; By including a nop call with sideeffects we can force a partial register spill of the
; relevant registers and check that the reload is correctly folded into the instruction.
+define <4 x i32> @stack_fold_extracti128(<8 x i32> %a0, <8 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_extracti128
+ ;CHECK: vextracti128 $1, {{%ymm[0-9][0-9]*}}, {{-?[0-9]*}}(%rsp) {{.*#+}} 16-byte Folded Spill
+ ; add forces execution domain
+ %1 = add <8 x i32> %a0, <i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1>
+ %2 = shufflevector <8 x i32> %1, <8 x i32> %a1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+ %3 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ ret <4 x i32> %2
+}
+
define <8 x i32> @stack_fold_inserti128(<4 x i32> %a0, <4 x i32> %a1) {
;CHECK-LABEL: stack_fold_inserti128
;CHECK: vinserti128 $1, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
}
declare <16 x i16> @llvm.x86.avx2.pavg.w(<16 x i16>, <16 x i16>) nounwind readnone
+define <4 x i32> @stack_fold_pblendd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pblendd
+ ;CHECK: vpblendd $7, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <4 x i32> %a0, <4 x i32> %a1, <4 x i32> <i32 4, i32 5, i32 6, i32 3>
+ ret <4 x i32> %2
+}
+
+define <8 x i32> @stack_fold_pblendd_ymm(<8 x i32> %a0, <8 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pblendd_ymm
+ ;CHECK: vpblendd $7, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <8 x i32> %a0, <8 x i32> %a1, <8 x i32> <i32 8, i32 9, i32 10, i32 3, i32 4, i32 5, i32 6, i32 7>
+ ret <8 x i32> %2
+}
+
+; TODO stack_fold_pblendvb
+declare <32 x i8> @llvm.x86.avx2.pblendvb(<32 x i8>, <32 x i8>, <32 x i8>) nounwind readnone
+
define <16 x i16> @stack_fold_pblendw(<16 x i16> %a0, <16 x i16> %a1) {
;CHECK-LABEL: stack_fold_pblendw
;CHECK: vpblendw $7, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
ret <16 x i16> %3
}
+define <8 x i32> @stack_fold_perm2i128(<8 x i32> %a0, <8 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_perm2i128
+ ;CHECK: vperm2i128 $33, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <8 x i32> %a0, <8 x i32> %a1, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11>
+ ; add forces execution domain
+ %3 = add <8 x i32> %2, <i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1>
+ ret <8 x i32> %3
+}
+
define <8 x i32> @stack_fold_phaddd(<8 x i32> %a0, <8 x i32> %a1) {
;CHECK-LABEL: stack_fold_phaddd
;CHECK: vphaddd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
}
declare <4 x i64> @llvm.x86.avx2.psll.q(<4 x i64>, <2 x i64>) nounwind readnone
+define <4 x i32> @stack_fold_psllvd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_psllvd
+ ;CHECK: vpsllvd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.avx2.psllv.d(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.avx2.psllv.d(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i32> @stack_fold_psllvd_ymm(<8 x i32> %a0, <8 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_psllvd_ymm
+ ;CHECK: vpsllvd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i32> @llvm.x86.avx2.psllv.d.256(<8 x i32> %a0, <8 x i32> %a1)
+ ret <8 x i32> %2
+}
+declare <8 x i32> @llvm.x86.avx2.psllv.d.256(<8 x i32>, <8 x i32>) nounwind readnone
+
+define <2 x i64> @stack_fold_psllvq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_psllvq
+ ;CHECK: vpsllvq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.avx2.psllv.q(<2 x i64> %a0, <2 x i64> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.avx2.psllv.q(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <4 x i64> @stack_fold_psllvq_ymm(<4 x i64> %a0, <4 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_psllvq_ymm
+ ;CHECK: vpsllvq {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i64> @llvm.x86.avx2.psllv.q.256(<4 x i64> %a0, <4 x i64> %a1)
+ ret <4 x i64> %2
+}
+declare <4 x i64> @llvm.x86.avx2.psllv.q.256(<4 x i64>, <4 x i64>) nounwind readnone
+
define <16 x i16> @stack_fold_psllw(<16 x i16> %a0, <8 x i16> %a1) {
;CHECK-LABEL: stack_fold_psllw
;CHECK: vpsllw {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
}
declare <8 x i32> @llvm.x86.avx2.psra.d(<8 x i32>, <4 x i32>) nounwind readnone
+define <4 x i32> @stack_fold_psravd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_psravd
+ ;CHECK: vpsravd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.avx2.psrav.d(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.avx2.psrav.d(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i32> @stack_fold_psravd_ymm(<8 x i32> %a0, <8 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_psravd_ymm
+ ;CHECK: vpsravd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i32> @llvm.x86.avx2.psrav.d.256(<8 x i32> %a0, <8 x i32> %a1)
+ ret <8 x i32> %2
+}
+declare <8 x i32> @llvm.x86.avx2.psrav.d.256(<8 x i32>, <8 x i32>) nounwind readnone
+
define <16 x i16> @stack_fold_psraw(<16 x i16> %a0, <8 x i16> %a1) {
;CHECK-LABEL: stack_fold_psraw
;CHECK: vpsraw {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
}
declare <4 x i64> @llvm.x86.avx2.psrl.q(<4 x i64>, <2 x i64>) nounwind readnone
+define <4 x i32> @stack_fold_psrlvd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_psrlvd
+ ;CHECK: vpsrlvd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.avx2.psrlv.d(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.avx2.psrlv.d(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i32> @stack_fold_psrlvd_ymm(<8 x i32> %a0, <8 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_psrlvd_ymm
+ ;CHECK: vpsrlvd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i32> @llvm.x86.avx2.psrlv.d.256(<8 x i32> %a0, <8 x i32> %a1)
+ ret <8 x i32> %2
+}
+declare <8 x i32> @llvm.x86.avx2.psrlv.d.256(<8 x i32>, <8 x i32>) nounwind readnone
+
+define <2 x i64> @stack_fold_psrlvq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_psrlvq
+ ;CHECK: vpsrlvq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.avx2.psrlv.q(<2 x i64> %a0, <2 x i64> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.avx2.psrlv.q(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <4 x i64> @stack_fold_psrlvq_ymm(<4 x i64> %a0, <4 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_psrlvq_ymm
+ ;CHECK: vpsrlvq {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i64> @llvm.x86.avx2.psrlv.q.256(<4 x i64> %a0, <4 x i64> %a1)
+ ret <4 x i64> %2
+}
+declare <4 x i64> @llvm.x86.avx2.psrlv.q.256(<4 x i64>, <4 x i64>) nounwind readnone
+
define <16 x i16> @stack_fold_psrlw(<16 x i16> %a0, <8 x i16> %a1) {
;CHECK-LABEL: stack_fold_psrlw
;CHECK: vpsrlw {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload