[llvm] df1033d - [NFC][X86][Codegen] Add exhaustive test coverage for PR50971
Roman Lebedev via llvm-commits
llvm-commits at lists.llvm.org
Wed Aug 18 05:03:29 PDT 2021
Author: Roman Lebedev
Date: 2021-08-18T15:02:52+03:00
New Revision: df1033d8dba13d54965fa17a89b3d7c9f1f3b04f
URL: https://github.com/llvm/llvm-project/commit/df1033d8dba13d54965fa17a89b3d7c9f1f3b04f
DIFF: https://github.com/llvm/llvm-project/commit/df1033d8dba13d54965fa17a89b3d7c9f1f3b04f.diff
LOG: [NFC][X86][Codegen] Add exhaustive test coverage for PR50971
Produced via https://godbolt.org/z/5hEdGY5x3
Added:
llvm/test/CodeGen/X86/copy-low-subvec-elt-to-high-subvec-elt.ll
Modified:
Removed:
################################################################################
diff --git a/llvm/test/CodeGen/X86/copy-low-subvec-elt-to-high-subvec-elt.ll b/llvm/test/CodeGen/X86/copy-low-subvec-elt-to-high-subvec-elt.ll
new file mode 100644
index 000000000000..39fed0690ece
--- /dev/null
+++ b/llvm/test/CodeGen/X86/copy-low-subvec-elt-to-high-subvec-elt.ll
@@ -0,0 +1,866 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefixes=CHECK,CHECK-SLOW
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2,+fast-variable-crosslane-shuffle | FileCheck %s --check-prefixes=CHECK,CHECK-FAST
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=CHECK,CHECK-SLOW
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2,+fast-variable-crosslane-shuffle,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=CHECK,CHECK-FAST
+
+; Shuffle lowest element of some subvector into highest element of some subvector.
+; Mainly this is testing how well we avoid subvector extractions/insertions.
+; https://bugs.llvm.org/show_bug.cgi?id=50971
+
+define <2 x double> @vec128_eltty_double_source_subvec_0_target_subvec_mask_1_unary(<2 x double> %x) nounwind {
+; CHECK-LABEL: vec128_eltty_double_source_subvec_0_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovddup {{.*#+}} xmm0 = xmm0[0,0]
+; CHECK-NEXT: retq
+ %r = shufflevector <2 x double> %x, <2 x double> poison, <2 x i32> zeroinitializer
+ ret <2 x double> %r
+}
+
+define <2 x double> @vec128_eltty_double_source_subvec_0_target_subvec_mask_1_binary(<2 x double> %x, <2 x double> %y) nounwind {
+; CHECK-LABEL: vec128_eltty_double_source_subvec_0_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovlhps {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; CHECK-NEXT: retq
+ %r = shufflevector <2 x double> %x, <2 x double> %y, <2 x i32> <i32 0, i32 2>
+ ret <2 x double> %r
+}
+
+define <2 x i64> @vec128_eltty_i64_source_subvec_0_target_subvec_mask_1_unary(<2 x i64> %x) nounwind {
+; CHECK-LABEL: vec128_eltty_i64_source_subvec_0_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovddup {{.*#+}} xmm0 = xmm0[0,0]
+; CHECK-NEXT: retq
+ %r = shufflevector <2 x i64> %x, <2 x i64> poison, <2 x i32> zeroinitializer
+ ret <2 x i64> %r
+}
+
+define <2 x i64> @vec128_eltty_i64_source_subvec_0_target_subvec_mask_1_binary(<2 x i64> %x, <2 x i64> %y) nounwind {
+; CHECK-LABEL: vec128_eltty_i64_source_subvec_0_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovlhps {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; CHECK-NEXT: retq
+ %r = shufflevector <2 x i64> %x, <2 x i64> %y, <2 x i32> <i32 0, i32 2>
+ ret <2 x i64> %r
+}
+
+define <4 x float> @vec128_eltty_float_source_subvec_0_target_subvec_mask_1_unary(<4 x float> %x) nounwind {
+; CHECK-LABEL: vec128_eltty_float_source_subvec_0_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermilps {{.*#+}} xmm0 = xmm0[0,1,2,0]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x float> %x, <4 x float> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 0>
+ ret <4 x float> %r
+}
+
+define <4 x float> @vec128_eltty_float_source_subvec_0_target_subvec_mask_1_binary(<4 x float> %x, <4 x float> %y) nounwind {
+; CHECK-LABEL: vec128_eltty_float_source_subvec_0_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0,1,2],xmm1[0]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x float> %x, <4 x float> %y, <4 x i32> <i32 0, i32 1, i32 2, i32 4>
+ ret <4 x float> %r
+}
+
+define <4 x i32> @vec128_eltty_i32_source_subvec_0_target_subvec_mask_1_unary(<4 x i32> %x) nounwind {
+; CHECK-LABEL: vec128_eltty_i32_source_subvec_0_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermilps {{.*#+}} xmm0 = xmm0[0,1,2,0]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x i32> %x, <4 x i32> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 0>
+ ret <4 x i32> %r
+}
+
+define <4 x i32> @vec128_eltty_i32_source_subvec_0_target_subvec_mask_1_binary(<4 x i32> %x, <4 x i32> %y) nounwind {
+; CHECK-LABEL: vec128_eltty_i32_source_subvec_0_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastss %xmm1, %xmm1
+; CHECK-NEXT: vblendps {{.*#+}} xmm0 = xmm0[0,1,2],xmm1[3]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x i32> %x, <4 x i32> %y, <4 x i32> <i32 0, i32 1, i32 2, i32 4>
+ ret <4 x i32> %r
+}
+
+define <8 x i16> @vec128_eltty_i16_source_subvec_0_target_subvec_mask_1_unary(<8 x i16> %x) nounwind {
+; CHECK-LABEL: vec128_eltty_i16_source_subvec_0_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13,0,1]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x i16> %x, <8 x i16> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 0>
+ ret <8 x i16> %r
+}
+
+define <8 x i16> @vec128_eltty_i16_source_subvec_0_target_subvec_mask_1_binary(<8 x i16> %x, <8 x i16> %y) nounwind {
+; CHECK-LABEL: vec128_eltty_i16_source_subvec_0_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpbroadcastw %xmm1, %xmm1
+; CHECK-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm1[7]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x i16> %x, <8 x i16> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 8>
+ ret <8 x i16> %r
+}
+
+define <16 x i8> @vec128_eltty_i8_source_subvec_0_target_subvec_mask_1_unary(<16 x i8> %x) nounwind {
+; CHECK-LABEL: vec128_eltty_i8_source_subvec_0_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,0]
+; CHECK-NEXT: retq
+ %r = shufflevector <16 x i8> %x, <16 x i8> poison, <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 0>
+ ret <16 x i8> %r
+}
+
+define <16 x i8> @vec128_eltty_i8_source_subvec_0_target_subvec_mask_1_binary(<16 x i8> %x, <16 x i8> %y) nounwind {
+; CHECK-LABEL: vec128_eltty_i8_source_subvec_0_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0]
+; CHECK-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14],zero
+; CHECK-NEXT: vpor %xmm1, %xmm0, %xmm0
+; CHECK-NEXT: retq
+ %r = shufflevector <16 x i8> %x, <16 x i8> %y, <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 16>
+ ret <16 x i8> %r
+}
+
+define <4 x double> @vec256_eltty_double_source_subvec_0_target_subvec_mask_1_unary(<4 x double> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_double_source_subvec_0_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[0,0,2,3]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x double> %x, <4 x double> poison, <4 x i32> <i32 0, i32 0, i32 2, i32 3>
+ ret <4 x double> %r
+}
+
+define <4 x double> @vec256_eltty_double_source_subvec_0_target_subvec_mask_1_binary(<4 x double> %x, <4 x double> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_double_source_subvec_0_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovddup {{.*#+}} xmm1 = xmm1[0,0]
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3],ymm0[4,5,6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x double> %x, <4 x double> %y, <4 x i32> <i32 0, i32 4, i32 2, i32 3>
+ ret <4 x double> %r
+}
+
+define <4 x double> @vec256_eltty_double_source_subvec_0_target_subvec_mask_2_unary(<4 x double> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_double_source_subvec_0_target_subvec_mask_2_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,1,2,0]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x double> %x, <4 x double> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 0>
+ ret <4 x double> %r
+}
+
+define <4 x double> @vec256_eltty_double_source_subvec_0_target_subvec_mask_2_binary(<4 x double> %x, <4 x double> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_double_source_subvec_0_target_subvec_mask_2_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm1
+; CHECK-NEXT: vshufpd {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[2]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x double> %x, <4 x double> %y, <4 x i32> <i32 0, i32 1, i32 2, i32 4>
+ ret <4 x double> %r
+}
+
+define <4 x double> @vec256_eltty_double_source_subvec_0_target_subvec_mask_3_unary(<4 x double> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_double_source_subvec_0_target_subvec_mask_3_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,0,2,0]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x double> %x, <4 x double> poison, <4 x i32> <i32 0, i32 0, i32 2, i32 0>
+ ret <4 x double> %r
+}
+
+define <4 x double> @vec256_eltty_double_source_subvec_0_target_subvec_mask_3_binary(<4 x double> %x, <4 x double> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_double_source_subvec_0_target_subvec_mask_3_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastsd %xmm1, %ymm1
+; CHECK-NEXT: vunpcklpd {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[2],ymm1[2]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x double> %x, <4 x double> %y, <4 x i32> <i32 0, i32 4, i32 2, i32 4>
+ ret <4 x double> %r
+}
+
+define <4 x double> @vec256_eltty_double_source_subvec_1_target_subvec_mask_1_unary(<4 x double> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_double_source_subvec_1_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,2,2,3]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x double> %x, <4 x double> poison, <4 x i32> <i32 0, i32 2, i32 2, i32 3>
+ ret <4 x double> %r
+}
+
+define <4 x double> @vec256_eltty_double_source_subvec_1_target_subvec_mask_1_binary(<4 x double> %x, <4 x double> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_double_source_subvec_1_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vperm2f128 {{.*#+}} ymm1 = ymm1[2,3],ymm0[2,3]
+; CHECK-NEXT: vshufpd {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[2],ymm1[3]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x double> %x, <4 x double> %y, <4 x i32> <i32 0, i32 6, i32 2, i32 3>
+ ret <4 x double> %r
+}
+
+define <4 x double> @vec256_eltty_double_source_subvec_1_target_subvec_mask_2_unary(<4 x double> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_double_source_subvec_1_target_subvec_mask_2_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[0,1,2,2]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x double> %x, <4 x double> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 2>
+ ret <4 x double> %r
+}
+
+define <4 x double> @vec256_eltty_double_source_subvec_1_target_subvec_mask_2_binary(<4 x double> %x, <4 x double> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_double_source_subvec_1_target_subvec_mask_2_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovddup {{.*#+}} ymm1 = ymm1[0,0,2,2]
+; CHECK-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1,2],ymm1[3]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x double> %x, <4 x double> %y, <4 x i32> <i32 0, i32 1, i32 2, i32 6>
+ ret <4 x double> %r
+}
+
+define <4 x double> @vec256_eltty_double_source_subvec_1_target_subvec_mask_3_unary(<4 x double> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_double_source_subvec_1_target_subvec_mask_3_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,2,2,2]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x double> %x, <4 x double> poison, <4 x i32> <i32 0, i32 2, i32 2, i32 2>
+ ret <4 x double> %r
+}
+
+define <4 x double> @vec256_eltty_double_source_subvec_1_target_subvec_mask_3_binary(<4 x double> %x, <4 x double> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_double_source_subvec_1_target_subvec_mask_3_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[2,2,2,2]
+; CHECK-NEXT: vunpcklpd {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[2],ymm1[2]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x double> %x, <4 x double> %y, <4 x i32> <i32 0, i32 6, i32 2, i32 6>
+ ret <4 x double> %r
+}
+
+define <4 x i64> @vec256_eltty_i64_source_subvec_0_target_subvec_mask_1_unary(<4 x i64> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i64_source_subvec_0_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,0,2,3]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x i64> %x, <4 x i64> poison, <4 x i32> <i32 0, i32 0, i32 2, i32 3>
+ ret <4 x i64> %r
+}
+
+define <4 x i64> @vec256_eltty_i64_source_subvec_0_target_subvec_mask_1_binary(<4 x i64> %x, <4 x i64> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i64_source_subvec_0_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovddup {{.*#+}} xmm1 = xmm1[0,0]
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3],ymm0[4,5,6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x i64> %x, <4 x i64> %y, <4 x i32> <i32 0, i32 4, i32 2, i32 3>
+ ret <4 x i64> %r
+}
+
+define <4 x i64> @vec256_eltty_i64_source_subvec_0_target_subvec_mask_2_unary(<4 x i64> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i64_source_subvec_0_target_subvec_mask_2_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,1,2,0]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x i64> %x, <4 x i64> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 0>
+ ret <4 x i64> %r
+}
+
+define <4 x i64> @vec256_eltty_i64_source_subvec_0_target_subvec_mask_2_binary(<4 x i64> %x, <4 x i64> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i64_source_subvec_0_target_subvec_mask_2_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastsd %xmm1, %ymm1
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5],ymm1[6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x i64> %x, <4 x i64> %y, <4 x i32> <i32 0, i32 1, i32 2, i32 4>
+ ret <4 x i64> %r
+}
+
+define <4 x i64> @vec256_eltty_i64_source_subvec_0_target_subvec_mask_3_unary(<4 x i64> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i64_source_subvec_0_target_subvec_mask_3_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,0,2,0]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x i64> %x, <4 x i64> poison, <4 x i32> <i32 0, i32 0, i32 2, i32 0>
+ ret <4 x i64> %r
+}
+
+define <4 x i64> @vec256_eltty_i64_source_subvec_0_target_subvec_mask_3_binary(<4 x i64> %x, <4 x i64> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i64_source_subvec_0_target_subvec_mask_3_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastsd %xmm1, %ymm1
+; CHECK-NEXT: vunpcklpd {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[2],ymm1[2]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x i64> %x, <4 x i64> %y, <4 x i32> <i32 0, i32 4, i32 2, i32 4>
+ ret <4 x i64> %r
+}
+
+define <4 x i64> @vec256_eltty_i64_source_subvec_1_target_subvec_mask_1_unary(<4 x i64> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i64_source_subvec_1_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,2,2,3]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x i64> %x, <4 x i64> poison, <4 x i32> <i32 0, i32 2, i32 2, i32 3>
+ ret <4 x i64> %r
+}
+
+define <4 x i64> @vec256_eltty_i64_source_subvec_1_target_subvec_mask_1_binary(<4 x i64> %x, <4 x i64> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i64_source_subvec_1_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[2,2,2,2]
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3],ymm0[4,5,6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x i64> %x, <4 x i64> %y, <4 x i32> <i32 0, i32 6, i32 2, i32 3>
+ ret <4 x i64> %r
+}
+
+define <4 x i64> @vec256_eltty_i64_source_subvec_1_target_subvec_mask_2_unary(<4 x i64> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i64_source_subvec_1_target_subvec_mask_2_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,1,2,2]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x i64> %x, <4 x i64> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 2>
+ ret <4 x i64> %r
+}
+
+define <4 x i64> @vec256_eltty_i64_source_subvec_1_target_subvec_mask_2_binary(<4 x i64> %x, <4 x i64> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i64_source_subvec_1_target_subvec_mask_2_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermilps {{.*#+}} ymm1 = ymm1[0,1,0,1,4,5,4,5]
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5],ymm1[6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x i64> %x, <4 x i64> %y, <4 x i32> <i32 0, i32 1, i32 2, i32 6>
+ ret <4 x i64> %r
+}
+
+define <4 x i64> @vec256_eltty_i64_source_subvec_1_target_subvec_mask_3_unary(<4 x i64> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i64_source_subvec_1_target_subvec_mask_3_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,2,2,2]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x i64> %x, <4 x i64> poison, <4 x i32> <i32 0, i32 2, i32 2, i32 2>
+ ret <4 x i64> %r
+}
+
+define <4 x i64> @vec256_eltty_i64_source_subvec_1_target_subvec_mask_3_binary(<4 x i64> %x, <4 x i64> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i64_source_subvec_1_target_subvec_mask_3_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[2,2,2,2]
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3],ymm0[4,5],ymm1[6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <4 x i64> %x, <4 x i64> %y, <4 x i32> <i32 0, i32 6, i32 2, i32 6>
+ ret <4 x i64> %r
+}
+
+define <8 x float> @vec256_eltty_float_source_subvec_0_target_subvec_mask_1_unary(<8 x float> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_float_source_subvec_0_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,1,2,0,4,5,6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x float> %x, <8 x float> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 0, i32 4, i32 5, i32 6, i32 7>
+ ret <8 x float> %r
+}
+
+define <8 x float> @vec256_eltty_float_source_subvec_0_target_subvec_mask_1_binary(<8 x float> %x, <8 x float> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_float_source_subvec_0_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastss %xmm1, %xmm1
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2],ymm1[3],ymm0[4,5,6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x float> %x, <8 x float> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 8, i32 4, i32 5, i32 6, i32 7>
+ ret <8 x float> %r
+}
+
+define <8 x float> @vec256_eltty_float_source_subvec_0_target_subvec_mask_2_unary(<8 x float> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_float_source_subvec_0_target_subvec_mask_2_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovaps {{.*#+}} ymm1 = [0,1,2,3,4,5,6,0]
+; CHECK-NEXT: vpermps %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x float> %x, <8 x float> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 0>
+ ret <8 x float> %r
+}
+
+define <8 x float> @vec256_eltty_float_source_subvec_0_target_subvec_mask_2_binary(<8 x float> %x, <8 x float> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_float_source_subvec_0_target_subvec_mask_2_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastss %xmm1, %ymm1
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm1[7]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x float> %x, <8 x float> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 8>
+ ret <8 x float> %r
+}
+
+define <8 x float> @vec256_eltty_float_source_subvec_0_target_subvec_mask_3_unary(<8 x float> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_float_source_subvec_0_target_subvec_mask_3_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovaps {{.*#+}} ymm1 = [0,1,2,0,4,5,6,0]
+; CHECK-NEXT: vpermps %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x float> %x, <8 x float> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 0, i32 4, i32 5, i32 6, i32 0>
+ ret <8 x float> %r
+}
+
+define <8 x float> @vec256_eltty_float_source_subvec_0_target_subvec_mask_3_binary(<8 x float> %x, <8 x float> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_float_source_subvec_0_target_subvec_mask_3_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[0,1,0,1]
+; CHECK-NEXT: vblendps {{.*#+}} ymm1 = ymm1[0,1],ymm0[2,3],ymm1[4,5],ymm0[6,7]
+; CHECK-NEXT: vshufps {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,0],ymm0[4,5],ymm1[6,4]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x float> %x, <8 x float> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 8, i32 4, i32 5, i32 6, i32 8>
+ ret <8 x float> %r
+}
+
+define <8 x float> @vec256_eltty_float_source_subvec_1_target_subvec_mask_1_unary(<8 x float> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_float_source_subvec_1_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovaps {{.*#+}} ymm1 = [0,1,2,4,4,5,6,7]
+; CHECK-NEXT: vpermps %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x float> %x, <8 x float> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 4, i32 4, i32 5, i32 6, i32 7>
+ ret <8 x float> %r
+}
+
+define <8 x float> @vec256_eltty_float_source_subvec_1_target_subvec_mask_1_binary(<8 x float> %x, <8 x float> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_float_source_subvec_1_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vextractf128 $1, %ymm1, %xmm1
+; CHECK-NEXT: vbroadcastss %xmm1, %xmm1
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2],ymm1[3],ymm0[4,5,6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x float> %x, <8 x float> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 12, i32 4, i32 5, i32 6, i32 7>
+ ret <8 x float> %r
+}
+
+define <8 x float> @vec256_eltty_float_source_subvec_1_target_subvec_mask_2_unary(<8 x float> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_float_source_subvec_1_target_subvec_mask_2_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6,4]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x float> %x, <8 x float> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 4>
+ ret <8 x float> %r
+}
+
+define <8 x float> @vec256_eltty_float_source_subvec_1_target_subvec_mask_2_binary(<8 x float> %x, <8 x float> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_float_source_subvec_1_target_subvec_mask_2_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vextractf128 $1, %ymm1, %xmm1
+; CHECK-NEXT: vbroadcastss %xmm1, %ymm1
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm1[7]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x float> %x, <8 x float> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 12>
+ ret <8 x float> %r
+}
+
+define <8 x float> @vec256_eltty_float_source_subvec_1_target_subvec_mask_3_unary(<8 x float> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_float_source_subvec_1_target_subvec_mask_3_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovaps {{.*#+}} ymm1 = [0,1,2,4,4,5,6,4]
+; CHECK-NEXT: vpermps %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x float> %x, <8 x float> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 4, i32 4, i32 5, i32 6, i32 4>
+ ret <8 x float> %r
+}
+
+define <8 x float> @vec256_eltty_float_source_subvec_1_target_subvec_mask_3_binary(<8 x float> %x, <8 x float> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_float_source_subvec_1_target_subvec_mask_3_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[2,3,2,3]
+; CHECK-NEXT: vblendps {{.*#+}} ymm1 = ymm1[0,1],ymm0[2,3],ymm1[4,5],ymm0[6,7]
+; CHECK-NEXT: vshufps {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,0],ymm0[4,5],ymm1[6,4]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x float> %x, <8 x float> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 12, i32 4, i32 5, i32 6, i32 12>
+ ret <8 x float> %r
+}
+
+define <8 x i32> @vec256_eltty_i32_source_subvec_0_target_subvec_mask_1_unary(<8 x i32> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i32_source_subvec_0_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovaps {{.*#+}} ymm1 = [0,1,2,0,4,5,6,7]
+; CHECK-NEXT: vpermps %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x i32> %x, <8 x i32> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 0, i32 4, i32 5, i32 6, i32 7>
+ ret <8 x i32> %r
+}
+
+define <8 x i32> @vec256_eltty_i32_source_subvec_0_target_subvec_mask_1_binary(<8 x i32> %x, <8 x i32> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i32_source_subvec_0_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastss %xmm1, %xmm1
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2],ymm1[3],ymm0[4,5,6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x i32> %x, <8 x i32> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 8, i32 4, i32 5, i32 6, i32 7>
+ ret <8 x i32> %r
+}
+
+define <8 x i32> @vec256_eltty_i32_source_subvec_0_target_subvec_mask_2_unary(<8 x i32> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i32_source_subvec_0_target_subvec_mask_2_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovaps {{.*#+}} ymm1 = [0,1,2,3,4,5,6,0]
+; CHECK-NEXT: vpermps %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x i32> %x, <8 x i32> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 0>
+ ret <8 x i32> %r
+}
+
+define <8 x i32> @vec256_eltty_i32_source_subvec_0_target_subvec_mask_2_binary(<8 x i32> %x, <8 x i32> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i32_source_subvec_0_target_subvec_mask_2_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastss %xmm1, %ymm1
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm1[7]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x i32> %x, <8 x i32> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 8>
+ ret <8 x i32> %r
+}
+
+define <8 x i32> @vec256_eltty_i32_source_subvec_0_target_subvec_mask_3_unary(<8 x i32> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i32_source_subvec_0_target_subvec_mask_3_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovaps {{.*#+}} ymm1 = [0,1,2,0,4,5,6,0]
+; CHECK-NEXT: vpermps %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x i32> %x, <8 x i32> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 0, i32 4, i32 5, i32 6, i32 0>
+ ret <8 x i32> %r
+}
+
+define <8 x i32> @vec256_eltty_i32_source_subvec_0_target_subvec_mask_3_binary(<8 x i32> %x, <8 x i32> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i32_source_subvec_0_target_subvec_mask_3_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastss %xmm1, %ymm1
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2],ymm1[3],ymm0[4,5,6],ymm1[7]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x i32> %x, <8 x i32> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 8, i32 4, i32 5, i32 6, i32 8>
+ ret <8 x i32> %r
+}
+
+define <8 x i32> @vec256_eltty_i32_source_subvec_1_target_subvec_mask_1_unary(<8 x i32> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i32_source_subvec_1_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovaps {{.*#+}} ymm1 = [0,1,2,4,4,5,6,7]
+; CHECK-NEXT: vpermps %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x i32> %x, <8 x i32> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 4, i32 4, i32 5, i32 6, i32 7>
+ ret <8 x i32> %r
+}
+
+define <8 x i32> @vec256_eltty_i32_source_subvec_1_target_subvec_mask_1_binary(<8 x i32> %x, <8 x i32> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i32_source_subvec_1_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vextractf128 $1, %ymm1, %xmm1
+; CHECK-NEXT: vbroadcastss %xmm1, %xmm1
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2],ymm1[3],ymm0[4,5,6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x i32> %x, <8 x i32> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 12, i32 4, i32 5, i32 6, i32 7>
+ ret <8 x i32> %r
+}
+
+define <8 x i32> @vec256_eltty_i32_source_subvec_1_target_subvec_mask_2_unary(<8 x i32> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i32_source_subvec_1_target_subvec_mask_2_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovaps {{.*#+}} ymm1 = [0,1,2,3,4,5,6,4]
+; CHECK-NEXT: vpermps %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x i32> %x, <8 x i32> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 4>
+ ret <8 x i32> %r
+}
+
+define <8 x i32> @vec256_eltty_i32_source_subvec_1_target_subvec_mask_2_binary(<8 x i32> %x, <8 x i32> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i32_source_subvec_1_target_subvec_mask_2_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vextractf128 $1, %ymm1, %xmm1
+; CHECK-NEXT: vbroadcastss %xmm1, %ymm1
+; CHECK-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm1[7]
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x i32> %x, <8 x i32> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 12>
+ ret <8 x i32> %r
+}
+
+define <8 x i32> @vec256_eltty_i32_source_subvec_1_target_subvec_mask_3_unary(<8 x i32> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i32_source_subvec_1_target_subvec_mask_3_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovaps {{.*#+}} ymm1 = [0,1,2,4,4,5,6,4]
+; CHECK-NEXT: vpermps %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <8 x i32> %x, <8 x i32> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 4, i32 4, i32 5, i32 6, i32 4>
+ ret <8 x i32> %r
+}
+
+define <8 x i32> @vec256_eltty_i32_source_subvec_1_target_subvec_mask_3_binary(<8 x i32> %x, <8 x i32> %y) nounwind {
+; CHECK-SLOW-LABEL: vec256_eltty_i32_source_subvec_1_target_subvec_mask_3_binary:
+; CHECK-SLOW: # %bb.0:
+; CHECK-SLOW-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[2,3,2,3]
+; CHECK-SLOW-NEXT: vpermilps {{.*#+}} ymm1 = ymm1[0,0,0,0,4,4,4,4]
+; CHECK-SLOW-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2],ymm1[3],ymm0[4,5,6],ymm1[7]
+; CHECK-SLOW-NEXT: retq
+;
+; CHECK-FAST-LABEL: vec256_eltty_i32_source_subvec_1_target_subvec_mask_3_binary:
+; CHECK-FAST: # %bb.0:
+; CHECK-FAST-NEXT: vbroadcastss {{.*#+}} ymm2 = [4,4,4,4,4,4,4,4]
+; CHECK-FAST-NEXT: vpermps %ymm1, %ymm2, %ymm1
+; CHECK-FAST-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2],ymm1[3],ymm0[4,5,6],ymm1[7]
+; CHECK-FAST-NEXT: retq
+ %r = shufflevector <8 x i32> %x, <8 x i32> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 12, i32 4, i32 5, i32 6, i32 12>
+ ret <8 x i32> %r
+}
+
+define <16 x i16> @vec256_eltty_i16_source_subvec_0_target_subvec_mask_1_unary(<16 x i16> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i16_source_subvec_0_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13,0,1,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31]
+; CHECK-NEXT: retq
+ %r = shufflevector <16 x i16> %x, <16 x i16> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 0, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
+ ret <16 x i16> %r
+}
+
+define <16 x i16> @vec256_eltty_i16_source_subvec_0_target_subvec_mask_1_binary(<16 x i16> %x, <16 x i16> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i16_source_subvec_0_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpbroadcastw %xmm1, %xmm1
+; CHECK-NEXT: vpblendw {{.*#+}} xmm1 = xmm0[0,1,2,3,4,5,6],xmm1[7]
+; CHECK-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5,6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <16 x i16> %x, <16 x i16> %y, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 16, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
+ ret <16 x i16> %r
+}
+
+define <16 x i16> @vec256_eltty_i16_source_subvec_0_target_subvec_mask_2_unary(<16 x i16> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i16_source_subvec_0_target_subvec_mask_2_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpbroadcastw %xmm0, %ymm1
+; CHECK-NEXT: vpblendw {{.*#+}} ymm1 = ymm0[0,1,2,3,4,5,6],ymm1[7],ymm0[8,9,10,11,12,13,14],ymm1[15]
+; CHECK-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <16 x i16> %x, <16 x i16> poison, <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 0>
+ ret <16 x i16> %r
+}
+
+define <16 x i16> @vec256_eltty_i16_source_subvec_0_target_subvec_mask_2_binary(<16 x i16> %x, <16 x i16> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i16_source_subvec_0_target_subvec_mask_2_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpbroadcastw %xmm1, %ymm1
+; CHECK-NEXT: vpblendw {{.*#+}} ymm1 = ymm0[0,1,2,3,4,5,6],ymm1[7],ymm0[8,9,10,11,12,13,14],ymm1[15]
+; CHECK-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <16 x i16> %x, <16 x i16> %y, <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 16>
+ ret <16 x i16> %r
+}
+
+define <16 x i16> @vec256_eltty_i16_source_subvec_0_target_subvec_mask_3_unary(<16 x i16> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i16_source_subvec_0_target_subvec_mask_3_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpbroadcastw %xmm0, %ymm1
+; CHECK-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm1[7],ymm0[8,9,10,11,12,13,14],ymm1[15]
+; CHECK-NEXT: retq
+ %r = shufflevector <16 x i16> %x, <16 x i16> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 0, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 0>
+ ret <16 x i16> %r
+}
+
+define <16 x i16> @vec256_eltty_i16_source_subvec_0_target_subvec_mask_3_binary(<16 x i16> %x, <16 x i16> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i16_source_subvec_0_target_subvec_mask_3_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpbroadcastw %xmm1, %ymm1
+; CHECK-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm1[7],ymm0[8,9,10,11,12,13,14],ymm1[15]
+; CHECK-NEXT: retq
+ %r = shufflevector <16 x i16> %x, <16 x i16> %y, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 16, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 16>
+ ret <16 x i16> %r
+}
+
+define <16 x i16> @vec256_eltty_i16_source_subvec_1_target_subvec_mask_1_unary(<16 x i16> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i16_source_subvec_1_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vextracti128 $1, %ymm0, %xmm1
+; CHECK-NEXT: vpbroadcastw %xmm1, %xmm1
+; CHECK-NEXT: vpblendw {{.*#+}} xmm1 = xmm0[0,1,2,3,4,5,6],xmm1[7]
+; CHECK-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5,6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <16 x i16> %x, <16 x i16> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 8, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
+ ret <16 x i16> %r
+}
+
+define <16 x i16> @vec256_eltty_i16_source_subvec_1_target_subvec_mask_1_binary(<16 x i16> %x, <16 x i16> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i16_source_subvec_1_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vextracti128 $1, %ymm1, %xmm1
+; CHECK-NEXT: vpbroadcastw %xmm1, %xmm1
+; CHECK-NEXT: vpblendw {{.*#+}} xmm1 = xmm0[0,1,2,3,4,5,6],xmm1[7]
+; CHECK-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5,6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <16 x i16> %x, <16 x i16> %y, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 24, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
+ ret <16 x i16> %r
+}
+
+define <16 x i16> @vec256_eltty_i16_source_subvec_1_target_subvec_mask_2_unary(<16 x i16> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i16_source_subvec_1_target_subvec_mask_2_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,16,17]
+; CHECK-NEXT: retq
+ %r = shufflevector <16 x i16> %x, <16 x i16> poison, <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 8>
+ ret <16 x i16> %r
+}
+
+define <16 x i16> @vec256_eltty_i16_source_subvec_1_target_subvec_mask_2_binary(<16 x i16> %x, <16 x i16> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i16_source_subvec_1_target_subvec_mask_2_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vextracti128 $1, %ymm1, %xmm1
+; CHECK-NEXT: vpbroadcastw %xmm1, %ymm1
+; CHECK-NEXT: vpblendw {{.*#+}} ymm1 = ymm0[0,1,2,3,4,5,6],ymm1[7],ymm0[8,9,10,11,12,13,14],ymm1[15]
+; CHECK-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
+; CHECK-NEXT: retq
+ %r = shufflevector <16 x i16> %x, <16 x i16> %y, <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 24>
+ ret <16 x i16> %r
+}
+
+define <16 x i16> @vec256_eltty_i16_source_subvec_1_target_subvec_mask_3_unary(<16 x i16> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i16_source_subvec_1_target_subvec_mask_3_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermq {{.*#+}} ymm1 = ymm0[2,3,2,3]
+; CHECK-NEXT: vpslldq {{.*#+}} ymm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm1[0,1],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm1[16,17]
+; CHECK-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm1[7],ymm0[8,9,10,11,12,13,14],ymm1[15]
+; CHECK-NEXT: retq
+ %r = shufflevector <16 x i16> %x, <16 x i16> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 8, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 8>
+ ret <16 x i16> %r
+}
+
+define <16 x i16> @vec256_eltty_i16_source_subvec_1_target_subvec_mask_3_binary(<16 x i16> %x, <16 x i16> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i16_source_subvec_1_target_subvec_mask_3_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
+; CHECK-NEXT: vpslldq {{.*#+}} ymm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm1[0,1],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm1[16,17]
+; CHECK-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm1[7],ymm0[8,9,10,11,12,13,14],ymm1[15]
+; CHECK-NEXT: retq
+ %r = shufflevector <16 x i16> %x, <16 x i16> %y, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 24, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 24>
+ ret <16 x i16> %r
+}
+
+define <32 x i8> @vec256_eltty_i8_source_subvec_0_target_subvec_mask_1_unary(<32 x i8> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i8_source_subvec_0_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,0,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31]
+; CHECK-NEXT: retq
+ %r = shufflevector <32 x i8> %x, <32 x i8> poison, <32 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 0, 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>
+ ret <32 x i8> %r
+}
+
+define <32 x i8> @vec256_eltty_i8_source_subvec_0_target_subvec_mask_1_binary(<32 x i8> %x, <32 x i8> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i8_source_subvec_0_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0]
+; CHECK-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
+; CHECK-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <32 x i8> %x, <32 x i8> %y, <32 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 32, 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>
+ ret <32 x i8> %r
+}
+
+define <32 x i8> @vec256_eltty_i8_source_subvec_0_target_subvec_mask_2_unary(<32 x i8> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i8_source_subvec_0_target_subvec_mask_2_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpbroadcastb %xmm0, %ymm1
+; CHECK-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0]
+; CHECK-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <32 x i8> %x, <32 x i8> poison, <32 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, 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 0>
+ ret <32 x i8> %r
+}
+
+define <32 x i8> @vec256_eltty_i8_source_subvec_0_target_subvec_mask_2_binary(<32 x i8> %x, <32 x i8> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i8_source_subvec_0_target_subvec_mask_2_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0]
+; CHECK-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
+; CHECK-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0]
+; CHECK-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <32 x i8> %x, <32 x i8> %y, <32 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, 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 32>
+ ret <32 x i8> %r
+}
+
+define <32 x i8> @vec256_eltty_i8_source_subvec_0_target_subvec_mask_3_unary(<32 x i8> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i8_source_subvec_0_target_subvec_mask_3_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermq {{.*#+}} ymm1 = ymm0[0,1,0,1]
+; CHECK-NEXT: vpslldq {{.*#+}} ymm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm1[0],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm1[16]
+; CHECK-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0]
+; CHECK-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <32 x i8> %x, <32 x i8> poison, <32 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 0, 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 0>
+ ret <32 x i8> %r
+}
+
+define <32 x i8> @vec256_eltty_i8_source_subvec_0_target_subvec_mask_3_binary(<32 x i8> %x, <32 x i8> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i8_source_subvec_0_target_subvec_mask_3_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,0,1]
+; CHECK-NEXT: vpslldq {{.*#+}} ymm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm1[0],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm1[16]
+; CHECK-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0]
+; CHECK-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <32 x i8> %x, <32 x i8> %y, <32 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 32, 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 32>
+ ret <32 x i8> %r
+}
+
+define <32 x i8> @vec256_eltty_i8_source_subvec_1_target_subvec_mask_1_unary(<32 x i8> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i8_source_subvec_1_target_subvec_mask_1_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vextracti128 $1, %ymm0, %xmm1
+; CHECK-NEXT: vpslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0]
+; CHECK-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
+; CHECK-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <32 x i8> %x, <32 x i8> poison, <32 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 16, 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>
+ ret <32 x i8> %r
+}
+
+define <32 x i8> @vec256_eltty_i8_source_subvec_1_target_subvec_mask_1_binary(<32 x i8> %x, <32 x i8> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i8_source_subvec_1_target_subvec_mask_1_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vextracti128 $1, %ymm1, %xmm1
+; CHECK-NEXT: vpslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0]
+; CHECK-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
+; CHECK-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <32 x i8> %x, <32 x i8> %y, <32 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 48, 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>
+ ret <32 x i8> %r
+}
+
+define <32 x i8> @vec256_eltty_i8_source_subvec_1_target_subvec_mask_2_unary(<32 x i8> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i8_source_subvec_1_target_subvec_mask_2_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,16]
+; CHECK-NEXT: retq
+ %r = shufflevector <32 x i8> %x, <32 x i8> poison, <32 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, 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 16>
+ ret <32 x i8> %r
+}
+
+define <32 x i8> @vec256_eltty_i8_source_subvec_1_target_subvec_mask_2_binary(<32 x i8> %x, <32 x i8> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i8_source_subvec_1_target_subvec_mask_2_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vextracti128 $1, %ymm1, %xmm1
+; CHECK-NEXT: vpbroadcastb %xmm1, %ymm1
+; CHECK-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0]
+; CHECK-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <32 x i8> %x, <32 x i8> %y, <32 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, 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 48>
+ ret <32 x i8> %r
+}
+
+define <32 x i8> @vec256_eltty_i8_source_subvec_1_target_subvec_mask_3_unary(<32 x i8> %x) nounwind {
+; CHECK-LABEL: vec256_eltty_i8_source_subvec_1_target_subvec_mask_3_unary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermq {{.*#+}} ymm1 = ymm0[2,3,2,3]
+; CHECK-NEXT: vpslldq {{.*#+}} ymm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm1[0],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm1[16]
+; CHECK-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0]
+; CHECK-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <32 x i8> %x, <32 x i8> poison, <32 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 16, 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 16>
+ ret <32 x i8> %r
+}
+
+define <32 x i8> @vec256_eltty_i8_source_subvec_1_target_subvec_mask_3_binary(<32 x i8> %x, <32 x i8> %y) nounwind {
+; CHECK-LABEL: vec256_eltty_i8_source_subvec_1_target_subvec_mask_3_binary:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
+; CHECK-NEXT: vpslldq {{.*#+}} ymm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm1[0],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm1[16]
+; CHECK-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0]
+; CHECK-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
+; CHECK-NEXT: retq
+ %r = shufflevector <32 x i8> %x, <32 x i8> %y, <32 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 48, 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 48>
+ ret <32 x i8> %r
+}
More information about the llvm-commits
mailing list