; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py ; RUN: llc < %s -mtriple=i686-apple-darwin -mattr=+avx2 | FileCheck %s --check-prefix=X32 ; RUN: llc < %s -mtriple=x86_64-apple-darwin -mattr=+avx2 | FileCheck %s --check-prefix=X64 define <16 x i16> @test_llvm_x86_avx2_pmovsxbw(<16 x i8>* %a) { ; X32-LABEL: test_llvm_x86_avx2_pmovsxbw: ; X32: ## BB#0: ; X32-NEXT: movl {{[0-9]+}}(%esp), %eax ; X32-NEXT: vpmovsxbw (%eax), %ymm0 ; X32-NEXT: retl ; ; X64-LABEL: test_llvm_x86_avx2_pmovsxbw: ; X64: ## BB#0: ; X64-NEXT: vpmovsxbw (%rdi), %ymm0 ; X64-NEXT: retq %1 = load <16 x i8>, <16 x i8>* %a, align 1 %2 = sext <16 x i8> %1 to <16 x i16> ret <16 x i16> %2 } define <8 x i32> @test_llvm_x86_avx2_pmovsxbd(<16 x i8>* %a) { ; X32-LABEL: test_llvm_x86_avx2_pmovsxbd: ; X32: ## BB#0: ; X32-NEXT: movl {{[0-9]+}}(%esp), %eax ; X32-NEXT: vpmovsxbd (%eax), %ymm0 ; X32-NEXT: retl ; ; X64-LABEL: test_llvm_x86_avx2_pmovsxbd: ; X64: ## BB#0: ; X64-NEXT: vpmovsxbd (%rdi), %ymm0 ; X64-NEXT: retq %1 = load <16 x i8>, <16 x i8>* %a, align 1 %2 = shufflevector <16 x i8> %1, <16 x i8> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7> %3 = sext <8 x i8> %2 to <8 x i32> ret <8 x i32> %3 } define <4 x i64> @test_llvm_x86_avx2_pmovsxbq(<16 x i8>* %a) { ; X32-LABEL: test_llvm_x86_avx2_pmovsxbq: ; X32: ## BB#0: ; X32-NEXT: movl {{[0-9]+}}(%esp), %eax ; X32-NEXT: vpmovsxbq (%eax), %ymm0 ; X32-NEXT: retl ; ; X64-LABEL: test_llvm_x86_avx2_pmovsxbq: ; X64: ## BB#0: ; X64-NEXT: vpmovsxbq (%rdi), %ymm0 ; X64-NEXT: retq %1 = load <16 x i8>, <16 x i8>* %a, align 1 %2 = shufflevector <16 x i8> %1, <16 x i8> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3> %3 = sext <4 x i8> %2 to <4 x i64> ret <4 x i64> %3 } define <8 x i32> @test_llvm_x86_avx2_pmovsxwd(<8 x i16>* %a) { ; X32-LABEL: test_llvm_x86_avx2_pmovsxwd: ; X32: ## BB#0: ; X32-NEXT: movl {{[0-9]+}}(%esp), %eax ; X32-NEXT: vpmovsxwd (%eax), %ymm0 ; X32-NEXT: retl ; ; X64-LABEL: test_llvm_x86_avx2_pmovsxwd: ; X64: ## BB#0: ; X64-NEXT: vpmovsxwd (%rdi), %ymm0 ; X64-NEXT: retq %1 = load <8 x i16>, <8 x i16>* %a, align 1 %2 = sext <8 x i16> %1 to <8 x i32> ret <8 x i32> %2 } define <4 x i64> @test_llvm_x86_avx2_pmovsxwq(<8 x i16>* %a) { ; X32-LABEL: test_llvm_x86_avx2_pmovsxwq: ; X32: ## BB#0: ; X32-NEXT: movl {{[0-9]+}}(%esp), %eax ; X32-NEXT: vpmovsxwq (%eax), %ymm0 ; X32-NEXT: retl ; ; X64-LABEL: test_llvm_x86_avx2_pmovsxwq: ; X64: ## BB#0: ; X64-NEXT: vpmovsxwq (%rdi), %ymm0 ; X64-NEXT: retq %1 = load <8 x i16>, <8 x i16>* %a, align 1 %2 = shufflevector <8 x i16> %1, <8 x i16> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3> %3 = sext <4 x i16> %2 to <4 x i64> ret <4 x i64> %3 } define <4 x i64> @test_llvm_x86_avx2_pmovsxdq(<4 x i32>* %a) { ; X32-LABEL: test_llvm_x86_avx2_pmovsxdq: ; X32: ## BB#0: ; X32-NEXT: movl {{[0-9]+}}(%esp), %eax ; X32-NEXT: vpmovsxdq (%eax), %ymm0 ; X32-NEXT: retl ; ; X64-LABEL: test_llvm_x86_avx2_pmovsxdq: ; X64: ## BB#0: ; X64-NEXT: vpmovsxdq (%rdi), %ymm0 ; X64-NEXT: retq %1 = load <4 x i32>, <4 x i32>* %a, align 1 %2 = sext <4 x i32> %1 to <4 x i64> ret <4 x i64> %2 } define <16 x i16> @test_llvm_x86_avx2_pmovzxbw(<16 x i8>* %a) { ; X32-LABEL: test_llvm_x86_avx2_pmovzxbw: ; X32: ## BB#0: ; X32-NEXT: movl {{[0-9]+}}(%esp), %eax ; X32-NEXT: vpmovzxbw {{.*#+}} ymm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero,mem[8],zero,mem[9],zero,mem[10],zero,mem[11],zero,mem[12],zero,mem[13],zero,mem[14],zero,mem[15],zero ; X32-NEXT: retl ; ; X64-LABEL: test_llvm_x86_avx2_pmovzxbw: ; X64: ## BB#0: ; X64-NEXT: vpmovzxbw {{.*#+}} ymm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero,mem[8],zero,mem[9],zero,mem[10],zero,mem[11],zero,mem[12],zero,mem[13],zero,mem[14],zero,mem[15],zero ; X64-NEXT: retq %1 = load <16 x i8>, <16 x i8>* %a, align 1 %2 = zext <16 x i8> %1 to <16 x i16> ret <16 x i16> %2 } define <8 x i32> @test_llvm_x86_avx2_pmovzxbd(<16 x i8>* %a) { ; X32-LABEL: test_llvm_x86_avx2_pmovzxbd: ; X32: ## BB#0: ; X32-NEXT: movl {{[0-9]+}}(%esp), %eax ; X32-NEXT: vpmovzxbd {{.*#+}} ymm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero ; X32-NEXT: retl ; ; X64-LABEL: test_llvm_x86_avx2_pmovzxbd: ; X64: ## BB#0: ; X64-NEXT: vpmovzxbd {{.*#+}} ymm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero ; X64-NEXT: retq %1 = load <16 x i8>, <16 x i8>* %a, align 1 %2 = shufflevector <16 x i8> %1, <16 x i8> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7> %3 = zext <8 x i8> %2 to <8 x i32> ret <8 x i32> %3 } define <4 x i64> @test_llvm_x86_avx2_pmovzxbq(<16 x i8>* %a) { ; X32-LABEL: test_llvm_x86_avx2_pmovzxbq: ; X32: ## BB#0: ; X32-NEXT: movl {{[0-9]+}}(%esp), %eax ; X32-NEXT: vpmovzxbq {{.*#+}} ymm0 = mem[0],zero,zero,zero,zero,zero,zero,zero,mem[1],zero,zero,zero,zero,zero,zero,zero,mem[2],zero,zero,zero,zero,zero,zero,zero,mem[3],zero,zero,zero,zero,zero,zero,zero ; X32-NEXT: retl ; ; X64-LABEL: test_llvm_x86_avx2_pmovzxbq: ; X64: ## BB#0: ; X64-NEXT: vpmovzxbq {{.*#+}} ymm0 = mem[0],zero,zero,zero,zero,zero,zero,zero,mem[1],zero,zero,zero,zero,zero,zero,zero,mem[2],zero,zero,zero,zero,zero,zero,zero,mem[3],zero,zero,zero,zero,zero,zero,zero ; X64-NEXT: retq %1 = load <16 x i8>, <16 x i8>* %a, align 1 %2 = shufflevector <16 x i8> %1, <16 x i8> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3> %3 = zext <4 x i8> %2 to <4 x i64> ret <4 x i64> %3 } define <8 x i32> @test_llvm_x86_avx2_pmovzxwd(<8 x i16>* %a) { ; X32-LABEL: test_llvm_x86_avx2_pmovzxwd: ; X32: ## BB#0: ; X32-NEXT: movl {{[0-9]+}}(%esp), %eax ; X32-NEXT: vpmovzxwd {{.*#+}} ymm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero ; X32-NEXT: retl ; ; X64-LABEL: test_llvm_x86_avx2_pmovzxwd: ; X64: ## BB#0: ; X64-NEXT: vpmovzxwd {{.*#+}} ymm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero ; X64-NEXT: retq %1 = load <8 x i16>, <8 x i16>* %a, align 1 %2 = zext <8 x i16> %1 to <8 x i32> ret <8 x i32> %2 } define <4 x i64> @test_llvm_x86_avx2_pmovzxwq(<8 x i16>* %a) { ; X32-LABEL: test_llvm_x86_avx2_pmovzxwq: ; X32: ## BB#0: ; X32-NEXT: movl {{[0-9]+}}(%esp), %eax ; X32-NEXT: vpmovzxwq {{.*#+}} ymm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero ; X32-NEXT: retl ; ; X64-LABEL: test_llvm_x86_avx2_pmovzxwq: ; X64: ## BB#0: ; X64-NEXT: vpmovzxwq {{.*#+}} ymm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero ; X64-NEXT: retq %1 = load <8 x i16>, <8 x i16>* %a, align 1 %2 = shufflevector <8 x i16> %1, <8 x i16> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3> %3 = zext <4 x i16> %2 to <4 x i64> ret <4 x i64> %3 } define <4 x i64> @test_llvm_x86_avx2_pmovzxdq(<4 x i32>* %a) { ; X32-LABEL: test_llvm_x86_avx2_pmovzxdq: ; X32: ## BB#0: ; X32-NEXT: movl {{[0-9]+}}(%esp), %eax ; X32-NEXT: vpmovzxdq {{.*#+}} ymm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero ; X32-NEXT: retl ; ; X64-LABEL: test_llvm_x86_avx2_pmovzxdq: ; X64: ## BB#0: ; X64-NEXT: vpmovzxdq {{.*#+}} ymm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero ; X64-NEXT: retq %1 = load <4 x i32>, <4 x i32>* %a, align 1 %2 = zext <4 x i32> %1 to <4 x i64> ret <4 x i64> %2 }