aboutsummaryrefslogtreecommitdiff
path: root/test/CodeGen/X86/insertelement-zero.ll
diff options
context:
space:
mode:
Diffstat (limited to 'test/CodeGen/X86/insertelement-zero.ll')
-rw-r--r--test/CodeGen/X86/insertelement-zero.ll180
1 files changed, 46 insertions, 134 deletions
diff --git a/test/CodeGen/X86/insertelement-zero.ll b/test/CodeGen/X86/insertelement-zero.ll
index ac27bb7d8af9..ea780a2fa68c 100644
--- a/test/CodeGen/X86/insertelement-zero.ll
+++ b/test/CodeGen/X86/insertelement-zero.ll
@@ -46,22 +46,22 @@ define <2 x double> @insert_v2f64_z1(<2 x double> %a) {
define <4 x double> @insert_v4f64_0zz3(<4 x double> %a) {
; SSE2-LABEL: insert_v4f64_0zz3:
; SSE2: # BB#0:
+; SSE2-NEXT: movq {{.*#+}} xmm0 = xmm0[0],zero
; SSE2-NEXT: xorpd %xmm2, %xmm2
-; SSE2-NEXT: unpcklpd {{.*#+}} xmm0 = xmm0[0],xmm2[0]
; SSE2-NEXT: movsd {{.*#+}} xmm1 = xmm2[0],xmm1[1]
; SSE2-NEXT: retq
;
; SSE3-LABEL: insert_v4f64_0zz3:
; SSE3: # BB#0:
+; SSE3-NEXT: movq {{.*#+}} xmm0 = xmm0[0],zero
; SSE3-NEXT: xorpd %xmm2, %xmm2
-; SSE3-NEXT: unpcklpd {{.*#+}} xmm0 = xmm0[0],xmm2[0]
; SSE3-NEXT: movsd {{.*#+}} xmm1 = xmm2[0],xmm1[1]
; SSE3-NEXT: retq
;
; SSSE3-LABEL: insert_v4f64_0zz3:
; SSSE3: # BB#0:
+; SSSE3-NEXT: movq {{.*#+}} xmm0 = xmm0[0],zero
; SSSE3-NEXT: xorpd %xmm2, %xmm2
-; SSSE3-NEXT: unpcklpd {{.*#+}} xmm0 = xmm0[0],xmm2[0]
; SSSE3-NEXT: movsd {{.*#+}} xmm1 = xmm2[0],xmm1[1]
; SSSE3-NEXT: retq
;
@@ -244,24 +244,21 @@ define <8 x float> @insert_v8f32_z12345z7(<8 x float> %a) {
define <4 x i32> @insert_v4i32_01z3(<4 x i32> %a) {
; SSE2-LABEL: insert_v4i32_01z3:
; SSE2: # BB#0:
-; SSE2-NEXT: xorl %eax, %eax
-; SSE2-NEXT: movd %eax, %xmm1
+; SSE2-NEXT: xorps %xmm1, %xmm1
; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[3,0]
; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,2]
; SSE2-NEXT: retq
;
; SSE3-LABEL: insert_v4i32_01z3:
; SSE3: # BB#0:
-; SSE3-NEXT: xorl %eax, %eax
-; SSE3-NEXT: movd %eax, %xmm1
+; SSE3-NEXT: xorps %xmm1, %xmm1
; SSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[3,0]
; SSE3-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,2]
; SSE3-NEXT: retq
;
; SSSE3-LABEL: insert_v4i32_01z3:
; SSSE3: # BB#0:
-; SSSE3-NEXT: xorl %eax, %eax
-; SSSE3-NEXT: movd %eax, %xmm1
+; SSSE3-NEXT: xorps %xmm1, %xmm1
; SSSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[3,0]
; SSSE3-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,2]
; SSSE3-NEXT: retq
@@ -292,8 +289,7 @@ define <8 x i32> @insert_v8i32_z12345z7(<8 x i32> %a) {
; SSE2: # BB#0:
; SSE2-NEXT: xorps %xmm2, %xmm2
; SSE2-NEXT: movss {{.*#+}} xmm0 = xmm2[0],xmm0[1,2,3]
-; SSE2-NEXT: xorl %eax, %eax
-; SSE2-NEXT: movd %eax, %xmm2
+; SSE2-NEXT: xorps %xmm2, %xmm2
; SSE2-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,0],xmm1[3,0]
; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm2[0,2]
; SSE2-NEXT: retq
@@ -302,8 +298,7 @@ define <8 x i32> @insert_v8i32_z12345z7(<8 x i32> %a) {
; SSE3: # BB#0:
; SSE3-NEXT: xorps %xmm2, %xmm2
; SSE3-NEXT: movss {{.*#+}} xmm0 = xmm2[0],xmm0[1,2,3]
-; SSE3-NEXT: xorl %eax, %eax
-; SSE3-NEXT: movd %eax, %xmm2
+; SSE3-NEXT: xorps %xmm2, %xmm2
; SSE3-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,0],xmm1[3,0]
; SSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm2[0,2]
; SSE3-NEXT: retq
@@ -312,8 +307,7 @@ define <8 x i32> @insert_v8i32_z12345z7(<8 x i32> %a) {
; SSSE3: # BB#0:
; SSSE3-NEXT: xorps %xmm2, %xmm2
; SSSE3-NEXT: movss {{.*#+}} xmm0 = xmm2[0],xmm0[1,2,3]
-; SSSE3-NEXT: xorl %eax, %eax
-; SSSE3-NEXT: movd %eax, %xmm2
+; SSSE3-NEXT: xorps %xmm2, %xmm2
; SSSE3-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,0],xmm1[3,0]
; SSSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm2[0,2]
; SSSE3-NEXT: retq
@@ -414,25 +408,21 @@ define <16 x i16> @insert_v16i16_z12345z789ABZDEz(<16 x i16> %a) {
; AVX1-LABEL: insert_v16i16_z12345z789ABZDEz:
; AVX1: # BB#0:
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX1-NEXT: vpblendw {{.*#+}} xmm2 = xmm1[0],xmm0[1,2,3,4,5,6,7]
-; AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm2[0,1,2,3],ymm0[4,5,6,7]
-; AVX1-NEXT: vpblendw {{.*#+}} xmm2 = xmm0[0,1,2,3,4,5],xmm1[6],xmm0[7]
-; AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm2[0,1,2,3],ymm0[4,5,6,7]
-; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm2
-; AVX1-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0,1,2,3,4,5,6],xmm1[7]
-; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT: vpblendw {{.*#+}} xmm2 = xmm1[0],xmm0[1,2,3,4,5],xmm1[6],xmm0[7]
+; AVX1-NEXT: vblendps {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm0[4,5,6,7]
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm1[7]
+; AVX1-NEXT: vinsertf128 $1, %xmm0, %ymm2, %ymm0
; AVX1-NEXT: retq
;
; AVX2-LABEL: insert_v16i16_z12345z789ABZDEz:
; AVX2: # BB#0:
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX2-NEXT: vpblendw {{.*#+}} xmm2 = xmm1[0],xmm0[1,2,3,4,5,6,7]
-; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm2[0,1,2,3],ymm0[4,5,6,7]
-; AVX2-NEXT: vpblendw {{.*#+}} xmm2 = xmm0[0,1,2,3,4,5],xmm1[6],xmm0[7]
-; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm2[0,1,2,3],ymm0[4,5,6,7]
-; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm2
-; AVX2-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0,1,2,3,4,5,6],xmm1[7]
-; AVX2-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm0
+; AVX2-NEXT: vpblendw {{.*#+}} xmm2 = xmm1[0],xmm0[1,2,3,4,5],xmm1[6],xmm0[7]
+; AVX2-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm0[4,5,6,7]
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm1[7]
+; AVX2-NEXT: vinserti128 $1, %xmm0, %ymm2, %ymm0
; AVX2-NEXT: retq
%1 = insertelement <16 x i16> %a, i16 0, i32 0
%2 = insertelement <16 x i16> %1, i16 0, i32 6
@@ -440,58 +430,30 @@ define <16 x i16> @insert_v16i16_z12345z789ABZDEz(<16 x i16> %a) {
ret <16 x i16> %3
}
-define <16 x i8> @insert_v16i8_z123456789ABZDEz(<16 x i8> %a) {
-; SSE2-LABEL: insert_v16i8_z123456789ABZDEz:
+define <16 x i8> @insert_v16i8_z123456789ABCDEz(<16 x i8> %a) {
+; SSE2-LABEL: insert_v16i8_z123456789ABCDEz:
; SSE2: # BB#0:
-; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
-; SSE2-NEXT: pand %xmm1, %xmm0
-; SSE2-NEXT: xorl %eax, %eax
-; SSE2-NEXT: movd %eax, %xmm2
-; SSE2-NEXT: pandn %xmm2, %xmm1
-; SSE2-NEXT: por %xmm1, %xmm0
-; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0]
-; SSE2-NEXT: pand %xmm1, %xmm0
-; SSE2-NEXT: pslldq {{.*#+}} xmm2 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm2[0]
-; SSE2-NEXT: pandn %xmm2, %xmm1
-; SSE2-NEXT: por %xmm1, %xmm0
+; SSE2-NEXT: andps {{.*}}(%rip), %xmm0
; SSE2-NEXT: retq
;
-; SSE3-LABEL: insert_v16i8_z123456789ABZDEz:
+; SSE3-LABEL: insert_v16i8_z123456789ABCDEz:
; SSE3: # BB#0:
-; SSE3-NEXT: movdqa {{.*#+}} xmm1 = [0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
-; SSE3-NEXT: pand %xmm1, %xmm0
-; SSE3-NEXT: xorl %eax, %eax
-; SSE3-NEXT: movd %eax, %xmm2
-; SSE3-NEXT: pandn %xmm2, %xmm1
-; SSE3-NEXT: por %xmm1, %xmm0
-; SSE3-NEXT: movdqa {{.*#+}} xmm1 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0]
-; SSE3-NEXT: pand %xmm1, %xmm0
-; SSE3-NEXT: pslldq {{.*#+}} xmm2 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm2[0]
-; SSE3-NEXT: pandn %xmm2, %xmm1
-; SSE3-NEXT: por %xmm1, %xmm0
+; SSE3-NEXT: andps {{.*}}(%rip), %xmm0
; SSE3-NEXT: retq
;
-; SSSE3-LABEL: insert_v16i8_z123456789ABZDEz:
+; SSSE3-LABEL: insert_v16i8_z123456789ABCDEz:
; SSSE3: # BB#0:
-; SSSE3-NEXT: pshufb {{.*#+}} xmm0 = zero,xmm0[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]
-; SSSE3-NEXT: xorl %eax, %eax
-; SSSE3-NEXT: movd %eax, %xmm1
-; SSSE3-NEXT: movdqa %xmm1, %xmm2
-; SSSE3-NEXT: pshufb {{.*#+}} xmm2 = xmm2[0],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
-; SSSE3-NEXT: por %xmm2, %xmm0
-; SSSE3-NEXT: pshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14],zero
-; SSSE3-NEXT: pslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0]
-; SSSE3-NEXT: por %xmm1, %xmm0
+; SSSE3-NEXT: andps {{.*}}(%rip), %xmm0
; SSSE3-NEXT: retq
;
-; SSE41-LABEL: insert_v16i8_z123456789ABZDEz:
+; SSE41-LABEL: insert_v16i8_z123456789ABCDEz:
; SSE41: # BB#0:
; SSE41-NEXT: xorl %eax, %eax
; SSE41-NEXT: pinsrb $0, %eax, %xmm0
; SSE41-NEXT: pinsrb $15, %eax, %xmm0
; SSE41-NEXT: retq
;
-; AVX-LABEL: insert_v16i8_z123456789ABZDEz:
+; AVX-LABEL: insert_v16i8_z123456789ABCDEz:
; AVX: # BB#0:
; AVX-NEXT: xorl %eax, %eax
; AVX-NEXT: vpinsrb $0, %eax, %xmm0, %xmm0
@@ -505,68 +467,20 @@ define <16 x i8> @insert_v16i8_z123456789ABZDEz(<16 x i8> %a) {
define <32 x i8> @insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz(<32 x i8> %a) {
; SSE2-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
; SSE2: # BB#0:
-; SSE2-NEXT: movdqa {{.*#+}} xmm2 = [0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
-; SSE2-NEXT: pand %xmm2, %xmm0
-; SSE2-NEXT: xorl %eax, %eax
-; SSE2-NEXT: movd %eax, %xmm3
-; SSE2-NEXT: pandn %xmm3, %xmm2
-; SSE2-NEXT: por %xmm2, %xmm0
-; SSE2-NEXT: movdqa {{.*#+}} xmm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0]
-; SSE2-NEXT: pand %xmm2, %xmm0
-; SSE2-NEXT: movdqa %xmm3, %xmm4
-; SSE2-NEXT: pslldq {{.*#+}} xmm4 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm4[0]
-; SSE2-NEXT: movdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,0,255]
-; SSE2-NEXT: pand %xmm5, %xmm1
-; SSE2-NEXT: pslldq {{.*#+}} xmm3 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm3[0,1]
-; SSE2-NEXT: pandn %xmm3, %xmm5
-; SSE2-NEXT: por %xmm5, %xmm1
-; SSE2-NEXT: pand %xmm2, %xmm1
-; SSE2-NEXT: pandn %xmm4, %xmm2
-; SSE2-NEXT: por %xmm2, %xmm0
-; SSE2-NEXT: por %xmm2, %xmm1
+; SSE2-NEXT: andps {{.*}}(%rip), %xmm0
+; SSE2-NEXT: andps {{.*}}(%rip), %xmm1
; SSE2-NEXT: retq
;
; SSE3-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
; SSE3: # BB#0:
-; SSE3-NEXT: movdqa {{.*#+}} xmm2 = [0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
-; SSE3-NEXT: pand %xmm2, %xmm0
-; SSE3-NEXT: xorl %eax, %eax
-; SSE3-NEXT: movd %eax, %xmm3
-; SSE3-NEXT: pandn %xmm3, %xmm2
-; SSE3-NEXT: por %xmm2, %xmm0
-; SSE3-NEXT: movdqa {{.*#+}} xmm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0]
-; SSE3-NEXT: pand %xmm2, %xmm0
-; SSE3-NEXT: movdqa %xmm3, %xmm4
-; SSE3-NEXT: pslldq {{.*#+}} xmm4 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm4[0]
-; SSE3-NEXT: movdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,0,255]
-; SSE3-NEXT: pand %xmm5, %xmm1
-; SSE3-NEXT: pslldq {{.*#+}} xmm3 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm3[0,1]
-; SSE3-NEXT: pandn %xmm3, %xmm5
-; SSE3-NEXT: por %xmm5, %xmm1
-; SSE3-NEXT: pand %xmm2, %xmm1
-; SSE3-NEXT: pandn %xmm4, %xmm2
-; SSE3-NEXT: por %xmm2, %xmm0
-; SSE3-NEXT: por %xmm2, %xmm1
+; SSE3-NEXT: andps {{.*}}(%rip), %xmm0
+; SSE3-NEXT: andps {{.*}}(%rip), %xmm1
; SSE3-NEXT: retq
;
; SSSE3-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
; SSSE3: # BB#0:
-; SSSE3-NEXT: pshufb {{.*#+}} xmm0 = zero,xmm0[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]
-; SSSE3-NEXT: xorl %eax, %eax
-; SSSE3-NEXT: movd %eax, %xmm2
-; SSSE3-NEXT: movdqa %xmm2, %xmm3
-; SSSE3-NEXT: pshufb {{.*#+}} xmm3 = xmm3[0],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
-; SSSE3-NEXT: por %xmm3, %xmm0
-; SSSE3-NEXT: movdqa {{.*#+}} xmm3 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,128]
-; SSSE3-NEXT: pshufb %xmm3, %xmm0
-; SSSE3-NEXT: movdqa %xmm2, %xmm4
-; SSSE3-NEXT: pslldq {{.*#+}} xmm4 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm4[0]
-; SSSE3-NEXT: por %xmm4, %xmm0
-; SSSE3-NEXT: pshufb {{.*#+}} xmm1 = xmm1[0,1,2,3,4,5,6,7,8,9,10,11,12,13],zero,xmm1[15]
-; SSSE3-NEXT: pshufb {{.*#+}} xmm2 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm2[0],zero
-; SSSE3-NEXT: por %xmm2, %xmm1
-; SSSE3-NEXT: pshufb %xmm3, %xmm1
-; SSSE3-NEXT: por %xmm4, %xmm1
+; SSSE3-NEXT: andps {{.*}}(%rip), %xmm0
+; SSSE3-NEXT: andps {{.*}}(%rip), %xmm1
; SSSE3-NEXT: retq
;
; SSE41-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
@@ -574,34 +488,32 @@ define <32 x i8> @insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz(<32 x i8> %a) {
; SSE41-NEXT: xorl %eax, %eax
; SSE41-NEXT: pinsrb $0, %eax, %xmm0
; SSE41-NEXT: pinsrb $15, %eax, %xmm0
-; SSE41-NEXT: pinsrb $14, %eax, %xmm1
-; SSE41-NEXT: pinsrb $15, %eax, %xmm1
+; SSE41-NEXT: pxor %xmm2, %xmm2
+; SSE41-NEXT: pblendw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,5,6],xmm2[7]
; SSE41-NEXT: retq
;
; AVX1-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
; AVX1: # BB#0:
; AVX1-NEXT: xorl %eax, %eax
; AVX1-NEXT: vpinsrb $0, %eax, %xmm0, %xmm1
-; AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5,6,7]
-; AVX1-NEXT: vpinsrb $15, %eax, %xmm0, %xmm1
-; AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5,6,7]
-; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT: vpinsrb $14, %eax, %xmm1, %xmm1
; AVX1-NEXT: vpinsrb $15, %eax, %xmm1, %xmm1
-; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT: vblendps {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm0[4,5,6,7]
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2
+; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm2[7]
+; AVX1-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm0
; AVX1-NEXT: retq
;
; AVX2-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
; AVX2: # BB#0:
; AVX2-NEXT: xorl %eax, %eax
; AVX2-NEXT: vpinsrb $0, %eax, %xmm0, %xmm1
-; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5,6,7]
-; AVX2-NEXT: vpinsrb $15, %eax, %xmm0, %xmm1
-; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5,6,7]
-; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT: vpinsrb $14, %eax, %xmm1, %xmm1
; AVX2-NEXT: vpinsrb $15, %eax, %xmm1, %xmm1
-; AVX2-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm0
+; AVX2-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm0[4,5,6,7]
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
+; AVX2-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm2[7]
+; AVX2-NEXT: vinserti128 $1, %xmm0, %ymm1, %ymm0
; AVX2-NEXT: retq
%1 = insertelement <32 x i8> %a, i8 0, i32 0
%2 = insertelement <32 x i8> %1, i8 0, i32 15
generated by cgit v1.2.3 (git 2.25.1) at 2025-03-09 20:29:46 +0000