Merge commit 'e7dfa2c456ba' from llvm.org/main into next

Author: git apple-llvm automerger

llvm/test/CodeGen/X86/shuffle-strided-with-offset-512.ll

@@ -1,9 +1,9 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f | FileCheck %s --check-prefixes=AVX512,AVX512F
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512vl,+fast-variable-crosslane-shuffle,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX512,AVX512VL
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512vl,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX512,AVX512VL
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw,+fast-variable-crosslane-shuffle,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX512,AVX512BW
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX512,AVX512BW
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512vl,+fast-variable-crosslane-shuffle,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX512,AVX512VL,AVX512VL-FAST-ALL
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512vl,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX512,AVX512VL,AVX512VL-FAST-PERLANE
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw,+fast-variable-crosslane-shuffle,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX512,AVX512BW,AVX512BW-FAST-ALL
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX512,AVX512BW,AVX512BW-FAST-PERLANE
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw,+avx512vl,+fast-variable-crosslane-shuffle,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX512,AVX512BWVL,AVX512BWVL-FAST-ALL
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw,+avx512vl,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX512,AVX512BWVL,AVX512BWVL-FAST-PERLANE
 
@@ -21,6 +21,31 @@ define void @shuffle_v64i8_to_v32i8_1(ptr %L, ptr %S) nounwind {
 ; AVX512F-NEXT:    vzeroupper
 ; AVX512F-NEXT:    retq
 ;
+; AVX512VL-FAST-ALL-LABEL: shuffle_v64i8_to_v32i8_1:
+; AVX512VL-FAST-ALL:       # %bb.0:
+; AVX512VL-FAST-ALL-NEXT:    vmovdqa (%rdi), %ymm0
+; AVX512VL-FAST-ALL-NEXT:    vmovdqa 32(%rdi), %ymm1
+; AVX512VL-FAST-ALL-NEXT:    vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,1,3,5,7,9,11,13,15,u,u,u,u,u,u,u,u,17,19,21,23,25,27,29,31]
+; AVX512VL-FAST-ALL-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[1,3,5,7,9,11,13,15,u,u,u,u,u,u,u,u,17,19,21,23,25,27,29,31,u,u,u,u,u,u,u,u]
+; AVX512VL-FAST-ALL-NEXT:    vpmovsxbq {{.*#+}} ymm2 = [0,2,5,7]
+; AVX512VL-FAST-ALL-NEXT:    vpermi2q %ymm1, %ymm0, %ymm2
+; AVX512VL-FAST-ALL-NEXT:    vmovdqa %ymm2, (%rsi)
+; AVX512VL-FAST-ALL-NEXT:    vzeroupper
+; AVX512VL-FAST-ALL-NEXT:    retq
+;
+; AVX512VL-FAST-PERLANE-LABEL: shuffle_v64i8_to_v32i8_1:
+; AVX512VL-FAST-PERLANE:       # %bb.0:
+; AVX512VL-FAST-PERLANE-NEXT:    vmovdqa (%rdi), %ymm0
+; AVX512VL-FAST-PERLANE-NEXT:    vmovdqa 32(%rdi), %ymm1
+; AVX512VL-FAST-PERLANE-NEXT:    vpbroadcastq {{.*#+}} ymm2 = [1,3,5,7,9,11,13,15,1,3,5,7,9,11,13,15,1,3,5,7,9,11,13,15,1,3,5,7,9,11,13,15]
+; AVX512VL-FAST-PERLANE-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
+; AVX512VL-FAST-PERLANE-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
+; AVX512VL-FAST-PERLANE-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3],ymm0[4,5],ymm1[6,7]
+; AVX512VL-FAST-PERLANE-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,1,3]
+; AVX512VL-FAST-PERLANE-NEXT:    vmovdqa %ymm0, (%rsi)
+; AVX512VL-FAST-PERLANE-NEXT:    vzeroupper
+; AVX512VL-FAST-PERLANE-NEXT:    retq
+;
 ; AVX512BW-LABEL: shuffle_v64i8_to_v32i8_1:
 ; AVX512BW:       # %bb.0:
 ; AVX512BW-NEXT:    vpsrlw $8, (%rdi), %zmm0
@@ -63,6 +88,40 @@ define void @shuffle_v16i32_to_v8i32_1(ptr %L, ptr %S) nounwind {
 ; AVX512F-NEXT:    vzeroupper
 ; AVX512F-NEXT:    retq
 ;
+; AVX512VL-FAST-ALL-LABEL: shuffle_v16i32_to_v8i32_1:
+; AVX512VL-FAST-ALL:       # %bb.0:
+; AVX512VL-FAST-ALL-NEXT:    vpmovsxbd {{.*#+}} ymm0 = [1,3,5,7,9,11,13,15]
+; AVX512VL-FAST-ALL-NEXT:    vpermps (%rdi), %zmm0, %zmm0
+; AVX512VL-FAST-ALL-NEXT:    vmovaps %ymm0, (%rsi)
+; AVX512VL-FAST-ALL-NEXT:    vzeroupper
+; AVX512VL-FAST-ALL-NEXT:    retq
+;
+; AVX512VL-FAST-PERLANE-LABEL: shuffle_v16i32_to_v8i32_1:
+; AVX512VL-FAST-PERLANE:       # %bb.0:
+; AVX512VL-FAST-PERLANE-NEXT:    vmovaps (%rdi), %ymm0
+; AVX512VL-FAST-PERLANE-NEXT:    vshufps {{.*#+}} ymm0 = ymm0[1,3],mem[1,3],ymm0[5,7],mem[5,7]
+; AVX512VL-FAST-PERLANE-NEXT:    vpermpd {{.*#+}} ymm0 = ymm0[0,2,1,3]
+; AVX512VL-FAST-PERLANE-NEXT:    vmovaps %ymm0, (%rsi)
+; AVX512VL-FAST-PERLANE-NEXT:    vzeroupper
+; AVX512VL-FAST-PERLANE-NEXT:    retq
+;
+; AVX512BW-FAST-ALL-LABEL: shuffle_v16i32_to_v8i32_1:
+; AVX512BW-FAST-ALL:       # %bb.0:
+; AVX512BW-FAST-ALL-NEXT:    vpmovsxbd {{.*#+}} ymm0 = [1,3,5,7,9,11,13,15]
+; AVX512BW-FAST-ALL-NEXT:    vpermps (%rdi), %zmm0, %zmm0
+; AVX512BW-FAST-ALL-NEXT:    vmovaps %ymm0, (%rsi)
+; AVX512BW-FAST-ALL-NEXT:    vzeroupper
+; AVX512BW-FAST-ALL-NEXT:    retq
+;
+; AVX512BW-FAST-PERLANE-LABEL: shuffle_v16i32_to_v8i32_1:
+; AVX512BW-FAST-PERLANE:       # %bb.0:
+; AVX512BW-FAST-PERLANE-NEXT:    vmovaps (%rdi), %ymm0
+; AVX512BW-FAST-PERLANE-NEXT:    vshufps {{.*#+}} ymm0 = ymm0[1,3],mem[1,3],ymm0[5,7],mem[5,7]
+; AVX512BW-FAST-PERLANE-NEXT:    vpermpd {{.*#+}} ymm0 = ymm0[0,2,1,3]
+; AVX512BW-FAST-PERLANE-NEXT:    vmovaps %ymm0, (%rsi)
+; AVX512BW-FAST-PERLANE-NEXT:    vzeroupper
+; AVX512BW-FAST-PERLANE-NEXT:    retq
+;
 ; AVX512BWVL-FAST-ALL-LABEL: shuffle_v16i32_to_v8i32_1:
 ; AVX512BWVL-FAST-ALL:       # %bb.0:
 ; AVX512BWVL-FAST-ALL-NEXT:    vpmovsxbd {{.*#+}} ymm0 = [1,3,5,7,9,11,13,15]