; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -disable-peephole -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=AVX --check-prefix=AVX1
; RUN: llc < %s -disable-peephole -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefix=AVX --check-prefix=AVX2
; RUN: llc < %s -disable-peephole -mtriple=x86_64-unknown-unknown -mattr=+avx512f | FileCheck %s --check-prefix=AVX --check-prefix=AVX512F
;
; Just one 32-bit run to make sure we do reasonable things.
; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=X86-AVX
define <4 x double> @merge_4f64_2f64_23(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_4f64_2f64_23:
; AVX: # %bb.0:
; AVX-NEXT: vmovups 32(%rdi), %ymm0
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_4f64_2f64_23:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovups 32(%eax), %ymm0
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds <2 x double>, ptr %ptr, i64 2
%ptr1 = getelementptr inbounds <2 x double>, ptr %ptr, i64 3
%val0 = load <2 x double>, ptr %ptr0
%val1 = load <2 x double>, ptr %ptr1
%res = shufflevector <2 x double> %val0, <2 x double> %val1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
ret <4 x double> %res
}
define <4 x double> @merge_4f64_2f64_2z(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_4f64_2f64_2z:
; AVX: # %bb.0:
; AVX-NEXT: vmovaps 32(%rdi), %xmm0
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_4f64_2f64_2z:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovaps 32(%eax), %xmm0
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds <2 x double>, ptr %ptr, i64 2
%val0 = load <2 x double>, ptr %ptr0
%res = shufflevector <2 x double> %val0, <2 x double> zeroinitializer, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
ret <4 x double> %res
}
define <4 x double> @merge_4f64_f64_2345(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_4f64_f64_2345:
; AVX: # %bb.0:
; AVX-NEXT: vmovups 16(%rdi), %ymm0
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_4f64_f64_2345:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovups 16(%eax), %ymm0
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds double, ptr %ptr, i64 2
%ptr1 = getelementptr inbounds double, ptr %ptr, i64 3
%ptr2 = getelementptr inbounds double, ptr %ptr, i64 4
%ptr3 = getelementptr inbounds double, ptr %ptr, i64 5
%val0 = load double, ptr %ptr0
%val1 = load double, ptr %ptr1
%val2 = load double, ptr %ptr2
%val3 = load double, ptr %ptr3
%res0 = insertelement <4 x double> undef, double %val0, i32 0
%res1 = insertelement <4 x double> %res0, double %val1, i32 1
%res2 = insertelement <4 x double> %res1, double %val2, i32 2
%res3 = insertelement <4 x double> %res2, double %val3, i32 3
ret <4 x double> %res3
}
define <4 x double> @merge_4f64_f64_3zuu(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_4f64_f64_3zuu:
; AVX: # %bb.0:
; AVX-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_4f64_f64_3zuu:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds double, ptr %ptr, i64 3
%val0 = load double, ptr %ptr0
%res0 = insertelement <4 x double> undef, double %val0, i32 0
%res1 = insertelement <4 x double> %res0, double 0.0, i32 1
ret <4 x double> %res1
}
define <4 x double> @merge_4f64_f64_34uu(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_4f64_f64_34uu:
; AVX: # %bb.0:
; AVX-NEXT: vmovups 24(%rdi), %xmm0
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_4f64_f64_34uu:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovups 24(%eax), %xmm0
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds double, ptr %ptr, i64 3
%ptr1 = getelementptr inbounds double, ptr %ptr, i64 4
%val0 = load double, ptr %ptr0
%val1 = load double, ptr %ptr1
%res0 = insertelement <4 x double> undef, double %val0, i32 0
%res1 = insertelement <4 x double> %res0, double %val1, i32 1
ret <4 x double> %res1
}
define <4 x double> @merge_4f64_f64_45zz(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_4f64_f64_45zz:
; AVX: # %bb.0:
; AVX-NEXT: vmovups 32(%rdi), %xmm0
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_4f64_f64_45zz:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovups 32(%eax), %xmm0
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds double, ptr %ptr, i64 4
%ptr1 = getelementptr inbounds double, ptr %ptr, i64 5
%val0 = load double, ptr %ptr0
%val1 = load double, ptr %ptr1
%res0 = insertelement <4 x double> zeroinitializer, double %val0, i32 0
%res1 = insertelement <4 x double> %res0, double %val1, i32 1
ret <4 x double> %res1
}
define <4 x double> @merge_4f64_f64_34z6(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_4f64_f64_34z6:
; AVX: # %bb.0:
; AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
; AVX-NEXT: vblendps {{.*#+}} ymm0 = mem[0,1,2,3],ymm0[4,5],mem[6,7]
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_4f64_f64_34z6:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
; X86-AVX-NEXT: vblendps {{.*#+}} ymm0 = mem[0,1,2,3],ymm0[4,5],mem[6,7]
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds double, ptr %ptr, i64 3
%ptr1 = getelementptr inbounds double, ptr %ptr, i64 4
%ptr3 = getelementptr inbounds double, ptr %ptr, i64 6
%val0 = load double, ptr %ptr0
%val1 = load double, ptr %ptr1
%val3 = load double, ptr %ptr3
%res0 = insertelement <4 x double> undef, double %val0, i32 0
%res1 = insertelement <4 x double> %res0, double %val1, i32 1
%res2 = insertelement <4 x double> %res1, double 0.0, i32 2
%res3 = insertelement <4 x double> %res2, double %val3, i32 3
ret <4 x double> %res3
}
define <4 x i64> @merge_4i64_2i64_3z(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_4i64_2i64_3z:
; AVX: # %bb.0:
; AVX-NEXT: vmovaps 48(%rdi), %xmm0
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_4i64_2i64_3z:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovaps 48(%eax), %xmm0
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds <2 x i64>, ptr %ptr, i64 3
%val0 = load <2 x i64>, ptr %ptr0
%res = shufflevector <2 x i64> %val0, <2 x i64> zeroinitializer, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
ret <4 x i64> %res
}
define <4 x i64> @merge_4i64_i64_1234(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_4i64_i64_1234:
; AVX: # %bb.0:
; AVX-NEXT: vmovups 8(%rdi), %ymm0
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_4i64_i64_1234:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovups 8(%eax), %ymm0
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds i64, ptr %ptr, i64 1
%ptr1 = getelementptr inbounds i64, ptr %ptr, i64 2
%ptr2 = getelementptr inbounds i64, ptr %ptr, i64 3
%ptr3 = getelementptr inbounds i64, ptr %ptr, i64 4
%val0 = load i64, ptr %ptr0
%val1 = load i64, ptr %ptr1
%val2 = load i64, ptr %ptr2
%val3 = load i64, ptr %ptr3
%res0 = insertelement <4 x i64> undef, i64 %val0, i32 0
%res1 = insertelement <4 x i64> %res0, i64 %val1, i32 1
%res2 = insertelement <4 x i64> %res1, i64 %val2, i32 2
%res3 = insertelement <4 x i64> %res2, i64 %val3, i32 3
ret <4 x i64> %res3
}
define <4 x i64> @merge_4i64_i64_1zzu(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_4i64_i64_1zzu:
; AVX: # %bb.0:
; AVX-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_4i64_i64_1zzu:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds i64, ptr %ptr, i64 1
%val0 = load i64, ptr %ptr0
%res0 = insertelement <4 x i64> undef, i64 %val0, i32 0
%res1 = insertelement <4 x i64> %res0, i64 0, i32 1
%res2 = insertelement <4 x i64> %res1, i64 0, i32 2
ret <4 x i64> %res2
}
define <4 x i64> @merge_4i64_i64_23zz(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_4i64_i64_23zz:
; AVX: # %bb.0:
; AVX-NEXT: vmovups 16(%rdi), %xmm0
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_4i64_i64_23zz:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovups 16(%eax), %xmm0
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds i64, ptr %ptr, i64 2
%ptr1 = getelementptr inbounds i64, ptr %ptr, i64 3
%val0 = load i64, ptr %ptr0
%val1 = load i64, ptr %ptr1
%res0 = insertelement <4 x i64> zeroinitializer, i64 %val0, i32 0
%res1 = insertelement <4 x i64> %res0, i64 %val1, i32 1
ret <4 x i64> %res1
}
define <8 x float> @merge_8f32_2f32_23z5(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_8f32_2f32_23z5:
; AVX: # %bb.0:
; AVX-NEXT: vmovups 16(%rdi), %xmm0
; AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
; AVX-NEXT: vmovhps {{.*#+}} xmm1 = xmm1[0,1],mem[0,1]
; AVX-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_8f32_2f32_23z5:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovups 16(%eax), %xmm0
; X86-AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
; X86-AVX-NEXT: vmovhps {{.*#+}} xmm1 = xmm1[0,1],mem[0,1]
; X86-AVX-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds <2 x float>, ptr %ptr, i64 2
%ptr1 = getelementptr inbounds <2 x float>, ptr %ptr, i64 3
%ptr3 = getelementptr inbounds <2 x float>, ptr %ptr, i64 5
%val0 = load <2 x float>, ptr %ptr0
%val1 = load <2 x float>, ptr %ptr1
%val3 = load <2 x float>, ptr %ptr3
%res01 = shufflevector <2 x float> %val0, <2 x float> %val1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
%res23 = shufflevector <2 x float> zeroinitializer, <2 x float> %val3, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
%res = shufflevector <4 x float> %res01, <4 x float> %res23, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
ret <8 x float> %res
}
define <8 x float> @merge_8f32_4f32_z2(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_8f32_4f32_z2:
; AVX: # %bb.0:
; AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
; AVX-NEXT: vinsertf128 $1, 32(%rdi), %ymm0, %ymm0
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_8f32_4f32_z2:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
; X86-AVX-NEXT: vinsertf128 $1, 32(%eax), %ymm0, %ymm0
; X86-AVX-NEXT: retl
%ptr1 = getelementptr inbounds <4 x float>, ptr %ptr, i64 2
%val1 = load <4 x float>, ptr %ptr1
%res = shufflevector <4 x float> zeroinitializer, <4 x float> %val1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
ret <8 x float> %res
}
define <8 x float> @merge_8f32_f32_12zzuuzz(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_8f32_f32_12zzuuzz:
; AVX: # %bb.0:
; AVX-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_8f32_f32_12zzuuzz:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds float, ptr %ptr, i64 1
%ptr1 = getelementptr inbounds float, ptr %ptr, i64 2
%val0 = load float, ptr %ptr0
%val1 = load float, ptr %ptr1
%res0 = insertelement <8 x float> undef, float %val0, i32 0
%res1 = insertelement <8 x float> %res0, float %val1, i32 1
%res2 = insertelement <8 x float> %res1, float 0.0, i32 2
%res3 = insertelement <8 x float> %res2, float 0.0, i32 3
%res6 = insertelement <8 x float> %res3, float 0.0, i32 6
%res7 = insertelement <8 x float> %res6, float 0.0, i32 7
ret <8 x float> %res7
}
define <8 x float> @merge_8f32_f32_1u3u5zu8(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_8f32_f32_1u3u5zu8:
; AVX: # %bb.0:
; AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
; AVX-NEXT: vblendps {{.*#+}} ymm0 = mem[0,1,2,3,4],ymm0[5],mem[6,7]
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_8f32_f32_1u3u5zu8:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
; X86-AVX-NEXT: vblendps {{.*#+}} ymm0 = mem[0,1,2,3,4],ymm0[5],mem[6,7]
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds float, ptr %ptr, i64 1
%ptr2 = getelementptr inbounds float, ptr %ptr, i64 3
%ptr4 = getelementptr inbounds float, ptr %ptr, i64 5
%ptr7 = getelementptr inbounds float, ptr %ptr, i64 8
%val0 = load float, ptr %ptr0
%val2 = load float, ptr %ptr2
%val4 = load float, ptr %ptr4
%val7 = load float, ptr %ptr7
%res0 = insertelement <8 x float> undef, float %val0, i32 0
%res2 = insertelement <8 x float> %res0, float %val2, i32 2
%res4 = insertelement <8 x float> %res2, float %val4, i32 4
%res5 = insertelement <8 x float> %res4, float 0.0, i32 5
%res7 = insertelement <8 x float> %res5, float %val7, i32 7
ret <8 x float> %res7
}
define <8 x i32> @merge_8i32_4i32_z3(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_8i32_4i32_z3:
; AVX: # %bb.0:
; AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
; AVX-NEXT: vinsertf128 $1, 48(%rdi), %ymm0, %ymm0
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_8i32_4i32_z3:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
; X86-AVX-NEXT: vinsertf128 $1, 48(%eax), %ymm0, %ymm0
; X86-AVX-NEXT: retl
%ptr1 = getelementptr inbounds <4 x i32>, ptr %ptr, i64 3
%val1 = load <4 x i32>, ptr %ptr1
%res = shufflevector <4 x i32> zeroinitializer, <4 x i32> %val1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
ret <8 x i32> %res
}
define <8 x i32> @merge_8i32_i32_56zz9uzz(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_8i32_i32_56zz9uzz:
; AVX: # %bb.0:
; AVX-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; AVX-NEXT: vmovss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; AVX-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_8i32_i32_56zz9uzz:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; X86-AVX-NEXT: vmovss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; X86-AVX-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds i32, ptr %ptr, i64 5
%ptr1 = getelementptr inbounds i32, ptr %ptr, i64 6
%ptr4 = getelementptr inbounds i32, ptr %ptr, i64 9
%val0 = load i32, ptr %ptr0
%val1 = load i32, ptr %ptr1
%val4 = load i32, ptr %ptr4
%res0 = insertelement <8 x i32> undef, i32 %val0, i32 0
%res1 = insertelement <8 x i32> %res0, i32 %val1, i32 1
%res2 = insertelement <8 x i32> %res1, i32 0, i32 2
%res3 = insertelement <8 x i32> %res2, i32 0, i32 3
%res4 = insertelement <8 x i32> %res3, i32 %val4, i32 4
%res6 = insertelement <8 x i32> %res4, i32 0, i32 6
%res7 = insertelement <8 x i32> %res6, i32 0, i32 7
ret <8 x i32> %res7
}
define <8 x i32> @merge_8i32_i32_1u3u5zu8(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_8i32_i32_1u3u5zu8:
; AVX: # %bb.0:
; AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
; AVX-NEXT: vblendps {{.*#+}} ymm0 = mem[0,1,2,3,4],ymm0[5],mem[6,7]
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_8i32_i32_1u3u5zu8:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
; X86-AVX-NEXT: vblendps {{.*#+}} ymm0 = mem[0,1,2,3,4],ymm0[5],mem[6,7]
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds i32, ptr %ptr, i64 1
%ptr2 = getelementptr inbounds i32, ptr %ptr, i64 3
%ptr4 = getelementptr inbounds i32, ptr %ptr, i64 5
%ptr7 = getelementptr inbounds i32, ptr %ptr, i64 8
%val0 = load i32, ptr %ptr0
%val2 = load i32, ptr %ptr2
%val4 = load i32, ptr %ptr4
%val7 = load i32, ptr %ptr7
%res0 = insertelement <8 x i32> undef, i32 %val0, i32 0
%res2 = insertelement <8 x i32> %res0, i32 %val2, i32 2
%res4 = insertelement <8 x i32> %res2, i32 %val4, i32 4
%res5 = insertelement <8 x i32> %res4, i32 0, i32 5
%res7 = insertelement <8 x i32> %res5, i32 %val7, i32 7
ret <8 x i32> %res7
}
define <16 x i16> @merge_16i16_i16_89zzzuuuuuuuuuuuz(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_16i16_i16_89zzzuuuuuuuuuuuz:
; AVX: # %bb.0:
; AVX-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_16i16_i16_89zzzuuuuuuuuuuuz:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds i16, ptr %ptr, i64 8
%ptr1 = getelementptr inbounds i16, ptr %ptr, i64 9
%val0 = load i16, ptr %ptr0
%val1 = load i16, ptr %ptr1
%res0 = insertelement <16 x i16> undef, i16 %val0, i16 0
%res1 = insertelement <16 x i16> %res0, i16 %val1, i16 1
%res2 = insertelement <16 x i16> %res1, i16 0, i16 2
%res3 = insertelement <16 x i16> %res2, i16 0, i16 3
%res4 = insertelement <16 x i16> %res3, i16 0, i16 4
%resF = insertelement <16 x i16> %res4, i16 0, i16 15
ret <16 x i16> %resF
}
define <16 x i16> @merge_16i16_i16_45u7uuuuuuuuuuuu(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_16i16_i16_45u7uuuuuuuuuuuu:
; AVX: # %bb.0:
; AVX-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_16i16_i16_45u7uuuuuuuuuuuu:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds i16, ptr %ptr, i64 4
%ptr1 = getelementptr inbounds i16, ptr %ptr, i64 5
%ptr3 = getelementptr inbounds i16, ptr %ptr, i64 7
%val0 = load i16, ptr %ptr0
%val1 = load i16, ptr %ptr1
%val3 = load i16, ptr %ptr3
%res0 = insertelement <16 x i16> undef, i16 %val0, i16 0
%res1 = insertelement <16 x i16> %res0, i16 %val1, i16 1
%res3 = insertelement <16 x i16> %res1, i16 %val3, i16 3
ret <16 x i16> %res3
}
define <16 x i16> @merge_16i16_i16_0uu3uuuuuuuuCuEF(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_16i16_i16_0uu3uuuuuuuuCuEF:
; AVX: # %bb.0:
; AVX-NEXT: vmovups (%rdi), %ymm0
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_16i16_i16_0uu3uuuuuuuuCuEF:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovups (%eax), %ymm0
; X86-AVX-NEXT: retl
%ptr3 = getelementptr inbounds i16, ptr %ptr, i64 3
%ptrC = getelementptr inbounds i16, ptr %ptr, i64 12
%ptrE = getelementptr inbounds i16, ptr %ptr, i64 14
%ptrF = getelementptr inbounds i16, ptr %ptr, i64 15
%val0 = load i16, ptr %ptr
%val3 = load i16, ptr %ptr3
%valC = load i16, ptr %ptrC
%valE = load i16, ptr %ptrE
%valF = load i16, ptr %ptrF
%res0 = insertelement <16 x i16> undef, i16 %val0, i16 0
%res3 = insertelement <16 x i16> %res0, i16 %val3, i16 3
%resC = insertelement <16 x i16> %res3, i16 %valC, i16 12
%resE = insertelement <16 x i16> %resC, i16 %valE, i16 14
%resF = insertelement <16 x i16> %resE, i16 %valF, i16 15
ret <16 x i16> %resF
}
define <16 x i16> @merge_16i16_i16_0uu3zzuuuuuzCuEF(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_16i16_i16_0uu3zzuuuuuzCuEF:
; AVX: # %bb.0:
; AVX-NEXT: vmovups (%rdi), %ymm0
; AVX-NEXT: vandps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm0, %ymm0
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_16i16_i16_0uu3zzuuuuuzCuEF:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovups (%eax), %ymm0
; X86-AVX-NEXT: vandps {{\.?LCPI[0-9]+_[0-9]+}}, %ymm0, %ymm0
; X86-AVX-NEXT: retl
%ptr3 = getelementptr inbounds i16, ptr %ptr, i64 3
%ptrC = getelementptr inbounds i16, ptr %ptr, i64 12
%ptrE = getelementptr inbounds i16, ptr %ptr, i64 14
%ptrF = getelementptr inbounds i16, ptr %ptr, i64 15
%val0 = load i16, ptr %ptr
%val3 = load i16, ptr %ptr3
%valC = load i16, ptr %ptrC
%valE = load i16, ptr %ptrE
%valF = load i16, ptr %ptrF
%res0 = insertelement <16 x i16> undef, i16 %val0, i16 0
%res3 = insertelement <16 x i16> %res0, i16 %val3, i16 3
%res4 = insertelement <16 x i16> %res3, i16 0, i16 4
%res5 = insertelement <16 x i16> %res4, i16 0, i16 5
%resC = insertelement <16 x i16> %res5, i16 %valC, i16 12
%resD = insertelement <16 x i16> %resC, i16 0, i16 13
%resE = insertelement <16 x i16> %resD, i16 %valE, i16 14
%resF = insertelement <16 x i16> %resE, i16 %valF, i16 15
ret <16 x i16> %resF
}
define <32 x i8> @merge_32i8_i8_45u7uuuuuuuuuuuuuuuuuuuuuuuuuuuu(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_32i8_i8_45u7uuuuuuuuuuuuuuuuuuuuuuuuuuuu:
; AVX: # %bb.0:
; AVX-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_32i8_i8_45u7uuuuuuuuuuuuuuuuuuuuuuuuuuuu:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds i8, ptr %ptr, i64 4
%ptr1 = getelementptr inbounds i8, ptr %ptr, i64 5
%ptr3 = getelementptr inbounds i8, ptr %ptr, i64 7
%val0 = load i8, ptr %ptr0
%val1 = load i8, ptr %ptr1
%val3 = load i8, ptr %ptr3
%res0 = insertelement <32 x i8> undef, i8 %val0, i8 0
%res1 = insertelement <32 x i8> %res0, i8 %val1, i8 1
%res3 = insertelement <32 x i8> %res1, i8 %val3, i8 3
ret <32 x i8> %res3
}
define <32 x i8> @merge_32i8_i8_23u5uuuuuuuuuuzzzzuuuuuuuuuuuuuu(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_32i8_i8_23u5uuuuuuuuuuzzzzuuuuuuuuuuuuuu:
; AVX: # %bb.0:
; AVX-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_32i8_i8_23u5uuuuuuuuuuzzzzuuuuuuuuuuuuuu:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds i8, ptr %ptr, i64 2
%ptr1 = getelementptr inbounds i8, ptr %ptr, i64 3
%ptr3 = getelementptr inbounds i8, ptr %ptr, i64 5
%val0 = load i8, ptr %ptr0
%val1 = load i8, ptr %ptr1
%val3 = load i8, ptr %ptr3
%res0 = insertelement <32 x i8> undef, i8 %val0, i8 0
%res1 = insertelement <32 x i8> %res0, i8 %val1, i8 1
%res3 = insertelement <32 x i8> %res1, i8 %val3, i8 3
%resE = insertelement <32 x i8> %res3, i8 0, i8 14
%resF = insertelement <32 x i8> %resE, i8 0, i8 15
%resG = insertelement <32 x i8> %resF, i8 0, i8 16
%resH = insertelement <32 x i8> %resG, i8 0, i8 17
ret <32 x i8> %resH
}
;
; consecutive loads including any/all volatiles may not be combined
;
define <4 x double> @merge_4f64_f64_34uz_volatile(ptr %ptr) nounwind uwtable noinline ssp {
; AVX-LABEL: merge_4f64_f64_34uz_volatile:
; AVX: # %bb.0:
; AVX-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; AVX-NEXT: vmovhps {{.*#+}} xmm0 = xmm0[0,1],mem[0,1]
; AVX-NEXT: retq
;
; X86-AVX-LABEL: merge_4f64_f64_34uz_volatile:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; X86-AVX-NEXT: vmovhps {{.*#+}} xmm0 = xmm0[0,1],mem[0,1]
; X86-AVX-NEXT: retl
%ptr0 = getelementptr inbounds double, ptr %ptr, i64 3
%ptr1 = getelementptr inbounds double, ptr %ptr, i64 4
%val0 = load volatile double, ptr %ptr0
%val1 = load volatile double, ptr %ptr1
%res0 = insertelement <4 x double> undef, double %val0, i32 0
%res1 = insertelement <4 x double> %res0, double %val1, i32 1
%res3 = insertelement <4 x double> %res1, double 0.0, i32 3
ret <4 x double> %res3
}
define <16 x i16> @merge_16i16_i16_0uu3zzuuuuuzCuEF_volatile(ptr %ptr) nounwind uwtable noinline ssp {
; AVX1-LABEL: merge_16i16_i16_0uu3zzuuuuuzCuEF_volatile:
; AVX1: # %bb.0:
; AVX1-NEXT: movzwl (%rdi), %eax
; AVX1-NEXT: vpxor %xmm0, %xmm0, %xmm0
; AVX1-NEXT: vpinsrw $4, 24(%rdi), %xmm0, %xmm0
; AVX1-NEXT: vpinsrw $6, 28(%rdi), %xmm0, %xmm0
; AVX1-NEXT: vpinsrw $7, 30(%rdi), %xmm0, %xmm0
; AVX1-NEXT: vmovd %eax, %xmm1
; AVX1-NEXT: vpinsrw $3, 6(%rdi), %xmm1, %xmm1
; AVX1-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm0
; AVX1-NEXT: retq
;
; AVX2-LABEL: merge_16i16_i16_0uu3zzuuuuuzCuEF_volatile:
; AVX2: # %bb.0:
; AVX2-NEXT: movzwl (%rdi), %eax
; AVX2-NEXT: vpxor %xmm0, %xmm0, %xmm0
; AVX2-NEXT: vpinsrw $4, 24(%rdi), %xmm0, %xmm0
; AVX2-NEXT: vpinsrw $6, 28(%rdi), %xmm0, %xmm0
; AVX2-NEXT: vpinsrw $7, 30(%rdi), %xmm0, %xmm0
; AVX2-NEXT: vmovd %eax, %xmm1
; AVX2-NEXT: vpinsrw $3, 6(%rdi), %xmm1, %xmm1
; AVX2-NEXT: vinserti128 $1, %xmm0, %ymm1, %ymm0
; AVX2-NEXT: retq
;
; AVX512F-LABEL: merge_16i16_i16_0uu3zzuuuuuzCuEF_volatile:
; AVX512F: # %bb.0:
; AVX512F-NEXT: movzwl (%rdi), %eax
; AVX512F-NEXT: vpxor %xmm0, %xmm0, %xmm0
; AVX512F-NEXT: vpinsrw $4, 24(%rdi), %xmm0, %xmm0
; AVX512F-NEXT: vpinsrw $6, 28(%rdi), %xmm0, %xmm0
; AVX512F-NEXT: vpinsrw $7, 30(%rdi), %xmm0, %xmm0
; AVX512F-NEXT: vmovd %eax, %xmm1
; AVX512F-NEXT: vpinsrw $3, 6(%rdi), %xmm1, %xmm1
; AVX512F-NEXT: vinserti128 $1, %xmm0, %ymm1, %ymm0
; AVX512F-NEXT: retq
;
; X86-AVX-LABEL: merge_16i16_i16_0uu3zzuuuuuzCuEF_volatile:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: movzwl (%eax), %ecx
; X86-AVX-NEXT: vpxor %xmm0, %xmm0, %xmm0
; X86-AVX-NEXT: vpinsrw $4, 24(%eax), %xmm0, %xmm0
; X86-AVX-NEXT: vpinsrw $6, 28(%eax), %xmm0, %xmm0
; X86-AVX-NEXT: vpinsrw $7, 30(%eax), %xmm0, %xmm0
; X86-AVX-NEXT: vmovd %ecx, %xmm1
; X86-AVX-NEXT: vpinsrw $3, 6(%eax), %xmm1, %xmm1
; X86-AVX-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm0
; X86-AVX-NEXT: retl
%ptr3 = getelementptr inbounds i16, ptr %ptr, i64 3
%ptrC = getelementptr inbounds i16, ptr %ptr, i64 12
%ptrE = getelementptr inbounds i16, ptr %ptr, i64 14
%ptrF = getelementptr inbounds i16, ptr %ptr, i64 15
%val0 = load volatile i16, ptr %ptr
%val3 = load i16, ptr %ptr3
%valC = load i16, ptr %ptrC
%valE = load i16, ptr %ptrE
%valF = load volatile i16, ptr %ptrF
%res0 = insertelement <16 x i16> undef, i16 %val0, i16 0
%res3 = insertelement <16 x i16> %res0, i16 %val3, i16 3
%res4 = insertelement <16 x i16> %res3, i16 0, i16 4
%res5 = insertelement <16 x i16> %res4, i16 0, i16 5
%resC = insertelement <16 x i16> %res5, i16 %valC, i16 12
%resD = insertelement <16 x i16> %resC, i16 0, i16 13
%resE = insertelement <16 x i16> %resD, i16 %valE, i16 14
%resF = insertelement <16 x i16> %resE, i16 %valF, i16 15
ret <16 x i16> %resF
}
;
; Volatile tests.
;
@l = external dso_local global <32 x i8>, align 32
define <2 x i8> @PR42846(ptr %j, <2 x i8> %k) {
; AVX-LABEL: PR42846:
; AVX: # %bb.0:
; AVX-NEXT: vmovdqa l(%rip), %ymm0
; AVX-NEXT: vpextrw $0, %xmm0, (%rdi)
; AVX-NEXT: # kill: def $xmm0 killed $xmm0 killed $ymm0
; AVX-NEXT: vzeroupper
; AVX-NEXT: retq
;
; X86-AVX-LABEL: PR42846:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-AVX-NEXT: vmovdqa l, %ymm0
; X86-AVX-NEXT: vpextrw $0, %xmm0, (%eax)
; X86-AVX-NEXT: # kill: def $xmm0 killed $xmm0 killed $ymm0
; X86-AVX-NEXT: vzeroupper
; X86-AVX-NEXT: retl
%t0 = load volatile <32 x i8>, ptr @l, align 32
%shuffle = shufflevector <32 x i8> %t0, <32 x i8> undef, <2 x i32> <i32 0, i32 1>
store <2 x i8> %shuffle, ptr %j, align 2
ret <2 x i8> %shuffle
}