+
+define <4 x float> @combine_insertps1(<4 x float> %a, <4 x float> %b) {
+; SSE2-LABEL: combine_insertps1:
+; SSE2: # BB#0:
+; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[2,0],xmm0[1,0]
+; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,2],xmm0[2,3]
+; SSE2-NEXT: movaps %xmm1, %xmm0
+; SSE2-NEXT: retq
+;
+; SSSE3-LABEL: combine_insertps1:
+; SSSE3: # BB#0:
+; SSSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[2,0],xmm0[1,0]
+; SSSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,2],xmm0[2,3]
+; SSSE3-NEXT: movaps %xmm1, %xmm0
+; SSSE3-NEXT: retq
+;
+; SSE41-LABEL: combine_insertps1:
+; SSE41: # BB#0:
+; SSE41-NEXT: insertps {{.*#+}} xmm0 = xmm1[2],xmm0[1,2,3]
+; SSE41-NEXT: retq
+;
+; AVX-LABEL: combine_insertps1:
+; AVX: # BB#0:
+; AVX-NEXT: vinsertps {{.*#+}} xmm0 = xmm1[2],xmm0[1,2,3]
+; AVX-NEXT: retq
+
+ %c = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32><i32 0, i32 6, i32 2, i32 4>
+ %d = shufflevector <4 x float> %a, <4 x float> %c, <4 x i32> <i32 5, i32 1, i32 6, i32 3>
+ ret <4 x float> %d
+}
+
+define <4 x float> @combine_insertps2(<4 x float> %a, <4 x float> %b) {
+; SSE2-LABEL: combine_insertps2:
+; SSE2: # BB#0:
+; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[2,0],xmm0[0,0]
+; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[2,0],xmm0[2,3]
+; SSE2-NEXT: movaps %xmm1, %xmm0
+; SSE2-NEXT: retq
+;
+; SSSE3-LABEL: combine_insertps2:
+; SSSE3: # BB#0:
+; SSSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[2,0],xmm0[0,0]
+; SSSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[2,0],xmm0[2,3]
+; SSSE3-NEXT: movaps %xmm1, %xmm0
+; SSSE3-NEXT: retq
+;
+; SSE41-LABEL: combine_insertps2:
+; SSE41: # BB#0:
+; SSE41-NEXT: insertps {{.*#+}} xmm0 = xmm0[0],xmm1[2],xmm0[2,3]
+; SSE41-NEXT: retq
+;
+; AVX-LABEL: combine_insertps2:
+; AVX: # BB#0:
+; AVX-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0],xmm1[2],xmm0[2,3]
+; AVX-NEXT: retq
+
+ %c = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32><i32 0, i32 1, i32 6, i32 7>
+ %d = shufflevector <4 x float> %a, <4 x float> %c, <4 x i32> <i32 4, i32 6, i32 2, i32 3>
+ ret <4 x float> %d
+}
+
+define <4 x float> @combine_insertps3(<4 x float> %a, <4 x float> %b) {
+; SSE2-LABEL: combine_insertps3:
+; SSE2: # BB#0:
+; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[3,0]
+; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,2]
+; SSE2-NEXT: retq
+;
+; SSSE3-LABEL: combine_insertps3:
+; SSSE3: # BB#0:
+; SSSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[3,0]
+; SSSE3-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,2]
+; SSSE3-NEXT: retq
+;
+; SSE41-LABEL: combine_insertps3:
+; SSE41: # BB#0:
+; SSE41-NEXT: insertps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0],xmm0[3]
+; SSE41-NEXT: retq
+;
+; AVX-LABEL: combine_insertps3:
+; AVX: # BB#0:
+; AVX-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0],xmm0[3]
+; AVX-NEXT: retq
+
+ %c = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32><i32 0, i32 4, i32 2, i32 5>
+ %d = shufflevector <4 x float> %a, <4 x float> %c, <4 x i32><i32 4, i32 1, i32 5, i32 3>
+ ret <4 x float> %d
+}
+
+define <4 x float> @combine_insertps4(<4 x float> %a, <4 x float> %b) {
+; SSE2-LABEL: combine_insertps4:
+; SSE2: # BB#0:
+; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[2,0]
+; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,0]
+; SSE2-NEXT: retq
+;
+; SSSE3-LABEL: combine_insertps4:
+; SSSE3: # BB#0:
+; SSSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[2,0]
+; SSSE3-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,0]
+; SSSE3-NEXT: retq
+;
+; SSE41-LABEL: combine_insertps4:
+; SSE41: # BB#0:
+; SSE41-NEXT: insertps {{.*#+}} xmm0 = xmm0[0,1,2],xmm1[0]
+; SSE41-NEXT: retq
+;
+; AVX-LABEL: combine_insertps4:
+; AVX: # BB#0:
+; AVX-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0,1,2],xmm1[0]
+; AVX-NEXT: retq
+
+ %c = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32><i32 0, i32 4, i32 2, i32 5>
+ %d = shufflevector <4 x float> %a, <4 x float> %c, <4 x i32><i32 4, i32 1, i32 6, i32 5>
+ ret <4 x float> %d
+}
+
+define <4 x float> @PR22377(<4 x float> %a, <4 x float> %b) {
+; SSE-LABEL: PR22377:
+; SSE: # BB#0: # %entry
+; SSE-NEXT: movaps %xmm0, %xmm1
+; SSE-NEXT: shufps {{.*#+}} xmm1 = xmm1[1,3,1,3]
+; SSE-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,2,0,2]
+; SSE-NEXT: addps %xmm0, %xmm1
+; SSE-NEXT: unpcklps {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; SSE-NEXT: retq
+;
+; AVX-LABEL: PR22377:
+; AVX: # BB#0: # %entry
+; AVX-NEXT: vpermilps {{.*#+}} xmm1 = xmm0[1,3,1,3]
+; AVX-NEXT: vpermilps {{.*#+}} xmm0 = xmm0[0,2,0,2]
+; AVX-NEXT: vaddps %xmm0, %xmm1, %xmm1
+; AVX-NEXT: vunpcklps {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; AVX-NEXT: retq
+entry:
+ %s1 = shufflevector <4 x float> %a, <4 x float> undef, <4 x i32> <i32 1, i32 3, i32 1, i32 3>
+ %s2 = shufflevector <4 x float> %a, <4 x float> undef, <4 x i32> <i32 0, i32 2, i32 0, i32 2>
+ %r2 = fadd <4 x float> %s1, %s2
+ %s3 = shufflevector <4 x float> %s2, <4 x float> %r2, <4 x i32> <i32 0, i32 4, i32 1, i32 5>
+ ret <4 x float> %s3
+}
+
+define <4 x float> @PR22390(<4 x float> %a, <4 x float> %b) {
+; SSE2-LABEL: PR22390:
+; SSE2: # BB#0: # %entry
+; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[3,0,1,2]
+; SSE2-NEXT: movaps %xmm0, %xmm2
+; SSE2-NEXT: movss {{.*#+}} xmm2 = xmm1[0],xmm2[1,2,3]
+; SSE2-NEXT: addps %xmm0, %xmm2
+; SSE2-NEXT: movaps %xmm2, %xmm0
+; SSE2-NEXT: retq
+;
+; SSSE3-LABEL: PR22390:
+; SSSE3: # BB#0: # %entry
+; SSSE3-NEXT: shufps {{.*#+}} xmm0 = xmm0[3,0,1,2]
+; SSSE3-NEXT: movaps %xmm0, %xmm2
+; SSSE3-NEXT: movss {{.*#+}} xmm2 = xmm1[0],xmm2[1,2,3]
+; SSSE3-NEXT: addps %xmm0, %xmm2
+; SSSE3-NEXT: movaps %xmm2, %xmm0
+; SSSE3-NEXT: retq
+;
+; SSE41-LABEL: PR22390:
+; SSE41: # BB#0: # %entry
+; SSE41-NEXT: shufps {{.*#+}} xmm0 = xmm0[3,0,1,2]
+; SSE41-NEXT: blendps {{.*#+}} xmm1 = xmm1[0],xmm0[1,2,3]
+; SSE41-NEXT: addps %xmm1, %xmm0
+; SSE41-NEXT: retq
+;
+; AVX-LABEL: PR22390:
+; AVX: # BB#0: # %entry
+; AVX-NEXT: vpermilps {{.*#+}} xmm0 = xmm0[3,0,1,2]
+; AVX-NEXT: vblendps {{.*#+}} xmm1 = xmm1[0],xmm0[1,2,3]
+; AVX-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX-NEXT: retq
+entry:
+ %s1 = shufflevector <4 x float> %a, <4 x float> undef, <4 x i32> <i32 3, i32 0, i32 1, i32 2>
+ %s2 = shufflevector <4 x float> %s1, <4 x float> %b, <4 x i32> <i32 4, i32 1, i32 2, i32 3>
+ %r2 = fadd <4 x float> %s1, %s2
+ ret <4 x float> %r2
+}