1
|
||||
; KNL-NEXT: vpmovsxwd (%rdi), %ymm1
|
||||
; KNL-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; KNL-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; KNL-NEXT: vmovdqa32 %zmm1, %zmm0 {%k1}
|
||||
; KNL-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<kill>
|
||||
; KNL-NEXT: retq
|
||||
@ -761,7 +761,7 @@ define <8 x i32> @zext_8x16_to_8x32mask(<8 x i16> %a , <8 x i1> %mask) nounwind
|
||||
; KNL-NEXT: vpsllq $63, %zmm1, %zmm1
|
||||
; KNL-NEXT: vptestmq %zmm1, %zmm1, %k1
|
||||
; KNL-NEXT: vpmovzxwd {{.*#+}} ymm1 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
|
||||
; KNL-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; KNL-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; KNL-NEXT: vmovdqa32 %zmm1, %zmm0 {%k1}
|
||||
; KNL-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<kill>
|
||||
; KNL-NEXT: retq
|
||||
|
||||
@ -1268,7 +1268,7 @@ define i32 @test_insertelement_v32i1(i32 %a, i32 %b, <32 x i32> %x , <32 x i32>
|
||||
; KNL-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm0
|
||||
; KNL-NEXT: vpsllw $7, %ymm0, %ymm0
|
||||
; KNL-NEXT: vpand {{.*}}(%rip), %ymm0, %ymm0
|
||||
; KNL-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; KNL-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; KNL-NEXT: vpcmpgtb %ymm0, %ymm1, %ymm0
|
||||
; KNL-NEXT: vextracti128 $1, %ymm0, %xmm1
|
||||
; KNL-NEXT: vpmovsxbd %xmm1, %zmm1
|
||||
|
||||
@ -1156,14 +1156,14 @@ define <8 x double> @test_mm512_zextpd256_pd512(<4 x double> %a0) nounwind {
|
||||
; X32-LABEL: test_mm512_zextpd256_pd512:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
|
||||
; X32-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vinsertf64x4 $1, %ymm1, %zmm0, %zmm0
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: test_mm512_zextpd256_pd512:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
|
||||
; X64-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vinsertf64x4 $1, %ymm1, %zmm0, %zmm0
|
||||
; X64-NEXT: retq
|
||||
%res = shufflevector <4 x double> %a0, <4 x double> zeroinitializer, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
|
||||
@ -1196,14 +1196,14 @@ define <16 x float> @test_mm512_zextps256_ps512(<8 x float> %a0) nounwind {
|
||||
; X32-LABEL: test_mm512_zextps256_ps512:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
|
||||
; X32-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vinsertf64x4 $1, %ymm1, %zmm0, %zmm0
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: test_mm512_zextps256_ps512:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
|
||||
; X64-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vinsertf64x4 $1, %ymm1, %zmm0, %zmm0
|
||||
; X64-NEXT: retq
|
||||
%res = shufflevector <8 x float> %a0, <8 x float> zeroinitializer, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
|
||||
@ -1236,14 +1236,14 @@ define <8 x i64> @test_mm512_zextsi256_si512(<4 x i64> %a0) nounwind {
|
||||
; X32-LABEL: test_mm512_zextsi256_si512:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
|
||||
; X32-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm0
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: test_mm512_zextsi256_si512:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
|
||||
; X64-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm0
|
||||
; X64-NEXT: retq
|
||||
%res = shufflevector <4 x i64> %a0, <4 x i64> zeroinitializer, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
|
||||
|
||||
@ -960,7 +960,7 @@ define <64 x i8> @test16(i64 %x) {
|
||||
; KNL-NEXT: vinserti128 $1, %xmm2, %ymm1, %ymm1
|
||||
; KNL-NEXT: vpsllw $7, %ymm0, %ymm0
|
||||
; KNL-NEXT: vpand {{.*}}(%rip), %ymm0, %ymm0
|
||||
; KNL-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; KNL-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; KNL-NEXT: vpcmpgtb %ymm0, %ymm2, %ymm0
|
||||
; KNL-NEXT: movq %rbp, %rsp
|
||||
; KNL-NEXT: popq %rbp
|
||||
@ -1030,7 +1030,7 @@ define <64 x i8> @test16(i64 %x) {
|
||||
; AVX512DQ-NEXT: vinserti128 $1, %xmm2, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vpsllw $7, %ymm0, %ymm0
|
||||
; AVX512DQ-NEXT: vpand {{.*}}(%rip), %ymm0, %ymm0
|
||||
; AVX512DQ-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX512DQ-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX512DQ-NEXT: vpcmpgtb %ymm0, %ymm2, %ymm0
|
||||
; AVX512DQ-NEXT: movq %rbp, %rsp
|
||||
; AVX512DQ-NEXT: popq %rbp
|
||||
@ -1078,7 +1078,7 @@ define <64 x i8> @test17(i64 %x, i32 %y, i32 %z) {
|
||||
; KNL-NEXT: vinserti128 $1, %xmm2, %ymm1, %ymm1
|
||||
; KNL-NEXT: vpsllw $7, %ymm0, %ymm0
|
||||
; KNL-NEXT: vpand {{.*}}(%rip), %ymm0, %ymm0
|
||||
; KNL-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; KNL-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; KNL-NEXT: vpcmpgtb %ymm0, %ymm2, %ymm0
|
||||
; KNL-NEXT: movq %rbp, %rsp
|
||||
; KNL-NEXT: popq %rbp
|
||||
@ -1152,7 +1152,7 @@ define <64 x i8> @test17(i64 %x, i32 %y, i32 %z) {
|
||||
; AVX512DQ-NEXT: vinserti128 $1, %xmm2, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vpsllw $7, %ymm0, %ymm0
|
||||
; AVX512DQ-NEXT: vpand {{.*}}(%rip), %ymm0, %ymm0
|
||||
; AVX512DQ-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX512DQ-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX512DQ-NEXT: vpcmpgtb %ymm0, %ymm2, %ymm0
|
||||
; AVX512DQ-NEXT: movq %rbp, %rsp
|
||||
; AVX512DQ-NEXT: popq %rbp
|
||||
|
||||
@ -125,7 +125,7 @@ define <8 x double> @_sd8xdouble_mask(double %a, <8 x double> %i, <8 x i32> %m
|
||||
; ALL-LABEL: _sd8xdouble_mask:
|
||||
; ALL: # BB#0:
|
||||
; ALL-NEXT: # kill: %YMM2<def> %YMM2<kill> %ZMM2<def>
|
||||
; ALL-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; ALL-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; ALL-NEXT: vpcmpneqd %zmm3, %zmm2, %k1
|
||||
; ALL-NEXT: vbroadcastsd %xmm0, %zmm1 {%k1}
|
||||
; ALL-NEXT: vmovapd %zmm1, %zmm0
|
||||
@ -141,7 +141,7 @@ define <8 x double> @_sd8xdouble_maskz(double %a, <8 x i32> %mask1) {
|
||||
; ALL-LABEL: _sd8xdouble_maskz:
|
||||
; ALL: # BB#0:
|
||||
; ALL-NEXT: # kill: %YMM1<def> %YMM1<kill> %ZMM1<def>
|
||||
; ALL-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; ALL-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; ALL-NEXT: vpcmpneqd %zmm2, %zmm1, %k1
|
||||
; ALL-NEXT: vbroadcastsd %xmm0, %zmm0 {%k1} {z}
|
||||
; ALL-NEXT: retq
|
||||
@ -167,7 +167,7 @@ define <8 x double> @_sd8xdouble_mask_load(double* %a.ptr, <8 x double> %i, <8
|
||||
; ALL-LABEL: _sd8xdouble_mask_load:
|
||||
; ALL: # BB#0:
|
||||
; ALL-NEXT: # kill: %YMM1<def> %YMM1<kill> %ZMM1<def>
|
||||
; ALL-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; ALL-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; ALL-NEXT: vpcmpneqd %zmm2, %zmm1, %k1
|
||||
; ALL-NEXT: vbroadcastsd (%rdi), %zmm0 {%k1}
|
||||
; ALL-NEXT: retq
|
||||
@ -183,7 +183,7 @@ define <8 x double> @_sd8xdouble_maskz_load(double* %a.ptr, <8 x i32> %mask1)
|
||||
; ALL-LABEL: _sd8xdouble_maskz_load:
|
||||
; ALL: # BB#0:
|
||||
; ALL-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
|
||||
; ALL-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; ALL-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; ALL-NEXT: vpcmpneqd %zmm1, %zmm0, %k1
|
||||
; ALL-NEXT: vbroadcastsd (%rdi), %zmm0 {%k1} {z}
|
||||
; ALL-NEXT: retq
|
||||
|
||||
@ -356,7 +356,7 @@ define i32 @v32i16(<32 x i16> %a, <32 x i16> %b, <32 x i16> %c, <32 x i16> %d) {
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpsllw $7, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpand {{.*}}(%rip), %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpcmpgtb %ymm0, %ymm1, %ymm0
|
||||
; AVX2-NEXT: vpmovmskb %ymm0, %eax
|
||||
; AVX2-NEXT: vzeroupper
|
||||
|
||||
@ -45,13 +45,13 @@ define <4 x i64> @_clearupper4xi64a(<4 x i64>) nounwind {
|
||||
;
|
||||
; AVX1-LABEL: _clearupper4xi64a:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3],ymm0[4],ymm1[5],ymm0[6],ymm1[7]
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
; AVX2-LABEL: _clearupper4xi64a:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3],ymm0[4],ymm1[5],ymm0[6],ymm1[7]
|
||||
; AVX2-NEXT: retq
|
||||
%x0 = extractelement <4 x i64> %0, i32 0
|
||||
@ -118,7 +118,7 @@ define <8 x i32> @_clearupper8xi32a(<8 x i32>) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: _clearupper8xi32a:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3],ymm0[4],ymm1[5],ymm0[6],ymm1[7],ymm0[8],ymm1[9],ymm0[10],ymm1[11],ymm0[12],ymm1[13],ymm0[14],ymm1[15]
|
||||
; AVX2-NEXT: retq
|
||||
%x0 = extractelement <8 x i32> %0, i32 0
|
||||
@ -666,13 +666,13 @@ define <4 x i64> @_clearupper4xi64b(<4 x i64>) nounwind {
|
||||
;
|
||||
; AVX1-LABEL: _clearupper4xi64b:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3],ymm0[4],ymm1[5],ymm0[6],ymm1[7]
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
; AVX2-LABEL: _clearupper4xi64b:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3],ymm0[4],ymm1[5],ymm0[6],ymm1[7]
|
||||
; AVX2-NEXT: retq
|
||||
%x32 = bitcast <4 x i64> %0 to <8 x i32>
|
||||
@ -719,7 +719,7 @@ define <8 x i32> @_clearupper8xi32b(<8 x i32>) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: _clearupper8xi32b:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3],ymm0[4],ymm1[5],ymm0[6],ymm1[7],ymm0[8],ymm1[9],ymm0[10],ymm1[11],ymm0[12],ymm1[13],ymm0[14],ymm1[15]
|
||||
; AVX2-NEXT: retq
|
||||
%x16 = bitcast <8 x i32> %0 to <16 x i16>
|
||||
@ -1468,13 +1468,13 @@ define <4 x i64> @_clearupper4xi64c(<4 x i64>) nounwind {
|
||||
;
|
||||
; AVX1-LABEL: _clearupper4xi64c:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3],ymm0[4],ymm1[5],ymm0[6],ymm1[7]
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
; AVX2-LABEL: _clearupper4xi64c:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3],ymm0[4],ymm1[5],ymm0[6],ymm1[7]
|
||||
; AVX2-NEXT: retq
|
||||
%r = and <4 x i64> <i64 4294967295, i64 4294967295, i64 4294967295, i64 4294967295>, %0
|
||||
@ -1511,7 +1511,7 @@ define <8 x i32> @_clearupper8xi32c(<8 x i32>) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: _clearupper8xi32c:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3],ymm0[4],ymm1[5],ymm0[6],ymm1[7],ymm0[8],ymm1[9],ymm0[10],ymm1[11],ymm0[12],ymm1[13],ymm0[14],ymm1[15]
|
||||
; AVX2-NEXT: retq
|
||||
%r = and <8 x i32> <i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535>, %0
|
||||
|
||||
@ -50,7 +50,7 @@ define <32 x i8> @combine_v32i8_abs_abs(<32 x i8> %a) {
|
||||
define <4 x i64> @combine_v4i64_abs_abs(<4 x i64> %a) {
|
||||
; AVX2-LABEL: combine_v4i64_abs_abs:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpcmpgtq %ymm0, %ymm1, %ymm2
|
||||
; AVX2-NEXT: vpaddq %ymm2, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %ymm2, %ymm0, %ymm0
|
||||
|
||||
@ -96,7 +96,7 @@ define <4 x i64> @combine_shuffle_zero_pmuludq_256(<8 x i32> %a0, <8 x i32> %a1)
|
||||
;
|
||||
; AVX-LABEL: combine_shuffle_zero_pmuludq_256:
|
||||
; AVX: # BB#0:
|
||||
; AVX-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm2[1],ymm0[2],ymm2[3],ymm0[4],ymm2[5],ymm0[6],ymm2[7]
|
||||
; AVX-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0],ymm2[1],ymm1[2],ymm2[3],ymm1[4],ymm2[5],ymm1[6],ymm2[7]
|
||||
; AVX-NEXT: vpmuludq %ymm1, %ymm0, %ymm0
|
||||
|
||||
@ -346,7 +346,7 @@ define <16 x double> @test16(double* %base, <16 x double> %src0, <16 x i32> %tri
|
||||
;
|
||||
; KNL-LABEL: test16:
|
||||
; KNL: # BB#0:
|
||||
; KNL-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; KNL-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; KNL-NEXT: vextracti64x4 $1, %zmm2, %ymm4
|
||||
; KNL-NEXT: vpcmpeqd %zmm3, %zmm4, %k1
|
||||
; KNL-NEXT: vpcmpeqd %zmm3, %zmm2, %k2
|
||||
|
||||
@ -1517,13 +1517,13 @@ define <4 x float> @test_v4f32_fneg_fmul(<4 x float> %x, <4 x float> %y) #0 {
|
||||
define <4 x double> @test_v4f64_fneg_fmul(<4 x double> %x, <4 x double> %y) #0 {
|
||||
; FMA-LABEL: test_v4f64_fneg_fmul:
|
||||
; FMA: # BB#0:
|
||||
; FMA-NEXT: vxorpd %ymm2, %ymm2, %ymm2
|
||||
; FMA-NEXT: vxorpd %xmm2, %xmm2, %xmm2
|
||||
; FMA-NEXT: vfnmsub213pd %ymm2, %ymm1, %ymm0
|
||||
; FMA-NEXT: retq
|
||||
;
|
||||
; FMA4-LABEL: test_v4f64_fneg_fmul:
|
||||
; FMA4: # BB#0:
|
||||
; FMA4-NEXT: vxorpd %ymm2, %ymm2, %ymm2
|
||||
; FMA4-NEXT: vxorpd %xmm2, %xmm2, %xmm2
|
||||
; FMA4-NEXT: vfnmsubpd %ymm2, %ymm1, %ymm0, %ymm0
|
||||
; FMA4-NEXT: retq
|
||||
;
|
||||
|
||||
@ -1091,14 +1091,14 @@ define <16 x float> @test_v16f32_fma_fmul_x_c1_c2_y(<16 x float> %x, <16 x float
|
||||
define <16 x float> @test_v16f32_fneg_fmul(<16 x float> %x, <16 x float> %y) #0 {
|
||||
; FMA-LABEL: test_v16f32_fneg_fmul:
|
||||
; FMA: # BB#0:
|
||||
; FMA-NEXT: vxorps %ymm4, %ymm4, %ymm4
|
||||
; FMA-NEXT: vxorps %xmm4, %xmm4, %xmm4
|
||||
; FMA-NEXT: vfnmsub213ps %ymm4, %ymm2, %ymm0
|
||||
; FMA-NEXT: vfnmsub213ps %ymm4, %ymm3, %ymm1
|
||||
; FMA-NEXT: retq
|
||||
;
|
||||
; FMA4-LABEL: test_v16f32_fneg_fmul:
|
||||
; FMA4: # BB#0:
|
||||
; FMA4-NEXT: vxorps %ymm4, %ymm4, %ymm4
|
||||
; FMA4-NEXT: vxorps %xmm4, %xmm4, %xmm4
|
||||
; FMA4-NEXT: vfnmsubps %ymm4, %ymm2, %ymm0, %ymm0
|
||||
; FMA4-NEXT: vfnmsubps %ymm4, %ymm3, %ymm1, %ymm1
|
||||
; FMA4-NEXT: retq
|
||||
@ -1116,14 +1116,14 @@ define <16 x float> @test_v16f32_fneg_fmul(<16 x float> %x, <16 x float> %y) #0
|
||||
define <8 x double> @test_v8f64_fneg_fmul(<8 x double> %x, <8 x double> %y) #0 {
|
||||
; FMA-LABEL: test_v8f64_fneg_fmul:
|
||||
; FMA: # BB#0:
|
||||
; FMA-NEXT: vxorpd %ymm4, %ymm4, %ymm4
|
||||
; FMA-NEXT: vxorpd %xmm4, %xmm4, %xmm4
|
||||
; FMA-NEXT: vfnmsub213pd %ymm4, %ymm2, %ymm0
|
||||
; FMA-NEXT: vfnmsub213pd %ymm4, %ymm3, %ymm1
|
||||
; FMA-NEXT: retq
|
||||
;
|
||||
; FMA4-LABEL: test_v8f64_fneg_fmul:
|
||||
; FMA4: # BB#0:
|
||||
; FMA4-NEXT: vxorpd %ymm4, %ymm4, %ymm4
|
||||
; FMA4-NEXT: vxorpd %xmm4, %xmm4, %xmm4
|
||||
; FMA4-NEXT: vfnmsubpd %ymm4, %ymm2, %ymm0, %ymm0
|
||||
; FMA4-NEXT: vfnmsubpd %ymm4, %ymm3, %ymm1, %ymm1
|
||||
; FMA4-NEXT: retq
|
||||
|
||||
@ -32,7 +32,7 @@ define void @PR15298(<4 x float>* nocapture %source, <8 x float>* nocapture %des
|
||||
; AVX-32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; AVX-32-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
||||
; AVX-32-NEXT: vbroadcastss 304(%ecx), %xmm0
|
||||
; AVX-32-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX-32-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX-32-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1,2],ymm1[3,4,5,6,7]
|
||||
; AVX-32-NEXT: vmovups %ymm0, 608(%eax)
|
||||
; AVX-32-NEXT: vzeroupper
|
||||
@ -41,7 +41,7 @@ define void @PR15298(<4 x float>* nocapture %source, <8 x float>* nocapture %des
|
||||
; AVX-64-LABEL: PR15298:
|
||||
; AVX-64: # BB#0: # %L.entry
|
||||
; AVX-64-NEXT: vbroadcastss 304(%rdi), %xmm0
|
||||
; AVX-64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX-64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX-64-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1,2],ymm1[3,4,5,6,7]
|
||||
; AVX-64-NEXT: vmovups %ymm0, 608(%rsi)
|
||||
; AVX-64-NEXT: vzeroupper
|
||||
|
||||
@ -71,7 +71,7 @@ define <4 x double> @insert_v4f64_0zz3(<4 x double> %a) {
|
||||
;
|
||||
; AVX-LABEL: insert_v4f64_0zz3:
|
||||
; AVX: # BB#0:
|
||||
; AVX-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2],ymm0[3]
|
||||
; AVX-NEXT: retq
|
||||
%1 = insertelement <4 x double> %a, double 0.0, i32 1
|
||||
@ -146,13 +146,13 @@ define <4 x i64> @insert_v4i64_01z3(<4 x i64> %a) {
|
||||
;
|
||||
; AVX1-LABEL: insert_v4i64_01z3:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2],ymm0[3]
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
; AVX2-LABEL: insert_v4i64_01z3:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5],ymm0[6,7]
|
||||
; AVX2-NEXT: retq
|
||||
%1 = insertelement <4 x i64> %a, i64 0, i32 2
|
||||
@ -230,7 +230,7 @@ define <8 x float> @insert_v8f32_z12345z7(<8 x float> %a) {
|
||||
;
|
||||
; AVX-LABEL: insert_v8f32_z12345z7:
|
||||
; AVX: # BB#0:
|
||||
; AVX-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1,2,3,4,5],ymm1[6],ymm0[7]
|
||||
; AVX-NEXT: retq
|
||||
%1 = insertelement <8 x float> %a, float 0.0, i32 0
|
||||
@ -318,13 +318,13 @@ define <8 x i32> @insert_v8i32_z12345z7(<8 x i32> %a) {
|
||||
;
|
||||
; AVX1-LABEL: insert_v8i32_z12345z7:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1,2,3,4,5],ymm1[6],ymm0[7]
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
; AVX2-LABEL: insert_v8i32_z12345z7:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0],ymm0[1,2,3,4,5],ymm1[6],ymm0[7]
|
||||
; AVX2-NEXT: retq
|
||||
%1 = insertelement <8 x i32> %a, i32 0, i32 0
|
||||
|
||||
@ -33,7 +33,7 @@ define i32 @_Z10test_shortPsS_i(i16* nocapture readonly, i16* nocapture readonly
|
||||
; AVX2-LABEL: _Z10test_shortPsS_i:
|
||||
; AVX2: # BB#0: # %entry
|
||||
; AVX2-NEXT: movl %edx, %eax
|
||||
; AVX2-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: .p2align 4, 0x90
|
||||
; AVX2-NEXT: .LBB0_1: # %vector.body
|
||||
@ -59,7 +59,7 @@ define i32 @_Z10test_shortPsS_i(i16* nocapture readonly, i16* nocapture readonly
|
||||
; AVX512-LABEL: _Z10test_shortPsS_i:
|
||||
; AVX512: # BB#0: # %entry
|
||||
; AVX512-NEXT: movl %edx, %eax
|
||||
; AVX512-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; AVX512-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; AVX512-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512-NEXT: .p2align 4, 0x90
|
||||
; AVX512-NEXT: .LBB0_1: # %vector.body
|
||||
@ -148,7 +148,7 @@ define i32 @test_unsigned_short(i16* nocapture readonly, i16* nocapture readonly
|
||||
; AVX2-LABEL: test_unsigned_short:
|
||||
; AVX2: # BB#0: # %entry
|
||||
; AVX2-NEXT: movl %edx, %eax
|
||||
; AVX2-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; AVX2-NEXT: .p2align 4, 0x90
|
||||
; AVX2-NEXT: .LBB1_1: # %vector.body
|
||||
; AVX2-NEXT: # =>This Inner Loop Header: Depth=1
|
||||
@ -173,7 +173,7 @@ define i32 @test_unsigned_short(i16* nocapture readonly, i16* nocapture readonly
|
||||
; AVX512-LABEL: test_unsigned_short:
|
||||
; AVX512: # BB#0: # %entry
|
||||
; AVX512-NEXT: movl %edx, %eax
|
||||
; AVX512-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; AVX512-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; AVX512-NEXT: .p2align 4, 0x90
|
||||
; AVX512-NEXT: .LBB1_1: # %vector.body
|
||||
; AVX512-NEXT: # =>This Inner Loop Header: Depth=1
|
||||
@ -281,8 +281,8 @@ define i32 @_Z9test_charPcS_i(i8* nocapture readonly, i8* nocapture readonly, i3
|
||||
; AVX2-LABEL: _Z9test_charPcS_i:
|
||||
; AVX2: # BB#0: # %entry
|
||||
; AVX2-NEXT: movl %edx, %eax
|
||||
; AVX2-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: .p2align 4, 0x90
|
||||
; AVX2-NEXT: .LBB2_1: # %vector.body
|
||||
; AVX2-NEXT: # =>This Inner Loop Header: Depth=1
|
||||
@ -309,7 +309,7 @@ define i32 @_Z9test_charPcS_i(i8* nocapture readonly, i8* nocapture readonly, i3
|
||||
; AVX512: # BB#0: # %entry
|
||||
; AVX512-NEXT: movl %edx, %eax
|
||||
; AVX512-NEXT: vpxord %zmm0, %zmm0, %zmm0
|
||||
; AVX512-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512-NEXT: .p2align 4, 0x90
|
||||
; AVX512-NEXT: .LBB2_1: # %vector.body
|
||||
; AVX512-NEXT: # =>This Inner Loop Header: Depth=1
|
||||
|
||||
@ -782,7 +782,7 @@ define <4 x float> @test15(float* %base, <4 x i32> %ind, <4 x i1> %mask) {
|
||||
; KNL_64: # BB#0:
|
||||
; KNL_64-NEXT: # kill: %XMM1<def> %XMM1<kill> %YMM1<def>
|
||||
; KNL_64-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; KNL_64-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; KNL_64-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; KNL_64-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm2[4,5,6,7]
|
||||
; KNL_64-NEXT: vpmovsxdq %ymm0, %zmm2
|
||||
; KNL_64-NEXT: vpslld $31, %ymm1, %ymm0
|
||||
@ -796,7 +796,7 @@ define <4 x float> @test15(float* %base, <4 x i32> %ind, <4 x i1> %mask) {
|
||||
; KNL_32: # BB#0:
|
||||
; KNL_32-NEXT: # kill: %XMM1<def> %XMM1<kill> %YMM1<def>
|
||||
; KNL_32-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; KNL_32-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; KNL_32-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; KNL_32-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm2[4,5,6,7]
|
||||
; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; KNL_32-NEXT: vpmovsxdq %ymm0, %zmm2
|
||||
@ -951,7 +951,7 @@ define void @test18(<4 x i32>%a1, <4 x i32*> %ptr, <4 x i1>%mask) {
|
||||
; KNL_64-NEXT: # kill: %XMM2<def> %XMM2<kill> %YMM2<def>
|
||||
; KNL_64-NEXT: # kill: %YMM1<def> %YMM1<kill> %ZMM1<def>
|
||||
; KNL_64-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; KNL_64-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; KNL_64-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; KNL_64-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm3[4,5,6,7]
|
||||
; KNL_64-NEXT: vpslld $31, %ymm2, %ymm2
|
||||
; KNL_64-NEXT: vptestmd %zmm2, %zmm2, %k1
|
||||
@ -964,7 +964,7 @@ define void @test18(<4 x i32>%a1, <4 x i32*> %ptr, <4 x i1>%mask) {
|
||||
; KNL_32-NEXT: # kill: %XMM2<def> %XMM2<kill> %YMM2<def>
|
||||
; KNL_32-NEXT: # kill: %XMM1<def> %XMM1<kill> %YMM1<def>
|
||||
; KNL_32-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; KNL_32-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; KNL_32-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; KNL_32-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm3[4,5,6,7]
|
||||
; KNL_32-NEXT: vpmovsxdq %ymm1, %zmm1
|
||||
; KNL_32-NEXT: vpslld $31, %ymm2, %ymm2
|
||||
@ -1051,7 +1051,7 @@ define void @test20(<2 x float>%a1, <2 x float*> %ptr, <2 x i1> %mask) {
|
||||
; KNL_64-NEXT: # kill: %XMM1<def> %XMM1<kill> %ZMM1<def>
|
||||
; KNL_64-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; KNL_64-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,2],zero,zero
|
||||
; KNL_64-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; KNL_64-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; KNL_64-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm3[4,5,6,7]
|
||||
; KNL_64-NEXT: vpslld $31, %ymm2, %ymm2
|
||||
; KNL_64-NEXT: vptestmd %zmm2, %zmm2, %k1
|
||||
@ -1063,7 +1063,7 @@ define void @test20(<2 x float>%a1, <2 x float*> %ptr, <2 x i1> %mask) {
|
||||
; KNL_32: # BB#0:
|
||||
; KNL_32-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; KNL_32-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,2],zero,zero
|
||||
; KNL_32-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; KNL_32-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; KNL_32-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm3[4,5,6,7]
|
||||
; KNL_32-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[0,2,2,3]
|
||||
; KNL_32-NEXT: vpmovsxdq %ymm1, %zmm1
|
||||
@ -1158,7 +1158,7 @@ define <2 x float> @test22(float* %base, <2 x i32> %ind, <2 x i1> %mask, <2 x fl
|
||||
; KNL_64: # BB#0:
|
||||
; KNL_64-NEXT: # kill: %XMM2<def> %XMM2<kill> %YMM2<def>
|
||||
; KNL_64-NEXT: vinsertps {{.*#+}} xmm1 = xmm1[0,2],zero,zero
|
||||
; KNL_64-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; KNL_64-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; KNL_64-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm3[4,5,6,7]
|
||||
; KNL_64-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
|
||||
; KNL_64-NEXT: vpmovsxdq %ymm0, %zmm0
|
||||
@ -1173,7 +1173,7 @@ define <2 x float> @test22(float* %base, <2 x i32> %ind, <2 x i1> %mask, <2 x fl
|
||||
; KNL_32: # BB#0:
|
||||
; KNL_32-NEXT: # kill: %XMM2<def> %XMM2<kill> %YMM2<def>
|
||||
; KNL_32-NEXT: vinsertps {{.*#+}} xmm1 = xmm1[0,2],zero,zero
|
||||
; KNL_32-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; KNL_32-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; KNL_32-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm3[4,5,6,7]
|
||||
; KNL_32-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
|
||||
; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
@ -1219,7 +1219,7 @@ define <2 x float> @test22a(float* %base, <2 x i64> %ind, <2 x i1> %mask, <2 x f
|
||||
; KNL_64-NEXT: # kill: %XMM2<def> %XMM2<kill> %YMM2<def>
|
||||
; KNL_64-NEXT: # kill: %XMM0<def> %XMM0<kill> %ZMM0<def>
|
||||
; KNL_64-NEXT: vinsertps {{.*#+}} xmm1 = xmm1[0,2],zero,zero
|
||||
; KNL_64-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; KNL_64-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; KNL_64-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm3[4,5,6,7]
|
||||
; KNL_64-NEXT: vpslld $31, %ymm1, %ymm1
|
||||
; KNL_64-NEXT: vptestmd %zmm1, %zmm1, %k1
|
||||
@ -1233,7 +1233,7 @@ define <2 x float> @test22a(float* %base, <2 x i64> %ind, <2 x i1> %mask, <2 x f
|
||||
; KNL_32-NEXT: # kill: %XMM2<def> %XMM2<kill> %YMM2<def>
|
||||
; KNL_32-NEXT: # kill: %XMM0<def> %XMM0<kill> %ZMM0<def>
|
||||
; KNL_32-NEXT: vinsertps {{.*#+}} xmm1 = xmm1[0,2],zero,zero
|
||||
; KNL_32-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; KNL_32-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; KNL_32-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm3[4,5,6,7]
|
||||
; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; KNL_32-NEXT: vpslld $31, %ymm1, %ymm1
|
||||
|
||||
@ -228,7 +228,7 @@ define <8 x float> @test11a(<8 x i32> %trigger, <8 x float>* %addr, <8 x float>
|
||||
;
|
||||
; AVX2-LABEL: test11a:
|
||||
; AVX2: ## BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX2-NEXT: vpcmpeqd %ymm2, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vmaskmovps (%rdi), %ymm0, %ymm2
|
||||
; AVX2-NEXT: vblendvps %ymm0, %ymm2, %ymm1, %ymm0
|
||||
@ -238,7 +238,7 @@ define <8 x float> @test11a(<8 x i32> %trigger, <8 x float>* %addr, <8 x float>
|
||||
; AVX512F: ## BB#0:
|
||||
; AVX512F-NEXT: ## kill: %YMM1<def> %YMM1<kill> %ZMM1<def>
|
||||
; AVX512F-NEXT: ## kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
|
||||
; AVX512F-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX512F-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX512F-NEXT: vpcmpeqd %zmm2, %zmm0, %k0
|
||||
; AVX512F-NEXT: kshiftlw $8, %k0, %k0
|
||||
; AVX512F-NEXT: kshiftrw $8, %k0, %k1
|
||||
@ -400,7 +400,7 @@ define void @test12(<8 x i32> %trigger, <8 x i32>* %addr, <8 x i32> %val) {
|
||||
;
|
||||
; AVX2-LABEL: test12:
|
||||
; AVX2: ## BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX2-NEXT: vpcmpeqd %ymm2, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpmaskmovd %ymm1, %ymm0, (%rdi)
|
||||
; AVX2-NEXT: vzeroupper
|
||||
@ -410,7 +410,7 @@ define void @test12(<8 x i32> %trigger, <8 x i32>* %addr, <8 x i32> %val) {
|
||||
; AVX512F: ## BB#0:
|
||||
; AVX512F-NEXT: ## kill: %YMM1<def> %YMM1<kill> %ZMM1<def>
|
||||
; AVX512F-NEXT: ## kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
|
||||
; AVX512F-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX512F-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX512F-NEXT: vpcmpeqd %zmm2, %zmm0, %k0
|
||||
; AVX512F-NEXT: kshiftlw $8, %k0, %k0
|
||||
; AVX512F-NEXT: kshiftrw $8, %k0, %k1
|
||||
|
||||
@ -43,7 +43,7 @@ define void @t() nounwind {
|
||||
; YMM-NEXT: movl %esp, %ebp
|
||||
; YMM-NEXT: andl $-32, %esp
|
||||
; YMM-NEXT: subl $96, %esp
|
||||
; YMM-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; YMM-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; YMM-NEXT: vmovaps %ymm0, {{[0-9]+}}(%esp)
|
||||
; YMM-NEXT: leal {{[0-9]+}}(%esp), %eax
|
||||
; YMM-NEXT: movl %eax, (%esp)
|
||||
|
||||
@ -137,14 +137,14 @@ define <4 x double> @merge_4f64_f64_45zz(double* %ptr) nounwind uwtable noinline
|
||||
define <4 x double> @merge_4f64_f64_34z6(double* %ptr) nounwind uwtable noinline ssp {
|
||||
; AVX-LABEL: merge_4f64_f64_34z6:
|
||||
; AVX: # BB#0:
|
||||
; AVX-NEXT: vxorpd %ymm0, %ymm0, %ymm0
|
||||
; AVX-NEXT: vxorpd %xmm0, %xmm0, %xmm0
|
||||
; AVX-NEXT: vblendpd {{.*#+}} ymm0 = mem[0,1],ymm0[2],mem[3]
|
||||
; AVX-NEXT: retq
|
||||
;
|
||||
; X32-AVX-LABEL: merge_4f64_f64_34z6:
|
||||
; X32-AVX: # BB#0:
|
||||
; X32-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX-NEXT: vxorpd %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vxorpd %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX-NEXT: vblendpd {{.*#+}} ymm0 = mem[0,1],ymm0[2],mem[3]
|
||||
; X32-AVX-NEXT: retl
|
||||
%ptr0 = getelementptr inbounds double, double* %ptr, i64 3
|
||||
@ -278,7 +278,7 @@ define <8 x float> @merge_8f32_2f32_23z5(<2 x float>* %ptr) nounwind uwtable noi
|
||||
; X32-AVX-LABEL: merge_8f32_2f32_23z5:
|
||||
; X32-AVX: # BB#0:
|
||||
; X32-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX-NEXT: vxorpd %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vxorpd %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX-NEXT: vblendpd {{.*#+}} ymm0 = mem[0,1],ymm0[2],mem[3]
|
||||
; X32-AVX-NEXT: retl
|
||||
%ptr0 = getelementptr inbounds <2 x float>, <2 x float>* %ptr, i64 2
|
||||
@ -339,14 +339,14 @@ define <8 x float> @merge_8f32_f32_12zzuuzz(float* %ptr) nounwind uwtable noinli
|
||||
define <8 x float> @merge_8f32_f32_1u3u5zu8(float* %ptr) nounwind uwtable noinline ssp {
|
||||
; AVX-LABEL: merge_8f32_f32_1u3u5zu8:
|
||||
; AVX: # BB#0:
|
||||
; AVX-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; AVX-NEXT: vblendps {{.*#+}} ymm0 = mem[0,1,2,3,4],ymm0[5],mem[6,7]
|
||||
; AVX-NEXT: retq
|
||||
;
|
||||
; X32-AVX-LABEL: merge_8f32_f32_1u3u5zu8:
|
||||
; X32-AVX: # BB#0:
|
||||
; X32-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX-NEXT: vblendps {{.*#+}} ymm0 = mem[0,1,2,3,4],ymm0[5],mem[6,7]
|
||||
; X32-AVX-NEXT: retl
|
||||
%ptr0 = getelementptr inbounds float, float* %ptr, i64 1
|
||||
@ -418,26 +418,26 @@ define <8 x i32> @merge_8i32_i32_56zz9uzz(i32* %ptr) nounwind uwtable noinline s
|
||||
define <8 x i32> @merge_8i32_i32_1u3u5zu8(i32* %ptr) nounwind uwtable noinline ssp {
|
||||
; AVX1-LABEL: merge_8i32_i32_1u3u5zu8:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; AVX1-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; AVX1-NEXT: vblendps {{.*#+}} ymm0 = mem[0,1,2,3,4],ymm0[5],mem[6,7]
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
; AVX2-LABEL: merge_8i32_i32_1u3u5zu8:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = mem[0,1,2,3,4],ymm0[5],mem[6,7]
|
||||
; AVX2-NEXT: retq
|
||||
;
|
||||
; AVX512F-LABEL: merge_8i32_i32_1u3u5zu8:
|
||||
; AVX512F: # BB#0:
|
||||
; AVX512F-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; AVX512F-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; AVX512F-NEXT: vpblendd {{.*#+}} ymm0 = mem[0,1,2,3,4],ymm0[5],mem[6,7]
|
||||
; AVX512F-NEXT: retq
|
||||
;
|
||||
; X32-AVX-LABEL: merge_8i32_i32_1u3u5zu8:
|
||||
; X32-AVX: # BB#0:
|
||||
; X32-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX-NEXT: vblendps {{.*#+}} ymm0 = mem[0,1,2,3,4],ymm0[5],mem[6,7]
|
||||
; X32-AVX-NEXT: retl
|
||||
%ptr0 = getelementptr inbounds i32, i32* %ptr, i64 1
|
||||
|
||||
@ -64,14 +64,14 @@ define <8 x double> @merge_8f64_2f64_23z5(<2 x double>* %ptr) nounwind uwtable n
|
||||
define <8 x double> @merge_8f64_4f64_z2(<4 x double>* %ptr) nounwind uwtable noinline ssp {
|
||||
; ALL-LABEL: merge_8f64_4f64_z2:
|
||||
; ALL: # BB#0:
|
||||
; ALL-NEXT: vxorpd %ymm0, %ymm0, %ymm0
|
||||
; ALL-NEXT: vxorpd %xmm0, %xmm0, %xmm0
|
||||
; ALL-NEXT: vinsertf64x4 $1, 64(%rdi), %zmm0, %zmm0
|
||||
; ALL-NEXT: retq
|
||||
;
|
||||
; X32-AVX512F-LABEL: merge_8f64_4f64_z2:
|
||||
; X32-AVX512F: # BB#0:
|
||||
; X32-AVX512F-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX512F-NEXT: vxorpd %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX512F-NEXT: vxorpd %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX512F-NEXT: vinsertf64x4 $1, 64(%eax), %zmm0, %zmm0
|
||||
; X32-AVX512F-NEXT: retl
|
||||
%ptr1 = getelementptr inbounds <4 x double>, <4 x double>* %ptr, i64 2
|
||||
@ -109,7 +109,7 @@ define <8 x double> @merge_8f64_f64_12zzuuzz(double* %ptr) nounwind uwtable noin
|
||||
; ALL-NEXT: vmovupd 8(%rdi), %xmm0
|
||||
; ALL-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; ALL-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
|
||||
; ALL-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; ALL-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; ALL-NEXT: vinsertf64x4 $1, %ymm1, %zmm0, %zmm0
|
||||
; ALL-NEXT: retq
|
||||
;
|
||||
@ -119,7 +119,7 @@ define <8 x double> @merge_8f64_f64_12zzuuzz(double* %ptr) nounwind uwtable noin
|
||||
; X32-AVX512F-NEXT: vmovupd 8(%eax), %xmm0
|
||||
; X32-AVX512F-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX512F-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
|
||||
; X32-AVX512F-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX512F-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX512F-NEXT: vinsertf64x4 $1, %ymm1, %zmm0, %zmm0
|
||||
; X32-AVX512F-NEXT: retl
|
||||
%ptr0 = getelementptr inbounds double, double* %ptr, i64 1
|
||||
@ -179,14 +179,14 @@ define <8 x double> @merge_8f64_f64_1u3u5zu8(double* %ptr) nounwind uwtable noin
|
||||
define <8 x i64> @merge_8i64_4i64_z3(<4 x i64>* %ptr) nounwind uwtable noinline ssp {
|
||||
; ALL-LABEL: merge_8i64_4i64_z3:
|
||||
; ALL: # BB#0:
|
||||
; ALL-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; ALL-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; ALL-NEXT: vinserti64x4 $1, 96(%rdi), %zmm0, %zmm0
|
||||
; ALL-NEXT: retq
|
||||
;
|
||||
; X32-AVX512F-LABEL: merge_8i64_4i64_z3:
|
||||
; X32-AVX512F: # BB#0:
|
||||
; X32-AVX512F-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX512F-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; X32-AVX512F-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; X32-AVX512F-NEXT: vinserti64x4 $1, 96(%eax), %zmm0, %zmm0
|
||||
; X32-AVX512F-NEXT: retl
|
||||
%ptr1 = getelementptr inbounds <4 x i64>, <4 x i64>* %ptr, i64 3
|
||||
@ -511,7 +511,7 @@ define <32 x i16> @merge_32i16_i16_12u4uuuuuuuuuuuuuuuuuuuuuuuuuuzz(i16* %ptr) n
|
||||
; AVX512F-LABEL: merge_32i16_i16_12u4uuuuuuuuuuuuuuuuuuuuuuuuuuzz:
|
||||
; AVX512F: # BB#0:
|
||||
; AVX512F-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
|
||||
; AVX512F-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX512F-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX512F-NEXT: retq
|
||||
;
|
||||
; AVX512BW-LABEL: merge_32i16_i16_12u4uuuuuuuuuuuuuuuuuuuuuuuuuuzz:
|
||||
@ -523,7 +523,7 @@ define <32 x i16> @merge_32i16_i16_12u4uuuuuuuuuuuuuuuuuuuuuuuuuuzz(i16* %ptr) n
|
||||
; X32-AVX512F: # BB#0:
|
||||
; X32-AVX512F-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX512F-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
|
||||
; X32-AVX512F-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX512F-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX512F-NEXT: retl
|
||||
%ptr0 = getelementptr inbounds i16, i16* %ptr, i64 1
|
||||
%ptr1 = getelementptr inbounds i16, i16* %ptr, i64 2
|
||||
@ -566,7 +566,7 @@ define <32 x i16> @merge_32i16_i16_23uzuuuuuuuuuuzzzzuuuuuuuuuuuuuu(i16* %ptr) n
|
||||
; AVX512F-LABEL: merge_32i16_i16_23uzuuuuuuuuuuzzzzuuuuuuuuuuuuuu:
|
||||
; AVX512F: # BB#0:
|
||||
; AVX512F-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
|
||||
; AVX512F-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX512F-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX512F-NEXT: retq
|
||||
;
|
||||
; AVX512BW-LABEL: merge_32i16_i16_23uzuuuuuuuuuuzzzzuuuuuuuuuuuuuu:
|
||||
@ -578,7 +578,7 @@ define <32 x i16> @merge_32i16_i16_23uzuuuuuuuuuuzzzzuuuuuuuuuuuuuu(i16* %ptr) n
|
||||
; X32-AVX512F: # BB#0:
|
||||
; X32-AVX512F-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX512F-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
|
||||
; X32-AVX512F-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX512F-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX512F-NEXT: retl
|
||||
%ptr0 = getelementptr inbounds i16, i16* %ptr, i64 2
|
||||
%ptr1 = getelementptr inbounds i16, i16* %ptr, i64 3
|
||||
@ -598,7 +598,7 @@ define <64 x i8> @merge_64i8_i8_12u4uuu8uuuuuuzzzzuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu
|
||||
; AVX512F-LABEL: merge_64i8_i8_12u4uuu8uuuuuuzzzzuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuz:
|
||||
; AVX512F: # BB#0:
|
||||
; AVX512F-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
|
||||
; AVX512F-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX512F-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX512F-NEXT: retq
|
||||
;
|
||||
; AVX512BW-LABEL: merge_64i8_i8_12u4uuu8uuuuuuzzzzuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuz:
|
||||
@ -610,7 +610,7 @@ define <64 x i8> @merge_64i8_i8_12u4uuu8uuuuuuzzzzuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu
|
||||
; X32-AVX512F: # BB#0:
|
||||
; X32-AVX512F-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX512F-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
|
||||
; X32-AVX512F-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX512F-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX512F-NEXT: retl
|
||||
%ptr0 = getelementptr inbounds i8, i8* %ptr, i64 1
|
||||
%ptr1 = getelementptr inbounds i8, i8* %ptr, i64 2
|
||||
@ -636,7 +636,7 @@ define <64 x i8> @merge_64i8_i8_12u4uuuuuuuuuuzzzzuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu
|
||||
; AVX512F-LABEL: merge_64i8_i8_12u4uuuuuuuuuuzzzzuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuz:
|
||||
; AVX512F: # BB#0:
|
||||
; AVX512F-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
|
||||
; AVX512F-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX512F-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX512F-NEXT: retq
|
||||
;
|
||||
; AVX512BW-LABEL: merge_64i8_i8_12u4uuuuuuuuuuzzzzuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuz:
|
||||
@ -648,7 +648,7 @@ define <64 x i8> @merge_64i8_i8_12u4uuuuuuuuuuzzzzuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu
|
||||
; X32-AVX512F: # BB#0:
|
||||
; X32-AVX512F-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX512F-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
|
||||
; X32-AVX512F-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX512F-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX512F-NEXT: retl
|
||||
%ptr0 = getelementptr inbounds i8, i8* %ptr, i64 1
|
||||
%ptr1 = getelementptr inbounds i8, i8* %ptr, i64 2
|
||||
|
||||
@ -246,7 +246,7 @@ define void @test_zero_v8f32(<8 x float>* %dst) {
|
||||
;
|
||||
; AVX-LABEL: test_zero_v8f32:
|
||||
; AVX: # BB#0:
|
||||
; AVX-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; AVX-NEXT: vmovntps %ymm0, (%rdi)
|
||||
; AVX-NEXT: vzeroupper
|
||||
; AVX-NEXT: retq
|
||||
@ -271,7 +271,7 @@ define void @test_zero_v8i32(<8 x i32>* %dst) {
|
||||
;
|
||||
; AVX-LABEL: test_zero_v8i32:
|
||||
; AVX: # BB#0:
|
||||
; AVX-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; AVX-NEXT: vmovntps %ymm0, (%rdi)
|
||||
; AVX-NEXT: vzeroupper
|
||||
; AVX-NEXT: retq
|
||||
@ -296,7 +296,7 @@ define void @test_zero_v4f64(<4 x double>* %dst) {
|
||||
;
|
||||
; AVX-LABEL: test_zero_v4f64:
|
||||
; AVX: # BB#0:
|
||||
; AVX-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; AVX-NEXT: vmovntps %ymm0, (%rdi)
|
||||
; AVX-NEXT: vzeroupper
|
||||
; AVX-NEXT: retq
|
||||
@ -321,7 +321,7 @@ define void @test_zero_v4i64(<4 x i64>* %dst) {
|
||||
;
|
||||
; AVX-LABEL: test_zero_v4i64:
|
||||
; AVX: # BB#0:
|
||||
; AVX-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; AVX-NEXT: vmovntps %ymm0, (%rdi)
|
||||
; AVX-NEXT: vzeroupper
|
||||
; AVX-NEXT: retq
|
||||
@ -346,7 +346,7 @@ define void @test_zero_v16i16(<16 x i16>* %dst) {
|
||||
;
|
||||
; AVX-LABEL: test_zero_v16i16:
|
||||
; AVX: # BB#0:
|
||||
; AVX-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; AVX-NEXT: vmovntps %ymm0, (%rdi)
|
||||
; AVX-NEXT: vzeroupper
|
||||
; AVX-NEXT: retq
|
||||
@ -371,7 +371,7 @@ define void @test_zero_v32i8(<32 x i8>* %dst) {
|
||||
;
|
||||
; AVX-LABEL: test_zero_v32i8:
|
||||
; AVX: # BB#0:
|
||||
; AVX-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; AVX-NEXT: vmovntps %ymm0, (%rdi)
|
||||
; AVX-NEXT: vzeroupper
|
||||
; AVX-NEXT: retq
|
||||
|
||||
@ -1886,7 +1886,7 @@ define <16 x i32> @test_masked_v16i32(i8 * %addr, <16 x i32> %old, <16 x i32> %m
|
||||
;
|
||||
; AVX2-LABEL: test_masked_v16i32:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; AVX2-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; AVX2-NEXT: vpcmpeqd %ymm4, %ymm3, %ymm3
|
||||
; AVX2-NEXT: vpcmpeqd %ymm5, %ymm5, %ymm5
|
||||
; AVX2-NEXT: vpxor %ymm5, %ymm3, %ymm3
|
||||
|
||||
@ -35,7 +35,7 @@ define <4 x i32> @trunc_ashr_v4i64(<4 x i64> %a) nounwind {
|
||||
;
|
||||
; X64-AVX2-LABEL: trunc_ashr_v4i64:
|
||||
; X64-AVX2: # BB#0:
|
||||
; X64-AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X64-AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X64-AVX2-NEXT: vpcmpgtq %ymm0, %ymm1, %ymm0
|
||||
; X64-AVX2-NEXT: vextracti128 $1, %ymm0, %xmm1
|
||||
; X64-AVX2-NEXT: vpacksswb %xmm1, %xmm0, %xmm0
|
||||
|
||||
@ -58,8 +58,8 @@ define void @test2(<4 x i8>* %in, <4 x i64>* %out) nounwind {
|
||||
; AVX1-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
|
||||
; AVX1-NEXT: vpmovsxdq %xmm0, %xmm0
|
||||
; AVX1-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm0
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vmovups %ymm1, (%rax)
|
||||
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vmovdqu %ymm1, (%rax)
|
||||
; AVX1-NEXT: vmovups %ymm0, (%rsi)
|
||||
; AVX1-NEXT: vzeroupper
|
||||
; AVX1-NEXT: retq
|
||||
@ -67,7 +67,7 @@ define void @test2(<4 x i8>* %in, <4 x i64>* %out) nounwind {
|
||||
; AVX2-LABEL: test2:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpmovsxbq (%rdi), %ymm0
|
||||
; AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vmovups %ymm1, (%rax)
|
||||
; AVX2-NEXT: vmovdqu %ymm0, (%rsi)
|
||||
; AVX2-NEXT: vzeroupper
|
||||
@ -78,7 +78,7 @@ define void @test2(<4 x i8>* %in, <4 x i64>* %out) nounwind {
|
||||
; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
||||
; X32-AVX2-NEXT: vpmovsxbq (%ecx), %ymm0
|
||||
; X32-AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX2-NEXT: vmovups %ymm1, (%eax)
|
||||
; X32-AVX2-NEXT: vmovdqu %ymm0, (%eax)
|
||||
; X32-AVX2-NEXT: vzeroupper
|
||||
@ -141,8 +141,8 @@ define void @test4(<8 x i8>* %in, <8 x i32>* %out) nounwind {
|
||||
; AVX1-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
|
||||
; AVX1-NEXT: vpmovsxwd %xmm0, %xmm0
|
||||
; AVX1-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm0
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vmovups %ymm1, (%rax)
|
||||
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vmovdqu %ymm1, (%rax)
|
||||
; AVX1-NEXT: vmovups %ymm0, (%rsi)
|
||||
; AVX1-NEXT: vzeroupper
|
||||
; AVX1-NEXT: retq
|
||||
@ -150,7 +150,7 @@ define void @test4(<8 x i8>* %in, <8 x i32>* %out) nounwind {
|
||||
; AVX2-LABEL: test4:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpmovsxbd (%rdi), %ymm0
|
||||
; AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vmovups %ymm1, (%rax)
|
||||
; AVX2-NEXT: vmovdqu %ymm0, (%rsi)
|
||||
; AVX2-NEXT: vzeroupper
|
||||
@ -161,7 +161,7 @@ define void @test4(<8 x i8>* %in, <8 x i32>* %out) nounwind {
|
||||
; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
||||
; X32-AVX2-NEXT: vpmovsxbd (%ecx), %ymm0
|
||||
; X32-AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX2-NEXT: vmovups %ymm1, (%eax)
|
||||
; X32-AVX2-NEXT: vmovdqu %ymm0, (%eax)
|
||||
; X32-AVX2-NEXT: vzeroupper
|
||||
@ -222,8 +222,8 @@ define void @test6(<16 x i8>* %in, <16 x i16>* %out) nounwind {
|
||||
; AVX1-NEXT: vpmovsxbw (%rdi), %xmm0
|
||||
; AVX1-NEXT: vpmovsxbw 8(%rdi), %xmm1
|
||||
; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vmovups %ymm1, (%rax)
|
||||
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vmovdqu %ymm1, (%rax)
|
||||
; AVX1-NEXT: vmovups %ymm0, (%rsi)
|
||||
; AVX1-NEXT: vzeroupper
|
||||
; AVX1-NEXT: retq
|
||||
@ -231,7 +231,7 @@ define void @test6(<16 x i8>* %in, <16 x i16>* %out) nounwind {
|
||||
; AVX2-LABEL: test6:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpmovsxbw (%rdi), %ymm0
|
||||
; AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vmovups %ymm1, (%rax)
|
||||
; AVX2-NEXT: vmovdqu %ymm0, (%rsi)
|
||||
; AVX2-NEXT: vzeroupper
|
||||
@ -242,7 +242,7 @@ define void @test6(<16 x i8>* %in, <16 x i16>* %out) nounwind {
|
||||
; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
||||
; X32-AVX2-NEXT: vpmovsxbw (%ecx), %ymm0
|
||||
; X32-AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX2-NEXT: vmovups %ymm1, (%eax)
|
||||
; X32-AVX2-NEXT: vmovdqu %ymm0, (%eax)
|
||||
; X32-AVX2-NEXT: vzeroupper
|
||||
@ -305,8 +305,8 @@ define void @test8(<4 x i16>* %in, <4 x i64>* %out) nounwind {
|
||||
; AVX1-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
|
||||
; AVX1-NEXT: vpmovsxdq %xmm0, %xmm0
|
||||
; AVX1-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm0
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vmovups %ymm1, (%rax)
|
||||
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vmovdqu %ymm1, (%rax)
|
||||
; AVX1-NEXT: vmovups %ymm0, (%rsi)
|
||||
; AVX1-NEXT: vzeroupper
|
||||
; AVX1-NEXT: retq
|
||||
@ -314,7 +314,7 @@ define void @test8(<4 x i16>* %in, <4 x i64>* %out) nounwind {
|
||||
; AVX2-LABEL: test8:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpmovsxwq (%rdi), %ymm0
|
||||
; AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vmovups %ymm1, (%rax)
|
||||
; AVX2-NEXT: vmovdqu %ymm0, (%rsi)
|
||||
; AVX2-NEXT: vzeroupper
|
||||
@ -325,7 +325,7 @@ define void @test8(<4 x i16>* %in, <4 x i64>* %out) nounwind {
|
||||
; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
||||
; X32-AVX2-NEXT: vpmovsxwq (%ecx), %ymm0
|
||||
; X32-AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX2-NEXT: vmovups %ymm1, (%eax)
|
||||
; X32-AVX2-NEXT: vmovdqu %ymm0, (%eax)
|
||||
; X32-AVX2-NEXT: vzeroupper
|
||||
@ -386,8 +386,8 @@ define void @test10(<8 x i16>* %in, <8 x i32>* %out) nounwind {
|
||||
; AVX1-NEXT: vpmovsxwd (%rdi), %xmm0
|
||||
; AVX1-NEXT: vpmovsxwd 8(%rdi), %xmm1
|
||||
; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vmovups %ymm1, (%rax)
|
||||
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vmovdqu %ymm1, (%rax)
|
||||
; AVX1-NEXT: vmovups %ymm0, (%rsi)
|
||||
; AVX1-NEXT: vzeroupper
|
||||
; AVX1-NEXT: retq
|
||||
@ -395,7 +395,7 @@ define void @test10(<8 x i16>* %in, <8 x i32>* %out) nounwind {
|
||||
; AVX2-LABEL: test10:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpmovsxwd (%rdi), %ymm0
|
||||
; AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vmovups %ymm1, (%rax)
|
||||
; AVX2-NEXT: vmovdqu %ymm0, (%rsi)
|
||||
; AVX2-NEXT: vzeroupper
|
||||
@ -406,7 +406,7 @@ define void @test10(<8 x i16>* %in, <8 x i32>* %out) nounwind {
|
||||
; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
||||
; X32-AVX2-NEXT: vpmovsxwd (%ecx), %ymm0
|
||||
; X32-AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX2-NEXT: vmovups %ymm1, (%eax)
|
||||
; X32-AVX2-NEXT: vmovdqu %ymm0, (%eax)
|
||||
; X32-AVX2-NEXT: vzeroupper
|
||||
@ -467,8 +467,8 @@ define void @test12(<4 x i32>* %in, <4 x i64>* %out) nounwind {
|
||||
; AVX1-NEXT: vpmovsxdq (%rdi), %xmm0
|
||||
; AVX1-NEXT: vpmovsxdq 8(%rdi), %xmm1
|
||||
; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vmovups %ymm1, (%rax)
|
||||
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vmovdqu %ymm1, (%rax)
|
||||
; AVX1-NEXT: vmovups %ymm0, (%rsi)
|
||||
; AVX1-NEXT: vzeroupper
|
||||
; AVX1-NEXT: retq
|
||||
@ -476,7 +476,7 @@ define void @test12(<4 x i32>* %in, <4 x i64>* %out) nounwind {
|
||||
; AVX2-LABEL: test12:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpmovsxdq (%rdi), %ymm0
|
||||
; AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vmovups %ymm1, (%rax)
|
||||
; AVX2-NEXT: vmovdqu %ymm0, (%rsi)
|
||||
; AVX2-NEXT: vzeroupper
|
||||
@ -487,7 +487,7 @@ define void @test12(<4 x i32>* %in, <4 x i64>* %out) nounwind {
|
||||
; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
||||
; X32-AVX2-NEXT: vpmovsxdq (%ecx), %ymm0
|
||||
; X32-AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX2-NEXT: vmovups %ymm1, (%eax)
|
||||
; X32-AVX2-NEXT: vmovdqu %ymm0, (%eax)
|
||||
; X32-AVX2-NEXT: vzeroupper
|
||||
|
||||
@ -5,14 +5,14 @@
|
||||
define <4 x double> @cmp4f64_domain(<4 x double> %a) {
|
||||
; X86-LABEL: cmp4f64_domain:
|
||||
; X86: # BB#0:
|
||||
; X86-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X86-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vaddpd %ymm1, %ymm0, %ymm0
|
||||
; X86-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: cmp4f64_domain:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vaddpd %ymm1, %ymm0, %ymm0
|
||||
; X64-NEXT: retq
|
||||
@ -26,14 +26,14 @@ define <4 x double> @cmp4f64_domain(<4 x double> %a) {
|
||||
define <4 x double> @cmp4f64_domain_optsize(<4 x double> %a) optsize {
|
||||
; X86-LABEL: cmp4f64_domain_optsize:
|
||||
; X86: # BB#0:
|
||||
; X86-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X86-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vaddpd %ymm1, %ymm0, %ymm0
|
||||
; X86-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: cmp4f64_domain_optsize:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vaddpd %ymm1, %ymm0, %ymm0
|
||||
; X64-NEXT: retq
|
||||
@ -47,14 +47,14 @@ define <4 x double> @cmp4f64_domain_optsize(<4 x double> %a) optsize {
|
||||
define <8 x float> @cmp8f32_domain(<8 x float> %a) {
|
||||
; X86-LABEL: cmp8f32_domain:
|
||||
; X86: # BB#0:
|
||||
; X86-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X86-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vaddps %ymm1, %ymm0, %ymm0
|
||||
; X86-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: cmp8f32_domain:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vaddps %ymm1, %ymm0, %ymm0
|
||||
; X64-NEXT: retq
|
||||
@ -68,14 +68,14 @@ define <8 x float> @cmp8f32_domain(<8 x float> %a) {
|
||||
define <8 x float> @cmp8f32_domain_optsize(<8 x float> %a) optsize {
|
||||
; X86-LABEL: cmp8f32_domain_optsize:
|
||||
; X86: # BB#0:
|
||||
; X86-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X86-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; X86-NEXT: vaddps %ymm1, %ymm0, %ymm0
|
||||
; X86-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: cmp8f32_domain_optsize:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vaddps %ymm1, %ymm0, %ymm0
|
||||
; X64-NEXT: retq
|
||||
|
||||
@ -35,9 +35,9 @@ define i32 @sad_16i8() nounwind {
|
||||
;
|
||||
; AVX2-LABEL: sad_16i8:
|
||||
; AVX2: # BB#0: # %entry
|
||||
; AVX2-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; AVX2-NEXT: movq $-1024, %rax # imm = 0xFC00
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: .p2align 4, 0x90
|
||||
; AVX2-NEXT: .LBB0_1: # %vector.body
|
||||
; AVX2-NEXT: # =>This Inner Loop Header: Depth=1
|
||||
@ -281,9 +281,9 @@ define i32 @sad_32i8() nounwind {
|
||||
;
|
||||
; AVX2-LABEL: sad_32i8:
|
||||
; AVX2: # BB#0: # %entry
|
||||
; AVX2-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; AVX2-NEXT: movq $-1024, %rax # imm = 0xFC00
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: .p2align 4, 0x90
|
||||
; AVX2-NEXT: .LBB1_1: # %vector.body
|
||||
; AVX2-NEXT: # =>This Inner Loop Header: Depth=1
|
||||
@ -684,15 +684,15 @@ define i32 @sad_avx64i8() nounwind {
|
||||
;
|
||||
; AVX2-LABEL: sad_avx64i8:
|
||||
; AVX2: # BB#0: # %entry
|
||||
; AVX2-NEXT: vpxor %ymm0, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %xmm0, %xmm0, %xmm0
|
||||
; AVX2-NEXT: movq $-1024, %rax # imm = 0xFC00
|
||||
; AVX2-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; AVX2-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX2-NEXT: vpxor %ymm6, %ymm6, %ymm6
|
||||
; AVX2-NEXT: vpxor %ymm5, %ymm5, %ymm5
|
||||
; AVX2-NEXT: vpxor %ymm7, %ymm7, %ymm7
|
||||
; AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; AVX2-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; AVX2-NEXT: vpxor %xmm6, %xmm6, %xmm6
|
||||
; AVX2-NEXT: vpxor %xmm5, %xmm5, %xmm5
|
||||
; AVX2-NEXT: vpxor %xmm7, %xmm7, %xmm7
|
||||
; AVX2-NEXT: .p2align 4, 0x90
|
||||
; AVX2-NEXT: .LBB2_1: # %vector.body
|
||||
; AVX2-NEXT: # =>This Inner Loop Header: Depth=1
|
||||
|
||||
@ -25,7 +25,7 @@ define <8 x i16> @pr25080(<8 x i32> %a) {
|
||||
; KNL-32: # BB#0: # %entry
|
||||
; KNL-32-NEXT: vpbroadcastd {{.*#+}} ymm1 = [8388607,8388607,8388607,8388607,8388607,8388607,8388607,8388607]
|
||||
; KNL-32-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; KNL-32-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; KNL-32-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; KNL-32-NEXT: vpcmpeqd %zmm1, %zmm0, %k0
|
||||
; KNL-32-NEXT: movb $15, %al
|
||||
; KNL-32-NEXT: kmovw %eax, %k1
|
||||
|
||||
@ -7,7 +7,7 @@ define <4 x i64> @autogen_SD88863() {
|
||||
; X32: # BB#0: # %BB
|
||||
; X32-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
|
||||
; X32-NEXT: vinsertf128 $1, %xmm0, %ymm0, %ymm0
|
||||
; X32-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vblendpd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2],ymm1[3]
|
||||
; X32-NEXT: movb $1, %al
|
||||
; X32-NEXT: .p2align 4, 0x90
|
||||
@ -22,7 +22,7 @@ define <4 x i64> @autogen_SD88863() {
|
||||
; X64: # BB#0: # %BB
|
||||
; X64-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
|
||||
; X64-NEXT: vinsertf128 $1, %xmm0, %ymm0, %ymm0
|
||||
; X64-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vblendpd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2],ymm1[3]
|
||||
; X64-NEXT: movb $1, %al
|
||||
; X64-NEXT: .p2align 4, 0x90
|
||||
|
||||
@ -65,7 +65,7 @@ define void @signum32b(<8 x float>*) {
|
||||
; AVX1-LABEL: signum32b:
|
||||
; AVX1: # BB#0: # %entry
|
||||
; AVX1-NEXT: vmovaps (%rdi), %ymm0
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vcmpltps %ymm1, %ymm0, %ymm2
|
||||
; AVX1-NEXT: vcvtdq2ps %ymm2, %ymm2
|
||||
; AVX1-NEXT: vcmpltps %ymm0, %ymm1, %ymm0
|
||||
@ -78,7 +78,7 @@ define void @signum32b(<8 x float>*) {
|
||||
; AVX2-LABEL: signum32b:
|
||||
; AVX2: # BB#0: # %entry
|
||||
; AVX2-NEXT: vmovaps (%rdi), %ymm0
|
||||
; AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vcmpltps %ymm1, %ymm0, %ymm2
|
||||
; AVX2-NEXT: vcvtdq2ps %ymm2, %ymm2
|
||||
; AVX2-NEXT: vcmpltps %ymm0, %ymm1, %ymm0
|
||||
@ -91,7 +91,7 @@ define void @signum32b(<8 x float>*) {
|
||||
; AVX512F-LABEL: signum32b:
|
||||
; AVX512F: # BB#0: # %entry
|
||||
; AVX512F-NEXT: vmovaps (%rdi), %ymm0
|
||||
; AVX512F-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX512F-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX512F-NEXT: vcmpltps %zmm1, %zmm0, %k1
|
||||
; AVX512F-NEXT: vpternlogq $255, %zmm2, %zmm2, %zmm2 {%k1} {z}
|
||||
; AVX512F-NEXT: vpmovqd %zmm2, %ymm2
|
||||
@ -119,7 +119,7 @@ define void @signum64b(<4 x double>*) {
|
||||
; AVX1-LABEL: signum64b:
|
||||
; AVX1: # BB#0: # %entry
|
||||
; AVX1-NEXT: vmovapd (%rdi), %ymm0
|
||||
; AVX1-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vcmpltpd %ymm1, %ymm0, %ymm2
|
||||
; AVX1-NEXT: vextractf128 $1, %ymm2, %xmm3
|
||||
; AVX1-NEXT: vpacksswb %xmm3, %xmm2, %xmm2
|
||||
@ -136,7 +136,7 @@ define void @signum64b(<4 x double>*) {
|
||||
; AVX2-LABEL: signum64b:
|
||||
; AVX2: # BB#0: # %entry
|
||||
; AVX2-NEXT: vmovapd (%rdi), %ymm0
|
||||
; AVX2-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vcmpltpd %ymm1, %ymm0, %ymm2
|
||||
; AVX2-NEXT: vextractf128 $1, %ymm2, %xmm3
|
||||
; AVX2-NEXT: vpacksswb %xmm3, %xmm2, %xmm2
|
||||
@ -153,7 +153,7 @@ define void @signum64b(<4 x double>*) {
|
||||
; AVX512F-LABEL: signum64b:
|
||||
; AVX512F: # BB#0: # %entry
|
||||
; AVX512F-NEXT: vmovapd (%rdi), %ymm0
|
||||
; AVX512F-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX512F-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX512F-NEXT: vcmpltpd %ymm1, %ymm0, %ymm2
|
||||
; AVX512F-NEXT: vpmovqd %zmm2, %ymm2
|
||||
; AVX512F-NEXT: vcvtdq2pd %xmm2, %ymm2
|
||||
@ -183,7 +183,7 @@ define void @signum32c(<8 x float>*) {
|
||||
; AVX-LABEL: signum32c:
|
||||
; AVX: # BB#0: # %entry
|
||||
; AVX-NEXT: vmovaps (%rdi), %ymm0
|
||||
; AVX-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX-NEXT: vcmpltps %ymm1, %ymm0, %ymm2
|
||||
; AVX-NEXT: vcvtdq2ps %ymm2, %ymm2
|
||||
; AVX-NEXT: vcmpltps %ymm0, %ymm1, %ymm0
|
||||
@ -209,7 +209,7 @@ define void @signum64c(<4 x double>*) {
|
||||
; AVX1-LABEL: signum64c:
|
||||
; AVX1: # BB#0: # %entry
|
||||
; AVX1-NEXT: vmovapd (%rdi), %ymm0
|
||||
; AVX1-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vcmpltpd %ymm1, %ymm0, %ymm2
|
||||
; AVX1-NEXT: vcmpltpd %ymm0, %ymm1, %ymm0
|
||||
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
|
||||
@ -225,7 +225,7 @@ define void @signum64c(<4 x double>*) {
|
||||
; AVX2-LABEL: signum64c:
|
||||
; AVX2: # BB#0: # %entry
|
||||
; AVX2-NEXT: vmovapd (%rdi), %ymm0
|
||||
; AVX2-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vcmpltpd %ymm1, %ymm0, %ymm2
|
||||
; AVX2-NEXT: vcmpltpd %ymm0, %ymm1, %ymm0
|
||||
; AVX2-NEXT: vpsubd %ymm0, %ymm2, %ymm0
|
||||
@ -239,7 +239,7 @@ define void @signum64c(<4 x double>*) {
|
||||
; AVX512F-LABEL: signum64c:
|
||||
; AVX512F: # BB#0: # %entry
|
||||
; AVX512F-NEXT: vmovapd (%rdi), %ymm0
|
||||
; AVX512F-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX512F-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX512F-NEXT: vcmpltpd %ymm1, %ymm0, %ymm2
|
||||
; AVX512F-NEXT: vcmpltpd %ymm0, %ymm1, %ymm0
|
||||
; AVX512F-NEXT: vpsubd %ymm0, %ymm2, %ymm0
|
||||
|
||||
@ -74,7 +74,7 @@ define <8 x i32> @d(<8 x i32> %a) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: d:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3],ymm0[4],ymm1[5],ymm0[6],ymm1[7],ymm0[8],ymm1[9],ymm0[10],ymm1[11],ymm0[12],ymm1[13],ymm0[14],ymm1[15]
|
||||
; AVX2-NEXT: retq
|
||||
%b = trunc <8 x i32> %a to <8 x i16>
|
||||
|
||||
@ -118,7 +118,7 @@ define void @legal_vzmovl_2i32_8i32(<2 x i32>* %in, <8 x i32>* %out) {
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
||||
; X32-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
|
||||
; X32-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3,4,5,6,7]
|
||||
; X32-NEXT: vmovaps %ymm0, (%eax)
|
||||
; X32-NEXT: vzeroupper
|
||||
@ -127,7 +127,7 @@ define void @legal_vzmovl_2i32_8i32(<2 x i32>* %in, <8 x i32>* %out) {
|
||||
; X64-LABEL: legal_vzmovl_2i32_8i32:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3,4,5,6,7]
|
||||
; X64-NEXT: vmovaps %ymm0, (%rsi)
|
||||
; X64-NEXT: vzeroupper
|
||||
@ -145,7 +145,7 @@ define void @legal_vzmovl_2i64_4i64(<2 x i64>* %in, <4 x i64>* %out) {
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
||||
; X32-NEXT: vmovupd (%ecx), %xmm0
|
||||
; X32-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3]
|
||||
; X32-NEXT: vmovapd %ymm0, (%eax)
|
||||
; X32-NEXT: vzeroupper
|
||||
@ -154,7 +154,7 @@ define void @legal_vzmovl_2i64_4i64(<2 x i64>* %in, <4 x i64>* %out) {
|
||||
; X64-LABEL: legal_vzmovl_2i64_4i64:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vmovupd (%rdi), %xmm0
|
||||
; X64-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3]
|
||||
; X64-NEXT: vmovapd %ymm0, (%rsi)
|
||||
; X64-NEXT: vzeroupper
|
||||
@ -179,7 +179,7 @@ define void @legal_vzmovl_2f32_8f32(<2 x float>* %in, <8 x float>* %out) {
|
||||
; X64-LABEL: legal_vzmovl_2f32_8f32:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
|
||||
; X64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3,4,5,6,7]
|
||||
; X64-NEXT: vmovaps %ymm0, (%rsi)
|
||||
; X64-NEXT: vzeroupper
|
||||
@ -197,7 +197,7 @@ define void @legal_vzmovl_2f64_4f64(<2 x double>* %in, <4 x double>* %out) {
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
|
||||
; X32-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
||||
; X32-NEXT: vmovupd (%ecx), %xmm0
|
||||
; X32-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3]
|
||||
; X32-NEXT: vmovapd %ymm0, (%eax)
|
||||
; X32-NEXT: vzeroupper
|
||||
@ -206,7 +206,7 @@ define void @legal_vzmovl_2f64_4f64(<2 x double>* %in, <4 x double>* %out) {
|
||||
; X64-LABEL: legal_vzmovl_2f64_4f64:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vmovupd (%rdi), %xmm0
|
||||
; X64-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3]
|
||||
; X64-NEXT: vmovapd %ymm0, (%rsi)
|
||||
; X64-NEXT: vzeroupper
|
||||
|
||||
@ -163,7 +163,7 @@ define <8 x float> @test_uitofp_v8i32_to_v8f32(<8 x i32> %arg) {
|
||||
; AVX2-NEXT: vcvtdq2ps %ymm1, %ymm1
|
||||
; AVX2-NEXT: vbroadcastss [[FPMASKCSTADDR_v8]](%rip), %ymm2
|
||||
; AVX2-NEXT: vmulps %ymm2, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vxorps %ymm2, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vxorps %xmm2, %xmm2, %xmm2
|
||||
; AVX2-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm2[1],ymm0[2],ymm2[3],ymm0[4],ymm2[5],ymm0[6],ymm2[7],ymm0[8],ymm2[9],ymm0[10],ymm2[11],ymm0[12],ymm2[13],ymm0[14],ymm2[15]
|
||||
; AVX2-NEXT: vcvtdq2ps %ymm0, %ymm0
|
||||
; AVX2-NEXT: vaddps %ymm0, %ymm1, %ymm0
|
||||
|
||||
@ -5903,7 +5903,7 @@ define <64 x i1> @test_cmp_v64i16(<64 x i16> %a0, <64 x i16> %a1) nounwind {
|
||||
; AVX512F-NEXT: vpsllw $7, %ymm2, %ymm2
|
||||
; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
|
||||
; AVX512F-NEXT: vpand %ymm3, %ymm2, %ymm2
|
||||
; AVX512F-NEXT: vpxor %ymm6, %ymm6, %ymm6
|
||||
; AVX512F-NEXT: vpxor %xmm6, %xmm6, %xmm6
|
||||
; AVX512F-NEXT: vpcmpgtb %ymm2, %ymm6, %ymm2
|
||||
; AVX512F-NEXT: vpcmpgtw %ymm5, %ymm1, %ymm1
|
||||
; AVX512F-NEXT: vpmovsxwd %ymm1, %zmm1
|
||||
@ -6190,7 +6190,7 @@ define <64 x i1> @test_cmp_v64i16(<64 x i16> %a0, <64 x i16> %a1) nounwind {
|
||||
; AVX512DQ-NEXT: vpsllw $7, %ymm2, %ymm2
|
||||
; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm3 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
|
||||
; AVX512DQ-NEXT: vpand %ymm3, %ymm2, %ymm2
|
||||
; AVX512DQ-NEXT: vpxor %ymm6, %ymm6, %ymm6
|
||||
; AVX512DQ-NEXT: vpxor %xmm6, %xmm6, %xmm6
|
||||
; AVX512DQ-NEXT: vpcmpgtb %ymm2, %ymm6, %ymm2
|
||||
; AVX512DQ-NEXT: vpcmpgtw %ymm5, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vpmovsxwd %ymm1, %zmm1
|
||||
|
||||
@ -72,7 +72,7 @@ define i64 @extract_any_extend_vector_inreg_v16i64(<16 x i64> %a0, i32 %a1) noun
|
||||
; X32-AVX-NEXT: subl $384, %esp # imm = 0x180
|
||||
; X32-AVX-NEXT: movl 40(%ebp), %ecx
|
||||
; X32-AVX-NEXT: vbroadcastsd 32(%ebp), %ymm0
|
||||
; X32-AVX-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3]
|
||||
; X32-AVX-NEXT: vmovapd %ymm1, {{[0-9]+}}(%esp)
|
||||
; X32-AVX-NEXT: vmovapd %ymm1, {{[0-9]+}}(%esp)
|
||||
@ -101,7 +101,7 @@ define i64 @extract_any_extend_vector_inreg_v16i64(<16 x i64> %a0, i32 %a1) noun
|
||||
; X64-AVX-NEXT: subq $256, %rsp # imm = 0x100
|
||||
; X64-AVX-NEXT: # kill: %EDI<def> %EDI<kill> %RDI<def>
|
||||
; X64-AVX-NEXT: vpermq {{.*#+}} ymm0 = ymm3[3,1,2,3]
|
||||
; X64-AVX-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X64-AVX-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X64-AVX-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3]
|
||||
; X64-AVX-NEXT: vmovapd %ymm1, {{[0-9]+}}(%rsp)
|
||||
; X64-AVX-NEXT: vmovapd %ymm1, {{[0-9]+}}(%rsp)
|
||||
|
||||
@ -73,7 +73,7 @@ define <4 x i64> @testv4i64(<4 x i64> %in) nounwind {
|
||||
; AVX2-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; AVX2-NEXT: vpsrlw $4, %ymm0, %ymm4
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm4, %ymm1
|
||||
; AVX2-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; AVX2-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; AVX2-NEXT: vpcmpeqb %ymm4, %ymm1, %ymm5
|
||||
; AVX2-NEXT: vpand %ymm5, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpshufb %ymm1, %ymm3, %ymm1
|
||||
@ -175,7 +175,7 @@ define <4 x i64> @testv4i64(<4 x i64> %in) nounwind {
|
||||
; X32-AVX-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; X32-AVX-NEXT: vpsrlw $4, %ymm0, %ymm4
|
||||
; X32-AVX-NEXT: vpand %ymm1, %ymm4, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; X32-AVX-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; X32-AVX-NEXT: vpcmpeqb %ymm4, %ymm1, %ymm5
|
||||
; X32-AVX-NEXT: vpand %ymm5, %ymm2, %ymm2
|
||||
; X32-AVX-NEXT: vpshufb %ymm1, %ymm3, %ymm1
|
||||
@ -265,7 +265,7 @@ define <4 x i64> @testv4i64u(<4 x i64> %in) nounwind {
|
||||
; AVX2-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; AVX2-NEXT: vpsrlw $4, %ymm0, %ymm4
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm4, %ymm1
|
||||
; AVX2-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; AVX2-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; AVX2-NEXT: vpcmpeqb %ymm4, %ymm1, %ymm5
|
||||
; AVX2-NEXT: vpand %ymm5, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpshufb %ymm1, %ymm3, %ymm1
|
||||
@ -367,7 +367,7 @@ define <4 x i64> @testv4i64u(<4 x i64> %in) nounwind {
|
||||
; X32-AVX-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; X32-AVX-NEXT: vpsrlw $4, %ymm0, %ymm4
|
||||
; X32-AVX-NEXT: vpand %ymm1, %ymm4, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; X32-AVX-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; X32-AVX-NEXT: vpcmpeqb %ymm4, %ymm1, %ymm5
|
||||
; X32-AVX-NEXT: vpand %ymm5, %ymm2, %ymm2
|
||||
; X32-AVX-NEXT: vpshufb %ymm1, %ymm3, %ymm1
|
||||
@ -447,7 +447,7 @@ define <8 x i32> @testv8i32(<8 x i32> %in) nounwind {
|
||||
; AVX2-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; AVX2-NEXT: vpsrlw $4, %ymm0, %ymm4
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm4, %ymm1
|
||||
; AVX2-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; AVX2-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; AVX2-NEXT: vpcmpeqb %ymm4, %ymm1, %ymm5
|
||||
; AVX2-NEXT: vpand %ymm5, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpshufb %ymm1, %ymm3, %ymm1
|
||||
@ -534,7 +534,7 @@ define <8 x i32> @testv8i32(<8 x i32> %in) nounwind {
|
||||
; X32-AVX-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; X32-AVX-NEXT: vpsrlw $4, %ymm0, %ymm4
|
||||
; X32-AVX-NEXT: vpand %ymm1, %ymm4, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; X32-AVX-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; X32-AVX-NEXT: vpcmpeqb %ymm4, %ymm1, %ymm5
|
||||
; X32-AVX-NEXT: vpand %ymm5, %ymm2, %ymm2
|
||||
; X32-AVX-NEXT: vpshufb %ymm1, %ymm3, %ymm1
|
||||
@ -609,7 +609,7 @@ define <8 x i32> @testv8i32u(<8 x i32> %in) nounwind {
|
||||
; AVX2-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; AVX2-NEXT: vpsrlw $4, %ymm0, %ymm4
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm4, %ymm1
|
||||
; AVX2-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; AVX2-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; AVX2-NEXT: vpcmpeqb %ymm4, %ymm1, %ymm5
|
||||
; AVX2-NEXT: vpand %ymm5, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpshufb %ymm1, %ymm3, %ymm1
|
||||
@ -696,7 +696,7 @@ define <8 x i32> @testv8i32u(<8 x i32> %in) nounwind {
|
||||
; X32-AVX-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; X32-AVX-NEXT: vpsrlw $4, %ymm0, %ymm4
|
||||
; X32-AVX-NEXT: vpand %ymm1, %ymm4, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; X32-AVX-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; X32-AVX-NEXT: vpcmpeqb %ymm4, %ymm1, %ymm5
|
||||
; X32-AVX-NEXT: vpand %ymm5, %ymm2, %ymm2
|
||||
; X32-AVX-NEXT: vpshufb %ymm1, %ymm3, %ymm1
|
||||
@ -761,7 +761,7 @@ define <16 x i16> @testv16i16(<16 x i16> %in) nounwind {
|
||||
; AVX2-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; AVX2-NEXT: vpsrlw $4, %ymm0, %ymm4
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm4, %ymm1
|
||||
; AVX2-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; AVX2-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; AVX2-NEXT: vpcmpeqb %ymm4, %ymm1, %ymm5
|
||||
; AVX2-NEXT: vpand %ymm5, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpshufb %ymm1, %ymm3, %ymm1
|
||||
@ -829,7 +829,7 @@ define <16 x i16> @testv16i16(<16 x i16> %in) nounwind {
|
||||
; X32-AVX-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; X32-AVX-NEXT: vpsrlw $4, %ymm0, %ymm4
|
||||
; X32-AVX-NEXT: vpand %ymm1, %ymm4, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; X32-AVX-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; X32-AVX-NEXT: vpcmpeqb %ymm4, %ymm1, %ymm5
|
||||
; X32-AVX-NEXT: vpand %ymm5, %ymm2, %ymm2
|
||||
; X32-AVX-NEXT: vpshufb %ymm1, %ymm3, %ymm1
|
||||
@ -888,7 +888,7 @@ define <16 x i16> @testv16i16u(<16 x i16> %in) nounwind {
|
||||
; AVX2-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; AVX2-NEXT: vpsrlw $4, %ymm0, %ymm4
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm4, %ymm1
|
||||
; AVX2-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; AVX2-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; AVX2-NEXT: vpcmpeqb %ymm4, %ymm1, %ymm5
|
||||
; AVX2-NEXT: vpand %ymm5, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpshufb %ymm1, %ymm3, %ymm1
|
||||
@ -956,7 +956,7 @@ define <16 x i16> @testv16i16u(<16 x i16> %in) nounwind {
|
||||
; X32-AVX-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; X32-AVX-NEXT: vpsrlw $4, %ymm0, %ymm4
|
||||
; X32-AVX-NEXT: vpand %ymm1, %ymm4, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %ymm4, %ymm4, %ymm4
|
||||
; X32-AVX-NEXT: vpxor %xmm4, %xmm4, %xmm4
|
||||
; X32-AVX-NEXT: vpcmpeqb %ymm4, %ymm1, %ymm5
|
||||
; X32-AVX-NEXT: vpand %ymm5, %ymm2, %ymm2
|
||||
; X32-AVX-NEXT: vpshufb %ymm1, %ymm3, %ymm1
|
||||
@ -1005,7 +1005,7 @@ define <32 x i8> @testv32i8(<32 x i8> %in) nounwind {
|
||||
; AVX2-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; AVX2-NEXT: vpsrlw $4, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpcmpeqb %ymm1, %ymm0, %ymm1
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm2, %ymm1
|
||||
; AVX2-NEXT: vpshufb %ymm0, %ymm3, %ymm0
|
||||
@ -1065,7 +1065,7 @@ define <32 x i8> @testv32i8(<32 x i8> %in) nounwind {
|
||||
; X32-AVX-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; X32-AVX-NEXT: vpsrlw $4, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX-NEXT: vpcmpeqb %ymm1, %ymm0, %ymm1
|
||||
; X32-AVX-NEXT: vpand %ymm1, %ymm2, %ymm1
|
||||
; X32-AVX-NEXT: vpshufb %ymm0, %ymm3, %ymm0
|
||||
@ -1109,7 +1109,7 @@ define <32 x i8> @testv32i8u(<32 x i8> %in) nounwind {
|
||||
; AVX2-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; AVX2-NEXT: vpsrlw $4, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpcmpeqb %ymm1, %ymm0, %ymm1
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm2, %ymm1
|
||||
; AVX2-NEXT: vpshufb %ymm0, %ymm3, %ymm0
|
||||
@ -1169,7 +1169,7 @@ define <32 x i8> @testv32i8u(<32 x i8> %in) nounwind {
|
||||
; X32-AVX-NEXT: vpshufb %ymm2, %ymm3, %ymm2
|
||||
; X32-AVX-NEXT: vpsrlw $4, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX-NEXT: vpcmpeqb %ymm1, %ymm0, %ymm1
|
||||
; X32-AVX-NEXT: vpand %ymm1, %ymm2, %ymm1
|
||||
; X32-AVX-NEXT: vpshufb %ymm0, %ymm3, %ymm0
|
||||
|
||||
@ -68,7 +68,7 @@ define <8 x i64> @testv8i64(<8 x i64> %in) nounwind {
|
||||
; AVX512DQ-NEXT: vpand %ymm2, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vpshufb %ymm1, %ymm4, %ymm1
|
||||
; AVX512DQ-NEXT: vpaddb %ymm3, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX512DQ-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; AVX512DQ-NEXT: vpsadbw %ymm3, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vpand %ymm2, %ymm0, %ymm5
|
||||
; AVX512DQ-NEXT: vpshufb %ymm5, %ymm4, %ymm5
|
||||
@ -147,7 +147,7 @@ define <8 x i64> @testv8i64u(<8 x i64> %in) nounwind {
|
||||
; AVX512DQ-NEXT: vpand %ymm2, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vpshufb %ymm1, %ymm4, %ymm1
|
||||
; AVX512DQ-NEXT: vpaddb %ymm3, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX512DQ-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; AVX512DQ-NEXT: vpsadbw %ymm3, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vpand %ymm2, %ymm0, %ymm5
|
||||
; AVX512DQ-NEXT: vpshufb %ymm5, %ymm4, %ymm5
|
||||
@ -226,7 +226,7 @@ define <16 x i32> @testv16i32(<16 x i32> %in) nounwind {
|
||||
; AVX512DQ-NEXT: vpand %ymm2, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vpshufb %ymm1, %ymm4, %ymm1
|
||||
; AVX512DQ-NEXT: vpaddb %ymm3, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX512DQ-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; AVX512DQ-NEXT: vpunpckhdq {{.*#+}} ymm5 = ymm1[2],ymm3[2],ymm1[3],ymm3[3],ymm1[6],ymm3[6],ymm1[7],ymm3[7]
|
||||
; AVX512DQ-NEXT: vpsadbw %ymm3, %ymm5, %ymm5
|
||||
; AVX512DQ-NEXT: vpunpckldq {{.*#+}} ymm1 = ymm1[0],ymm3[0],ymm1[1],ymm3[1],ymm1[4],ymm3[4],ymm1[5],ymm3[5]
|
||||
@ -313,7 +313,7 @@ define <16 x i32> @testv16i32u(<16 x i32> %in) nounwind {
|
||||
; AVX512DQ-NEXT: vpand %ymm2, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vpshufb %ymm1, %ymm4, %ymm1
|
||||
; AVX512DQ-NEXT: vpaddb %ymm3, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX512DQ-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; AVX512DQ-NEXT: vpunpckhdq {{.*#+}} ymm5 = ymm1[2],ymm3[2],ymm1[3],ymm3[3],ymm1[6],ymm3[6],ymm1[7],ymm3[7]
|
||||
; AVX512DQ-NEXT: vpsadbw %ymm3, %ymm5, %ymm5
|
||||
; AVX512DQ-NEXT: vpunpckldq {{.*#+}} ymm1 = ymm1[0],ymm3[0],ymm1[1],ymm3[1],ymm1[4],ymm3[4],ymm1[5],ymm3[5]
|
||||
@ -398,7 +398,7 @@ define <32 x i16> @testv32i16(<32 x i16> %in) nounwind {
|
||||
; AVX512DQ-NEXT: vpshufb %ymm3, %ymm4, %ymm3
|
||||
; AVX512DQ-NEXT: vpsrlw $4, %ymm0, %ymm5
|
||||
; AVX512DQ-NEXT: vpand %ymm2, %ymm5, %ymm5
|
||||
; AVX512DQ-NEXT: vpxor %ymm6, %ymm6, %ymm6
|
||||
; AVX512DQ-NEXT: vpxor %xmm6, %xmm6, %xmm6
|
||||
; AVX512DQ-NEXT: vpcmpeqb %ymm6, %ymm5, %ymm7
|
||||
; AVX512DQ-NEXT: vpand %ymm7, %ymm3, %ymm3
|
||||
; AVX512DQ-NEXT: vpshufb %ymm5, %ymm4, %ymm5
|
||||
@ -488,7 +488,7 @@ define <32 x i16> @testv32i16u(<32 x i16> %in) nounwind {
|
||||
; AVX512DQ-NEXT: vpshufb %ymm3, %ymm4, %ymm3
|
||||
; AVX512DQ-NEXT: vpsrlw $4, %ymm0, %ymm5
|
||||
; AVX512DQ-NEXT: vpand %ymm2, %ymm5, %ymm5
|
||||
; AVX512DQ-NEXT: vpxor %ymm6, %ymm6, %ymm6
|
||||
; AVX512DQ-NEXT: vpxor %xmm6, %xmm6, %xmm6
|
||||
; AVX512DQ-NEXT: vpcmpeqb %ymm6, %ymm5, %ymm7
|
||||
; AVX512DQ-NEXT: vpand %ymm7, %ymm3, %ymm3
|
||||
; AVX512DQ-NEXT: vpshufb %ymm5, %ymm4, %ymm5
|
||||
@ -600,7 +600,7 @@ define <64 x i8> @testv64i8(<64 x i8> %in) nounwind {
|
||||
; AVX512DQ-NEXT: vpshufb %ymm3, %ymm4, %ymm3
|
||||
; AVX512DQ-NEXT: vpsrlw $4, %ymm0, %ymm0
|
||||
; AVX512DQ-NEXT: vpand %ymm2, %ymm0, %ymm0
|
||||
; AVX512DQ-NEXT: vpxor %ymm5, %ymm5, %ymm5
|
||||
; AVX512DQ-NEXT: vpxor %xmm5, %xmm5, %xmm5
|
||||
; AVX512DQ-NEXT: vpcmpeqb %ymm5, %ymm0, %ymm6
|
||||
; AVX512DQ-NEXT: vpand %ymm6, %ymm3, %ymm3
|
||||
; AVX512DQ-NEXT: vpshufb %ymm0, %ymm4, %ymm0
|
||||
@ -702,7 +702,7 @@ define <64 x i8> @testv64i8u(<64 x i8> %in) nounwind {
|
||||
; AVX512DQ-NEXT: vpshufb %ymm3, %ymm4, %ymm3
|
||||
; AVX512DQ-NEXT: vpsrlw $4, %ymm0, %ymm0
|
||||
; AVX512DQ-NEXT: vpand %ymm2, %ymm0, %ymm0
|
||||
; AVX512DQ-NEXT: vpxor %ymm5, %ymm5, %ymm5
|
||||
; AVX512DQ-NEXT: vpxor %xmm5, %xmm5, %xmm5
|
||||
; AVX512DQ-NEXT: vpcmpeqb %ymm5, %ymm0, %ymm6
|
||||
; AVX512DQ-NEXT: vpand %ymm6, %ymm3, %ymm3
|
||||
; AVX512DQ-NEXT: vpshufb %ymm0, %ymm4, %ymm0
|
||||
|
||||
@ -148,7 +148,7 @@ define <32 x i8> @test_pcmpgtb_256(<32 x i8> %x) {
|
||||
; AVX1-NEXT: vpcmpgtb %xmm1, %xmm2, %xmm1
|
||||
; AVX1-NEXT: vpcmpgtb %xmm0, %xmm2, %xmm0
|
||||
; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm0, %ymm0
|
||||
; AVX1-NEXT: retq
|
||||
@ -177,7 +177,7 @@ define <16 x i16> @test_pcmpgtw_256(<16 x i16> %x) {
|
||||
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
|
||||
; AVX1-NEXT: vpsraw $15, %xmm0, %xmm0
|
||||
; AVX1-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm0
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm0, %ymm0
|
||||
; AVX1-NEXT: retq
|
||||
@ -206,7 +206,7 @@ define <8 x i32> @test_pcmpgtd_256(<8 x i32> %x) {
|
||||
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
|
||||
; AVX1-NEXT: vpsrad $31, %xmm0, %xmm0
|
||||
; AVX1-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm0
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm0, %ymm0
|
||||
; AVX1-NEXT: retq
|
||||
@ -247,7 +247,7 @@ define <4 x i64> @test_pcmpgtq_256(<4 x i64> %x) {
|
||||
; AVX1-NEXT: vpcmpgtq %xmm1, %xmm2, %xmm1
|
||||
; AVX1-NEXT: vpcmpgtq %xmm0, %xmm2, %xmm0
|
||||
; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm0, %ymm0
|
||||
; AVX1-NEXT: retq
|
||||
|
||||
@ -37,7 +37,7 @@ define <4 x i64> @testv4i64(<4 x i64> %in) nounwind {
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpshufb %ymm0, %ymm3, %ymm0
|
||||
; AVX2-NEXT: vpaddb %ymm2, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpsadbw %ymm1, %ymm0, %ymm0
|
||||
; AVX2-NEXT: retq
|
||||
;
|
||||
@ -93,7 +93,7 @@ define <8 x i32> @testv8i32(<8 x i32> %in) nounwind {
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpshufb %ymm0, %ymm3, %ymm0
|
||||
; AVX2-NEXT: vpaddb %ymm2, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpunpckhdq {{.*#+}} ymm2 = ymm0[2],ymm1[2],ymm0[3],ymm1[3],ymm0[6],ymm1[6],ymm0[7],ymm1[7]
|
||||
; AVX2-NEXT: vpsadbw %ymm1, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpunpckldq {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[1],ymm1[1],ymm0[4],ymm1[4],ymm0[5],ymm1[5]
|
||||
|
||||
@ -16,7 +16,7 @@ define <8 x i64> @testv8i64(<8 x i64> %in) nounwind {
|
||||
; AVX512F-NEXT: vpand %ymm2, %ymm1, %ymm1
|
||||
; AVX512F-NEXT: vpshufb %ymm1, %ymm4, %ymm1
|
||||
; AVX512F-NEXT: vpaddb %ymm3, %ymm1, %ymm1
|
||||
; AVX512F-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX512F-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; AVX512F-NEXT: vpsadbw %ymm3, %ymm1, %ymm1
|
||||
; AVX512F-NEXT: vpand %ymm2, %ymm0, %ymm5
|
||||
; AVX512F-NEXT: vpshufb %ymm5, %ymm4, %ymm5
|
||||
@ -62,7 +62,7 @@ define <16 x i32> @testv16i32(<16 x i32> %in) nounwind {
|
||||
; AVX512F-NEXT: vpand %ymm2, %ymm1, %ymm1
|
||||
; AVX512F-NEXT: vpshufb %ymm1, %ymm4, %ymm1
|
||||
; AVX512F-NEXT: vpaddb %ymm3, %ymm1, %ymm1
|
||||
; AVX512F-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX512F-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; AVX512F-NEXT: vpunpckhdq {{.*#+}} ymm5 = ymm1[2],ymm3[2],ymm1[3],ymm3[3],ymm1[6],ymm3[6],ymm1[7],ymm3[7]
|
||||
; AVX512F-NEXT: vpsadbw %ymm3, %ymm5, %ymm5
|
||||
; AVX512F-NEXT: vpunpckldq {{.*#+}} ymm1 = ymm1[0],ymm3[0],ymm1[1],ymm3[1],ymm1[4],ymm3[4],ymm1[5],ymm3[5]
|
||||
|
||||
@ -249,7 +249,7 @@ define <16 x i16> @var_rotate_v16i16(<16 x i16> %a, <16 x i16> %b) nounwind {
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vmovdqa {{.*#+}} ymm2 = [16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
|
||||
; AVX2-NEXT: vpsubw %ymm1, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX2-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; AVX2-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm1[4],ymm3[4],ymm1[5],ymm3[5],ymm1[6],ymm3[6],ymm1[7],ymm3[7],ymm1[12],ymm3[12],ymm1[13],ymm3[13],ymm1[14],ymm3[14],ymm1[15],ymm3[15]
|
||||
; AVX2-NEXT: vpunpckhwd {{.*#+}} ymm5 = ymm3[4],ymm0[4],ymm3[5],ymm0[5],ymm3[6],ymm0[6],ymm3[7],ymm0[7],ymm3[12],ymm0[12],ymm3[13],ymm0[13],ymm3[14],ymm0[14],ymm3[15],ymm0[15]
|
||||
; AVX2-NEXT: vpsllvd %ymm4, %ymm5, %ymm4
|
||||
@ -621,7 +621,7 @@ define <16 x i16> @constant_rotate_v16i16(<16 x i16> %a) nounwind {
|
||||
; AVX2-LABEL: constant_rotate_v16i16:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpmullw {{.*}}(%rip), %ymm0, %ymm1
|
||||
; AVX2-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX2-NEXT: vmovdqa {{.*#+}} ymm3 = [16,15,14,13,12,11,10,9,8,7,6,5,4,3,2,1]
|
||||
; AVX2-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm3[4],ymm2[4],ymm3[5],ymm2[5],ymm3[6],ymm2[6],ymm3[7],ymm2[7],ymm3[12],ymm2[12],ymm3[13],ymm2[13],ymm3[14],ymm2[14],ymm3[15],ymm2[15]
|
||||
; AVX2-NEXT: vpunpckhwd {{.*#+}} ymm5 = ymm2[4],ymm0[4],ymm2[5],ymm0[5],ymm2[6],ymm0[6],ymm2[7],ymm0[7],ymm2[12],ymm0[12],ymm2[13],ymm0[13],ymm2[14],ymm0[14],ymm2[15],ymm0[15]
|
||||
|
||||
@ -263,7 +263,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: var_shift_v16i16:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX2-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm1[4],ymm2[4],ymm1[5],ymm2[5],ymm1[6],ymm2[6],ymm1[7],ymm2[7],ymm1[12],ymm2[12],ymm1[13],ymm2[13],ymm1[14],ymm2[14],ymm1[15],ymm2[15]
|
||||
; AVX2-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm2[4],ymm0[4],ymm2[5],ymm0[5],ymm2[6],ymm0[6],ymm2[7],ymm0[7],ymm2[12],ymm0[12],ymm2[13],ymm0[13],ymm2[14],ymm0[14],ymm2[15],ymm0[15]
|
||||
; AVX2-NEXT: vpsravd %ymm3, %ymm4, %ymm3
|
||||
@ -365,7 +365,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) nounwind {
|
||||
;
|
||||
; X32-AVX2-LABEL: var_shift_v16i16:
|
||||
; X32-AVX2: # BB#0:
|
||||
; X32-AVX2-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; X32-AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; X32-AVX2-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm1[4],ymm2[4],ymm1[5],ymm2[5],ymm1[6],ymm2[6],ymm1[7],ymm2[7],ymm1[12],ymm2[12],ymm1[13],ymm2[13],ymm1[14],ymm2[14],ymm1[15],ymm2[15]
|
||||
; X32-AVX2-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm2[4],ymm0[4],ymm2[5],ymm0[5],ymm2[6],ymm0[6],ymm2[7],ymm0[7],ymm2[12],ymm0[12],ymm2[13],ymm0[13],ymm2[14],ymm0[14],ymm2[15],ymm0[15]
|
||||
; X32-AVX2-NEXT: vpsravd %ymm3, %ymm4, %ymm3
|
||||
@ -1316,7 +1316,7 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: constant_shift_v16i16:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]
|
||||
; AVX2-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm2[4],ymm1[4],ymm2[5],ymm1[5],ymm2[6],ymm1[6],ymm2[7],ymm1[7],ymm2[12],ymm1[12],ymm2[13],ymm1[13],ymm2[14],ymm1[14],ymm2[15],ymm1[15]
|
||||
; AVX2-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm1[4],ymm0[4],ymm1[5],ymm0[5],ymm1[6],ymm0[6],ymm1[7],ymm0[7],ymm1[12],ymm0[12],ymm1[13],ymm0[13],ymm1[14],ymm0[14],ymm1[15],ymm0[15]
|
||||
@ -1399,7 +1399,7 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) nounwind {
|
||||
;
|
||||
; X32-AVX2-LABEL: constant_shift_v16i16:
|
||||
; X32-AVX2: # BB#0:
|
||||
; X32-AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX2-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]
|
||||
; X32-AVX2-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm2[4],ymm1[4],ymm2[5],ymm1[5],ymm2[6],ymm1[6],ymm2[7],ymm1[7],ymm2[12],ymm1[12],ymm2[13],ymm1[13],ymm2[14],ymm1[14],ymm2[15],ymm1[15]
|
||||
; X32-AVX2-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm1[4],ymm0[4],ymm1[5],ymm0[5],ymm1[6],ymm0[6],ymm1[7],ymm0[7],ymm1[12],ymm0[12],ymm1[13],ymm0[13],ymm1[14],ymm0[14],ymm1[15],ymm0[15]
|
||||
|
||||
@ -456,7 +456,7 @@ define <64 x i8> @splatconstant_shift_v64i8(<64 x i8> %a) nounwind {
|
||||
define <64 x i8> @ashr_const7_v64i8(<64 x i8> %a) {
|
||||
; AVX512DQ-LABEL: ashr_const7_v64i8:
|
||||
; AVX512DQ: # BB#0:
|
||||
; AVX512DQ-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX512DQ-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX512DQ-NEXT: vpcmpgtb %ymm0, %ymm2, %ymm0
|
||||
; AVX512DQ-NEXT: vpcmpgtb %ymm1, %ymm2, %ymm1
|
||||
; AVX512DQ-NEXT: retq
|
||||
|
||||
@ -226,7 +226,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: var_shift_v16i16:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX2-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm1[4],ymm2[4],ymm1[5],ymm2[5],ymm1[6],ymm2[6],ymm1[7],ymm2[7],ymm1[12],ymm2[12],ymm1[13],ymm2[13],ymm1[14],ymm2[14],ymm1[15],ymm2[15]
|
||||
; AVX2-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm2[4],ymm0[4],ymm2[5],ymm0[5],ymm2[6],ymm0[6],ymm2[7],ymm0[7],ymm2[12],ymm0[12],ymm2[13],ymm0[13],ymm2[14],ymm0[14],ymm2[15],ymm0[15]
|
||||
; AVX2-NEXT: vpsrlvd %ymm3, %ymm4, %ymm3
|
||||
@ -328,7 +328,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) nounwind {
|
||||
;
|
||||
; X32-AVX2-LABEL: var_shift_v16i16:
|
||||
; X32-AVX2: # BB#0:
|
||||
; X32-AVX2-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; X32-AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; X32-AVX2-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm1[4],ymm2[4],ymm1[5],ymm2[5],ymm1[6],ymm2[6],ymm1[7],ymm2[7],ymm1[12],ymm2[12],ymm1[13],ymm2[13],ymm1[14],ymm2[14],ymm1[15],ymm2[15]
|
||||
; X32-AVX2-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm2[4],ymm0[4],ymm2[5],ymm0[5],ymm2[6],ymm0[6],ymm2[7],ymm0[7],ymm2[12],ymm0[12],ymm2[13],ymm0[13],ymm2[14],ymm0[14],ymm2[15],ymm0[15]
|
||||
; X32-AVX2-NEXT: vpsrlvd %ymm3, %ymm4, %ymm3
|
||||
@ -1047,7 +1047,7 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: constant_shift_v16i16:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]
|
||||
; AVX2-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm2[4],ymm1[4],ymm2[5],ymm1[5],ymm2[6],ymm1[6],ymm2[7],ymm1[7],ymm2[12],ymm1[12],ymm2[13],ymm1[13],ymm2[14],ymm1[14],ymm2[15],ymm1[15]
|
||||
; AVX2-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm1[4],ymm0[4],ymm1[5],ymm0[5],ymm1[6],ymm0[6],ymm1[7],ymm0[7],ymm1[12],ymm0[12],ymm1[13],ymm0[13],ymm1[14],ymm0[14],ymm1[15],ymm0[15]
|
||||
@ -1130,7 +1130,7 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) nounwind {
|
||||
;
|
||||
; X32-AVX2-LABEL: constant_shift_v16i16:
|
||||
; X32-AVX2: # BB#0:
|
||||
; X32-AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX2-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]
|
||||
; X32-AVX2-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm2[4],ymm1[4],ymm2[5],ymm1[5],ymm2[6],ymm1[6],ymm2[7],ymm1[7],ymm2[12],ymm1[12],ymm2[13],ymm1[13],ymm2[14],ymm1[14],ymm2[15],ymm1[15]
|
||||
; X32-AVX2-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm1[4],ymm0[4],ymm1[5],ymm0[5],ymm1[6],ymm0[6],ymm1[7],ymm0[7],ymm1[12],ymm0[12],ymm1[13],ymm0[13],ymm1[14],ymm0[14],ymm1[15],ymm0[15]
|
||||
|
||||
@ -192,7 +192,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: var_shift_v16i16:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX2-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm1[4],ymm2[4],ymm1[5],ymm2[5],ymm1[6],ymm2[6],ymm1[7],ymm2[7],ymm1[12],ymm2[12],ymm1[13],ymm2[13],ymm1[14],ymm2[14],ymm1[15],ymm2[15]
|
||||
; AVX2-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm2[4],ymm0[4],ymm2[5],ymm0[5],ymm2[6],ymm0[6],ymm2[7],ymm0[7],ymm2[12],ymm0[12],ymm2[13],ymm0[13],ymm2[14],ymm0[14],ymm2[15],ymm0[15]
|
||||
; AVX2-NEXT: vpsllvd %ymm3, %ymm4, %ymm3
|
||||
@ -288,7 +288,7 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) nounwind {
|
||||
;
|
||||
; X32-AVX2-LABEL: var_shift_v16i16:
|
||||
; X32-AVX2: # BB#0:
|
||||
; X32-AVX2-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; X32-AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; X32-AVX2-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm1[4],ymm2[4],ymm1[5],ymm2[5],ymm1[6],ymm2[6],ymm1[7],ymm2[7],ymm1[12],ymm2[12],ymm1[13],ymm2[13],ymm1[14],ymm2[14],ymm1[15],ymm2[15]
|
||||
; X32-AVX2-NEXT: vpunpckhwd {{.*#+}} ymm4 = ymm2[4],ymm0[4],ymm2[5],ymm0[5],ymm2[6],ymm0[6],ymm2[7],ymm0[7],ymm2[12],ymm0[12],ymm2[13],ymm0[13],ymm2[14],ymm0[14],ymm2[15],ymm0[15]
|
||||
; X32-AVX2-NEXT: vpsllvd %ymm3, %ymm4, %ymm3
|
||||
|
||||
@ -1633,7 +1633,7 @@ define <16 x i16> @shuffle_v16i16_01_zz_03_zz_05_zz_07_zz_09_zz_11_zz_13_zz_15_z
|
||||
define <16 x i16> @shuffle_v16i16_02_03_zz_zz_06_07_zz_zz_10_11_zz_zz_14_15_zz_zz(<16 x i16> %a) {
|
||||
; AVX1-LABEL: shuffle_v16i16_02_03_zz_zz_06_07_zz_zz_10_11_zz_zz_14_15_zz_zz:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vshufps {{.*#+}} ymm0 = ymm0[1,3],ymm1[1,3],ymm0[5,7],ymm1[5,7]
|
||||
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,2,1,3,4,6,5,7]
|
||||
; AVX1-NEXT: retq
|
||||
@ -1699,7 +1699,7 @@ define <16 x i16> @shuffle_v16i16_28_zz_zz_zz_29_zz_zz_zz_30_zz_zz_zz_31_zz_zz_z
|
||||
; AVX2-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
|
||||
; AVX2-NEXT: vpmovzxwq {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
|
||||
; AVX2-NEXT: vinserti128 $1, %xmm0, %ymm1, %ymm0
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3],ymm0[4],ymm1[5,6,7],ymm0[8],ymm1[9,10,11],ymm0[12],ymm1[13,14,15]
|
||||
; AVX2-NEXT: retq
|
||||
;
|
||||
|
||||
@ -305,7 +305,7 @@ define <32 x i8> @shuffle_v32i8_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_16_
|
||||
; AVX2-LABEL: shuffle_v32i8_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_16_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vperm2i128 {{.*#+}} ymm1 = ymm0[2,3,0,1]
|
||||
; AVX2-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX2-NEXT: vpshufb %ymm2, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpbroadcastb %xmm0, %xmm0
|
||||
; AVX2-NEXT: vmovdqa {{.*#+}} ymm2 = [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
|
||||
@ -744,11 +744,17 @@ define <32 x i8> @shuffle_v32i8_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_
|
||||
; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
; AVX2OR512VL-LABEL: shuffle_v32i8_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_16_16_16_16_16_16_16_16_16_16_16_16_16_16_16_16:
|
||||
; AVX2OR512VL: # BB#0:
|
||||
; AVX2OR512VL-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2OR512VL-NEXT: vpshufb %ymm1, %ymm0, %ymm0
|
||||
; AVX2OR512VL-NEXT: retq
|
||||
; AVX2-LABEL: shuffle_v32i8_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_16_16_16_16_16_16_16_16_16_16_16_16_16_16_16_16:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpshufb %ymm1, %ymm0, %ymm0
|
||||
; AVX2-NEXT: retq
|
||||
;
|
||||
; AVX512VL-LABEL: shuffle_v32i8_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_16_16_16_16_16_16_16_16_16_16_16_16_16_16_16_16:
|
||||
; AVX512VL: # BB#0:
|
||||
; AVX512VL-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512VL-NEXT: vpshufb %ymm1, %ymm0, %ymm0
|
||||
; AVX512VL-NEXT: retq
|
||||
%shuffle = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16>
|
||||
ret <32 x i8> %shuffle
|
||||
}
|
||||
@ -1132,7 +1138,7 @@ define <32 x i8> @shuffle_v32i8_00_32_00_32_00_32_00_32_00_32_00_32_00_32_00_32_
|
||||
;
|
||||
; AVX2-LABEL: shuffle_v32i8_00_32_00_32_00_32_00_32_00_32_00_32_00_32_00_32_16_48_16_48_16_48_16_48_16_48_16_48_16_48_16_48:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX2-NEXT: vpshufb %ymm2, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpshuflw {{.*#+}} ymm0 = ymm0[0,0,0,0,4,5,6,7,8,8,8,8,12,13,14,15]
|
||||
; AVX2-NEXT: vpshufd {{.*#+}} ymm0 = ymm0[0,0,1,1,4,4,5,5]
|
||||
@ -1167,12 +1173,19 @@ define <32 x i8> @shuffle_v32i8_32_32_32_32_32_32_32_32_08_09_10_11_12_13_14_15_
|
||||
; AVX1-NEXT: vinsertf128 $1, %xmm2, %ymm0, %ymm0
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
; AVX2OR512VL-LABEL: shuffle_v32i8_32_32_32_32_32_32_32_32_08_09_10_11_12_13_14_15_48_48_48_48_48_48_48_48_24_25_26_27_28_29_30_31:
|
||||
; AVX2OR512VL: # BB#0:
|
||||
; AVX2OR512VL-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX2OR512VL-NEXT: vpshufb %ymm2, %ymm1, %ymm1
|
||||
; AVX2OR512VL-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,3],ymm1[4,5],ymm0[6,7]
|
||||
; AVX2OR512VL-NEXT: retq
|
||||
; AVX2-LABEL: shuffle_v32i8_32_32_32_32_32_32_32_32_08_09_10_11_12_13_14_15_48_48_48_48_48_48_48_48_24_25_26_27_28_29_30_31:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX2-NEXT: vpshufb %ymm2, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,3],ymm1[4,5],ymm0[6,7]
|
||||
; AVX2-NEXT: retq
|
||||
;
|
||||
; AVX512VL-LABEL: shuffle_v32i8_32_32_32_32_32_32_32_32_08_09_10_11_12_13_14_15_48_48_48_48_48_48_48_48_24_25_26_27_28_29_30_31:
|
||||
; AVX512VL: # BB#0:
|
||||
; AVX512VL-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX512VL-NEXT: vpshufb %ymm2, %ymm1, %ymm1
|
||||
; AVX512VL-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,3],ymm1[4,5],ymm0[6,7]
|
||||
; AVX512VL-NEXT: retq
|
||||
%shuffle = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 32, i32 32, i32 32, i32 32, i32 32, i32 32, i32 32, i32 32, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 48, i32 48, i32 48, i32 48, i32 48, i32 48, i32 48, i32 48, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
|
||||
ret <32 x i8> %shuffle
|
||||
}
|
||||
|
||||
@ -521,14 +521,14 @@ define <4 x double> @shuffle_v4f64_0z3z(<4 x double> %a, <4 x double> %b) {
|
||||
; AVX1-LABEL: shuffle_v4f64_0z3z:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[0,0,3,2]
|
||||
; AVX1-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3]
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
; AVX2-LABEL: shuffle_v4f64_0z3z:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[0,0,3,2]
|
||||
; AVX2-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3]
|
||||
; AVX2-NEXT: retq
|
||||
;
|
||||
@ -545,7 +545,7 @@ define <4 x double> @shuffle_v4f64_0z3z(<4 x double> %a, <4 x double> %b) {
|
||||
define <4 x double> @shuffle_v4f64_1z2z(<4 x double> %a, <4 x double> %b) {
|
||||
; AVX1-LABEL: shuffle_v4f64_1z2z:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vblendpd {{.*#+}} ymm0 = ymm1[0],ymm0[1,2,3]
|
||||
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
|
||||
; AVX1-NEXT: vunpcklpd {{.*#+}} xmm1 = xmm1[0],xmm0[0]
|
||||
@ -555,7 +555,7 @@ define <4 x double> @shuffle_v4f64_1z2z(<4 x double> %a, <4 x double> %b) {
|
||||
;
|
||||
; AVX2-LABEL: shuffle_v4f64_1z2z:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vblendpd {{.*#+}} ymm0 = ymm1[0],ymm0[1,2,3]
|
||||
; AVX2-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[1,0,2,0]
|
||||
; AVX2-NEXT: retq
|
||||
@ -1110,7 +1110,7 @@ define <4 x i64> @shuffle_v4i64_0415(<4 x i64> %a, <4 x i64> %b) {
|
||||
define <4 x i64> @shuffle_v4i64_z4z6(<4 x i64> %a) {
|
||||
; AVX1-LABEL: shuffle_v4i64_z4z6:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vunpcklpd {{.*#+}} ymm0 = ymm1[0],ymm0[0],ymm1[2],ymm0[2]
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
@ -1130,7 +1130,7 @@ define <4 x i64> @shuffle_v4i64_z4z6(<4 x i64> %a) {
|
||||
define <4 x i64> @shuffle_v4i64_5zuz(<4 x i64> %a) {
|
||||
; AVX1-LABEL: shuffle_v4i64_5zuz:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vunpckhpd {{.*#+}} ymm0 = ymm0[1],ymm1[1],ymm0[3],ymm1[3]
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
@ -1227,7 +1227,7 @@ define <4 x i64> @shuffle_v4i64_3333(<4 x i64> %a, <4 x i64> %b) {
|
||||
define <4 x i64> @shuffle_v4i64_1z3z(<4 x i64> %a, <4 x i64> %b) {
|
||||
; AVX1-LABEL: shuffle_v4i64_1z3z:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vblendpd {{.*#+}} ymm0 = ymm1[0],ymm0[1,2,3]
|
||||
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
|
||||
; AVX1-NEXT: vshufpd {{.*#+}} xmm1 = xmm1[1],xmm0[0]
|
||||
@ -1284,14 +1284,14 @@ define <4 x double> @insert_reg_and_zero_v4f64(double %a) {
|
||||
; AVX1-LABEL: insert_reg_and_zero_v4f64:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; AVX1-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3]
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
; AVX2-LABEL: insert_reg_and_zero_v4f64:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; AVX2-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3]
|
||||
; AVX2-NEXT: retq
|
||||
;
|
||||
@ -1567,14 +1567,14 @@ define <4 x i64> @shuffle_v4i64_z0z3(<4 x i64> %a, <4 x i64> %b) {
|
||||
; AVX1-LABEL: shuffle_v4i64_z0z3:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[0,0,2,3]
|
||||
; AVX1-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vblendpd {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3]
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
; AVX2-LABEL: shuffle_v4i64_z0z3:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,2,3]
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,3],ymm1[4,5],ymm0[6,7]
|
||||
; AVX2-NEXT: retq
|
||||
;
|
||||
@ -1591,7 +1591,7 @@ define <4 x i64> @shuffle_v4i64_z0z3(<4 x i64> %a, <4 x i64> %b) {
|
||||
define <4 x i64> @shuffle_v4i64_1z2z(<4 x i64> %a, <4 x i64> %b) {
|
||||
; AVX1-LABEL: shuffle_v4i64_1z2z:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vblendpd {{.*#+}} ymm0 = ymm1[0],ymm0[1,2,3]
|
||||
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
|
||||
; AVX1-NEXT: vunpcklpd {{.*#+}} xmm1 = xmm1[0],xmm0[0]
|
||||
@ -1601,7 +1601,7 @@ define <4 x i64> @shuffle_v4i64_1z2z(<4 x i64> %a, <4 x i64> %b) {
|
||||
;
|
||||
; AVX2-LABEL: shuffle_v4i64_1z2z:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,3,4,5,6,7]
|
||||
; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[1,0,2,0]
|
||||
; AVX2-NEXT: retq
|
||||
|
||||
@ -1938,7 +1938,7 @@ define <8 x i32> @shuffle_v8i32_ba983210(<8 x i32> %a, <8 x i32> %b) {
|
||||
define <8 x i32> @shuffle_v8i32_zuu8zuuc(<8 x i32> %a) {
|
||||
; AVX1-LABEL: shuffle_v8i32_zuu8zuuc:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vshufps {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,0],ymm1[4,5],ymm0[6,4]
|
||||
; AVX1-NEXT: retq
|
||||
;
|
||||
@ -1953,7 +1953,7 @@ define <8 x i32> @shuffle_v8i32_zuu8zuuc(<8 x i32> %a) {
|
||||
define <8 x i32> @shuffle_v8i32_9ubzdefz(<8 x i32> %a) {
|
||||
; AVX1-LABEL: shuffle_v8i32_9ubzdefz:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vshufps {{.*#+}} ymm1 = ymm1[3,0],ymm0[3,0],ymm1[7,4],ymm0[7,4]
|
||||
; AVX1-NEXT: vshufps {{.*#+}} ymm0 = ymm0[1,2],ymm1[2,0],ymm0[5,6],ymm1[6,4]
|
||||
; AVX1-NEXT: retq
|
||||
@ -2118,7 +2118,7 @@ define <8 x float> @splat_v8f32(<4 x float> %r) {
|
||||
define <8 x i32> @shuffle_v8i32_z0U2zUz6(<8 x i32> %a) {
|
||||
; AVX1-LABEL: shuffle_v8i32_z0U2zUz6:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vshufps {{.*#+}} ymm0 = ymm0[0,2],ymm1[0,2],ymm0[4,6],ymm1[4,6]
|
||||
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[2,0,3,1,6,4,7,5]
|
||||
; AVX1-NEXT: retq
|
||||
@ -2134,7 +2134,7 @@ define <8 x i32> @shuffle_v8i32_z0U2zUz6(<8 x i32> %a) {
|
||||
define <8 x i32> @shuffle_v8i32_1U3z5zUU(<8 x i32> %a) {
|
||||
; AVX1-LABEL: shuffle_v8i32_1U3z5zUU:
|
||||
; AVX1: # BB#0:
|
||||
; AVX1-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; AVX1-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; AVX1-NEXT: vshufps {{.*#+}} ymm0 = ymm0[1,3],ymm1[1,3],ymm0[5,7],ymm1[5,7]
|
||||
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,2,1,3,4,6,5,7]
|
||||
; AVX1-NEXT: retq
|
||||
|
||||
@ -205,7 +205,7 @@ define <32 x i16> @shuffle_v32i16_0zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz(<32 x i16> %a
|
||||
; KNL-NEXT: movl $65535, %eax ## imm = 0xFFFF
|
||||
; KNL-NEXT: vmovd %eax, %xmm1
|
||||
; KNL-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; KNL-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; KNL-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; KNL-NEXT: retq
|
||||
;
|
||||
; SKX-LABEL: shuffle_v32i16_0zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz:
|
||||
|
||||
@ -115,7 +115,7 @@ define <64 x i8> @shuffle_v64i8_0zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz(<64 x i8> %a) {
|
||||
; AVX512F-NEXT: movl $255, %eax
|
||||
; AVX512F-NEXT: vmovd %eax, %xmm1
|
||||
; AVX512F-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512F-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512F-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512F-NEXT: retq
|
||||
;
|
||||
; AVX512BW-LABEL: shuffle_v64i8_0zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz:
|
||||
@ -128,7 +128,7 @@ define <64 x i8> @shuffle_v64i8_0zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz(<64 x i8> %a) {
|
||||
; AVX512DQ-NEXT: movl $255, %eax
|
||||
; AVX512DQ-NEXT: vmovd %eax, %xmm1
|
||||
; AVX512DQ-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512DQ-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512DQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512DQ-NEXT: retq
|
||||
;
|
||||
; AVX512VBMI-LABEL: shuffle_v64i8_0zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz:
|
||||
|
||||
@ -17,7 +17,7 @@ define <8 x float> @expand(<4 x float> %a) {
|
||||
; KNL64-LABEL: expand:
|
||||
; KNL64: # BB#0:
|
||||
; KNL64-NEXT: vpermilps {{.*#+}} xmm0 = xmm0[0,1,1,3]
|
||||
; KNL64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; KNL64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; KNL64-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3,4,5,6,7]
|
||||
; KNL64-NEXT: retq
|
||||
;
|
||||
@ -32,7 +32,7 @@ define <8 x float> @expand(<4 x float> %a) {
|
||||
; KNL32-LABEL: expand:
|
||||
; KNL32: # BB#0:
|
||||
; KNL32-NEXT: vpermilps {{.*#+}} xmm0 = xmm0[0,1,1,3]
|
||||
; KNL32-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; KNL32-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; KNL32-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3,4,5,6,7]
|
||||
; KNL32-NEXT: retl
|
||||
%res = shufflevector <4 x float> %a, <4 x float> zeroinitializer, <8 x i32> <i32 0, i32 5, i32 1, i32 5, i32 5, i32 5, i32 5, i32 5>
|
||||
@ -53,7 +53,7 @@ define <8 x float> @expand1(<4 x float> %a ) {
|
||||
; KNL64-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; KNL64-NEXT: vmovaps {{.*#+}} ymm1 = <u,0,u,1,u,2,u,3>
|
||||
; KNL64-NEXT: vpermps %ymm0, %ymm1, %ymm0
|
||||
; KNL64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; KNL64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; KNL64-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4],ymm0[5],ymm1[6],ymm0[7]
|
||||
; KNL64-NEXT: retq
|
||||
;
|
||||
@ -70,7 +70,7 @@ define <8 x float> @expand1(<4 x float> %a ) {
|
||||
; KNL32-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; KNL32-NEXT: vmovaps {{.*#+}} ymm1 = <u,0,u,1,u,2,u,3>
|
||||
; KNL32-NEXT: vpermps %ymm0, %ymm1, %ymm0
|
||||
; KNL32-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; KNL32-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; KNL32-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4],ymm0[5],ymm1[6],ymm0[7]
|
||||
; KNL32-NEXT: retl
|
||||
%res = shufflevector <4 x float> zeroinitializer, <4 x float> %a, <8 x i32> <i32 0, i32 4, i32 1, i32 5, i32 2, i32 6, i32 3, i32 7>
|
||||
@ -91,7 +91,7 @@ define <4 x double> @expand2(<2 x double> %a) {
|
||||
; KNL64: # BB#0:
|
||||
; KNL64-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; KNL64-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,1,2,1]
|
||||
; KNL64-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; KNL64-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; KNL64-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2],ymm0[3]
|
||||
; KNL64-NEXT: retq
|
||||
;
|
||||
@ -107,7 +107,7 @@ define <4 x double> @expand2(<2 x double> %a) {
|
||||
; KNL32: # BB#0:
|
||||
; KNL32-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; KNL32-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,1,2,1]
|
||||
; KNL32-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; KNL32-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; KNL32-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2],ymm0[3]
|
||||
; KNL32-NEXT: retl
|
||||
%res = shufflevector <2 x double> %a, <2 x double> zeroinitializer, <4 x i32> <i32 0, i32 2, i32 2, i32 1>
|
||||
@ -127,7 +127,7 @@ define <8 x i32> @expand3(<4 x i32> %a ) {
|
||||
; KNL64-LABEL: expand3:
|
||||
; KNL64: # BB#0:
|
||||
; KNL64-NEXT: vpbroadcastq %xmm0, %ymm0
|
||||
; KNL64-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; KNL64-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; KNL64-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3,4,5,6],ymm0[7]
|
||||
; KNL64-NEXT: retq
|
||||
;
|
||||
@ -142,7 +142,7 @@ define <8 x i32> @expand3(<4 x i32> %a ) {
|
||||
; KNL32-LABEL: expand3:
|
||||
; KNL32: # BB#0:
|
||||
; KNL32-NEXT: vpbroadcastq %xmm0, %ymm0
|
||||
; KNL32-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; KNL32-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; KNL32-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3,4,5,6],ymm0[7]
|
||||
; KNL32-NEXT: retl
|
||||
%res = shufflevector <4 x i32> zeroinitializer, <4 x i32> %a, <8 x i32> <i32 4, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0,i32 5>
|
||||
@ -163,7 +163,7 @@ define <4 x i64> @expand4(<2 x i64> %a ) {
|
||||
; KNL64: # BB#0:
|
||||
; KNL64-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; KNL64-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,2,1]
|
||||
; KNL64-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; KNL64-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; KNL64-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5],ymm0[6,7]
|
||||
; KNL64-NEXT: retq
|
||||
;
|
||||
@ -179,7 +179,7 @@ define <4 x i64> @expand4(<2 x i64> %a ) {
|
||||
; KNL32: # BB#0:
|
||||
; KNL32-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def>
|
||||
; KNL32-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,2,1]
|
||||
; KNL32-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; KNL32-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; KNL32-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5],ymm0[6,7]
|
||||
; KNL32-NEXT: retl
|
||||
%res = shufflevector <2 x i64> zeroinitializer, <2 x i64> %a, <4 x i32> <i32 2, i32 0, i32 0, i32 3>
|
||||
@ -198,7 +198,7 @@ define <8 x float> @expand5(<4 x float> %a ) {
|
||||
; KNL64-LABEL: expand5:
|
||||
; KNL64: # BB#0:
|
||||
; KNL64-NEXT: vbroadcastss %xmm0, %ymm0
|
||||
; KNL64-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; KNL64-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; KNL64-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4],ymm0[5],ymm1[6],ymm0[7]
|
||||
; KNL64-NEXT: retq
|
||||
;
|
||||
@ -212,7 +212,7 @@ define <8 x float> @expand5(<4 x float> %a ) {
|
||||
; KNL32-LABEL: expand5:
|
||||
; KNL32: # BB#0:
|
||||
; KNL32-NEXT: vbroadcastss %xmm0, %ymm0
|
||||
; KNL32-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; KNL32-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; KNL32-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4],ymm0[5],ymm1[6],ymm0[7]
|
||||
; KNL32-NEXT: retl
|
||||
%res = shufflevector <4 x float> zeroinitializer, <4 x float> %a, <8 x i32> <i32 0, i32 4, i32 1, i32 4, i32 2, i32 4, i32 3, i32 4>
|
||||
@ -479,7 +479,7 @@ define <16 x float> @expand13(<8 x float> %a ) {
|
||||
;
|
||||
; KNL64-LABEL: expand13:
|
||||
; KNL64: # BB#0:
|
||||
; KNL64-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; KNL64-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; KNL64-NEXT: vinsertf64x4 $1, %ymm0, %zmm1, %zmm0
|
||||
; KNL64-NEXT: retq
|
||||
;
|
||||
@ -491,7 +491,7 @@ define <16 x float> @expand13(<8 x float> %a ) {
|
||||
;
|
||||
; KNL32-LABEL: expand13:
|
||||
; KNL32: # BB#0:
|
||||
; KNL32-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; KNL32-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; KNL32-NEXT: vinsertf64x4 $1, %ymm0, %zmm1, %zmm0
|
||||
; KNL32-NEXT: retl
|
||||
%res = shufflevector <8 x float> zeroinitializer, <8 x float> %a, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7,i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
@ -12,12 +12,12 @@ declare <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8>, <32 x i8>)
|
||||
define <32 x i8> @combine_pshufb_pslldq(<32 x i8> %a0) {
|
||||
; X32-LABEL: combine_pshufb_pslldq:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: combine_pshufb_pslldq:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-NEXT: retq
|
||||
%1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %a0, <32 x i8> <i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7>)
|
||||
%2 = shufflevector <32 x i8> %1, <32 x i8> zeroinitializer, <32 x i32> <i32 32, i32 32, i32 32, i32 32, i32 32, i32 32, i32 32, i32 32, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 32, i32 32, i32 32, i32 32, i32 32, i32 32, i32 32, i32 32, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23>
|
||||
@ -27,12 +27,12 @@ define <32 x i8> @combine_pshufb_pslldq(<32 x i8> %a0) {
|
||||
define <32 x i8> @combine_pshufb_psrldq(<32 x i8> %a0) {
|
||||
; X32-LABEL: combine_pshufb_psrldq:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: combine_pshufb_psrldq:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-NEXT: retq
|
||||
%1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %a0, <32 x i8> <i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128>)
|
||||
%2 = shufflevector <32 x i8> %1, <32 x i8> zeroinitializer, <32 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 32, i32 32, i32 32, i32 32, i32 32, i32 32, i32 32, i32 32, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31, i32 32, i32 32, i32 32, i32 32, i32 32, i32 32, i32 32, i32 32>
|
||||
@ -74,13 +74,13 @@ define <32 x i8> @combine_pshufb_vpermps(<8 x float> %a) {
|
||||
define <32 x i8> @combine_and_pshufb(<32 x i8> %a0) {
|
||||
; X32-LABEL: combine_and_pshufb:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3],ymm0[4],ymm1[5,6,7],ymm0[8],ymm1[9,10,11],ymm0[12],ymm1[13,14,15]
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: combine_and_pshufb:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3],ymm0[4],ymm1[5,6,7],ymm0[8],ymm1[9,10,11],ymm0[12],ymm1[13,14,15]
|
||||
; X64-NEXT: retq
|
||||
%1 = shufflevector <32 x i8> %a0, <32 x i8> zeroinitializer, <32 x i32> <i32 0, i32 1, i32 32, i32 32, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
|
||||
@ -91,13 +91,13 @@ define <32 x i8> @combine_and_pshufb(<32 x i8> %a0) {
|
||||
define <32 x i8> @combine_pshufb_and(<32 x i8> %a0) {
|
||||
; X32-LABEL: combine_pshufb_and:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3],ymm0[4],ymm1[5,6,7],ymm0[8],ymm1[9,10,11],ymm0[12],ymm1[13,14,15]
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: combine_pshufb_and:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3],ymm0[4],ymm1[5,6,7],ymm0[8],ymm1[9,10,11],ymm0[12],ymm1[13,14,15]
|
||||
; X64-NEXT: retq
|
||||
%1 = call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %a0, <32 x i8> <i8 0, i8 1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 8, i8 9, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 0, i8 1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 8, i8 9, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>)
|
||||
@ -164,13 +164,13 @@ define <8 x float> @combine_as_vpermps(<8 x float> %a0) {
|
||||
define <32 x i8> @combine_permq_pshufb_as_vpblendd(<4 x i64> %a0) {
|
||||
; X32-LABEL: combine_permq_pshufb_as_vpblendd:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: combine_permq_pshufb_as_vpblendd:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
|
||||
; X64-NEXT: retq
|
||||
%1 = shufflevector <4 x i64> %a0, <4 x i64> undef, <4 x i32> <i32 1, i32 0, i32 3, i32 2>
|
||||
@ -522,7 +522,7 @@ define <4 x i64> @combine_pshufb_as_zext128(<32 x i8> %a0) {
|
||||
define <4 x double> @combine_pshufb_as_vzmovl_64(<4 x double> %a0) {
|
||||
; X32-AVX2-LABEL: combine_pshufb_as_vzmovl_64:
|
||||
; X32-AVX2: # BB#0:
|
||||
; X32-AVX2-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX2-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX2-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3]
|
||||
; X32-AVX2-NEXT: retl
|
||||
;
|
||||
@ -534,7 +534,7 @@ define <4 x double> @combine_pshufb_as_vzmovl_64(<4 x double> %a0) {
|
||||
;
|
||||
; X64-AVX2-LABEL: combine_pshufb_as_vzmovl_64:
|
||||
; X64-AVX2: # BB#0:
|
||||
; X64-AVX2-NEXT: vxorpd %ymm1, %ymm1, %ymm1
|
||||
; X64-AVX2-NEXT: vxorpd %xmm1, %xmm1, %xmm1
|
||||
; X64-AVX2-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3]
|
||||
; X64-AVX2-NEXT: retq
|
||||
;
|
||||
@ -552,7 +552,7 @@ define <4 x double> @combine_pshufb_as_vzmovl_64(<4 x double> %a0) {
|
||||
define <8 x float> @combine_pshufb_as_vzmovl_32(<8 x float> %a0) {
|
||||
; X32-AVX2-LABEL: combine_pshufb_as_vzmovl_32:
|
||||
; X32-AVX2: # BB#0:
|
||||
; X32-AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3,4,5,6,7]
|
||||
; X32-AVX2-NEXT: retl
|
||||
;
|
||||
@ -564,7 +564,7 @@ define <8 x float> @combine_pshufb_as_vzmovl_32(<8 x float> %a0) {
|
||||
;
|
||||
; X64-AVX2-LABEL: combine_pshufb_as_vzmovl_32:
|
||||
; X64-AVX2: # BB#0:
|
||||
; X64-AVX2-NEXT: vxorps %ymm1, %ymm1, %ymm1
|
||||
; X64-AVX2-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
||||
; X64-AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3,4,5,6,7]
|
||||
; X64-AVX2-NEXT: retq
|
||||
;
|
||||
@ -708,13 +708,13 @@ define <32 x i8> @combine_pshufb_as_unpacklo_undef(<32 x i8> %a0) {
|
||||
define <32 x i8> @combine_pshufb_as_unpacklo_zero(<32 x i8> %a0) {
|
||||
; X32-LABEL: combine_pshufb_as_unpacklo_zero:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vpunpcklwd {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[1],ymm1[1],ymm0[2],ymm1[2],ymm0[3],ymm1[3],ymm0[8],ymm1[8],ymm0[9],ymm1[9],ymm0[10],ymm1[10],ymm0[11],ymm1[11]
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: combine_pshufb_as_unpacklo_zero:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vpunpcklwd {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[1],ymm1[1],ymm0[2],ymm1[2],ymm0[3],ymm1[3],ymm0[8],ymm1[8],ymm0[9],ymm1[9],ymm0[10],ymm1[10],ymm0[11],ymm1[11]
|
||||
; X64-NEXT: retq
|
||||
%1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %a0, <32 x i8> <i8 0, i8 1, i8 -1, i8 -1, i8 2, i8 3, i8 -1, i8 -1, i8 4, i8 5, i8 -1, i8 -1, i8 6, i8 7, i8 -1, i8 -1, i8 16, i8 17, i8 -1, i8 -1, i8 18, i8 19, i8 -1, i8 -1, i8 20, i8 21, i8 -1, i8 -1, i8 22, i8 23, i8 -1, i8 -1>)
|
||||
@ -724,13 +724,13 @@ define <32 x i8> @combine_pshufb_as_unpacklo_zero(<32 x i8> %a0) {
|
||||
define <32 x i8> @combine_pshufb_as_unpackhi_zero(<32 x i8> %a0) {
|
||||
; X32-LABEL: combine_pshufb_as_unpackhi_zero:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-NEXT: vpunpckhbw {{.*#+}} ymm0 = ymm1[8],ymm0[8],ymm1[9],ymm0[9],ymm1[10],ymm0[10],ymm1[11],ymm0[11],ymm1[12],ymm0[12],ymm1[13],ymm0[13],ymm1[14],ymm0[14],ymm1[15],ymm0[15],ymm1[24],ymm0[24],ymm1[25],ymm0[25],ymm1[26],ymm0[26],ymm1[27],ymm0[27],ymm1[28],ymm0[28],ymm1[29],ymm0[29],ymm1[30],ymm0[30],ymm1[31],ymm0[31]
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: combine_pshufb_as_unpackhi_zero:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X64-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X64-NEXT: vpunpckhbw {{.*#+}} ymm0 = ymm1[8],ymm0[8],ymm1[9],ymm0[9],ymm1[10],ymm0[10],ymm1[11],ymm0[11],ymm1[12],ymm0[12],ymm1[13],ymm0[13],ymm1[14],ymm0[14],ymm1[15],ymm0[15],ymm1[24],ymm0[24],ymm1[25],ymm0[25],ymm1[26],ymm0[26],ymm1[27],ymm0[27],ymm1[28],ymm0[28],ymm1[29],ymm0[29],ymm1[30],ymm0[30],ymm1[31],ymm0[31]
|
||||
; X64-NEXT: retq
|
||||
%1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %a0, <32 x i8> <i8 -1, i8 8, i8 -1, i8 9, i8 -1, i8 10, i8 -1, i8 11, i8 -1, i8 12, i8 -1, i8 13, i8 -1, i8 14, i8 -1, i8 15, i8 -1, i8 24, i8 -1, i8 25, i8 -1, i8 26, i8 -1, i8 27, i8 -1, i8 28, i8 -1, i8 29, i8 -1, i8 30, i8 -1, i8 31>)
|
||||
|
||||
@ -138,12 +138,12 @@ define <8 x float> @combine_vpermil2ps256_08z945Az(<8 x float> %a0, <8 x float>
|
||||
define <8 x float> @combine_vpermil2ps256_zero(<8 x float> %a0, <8 x float> %a1) {
|
||||
; X32-LABEL: combine_vpermil2ps256_zero:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X32-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X32-NEXT: retl
|
||||
;
|
||||
; X64-LABEL: combine_vpermil2ps256_zero:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm0, %ymm0, %ymm0
|
||||
; X64-NEXT: vxorps %xmm0, %xmm0, %xmm0
|
||||
; X64-NEXT: retq
|
||||
%res0 = call <8 x float> @llvm.x86.xop.vpermil2ps.256(<8 x float> %a1, <8 x float> %a0, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 8, i32 9, i32 10, i32 11>, i8 2)
|
||||
ret <8 x float> %res0
|
||||
|
||||
@ -41,7 +41,7 @@ define <4 x i64> @testv4i64(<4 x i64> %in) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: testv4i64:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpsubq %ymm0, %ymm1, %ymm2
|
||||
; AVX2-NEXT: vpand %ymm2, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpcmpeqd %ymm2, %ymm2, %ymm2
|
||||
@ -77,7 +77,7 @@ define <4 x i64> @testv4i64(<4 x i64> %in) nounwind {
|
||||
;
|
||||
; AVX512CD-LABEL: testv4i64:
|
||||
; AVX512CD: # BB#0:
|
||||
; AVX512CD-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512CD-NEXT: vpsubq %ymm0, %ymm1, %ymm2
|
||||
; AVX512CD-NEXT: vpand %ymm2, %ymm0, %ymm0
|
||||
; AVX512CD-NEXT: vpcmpeqd %ymm2, %ymm2, %ymm2
|
||||
@ -95,7 +95,7 @@ define <4 x i64> @testv4i64(<4 x i64> %in) nounwind {
|
||||
;
|
||||
; AVX512VPOPCNTDQ-LABEL: testv4i64:
|
||||
; AVX512VPOPCNTDQ: # BB#0:
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpsubq %ymm0, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -106,7 +106,7 @@ define <4 x i64> @testv4i64(<4 x i64> %in) nounwind {
|
||||
;
|
||||
; X32-AVX-LABEL: testv4i64:
|
||||
; X32-AVX: # BB#0:
|
||||
; X32-AVX-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX-NEXT: vpsubq %ymm0, %ymm1, %ymm2
|
||||
; X32-AVX-NEXT: vpand %ymm2, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vpsubq {{\.LCPI.*}}, %ymm0, %ymm0
|
||||
@ -157,7 +157,7 @@ define <4 x i64> @testv4i64u(<4 x i64> %in) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: testv4i64u:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpsubq %ymm0, %ymm1, %ymm2
|
||||
; AVX2-NEXT: vpand %ymm2, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpcmpeqd %ymm2, %ymm2, %ymm2
|
||||
@ -185,7 +185,7 @@ define <4 x i64> @testv4i64u(<4 x i64> %in) nounwind {
|
||||
;
|
||||
; AVX512CD-LABEL: testv4i64u:
|
||||
; AVX512CD: # BB#0:
|
||||
; AVX512CD-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512CD-NEXT: vpsubq %ymm0, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512CD-NEXT: vplzcntq %zmm0, %zmm0
|
||||
@ -195,7 +195,7 @@ define <4 x i64> @testv4i64u(<4 x i64> %in) nounwind {
|
||||
;
|
||||
; AVX512VPOPCNTDQ-LABEL: testv4i64u:
|
||||
; AVX512VPOPCNTDQ: # BB#0:
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpsubq %ymm0, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -206,7 +206,7 @@ define <4 x i64> @testv4i64u(<4 x i64> %in) nounwind {
|
||||
;
|
||||
; X32-AVX-LABEL: testv4i64u:
|
||||
; X32-AVX: # BB#0:
|
||||
; X32-AVX-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX-NEXT: vpsubq %ymm0, %ymm1, %ymm2
|
||||
; X32-AVX-NEXT: vpand %ymm2, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vpsubq {{\.LCPI.*}}, %ymm0, %ymm0
|
||||
@ -265,7 +265,7 @@ define <8 x i32> @testv8i32(<8 x i32> %in) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: testv8i32:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpsubd %ymm0, %ymm1, %ymm2
|
||||
; AVX2-NEXT: vpand %ymm2, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpcmpeqd %ymm2, %ymm2, %ymm2
|
||||
@ -309,7 +309,7 @@ define <8 x i32> @testv8i32(<8 x i32> %in) nounwind {
|
||||
;
|
||||
; AVX512CD-LABEL: testv8i32:
|
||||
; AVX512CD: # BB#0:
|
||||
; AVX512CD-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512CD-NEXT: vpsubd %ymm0, %ymm1, %ymm2
|
||||
; AVX512CD-NEXT: vpand %ymm2, %ymm0, %ymm0
|
||||
; AVX512CD-NEXT: vpcmpeqd %ymm2, %ymm2, %ymm2
|
||||
@ -331,7 +331,7 @@ define <8 x i32> @testv8i32(<8 x i32> %in) nounwind {
|
||||
;
|
||||
; AVX512VPOPCNTDQ-LABEL: testv8i32:
|
||||
; AVX512VPOPCNTDQ: # BB#0:
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpsubd %ymm0, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -342,7 +342,7 @@ define <8 x i32> @testv8i32(<8 x i32> %in) nounwind {
|
||||
;
|
||||
; X32-AVX-LABEL: testv8i32:
|
||||
; X32-AVX: # BB#0:
|
||||
; X32-AVX-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX-NEXT: vpsubd %ymm0, %ymm1, %ymm2
|
||||
; X32-AVX-NEXT: vpand %ymm2, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vpcmpeqd %ymm2, %ymm2, %ymm2
|
||||
@ -406,7 +406,7 @@ define <8 x i32> @testv8i32u(<8 x i32> %in) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: testv8i32u:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpsubd %ymm0, %ymm1, %ymm2
|
||||
; AVX2-NEXT: vpand %ymm2, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpcmpeqd %ymm2, %ymm2, %ymm2
|
||||
@ -438,7 +438,7 @@ define <8 x i32> @testv8i32u(<8 x i32> %in) nounwind {
|
||||
;
|
||||
; AVX512CD-LABEL: testv8i32u:
|
||||
; AVX512CD: # BB#0:
|
||||
; AVX512CD-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512CD-NEXT: vpsubd %ymm0, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512CD-NEXT: vplzcntd %zmm0, %zmm0
|
||||
@ -448,7 +448,7 @@ define <8 x i32> @testv8i32u(<8 x i32> %in) nounwind {
|
||||
;
|
||||
; AVX512VPOPCNTDQ-LABEL: testv8i32u:
|
||||
; AVX512VPOPCNTDQ: # BB#0:
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpsubd %ymm0, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -459,7 +459,7 @@ define <8 x i32> @testv8i32u(<8 x i32> %in) nounwind {
|
||||
;
|
||||
; X32-AVX-LABEL: testv8i32u:
|
||||
; X32-AVX: # BB#0:
|
||||
; X32-AVX-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX-NEXT: vpsubd %ymm0, %ymm1, %ymm2
|
||||
; X32-AVX-NEXT: vpand %ymm2, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vpcmpeqd %ymm2, %ymm2, %ymm2
|
||||
@ -519,7 +519,7 @@ define <16 x i16> @testv16i16(<16 x i16> %in) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: testv16i16:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpsubw %ymm0, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -559,7 +559,7 @@ define <16 x i16> @testv16i16(<16 x i16> %in) nounwind {
|
||||
;
|
||||
; AVX512CD-LABEL: testv16i16:
|
||||
; AVX512CD: # BB#0:
|
||||
; AVX512CD-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512CD-NEXT: vpsubw %ymm0, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512CD-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -579,7 +579,7 @@ define <16 x i16> @testv16i16(<16 x i16> %in) nounwind {
|
||||
;
|
||||
; AVX512VPOPCNTDQ-LABEL: testv16i16:
|
||||
; AVX512VPOPCNTDQ: # BB#0:
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpsubw %ymm0, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -591,7 +591,7 @@ define <16 x i16> @testv16i16(<16 x i16> %in) nounwind {
|
||||
;
|
||||
; X32-AVX-LABEL: testv16i16:
|
||||
; X32-AVX: # BB#0:
|
||||
; X32-AVX-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX-NEXT: vpsubw %ymm0, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -649,7 +649,7 @@ define <16 x i16> @testv16i16u(<16 x i16> %in) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: testv16i16u:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpsubw %ymm0, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -689,7 +689,7 @@ define <16 x i16> @testv16i16u(<16 x i16> %in) nounwind {
|
||||
;
|
||||
; AVX512CD-LABEL: testv16i16u:
|
||||
; AVX512CD: # BB#0:
|
||||
; AVX512CD-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512CD-NEXT: vpsubw %ymm0, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512CD-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -709,7 +709,7 @@ define <16 x i16> @testv16i16u(<16 x i16> %in) nounwind {
|
||||
;
|
||||
; AVX512VPOPCNTDQ-LABEL: testv16i16u:
|
||||
; AVX512VPOPCNTDQ: # BB#0:
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpsubw %ymm0, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -721,7 +721,7 @@ define <16 x i16> @testv16i16u(<16 x i16> %in) nounwind {
|
||||
;
|
||||
; X32-AVX-LABEL: testv16i16u:
|
||||
; X32-AVX: # BB#0:
|
||||
; X32-AVX-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX-NEXT: vpsubw %ymm0, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -773,7 +773,7 @@ define <32 x i8> @testv32i8(<32 x i8> %in) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: testv32i8:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpsubb %ymm0, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -807,7 +807,7 @@ define <32 x i8> @testv32i8(<32 x i8> %in) nounwind {
|
||||
;
|
||||
; AVX512CD-LABEL: testv32i8:
|
||||
; AVX512CD: # BB#0:
|
||||
; AVX512CD-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512CD-NEXT: vpsubb %ymm0, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512CD-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -824,7 +824,7 @@ define <32 x i8> @testv32i8(<32 x i8> %in) nounwind {
|
||||
;
|
||||
; AVX512VPOPCNTDQ-LABEL: testv32i8:
|
||||
; AVX512VPOPCNTDQ: # BB#0:
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpsubb %ymm0, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -841,7 +841,7 @@ define <32 x i8> @testv32i8(<32 x i8> %in) nounwind {
|
||||
;
|
||||
; X32-AVX-LABEL: testv32i8:
|
||||
; X32-AVX: # BB#0:
|
||||
; X32-AVX-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX-NEXT: vpsubb %ymm0, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -890,7 +890,7 @@ define <32 x i8> @testv32i8u(<32 x i8> %in) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: testv32i8u:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpsubb %ymm0, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -924,7 +924,7 @@ define <32 x i8> @testv32i8u(<32 x i8> %in) nounwind {
|
||||
;
|
||||
; AVX512CD-LABEL: testv32i8u:
|
||||
; AVX512CD: # BB#0:
|
||||
; AVX512CD-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512CD-NEXT: vpsubb %ymm0, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512CD-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -941,7 +941,7 @@ define <32 x i8> @testv32i8u(<32 x i8> %in) nounwind {
|
||||
;
|
||||
; AVX512VPOPCNTDQ-LABEL: testv32i8u:
|
||||
; AVX512VPOPCNTDQ: # BB#0:
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpsubb %ymm0, %ymm1, %ymm1
|
||||
; AVX512VPOPCNTDQ-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
@ -958,7 +958,7 @@ define <32 x i8> @testv32i8u(<32 x i8> %in) nounwind {
|
||||
;
|
||||
; X32-AVX-LABEL: testv32i8u:
|
||||
; X32-AVX: # BB#0:
|
||||
; X32-AVX-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; X32-AVX-NEXT: vpsubb %ymm0, %ymm1, %ymm1
|
||||
; X32-AVX-NEXT: vpand %ymm1, %ymm0, %ymm0
|
||||
; X32-AVX-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
||||
|
||||
@ -21,7 +21,7 @@ define <8 x i64> @testv8i64(<8 x i64> %in) nounwind {
|
||||
; AVX512CD-NEXT: vpand %ymm2, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpshufb %ymm1, %ymm4, %ymm1
|
||||
; AVX512CD-NEXT: vpaddb %ymm3, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX512CD-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; AVX512CD-NEXT: vpsadbw %ymm3, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpand %ymm2, %ymm0, %ymm5
|
||||
; AVX512CD-NEXT: vpshufb %ymm5, %ymm4, %ymm5
|
||||
@ -151,7 +151,7 @@ define <16 x i32> @testv16i32(<16 x i32> %in) nounwind {
|
||||
; AVX512CD-NEXT: vpand %ymm2, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpshufb %ymm1, %ymm4, %ymm1
|
||||
; AVX512CD-NEXT: vpaddb %ymm3, %ymm1, %ymm1
|
||||
; AVX512CD-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX512CD-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; AVX512CD-NEXT: vpunpckhdq {{.*#+}} ymm5 = ymm1[2],ymm3[2],ymm1[3],ymm3[3],ymm1[6],ymm3[6],ymm1[7],ymm3[7]
|
||||
; AVX512CD-NEXT: vpsadbw %ymm3, %ymm5, %ymm5
|
||||
; AVX512CD-NEXT: vpunpckldq {{.*#+}} ymm1 = ymm1[0],ymm3[0],ymm1[1],ymm3[1],ymm1[4],ymm3[4],ymm1[5],ymm3[5]
|
||||
@ -287,7 +287,7 @@ define <16 x i32> @testv16i32u(<16 x i32> %in) nounwind {
|
||||
define <32 x i16> @testv32i16(<32 x i16> %in) nounwind {
|
||||
; AVX512CD-LABEL: testv32i16:
|
||||
; AVX512CD: # BB#0:
|
||||
; AVX512CD-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX512CD-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX512CD-NEXT: vpsubw %ymm0, %ymm2, %ymm3
|
||||
; AVX512CD-NEXT: vpand %ymm3, %ymm0, %ymm0
|
||||
; AVX512CD-NEXT: vpcmpeqd %ymm3, %ymm3, %ymm3
|
||||
@ -359,7 +359,7 @@ define <32 x i16> @testv32i16(<32 x i16> %in) nounwind {
|
||||
;
|
||||
; AVX512VPOPCNTDQ-LABEL: testv32i16:
|
||||
; AVX512VPOPCNTDQ: # BB#0:
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX512VPOPCNTDQ-NEXT: vpsubw %ymm0, %ymm2, %ymm3
|
||||
; AVX512VPOPCNTDQ-NEXT: vpand %ymm3, %ymm0, %ymm0
|
||||
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %ymm3, %ymm3, %ymm3
|
||||
@ -381,7 +381,7 @@ define <32 x i16> @testv32i16(<32 x i16> %in) nounwind {
|
||||
define <32 x i16> @testv32i16u(<32 x i16> %in) nounwind {
|
||||
; AVX512CD-LABEL: testv32i16u:
|
||||
; AVX512CD: # BB#0:
|
||||
; AVX512CD-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX512CD-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX512CD-NEXT: vpsubw %ymm0, %ymm2, %ymm3
|
||||
; AVX512CD-NEXT: vpand %ymm3, %ymm0, %ymm0
|
||||
; AVX512CD-NEXT: vpcmpeqd %ymm3, %ymm3, %ymm3
|
||||
@ -453,7 +453,7 @@ define <32 x i16> @testv32i16u(<32 x i16> %in) nounwind {
|
||||
;
|
||||
; AVX512VPOPCNTDQ-LABEL: testv32i16u:
|
||||
; AVX512VPOPCNTDQ: # BB#0:
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX512VPOPCNTDQ-NEXT: vpsubw %ymm0, %ymm2, %ymm3
|
||||
; AVX512VPOPCNTDQ-NEXT: vpand %ymm3, %ymm0, %ymm0
|
||||
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %ymm3, %ymm3, %ymm3
|
||||
@ -475,7 +475,7 @@ define <32 x i16> @testv32i16u(<32 x i16> %in) nounwind {
|
||||
define <64 x i8> @testv64i8(<64 x i8> %in) nounwind {
|
||||
; AVX512CD-LABEL: testv64i8:
|
||||
; AVX512CD: # BB#0:
|
||||
; AVX512CD-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX512CD-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX512CD-NEXT: vpsubb %ymm0, %ymm2, %ymm3
|
||||
; AVX512CD-NEXT: vpand %ymm3, %ymm0, %ymm0
|
||||
; AVX512CD-NEXT: vpcmpeqd %ymm3, %ymm3, %ymm3
|
||||
@ -535,7 +535,7 @@ define <64 x i8> @testv64i8(<64 x i8> %in) nounwind {
|
||||
;
|
||||
; AVX512VPOPCNTDQ-LABEL: testv64i8:
|
||||
; AVX512VPOPCNTDQ: # BB#0:
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX512VPOPCNTDQ-NEXT: vpsubb %ymm0, %ymm2, %ymm3
|
||||
; AVX512VPOPCNTDQ-NEXT: vpand %ymm3, %ymm0, %ymm0
|
||||
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %ymm3, %ymm3, %ymm3
|
||||
@ -565,7 +565,7 @@ define <64 x i8> @testv64i8(<64 x i8> %in) nounwind {
|
||||
define <64 x i8> @testv64i8u(<64 x i8> %in) nounwind {
|
||||
; AVX512CD-LABEL: testv64i8u:
|
||||
; AVX512CD: # BB#0:
|
||||
; AVX512CD-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX512CD-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX512CD-NEXT: vpsubb %ymm0, %ymm2, %ymm3
|
||||
; AVX512CD-NEXT: vpand %ymm3, %ymm0, %ymm0
|
||||
; AVX512CD-NEXT: vpcmpeqd %ymm3, %ymm3, %ymm3
|
||||
@ -625,7 +625,7 @@ define <64 x i8> @testv64i8u(<64 x i8> %in) nounwind {
|
||||
;
|
||||
; AVX512VPOPCNTDQ-LABEL: testv64i8u:
|
||||
; AVX512VPOPCNTDQ: # BB#0:
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX512VPOPCNTDQ-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX512VPOPCNTDQ-NEXT: vpsubb %ymm0, %ymm2, %ymm3
|
||||
; AVX512VPOPCNTDQ-NEXT: vpand %ymm3, %ymm0, %ymm0
|
||||
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %ymm3, %ymm3, %ymm3
|
||||
|
||||
@ -459,7 +459,7 @@ define <4 x i64> @test_abs_gt_v4i64(<4 x i64> %a) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: test_abs_gt_v4i64:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
||||
; AVX2-NEXT: vpcmpgtq %ymm0, %ymm1, %ymm1
|
||||
; AVX2-NEXT: vpaddq %ymm1, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %ymm1, %ymm0, %ymm0
|
||||
@ -523,7 +523,7 @@ define <8 x i64> @test_abs_le_v8i64(<8 x i64> %a) nounwind {
|
||||
;
|
||||
; AVX2-LABEL: test_abs_le_v8i64:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX2-NEXT: vpcmpgtq %ymm0, %ymm2, %ymm3
|
||||
; AVX2-NEXT: vpaddq %ymm3, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %ymm3, %ymm0, %ymm0
|
||||
@ -598,7 +598,7 @@ define <8 x i64> @test_abs_le_v8i64_fold(<8 x i64>* %a.ptr) nounwind {
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vmovdqu (%rdi), %ymm0
|
||||
; AVX2-NEXT: vmovdqu 32(%rdi), %ymm1
|
||||
; AVX2-NEXT: vpxor %ymm2, %ymm2, %ymm2
|
||||
; AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
|
||||
; AVX2-NEXT: vpcmpgtq %ymm0, %ymm2, %ymm3
|
||||
; AVX2-NEXT: vpaddq %ymm3, %ymm0, %ymm0
|
||||
; AVX2-NEXT: vpxor %ymm3, %ymm0, %ymm0
|
||||
|
||||
@ -130,12 +130,19 @@ define <32 x i8> @signbit_sel_v32i8(<32 x i8> %x, <32 x i8> %y, <32 x i8> %mask)
|
||||
; AVX2-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
|
||||
; AVX2-NEXT: retq
|
||||
;
|
||||
; AVX512-LABEL: signbit_sel_v32i8:
|
||||
; AVX512: # BB#0:
|
||||
; AVX512-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX512-NEXT: vpcmpgtb %ymm2, %ymm3, %ymm2
|
||||
; AVX512-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
|
||||
; AVX512-NEXT: retq
|
||||
; AVX512F-LABEL: signbit_sel_v32i8:
|
||||
; AVX512F: # BB#0:
|
||||
; AVX512F-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; AVX512F-NEXT: vpcmpgtb %ymm2, %ymm3, %ymm2
|
||||
; AVX512F-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
|
||||
; AVX512F-NEXT: retq
|
||||
;
|
||||
; AVX512VL-LABEL: signbit_sel_v32i8:
|
||||
; AVX512VL: # BB#0:
|
||||
; AVX512VL-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX512VL-NEXT: vpcmpgtb %ymm2, %ymm3, %ymm2
|
||||
; AVX512VL-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
|
||||
; AVX512VL-NEXT: retq
|
||||
%tr = icmp slt <32 x i8> %mask, zeroinitializer
|
||||
%z = select <32 x i1> %tr, <32 x i8> %x, <32 x i8> %y
|
||||
ret <32 x i8> %z
|
||||
@ -158,17 +165,24 @@ define <16 x i16> @signbit_sel_v16i16(<16 x i16> %x, <16 x i16> %y, <16 x i16> %
|
||||
;
|
||||
; AVX2-LABEL: signbit_sel_v16i16:
|
||||
; AVX2: # BB#0:
|
||||
; AVX2-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX2-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; AVX2-NEXT: vpcmpgtw %ymm2, %ymm3, %ymm2
|
||||
; AVX2-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
|
||||
; AVX2-NEXT: retq
|
||||
;
|
||||
; AVX512-LABEL: signbit_sel_v16i16:
|
||||
; AVX512: # BB#0:
|
||||
; AVX512-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX512-NEXT: vpcmpgtw %ymm2, %ymm3, %ymm2
|
||||
; AVX512-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
|
||||
; AVX512-NEXT: retq
|
||||
; AVX512F-LABEL: signbit_sel_v16i16:
|
||||
; AVX512F: # BB#0:
|
||||
; AVX512F-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; AVX512F-NEXT: vpcmpgtw %ymm2, %ymm3, %ymm2
|
||||
; AVX512F-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
|
||||
; AVX512F-NEXT: retq
|
||||
;
|
||||
; AVX512VL-LABEL: signbit_sel_v16i16:
|
||||
; AVX512VL: # BB#0:
|
||||
; AVX512VL-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX512VL-NEXT: vpcmpgtw %ymm2, %ymm3, %ymm2
|
||||
; AVX512VL-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
|
||||
; AVX512VL-NEXT: retq
|
||||
%tr = icmp slt <16 x i16> %mask, zeroinitializer
|
||||
%z = select <16 x i1> %tr, <16 x i16> %x, <16 x i16> %y
|
||||
ret <16 x i16> %z
|
||||
@ -185,7 +199,7 @@ define <8 x i32> @signbit_sel_v8i32(<8 x i32> %x, <8 x i32> %y, <8 x i32> %mask)
|
||||
; AVX512F-NEXT: # kill: %YMM2<def> %YMM2<kill> %ZMM2<def>
|
||||
; AVX512F-NEXT: # kill: %YMM1<def> %YMM1<kill> %ZMM1<def>
|
||||
; AVX512F-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
|
||||
; AVX512F-NEXT: vpxor %ymm3, %ymm3, %ymm3
|
||||
; AVX512F-NEXT: vpxor %xmm3, %xmm3, %xmm3
|
||||
; AVX512F-NEXT: vpcmpgtd %zmm2, %zmm3, %k1
|
||||
; AVX512F-NEXT: vpblendmd %zmm0, %zmm1, %zmm0 {%k1}
|
||||
; AVX512F-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<kill>
|
||||
|
||||
@ -499,7 +499,7 @@ declare <2 x i64> @llvm.x86.xop.vpcmov(<2 x i64>, <2 x i64>, <2 x i64>) nounwind
|
||||
define <4 x i64> @test_mm256_cmov_si256(<4 x i64> %a0, <4 x i64> %a1, <4 x i64> %a2) {
|
||||
; X32-LABEL: test_mm256_cmov_si256:
|
||||
; X32: # BB#0:
|
||||
; X32-NEXT: vxorps %ymm3, %ymm3, %ymm3
|
||||
; X32-NEXT: vxorps %xmm3, %xmm3, %xmm3
|
||||
; X32-NEXT: vcmptrueps %ymm3, %ymm3, %ymm3
|
||||
; X32-NEXT: vxorps %ymm3, %ymm2, %ymm3
|
||||
; X32-NEXT: vandps %ymm2, %ymm0, %ymm0
|
||||
@ -509,7 +509,7 @@ define <4 x i64> @test_mm256_cmov_si256(<4 x i64> %a0, <4 x i64> %a1, <4 x i64>
|
||||
;
|
||||
; X64-LABEL: test_mm256_cmov_si256:
|
||||
; X64: # BB#0:
|
||||
; X64-NEXT: vxorps %ymm3, %ymm3, %ymm3
|
||||
; X64-NEXT: vxorps %xmm3, %xmm3, %xmm3
|
||||
; X64-NEXT: vcmptrueps %ymm3, %ymm3, %ymm3
|
||||
; X64-NEXT: vxorps %ymm3, %ymm2, %ymm3
|
||||
; X64-NEXT: vandps %ymm2, %ymm0, %ymm0
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user