diff options
Diffstat (limited to 'llvm/test/CodeGen/X86/stack-folding-int-avx512.ll')
| -rw-r--r-- | llvm/test/CodeGen/X86/stack-folding-int-avx512.ll | 20 |
1 files changed, 19 insertions, 1 deletions
diff --git a/llvm/test/CodeGen/X86/stack-folding-int-avx512.ll b/llvm/test/CodeGen/X86/stack-folding-int-avx512.ll index 6bde51286dc..284021ac504 100644 --- a/llvm/test/CodeGen/X86/stack-folding-int-avx512.ll +++ b/llvm/test/CodeGen/X86/stack-folding-int-avx512.ll @@ -1,4 +1,4 @@ -; RUN: llc -O3 -disable-peephole -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw,+avx512dq,+avx512vbmi,+avx512cd < %s | FileCheck %s +; RUN: llc -O3 -disable-peephole -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw,+avx512dq,+avx512vbmi,+avx512cd,+avx512vpopcntdq < %s | FileCheck %s target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128" target triple = "x86_64-unknown-unknown" @@ -1214,6 +1214,24 @@ define <8 x i64> @stack_fold_pmovzxwq_maskz_zmm(<8 x i16> %a0, i8 %mask) { ret <8 x i64> %4 } +define <16 x i32> @stack_fold_vpopcntd(<16 x i32> %a0) { + ;CHECK-LABEL: stack_fold_vpopcntd + ;CHECK: vpopcntd {{-?[0-9]*}}(%rsp), {{%zmm[0-9][0-9]*}} {{.*#+}} 64-byte Folded Reload + %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"() + %2 = call <16 x i32> @llvm.ctpop.v16i32(<16 x i32> %a0) + ret <16 x i32> %2 +} +declare <16 x i32> @llvm.ctpop.v16i32(<16 x i32>) nounwind readonly + +define <8 x i64> @stack_fold_vpopcntq(<8 x i64> %a0) { + ;CHECK-LABEL: stack_fold_vpopcntq + ;CHECK: vpopcntq {{-?[0-9]*}}(%rsp), {{%zmm[0-9][0-9]*}} {{.*#+}} 64-byte Folded Reload + %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"() + %2 = call <8 x i64> @llvm.ctpop.v8i64(<8 x i64> %a0) + ret <8 x i64> %2 +} +declare <8 x i64> @llvm.ctpop.v8i64(<8 x i64>) nounwind readnone + define <8 x i64> @stack_fold_psadbw(<64 x i8> %a0, <64 x i8> %a1) { ;CHECK-LABEL: stack_fold_psadbw ;CHECK: vpsadbw {{-?[0-9]*}}(%rsp), {{%zmm[0-9][0-9]*}}, {{%zmm[0-9][0-9]*}} {{.*#+}} 64-byte Folded Reload |
