mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-26 04:32:44 +01:00
18451cc4a4
The motivation is that the update script has at least two deviations (`<...>@GOT`/`<...>@PLT`/ and not hiding pointer arithmetics) from what pretty much all the checklines were generated with, and most of the tests are still not updated, so each time one of the non-up-to-date tests is updated to see the effect of the code change, there is a lot of noise. Instead of having to deal with that each time, let's just deal with everything at once. This has been done via: ``` cd llvm-project/llvm/test/CodeGen/X86 grep -rl "; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py" | xargs -L1 <...>/llvm-project/llvm/utils/update_llc_test_checks.py --llc-binary <...>/llvm-project/build/bin/llc ``` Not all tests were regenerated, however.
1362 lines
51 KiB
LLVM
1362 lines
51 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown | FileCheck %s --check-prefixes=SSE,SSE2
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse3 | FileCheck %s --check-prefixes=SSE,SSE3
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+ssse3 | FileCheck %s --check-prefixes=SSE,SSSE3
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefixes=SSE,SSE41
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefixes=AVX,AVX1
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefixes=AVX,AVX2
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512vpopcntdq | FileCheck %s --check-prefixes=AVX,AVX512VPOPCNTDQ
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512vpopcntdq,+avx512vl | FileCheck %s --check-prefixes=AVX,AVX512VPOPCNTDQVL
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bitalg | FileCheck %s --check-prefix=BITALG_NOVLX
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bitalg,+avx512vl | FileCheck %s --check-prefix=BITALG
|
|
|
|
define <2 x i64> @testv2i64(<2 x i64> %in) nounwind {
|
|
; SSE2-LABEL: testv2i64:
|
|
; SSE2: # %bb.0:
|
|
; SSE2-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE2-NEXT: psrlw $1, %xmm1
|
|
; SSE2-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE2-NEXT: psubb %xmm1, %xmm0
|
|
; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
|
|
; SSE2-NEXT: movdqa %xmm0, %xmm2
|
|
; SSE2-NEXT: pand %xmm1, %xmm2
|
|
; SSE2-NEXT: psrlw $2, %xmm0
|
|
; SSE2-NEXT: pand %xmm1, %xmm0
|
|
; SSE2-NEXT: paddb %xmm2, %xmm0
|
|
; SSE2-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE2-NEXT: psrlw $4, %xmm1
|
|
; SSE2-NEXT: paddb %xmm0, %xmm1
|
|
; SSE2-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE2-NEXT: pxor %xmm0, %xmm0
|
|
; SSE2-NEXT: psadbw %xmm0, %xmm1
|
|
; SSE2-NEXT: movdqa %xmm1, %xmm0
|
|
; SSE2-NEXT: retq
|
|
;
|
|
; SSE3-LABEL: testv2i64:
|
|
; SSE3: # %bb.0:
|
|
; SSE3-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE3-NEXT: psrlw $1, %xmm1
|
|
; SSE3-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE3-NEXT: psubb %xmm1, %xmm0
|
|
; SSE3-NEXT: movdqa {{.*#+}} xmm1 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
|
|
; SSE3-NEXT: movdqa %xmm0, %xmm2
|
|
; SSE3-NEXT: pand %xmm1, %xmm2
|
|
; SSE3-NEXT: psrlw $2, %xmm0
|
|
; SSE3-NEXT: pand %xmm1, %xmm0
|
|
; SSE3-NEXT: paddb %xmm2, %xmm0
|
|
; SSE3-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE3-NEXT: psrlw $4, %xmm1
|
|
; SSE3-NEXT: paddb %xmm0, %xmm1
|
|
; SSE3-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE3-NEXT: pxor %xmm0, %xmm0
|
|
; SSE3-NEXT: psadbw %xmm0, %xmm1
|
|
; SSE3-NEXT: movdqa %xmm1, %xmm0
|
|
; SSE3-NEXT: retq
|
|
;
|
|
; SSSE3-LABEL: testv2i64:
|
|
; SSSE3: # %bb.0:
|
|
; SSSE3-NEXT: movdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; SSSE3-NEXT: movdqa %xmm0, %xmm2
|
|
; SSSE3-NEXT: pand %xmm1, %xmm2
|
|
; SSSE3-NEXT: movdqa {{.*#+}} xmm3 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; SSSE3-NEXT: movdqa %xmm3, %xmm4
|
|
; SSSE3-NEXT: pshufb %xmm2, %xmm4
|
|
; SSSE3-NEXT: psrlw $4, %xmm0
|
|
; SSSE3-NEXT: pand %xmm1, %xmm0
|
|
; SSSE3-NEXT: pshufb %xmm0, %xmm3
|
|
; SSSE3-NEXT: paddb %xmm4, %xmm3
|
|
; SSSE3-NEXT: pxor %xmm0, %xmm0
|
|
; SSSE3-NEXT: psadbw %xmm3, %xmm0
|
|
; SSSE3-NEXT: retq
|
|
;
|
|
; SSE41-LABEL: testv2i64:
|
|
; SSE41: # %bb.0:
|
|
; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; SSE41-NEXT: movdqa %xmm0, %xmm2
|
|
; SSE41-NEXT: pand %xmm1, %xmm2
|
|
; SSE41-NEXT: movdqa {{.*#+}} xmm3 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; SSE41-NEXT: movdqa %xmm3, %xmm4
|
|
; SSE41-NEXT: pshufb %xmm2, %xmm4
|
|
; SSE41-NEXT: psrlw $4, %xmm0
|
|
; SSE41-NEXT: pand %xmm1, %xmm0
|
|
; SSE41-NEXT: pshufb %xmm0, %xmm3
|
|
; SSE41-NEXT: paddb %xmm4, %xmm3
|
|
; SSE41-NEXT: pxor %xmm0, %xmm0
|
|
; SSE41-NEXT: psadbw %xmm3, %xmm0
|
|
; SSE41-NEXT: retq
|
|
;
|
|
; AVX1-LABEL: testv2i64:
|
|
; AVX1: # %bb.0:
|
|
; AVX1-NEXT: vmovdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; AVX1-NEXT: vpand %xmm1, %xmm0, %xmm2
|
|
; AVX1-NEXT: vmovdqa {{.*#+}} xmm3 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; AVX1-NEXT: vpshufb %xmm2, %xmm3, %xmm2
|
|
; AVX1-NEXT: vpsrlw $4, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpand %xmm1, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpshufb %xmm0, %xmm3, %xmm0
|
|
; AVX1-NEXT: vpaddb %xmm2, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX1-NEXT: vpsadbw %xmm1, %xmm0, %xmm0
|
|
; AVX1-NEXT: retq
|
|
;
|
|
; AVX2-LABEL: testv2i64:
|
|
; AVX2: # %bb.0:
|
|
; AVX2-NEXT: vmovdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; AVX2-NEXT: vpand %xmm1, %xmm0, %xmm2
|
|
; AVX2-NEXT: vmovdqa {{.*#+}} xmm3 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; AVX2-NEXT: vpshufb %xmm2, %xmm3, %xmm2
|
|
; AVX2-NEXT: vpsrlw $4, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpand %xmm1, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpshufb %xmm0, %xmm3, %xmm0
|
|
; AVX2-NEXT: vpaddb %xmm2, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX2-NEXT: vpsadbw %xmm1, %xmm0, %xmm0
|
|
; AVX2-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQ-LABEL: testv2i64:
|
|
; AVX512VPOPCNTDQ: # %bb.0:
|
|
; AVX512VPOPCNTDQ-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpopcntq %zmm0, %zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: # kill: def $xmm0 killed $xmm0 killed $zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vzeroupper
|
|
; AVX512VPOPCNTDQ-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQVL-LABEL: testv2i64:
|
|
; AVX512VPOPCNTDQVL: # %bb.0:
|
|
; AVX512VPOPCNTDQVL-NEXT: vpopcntq %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: testv2i64:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
|
|
; BITALG_NOVLX-NEXT: vpopcntb %zmm0, %zmm0
|
|
; BITALG_NOVLX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; BITALG_NOVLX-NEXT: vpsadbw %xmm1, %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vzeroupper
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: testv2i64:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vpopcntb %xmm0, %xmm0
|
|
; BITALG-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; BITALG-NEXT: vpsadbw %xmm1, %xmm0, %xmm0
|
|
; BITALG-NEXT: retq
|
|
%out = call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> %in)
|
|
ret <2 x i64> %out
|
|
}
|
|
|
|
define <4 x i32> @testv4i32(<4 x i32> %in) nounwind {
|
|
; SSE2-LABEL: testv4i32:
|
|
; SSE2: # %bb.0:
|
|
; SSE2-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE2-NEXT: psrlw $1, %xmm1
|
|
; SSE2-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE2-NEXT: psubb %xmm1, %xmm0
|
|
; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
|
|
; SSE2-NEXT: movdqa %xmm0, %xmm2
|
|
; SSE2-NEXT: pand %xmm1, %xmm2
|
|
; SSE2-NEXT: psrlw $2, %xmm0
|
|
; SSE2-NEXT: pand %xmm1, %xmm0
|
|
; SSE2-NEXT: paddb %xmm2, %xmm0
|
|
; SSE2-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE2-NEXT: psrlw $4, %xmm1
|
|
; SSE2-NEXT: paddb %xmm0, %xmm1
|
|
; SSE2-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE2-NEXT: pxor %xmm0, %xmm0
|
|
; SSE2-NEXT: movdqa %xmm1, %xmm2
|
|
; SSE2-NEXT: punpckhdq {{.*#+}} xmm2 = xmm2[2],xmm0[2],xmm2[3],xmm0[3]
|
|
; SSE2-NEXT: psadbw %xmm0, %xmm2
|
|
; SSE2-NEXT: punpckldq {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1]
|
|
; SSE2-NEXT: psadbw %xmm0, %xmm1
|
|
; SSE2-NEXT: packuswb %xmm2, %xmm1
|
|
; SSE2-NEXT: movdqa %xmm1, %xmm0
|
|
; SSE2-NEXT: retq
|
|
;
|
|
; SSE3-LABEL: testv4i32:
|
|
; SSE3: # %bb.0:
|
|
; SSE3-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE3-NEXT: psrlw $1, %xmm1
|
|
; SSE3-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE3-NEXT: psubb %xmm1, %xmm0
|
|
; SSE3-NEXT: movdqa {{.*#+}} xmm1 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
|
|
; SSE3-NEXT: movdqa %xmm0, %xmm2
|
|
; SSE3-NEXT: pand %xmm1, %xmm2
|
|
; SSE3-NEXT: psrlw $2, %xmm0
|
|
; SSE3-NEXT: pand %xmm1, %xmm0
|
|
; SSE3-NEXT: paddb %xmm2, %xmm0
|
|
; SSE3-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE3-NEXT: psrlw $4, %xmm1
|
|
; SSE3-NEXT: paddb %xmm0, %xmm1
|
|
; SSE3-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE3-NEXT: pxor %xmm0, %xmm0
|
|
; SSE3-NEXT: movdqa %xmm1, %xmm2
|
|
; SSE3-NEXT: punpckhdq {{.*#+}} xmm2 = xmm2[2],xmm0[2],xmm2[3],xmm0[3]
|
|
; SSE3-NEXT: psadbw %xmm0, %xmm2
|
|
; SSE3-NEXT: punpckldq {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1]
|
|
; SSE3-NEXT: psadbw %xmm0, %xmm1
|
|
; SSE3-NEXT: packuswb %xmm2, %xmm1
|
|
; SSE3-NEXT: movdqa %xmm1, %xmm0
|
|
; SSE3-NEXT: retq
|
|
;
|
|
; SSSE3-LABEL: testv4i32:
|
|
; SSSE3: # %bb.0:
|
|
; SSSE3-NEXT: movdqa {{.*#+}} xmm2 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; SSSE3-NEXT: movdqa %xmm0, %xmm3
|
|
; SSSE3-NEXT: pand %xmm2, %xmm3
|
|
; SSSE3-NEXT: movdqa {{.*#+}} xmm1 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; SSSE3-NEXT: movdqa %xmm1, %xmm4
|
|
; SSSE3-NEXT: pshufb %xmm3, %xmm4
|
|
; SSSE3-NEXT: psrlw $4, %xmm0
|
|
; SSSE3-NEXT: pand %xmm2, %xmm0
|
|
; SSSE3-NEXT: pshufb %xmm0, %xmm1
|
|
; SSSE3-NEXT: paddb %xmm4, %xmm1
|
|
; SSSE3-NEXT: pxor %xmm0, %xmm0
|
|
; SSSE3-NEXT: movdqa %xmm1, %xmm2
|
|
; SSSE3-NEXT: punpckhdq {{.*#+}} xmm2 = xmm2[2],xmm0[2],xmm2[3],xmm0[3]
|
|
; SSSE3-NEXT: psadbw %xmm0, %xmm2
|
|
; SSSE3-NEXT: punpckldq {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1]
|
|
; SSSE3-NEXT: psadbw %xmm0, %xmm1
|
|
; SSSE3-NEXT: packuswb %xmm2, %xmm1
|
|
; SSSE3-NEXT: movdqa %xmm1, %xmm0
|
|
; SSSE3-NEXT: retq
|
|
;
|
|
; SSE41-LABEL: testv4i32:
|
|
; SSE41: # %bb.0:
|
|
; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; SSE41-NEXT: movdqa %xmm0, %xmm2
|
|
; SSE41-NEXT: pand %xmm1, %xmm2
|
|
; SSE41-NEXT: movdqa {{.*#+}} xmm3 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; SSE41-NEXT: movdqa %xmm3, %xmm4
|
|
; SSE41-NEXT: pshufb %xmm2, %xmm4
|
|
; SSE41-NEXT: psrlw $4, %xmm0
|
|
; SSE41-NEXT: pand %xmm1, %xmm0
|
|
; SSE41-NEXT: pshufb %xmm0, %xmm3
|
|
; SSE41-NEXT: paddb %xmm4, %xmm3
|
|
; SSE41-NEXT: pxor %xmm1, %xmm1
|
|
; SSE41-NEXT: pmovzxdq {{.*#+}} xmm0 = xmm3[0],zero,xmm3[1],zero
|
|
; SSE41-NEXT: punpckhdq {{.*#+}} xmm3 = xmm3[2],xmm1[2],xmm3[3],xmm1[3]
|
|
; SSE41-NEXT: psadbw %xmm1, %xmm3
|
|
; SSE41-NEXT: psadbw %xmm1, %xmm0
|
|
; SSE41-NEXT: packuswb %xmm3, %xmm0
|
|
; SSE41-NEXT: retq
|
|
;
|
|
; AVX1-LABEL: testv4i32:
|
|
; AVX1: # %bb.0:
|
|
; AVX1-NEXT: vmovdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; AVX1-NEXT: vpand %xmm1, %xmm0, %xmm2
|
|
; AVX1-NEXT: vmovdqa {{.*#+}} xmm3 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; AVX1-NEXT: vpshufb %xmm2, %xmm3, %xmm2
|
|
; AVX1-NEXT: vpsrlw $4, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpand %xmm1, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpshufb %xmm0, %xmm3, %xmm0
|
|
; AVX1-NEXT: vpaddb %xmm2, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX1-NEXT: vpunpckhdq {{.*#+}} xmm2 = xmm0[2],xmm1[2],xmm0[3],xmm1[3]
|
|
; AVX1-NEXT: vpsadbw %xmm1, %xmm2, %xmm2
|
|
; AVX1-NEXT: vpmovzxdq {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero
|
|
; AVX1-NEXT: vpsadbw %xmm1, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpackuswb %xmm2, %xmm0, %xmm0
|
|
; AVX1-NEXT: retq
|
|
;
|
|
; AVX2-LABEL: testv4i32:
|
|
; AVX2: # %bb.0:
|
|
; AVX2-NEXT: vmovdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; AVX2-NEXT: vpand %xmm1, %xmm0, %xmm2
|
|
; AVX2-NEXT: vmovdqa {{.*#+}} xmm3 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; AVX2-NEXT: vpshufb %xmm2, %xmm3, %xmm2
|
|
; AVX2-NEXT: vpsrlw $4, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpand %xmm1, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpshufb %xmm0, %xmm3, %xmm0
|
|
; AVX2-NEXT: vpaddb %xmm2, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX2-NEXT: vpunpckhdq {{.*#+}} xmm2 = xmm0[2],xmm1[2],xmm0[3],xmm1[3]
|
|
; AVX2-NEXT: vpsadbw %xmm1, %xmm2, %xmm2
|
|
; AVX2-NEXT: vpmovzxdq {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero
|
|
; AVX2-NEXT: vpsadbw %xmm1, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpackuswb %xmm2, %xmm0, %xmm0
|
|
; AVX2-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQ-LABEL: testv4i32:
|
|
; AVX512VPOPCNTDQ: # %bb.0:
|
|
; AVX512VPOPCNTDQ-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpopcntd %zmm0, %zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: # kill: def $xmm0 killed $xmm0 killed $zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vzeroupper
|
|
; AVX512VPOPCNTDQ-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQVL-LABEL: testv4i32:
|
|
; AVX512VPOPCNTDQVL: # %bb.0:
|
|
; AVX512VPOPCNTDQVL-NEXT: vpopcntd %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: testv4i32:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
|
|
; BITALG_NOVLX-NEXT: vpopcntb %zmm0, %zmm0
|
|
; BITALG_NOVLX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; BITALG_NOVLX-NEXT: vpunpckhdq {{.*#+}} xmm2 = xmm0[2],xmm1[2],xmm0[3],xmm1[3]
|
|
; BITALG_NOVLX-NEXT: vpsadbw %xmm1, %xmm2, %xmm2
|
|
; BITALG_NOVLX-NEXT: vpmovzxdq {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero
|
|
; BITALG_NOVLX-NEXT: vpsadbw %xmm1, %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vpackuswb %xmm2, %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vzeroupper
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: testv4i32:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vpopcntb %xmm0, %xmm0
|
|
; BITALG-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; BITALG-NEXT: vpunpckhdq {{.*#+}} xmm2 = xmm0[2],xmm1[2],xmm0[3],xmm1[3]
|
|
; BITALG-NEXT: vpsadbw %xmm1, %xmm2, %xmm2
|
|
; BITALG-NEXT: vpmovzxdq {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero
|
|
; BITALG-NEXT: vpsadbw %xmm1, %xmm0, %xmm0
|
|
; BITALG-NEXT: vpackuswb %xmm2, %xmm0, %xmm0
|
|
; BITALG-NEXT: retq
|
|
%out = call <4 x i32> @llvm.ctpop.v4i32(<4 x i32> %in)
|
|
ret <4 x i32> %out
|
|
}
|
|
|
|
define <8 x i16> @testv8i16(<8 x i16> %in) nounwind {
|
|
; SSE2-LABEL: testv8i16:
|
|
; SSE2: # %bb.0:
|
|
; SSE2-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE2-NEXT: psrlw $1, %xmm1
|
|
; SSE2-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE2-NEXT: psubb %xmm1, %xmm0
|
|
; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
|
|
; SSE2-NEXT: movdqa %xmm0, %xmm2
|
|
; SSE2-NEXT: pand %xmm1, %xmm2
|
|
; SSE2-NEXT: psrlw $2, %xmm0
|
|
; SSE2-NEXT: pand %xmm1, %xmm0
|
|
; SSE2-NEXT: paddb %xmm2, %xmm0
|
|
; SSE2-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE2-NEXT: psrlw $4, %xmm1
|
|
; SSE2-NEXT: paddb %xmm0, %xmm1
|
|
; SSE2-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE2-NEXT: movdqa %xmm1, %xmm0
|
|
; SSE2-NEXT: psllw $8, %xmm0
|
|
; SSE2-NEXT: paddb %xmm1, %xmm0
|
|
; SSE2-NEXT: psrlw $8, %xmm0
|
|
; SSE2-NEXT: retq
|
|
;
|
|
; SSE3-LABEL: testv8i16:
|
|
; SSE3: # %bb.0:
|
|
; SSE3-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE3-NEXT: psrlw $1, %xmm1
|
|
; SSE3-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE3-NEXT: psubb %xmm1, %xmm0
|
|
; SSE3-NEXT: movdqa {{.*#+}} xmm1 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
|
|
; SSE3-NEXT: movdqa %xmm0, %xmm2
|
|
; SSE3-NEXT: pand %xmm1, %xmm2
|
|
; SSE3-NEXT: psrlw $2, %xmm0
|
|
; SSE3-NEXT: pand %xmm1, %xmm0
|
|
; SSE3-NEXT: paddb %xmm2, %xmm0
|
|
; SSE3-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE3-NEXT: psrlw $4, %xmm1
|
|
; SSE3-NEXT: paddb %xmm0, %xmm1
|
|
; SSE3-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE3-NEXT: movdqa %xmm1, %xmm0
|
|
; SSE3-NEXT: psllw $8, %xmm0
|
|
; SSE3-NEXT: paddb %xmm1, %xmm0
|
|
; SSE3-NEXT: psrlw $8, %xmm0
|
|
; SSE3-NEXT: retq
|
|
;
|
|
; SSSE3-LABEL: testv8i16:
|
|
; SSSE3: # %bb.0:
|
|
; SSSE3-NEXT: movdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; SSSE3-NEXT: movdqa %xmm0, %xmm2
|
|
; SSSE3-NEXT: pand %xmm1, %xmm2
|
|
; SSSE3-NEXT: movdqa {{.*#+}} xmm3 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; SSSE3-NEXT: movdqa %xmm3, %xmm4
|
|
; SSSE3-NEXT: pshufb %xmm2, %xmm4
|
|
; SSSE3-NEXT: psrlw $4, %xmm0
|
|
; SSSE3-NEXT: pand %xmm1, %xmm0
|
|
; SSSE3-NEXT: pshufb %xmm0, %xmm3
|
|
; SSSE3-NEXT: paddb %xmm4, %xmm3
|
|
; SSSE3-NEXT: movdqa %xmm3, %xmm0
|
|
; SSSE3-NEXT: psllw $8, %xmm0
|
|
; SSSE3-NEXT: paddb %xmm3, %xmm0
|
|
; SSSE3-NEXT: psrlw $8, %xmm0
|
|
; SSSE3-NEXT: retq
|
|
;
|
|
; SSE41-LABEL: testv8i16:
|
|
; SSE41: # %bb.0:
|
|
; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; SSE41-NEXT: movdqa %xmm0, %xmm2
|
|
; SSE41-NEXT: pand %xmm1, %xmm2
|
|
; SSE41-NEXT: movdqa {{.*#+}} xmm3 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; SSE41-NEXT: movdqa %xmm3, %xmm4
|
|
; SSE41-NEXT: pshufb %xmm2, %xmm4
|
|
; SSE41-NEXT: psrlw $4, %xmm0
|
|
; SSE41-NEXT: pand %xmm1, %xmm0
|
|
; SSE41-NEXT: pshufb %xmm0, %xmm3
|
|
; SSE41-NEXT: paddb %xmm4, %xmm3
|
|
; SSE41-NEXT: movdqa %xmm3, %xmm0
|
|
; SSE41-NEXT: psllw $8, %xmm0
|
|
; SSE41-NEXT: paddb %xmm3, %xmm0
|
|
; SSE41-NEXT: psrlw $8, %xmm0
|
|
; SSE41-NEXT: retq
|
|
;
|
|
; AVX1-LABEL: testv8i16:
|
|
; AVX1: # %bb.0:
|
|
; AVX1-NEXT: vmovdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; AVX1-NEXT: vpand %xmm1, %xmm0, %xmm2
|
|
; AVX1-NEXT: vmovdqa {{.*#+}} xmm3 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; AVX1-NEXT: vpshufb %xmm2, %xmm3, %xmm2
|
|
; AVX1-NEXT: vpsrlw $4, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpand %xmm1, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpshufb %xmm0, %xmm3, %xmm0
|
|
; AVX1-NEXT: vpaddb %xmm2, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpsllw $8, %xmm0, %xmm1
|
|
; AVX1-NEXT: vpaddb %xmm0, %xmm1, %xmm0
|
|
; AVX1-NEXT: vpsrlw $8, %xmm0, %xmm0
|
|
; AVX1-NEXT: retq
|
|
;
|
|
; AVX2-LABEL: testv8i16:
|
|
; AVX2: # %bb.0:
|
|
; AVX2-NEXT: vmovdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; AVX2-NEXT: vpand %xmm1, %xmm0, %xmm2
|
|
; AVX2-NEXT: vmovdqa {{.*#+}} xmm3 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; AVX2-NEXT: vpshufb %xmm2, %xmm3, %xmm2
|
|
; AVX2-NEXT: vpsrlw $4, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpand %xmm1, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpshufb %xmm0, %xmm3, %xmm0
|
|
; AVX2-NEXT: vpaddb %xmm2, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpsllw $8, %xmm0, %xmm1
|
|
; AVX2-NEXT: vpaddb %xmm0, %xmm1, %xmm0
|
|
; AVX2-NEXT: vpsrlw $8, %xmm0, %xmm0
|
|
; AVX2-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQ-LABEL: testv8i16:
|
|
; AVX512VPOPCNTDQ: # %bb.0:
|
|
; AVX512VPOPCNTDQ-NEXT: vpmovzxwd {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
|
|
; AVX512VPOPCNTDQ-NEXT: vpopcntd %zmm0, %zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpmovdw %zmm0, %ymm0
|
|
; AVX512VPOPCNTDQ-NEXT: # kill: def $xmm0 killed $xmm0 killed $ymm0
|
|
; AVX512VPOPCNTDQ-NEXT: vzeroupper
|
|
; AVX512VPOPCNTDQ-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQVL-LABEL: testv8i16:
|
|
; AVX512VPOPCNTDQVL: # %bb.0:
|
|
; AVX512VPOPCNTDQVL-NEXT: vpmovzxwd {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
|
|
; AVX512VPOPCNTDQVL-NEXT: vpopcntd %ymm0, %ymm0
|
|
; AVX512VPOPCNTDQVL-NEXT: vpmovdw %ymm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: vzeroupper
|
|
; AVX512VPOPCNTDQVL-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: testv8i16:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
|
|
; BITALG_NOVLX-NEXT: vpopcntw %zmm0, %zmm0
|
|
; BITALG_NOVLX-NEXT: # kill: def $xmm0 killed $xmm0 killed $zmm0
|
|
; BITALG_NOVLX-NEXT: vzeroupper
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: testv8i16:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vpopcntw %xmm0, %xmm0
|
|
; BITALG-NEXT: retq
|
|
%out = call <8 x i16> @llvm.ctpop.v8i16(<8 x i16> %in)
|
|
ret <8 x i16> %out
|
|
}
|
|
|
|
define <16 x i8> @testv16i8(<16 x i8> %in) nounwind {
|
|
; SSE2-LABEL: testv16i8:
|
|
; SSE2: # %bb.0:
|
|
; SSE2-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE2-NEXT: psrlw $1, %xmm1
|
|
; SSE2-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE2-NEXT: psubb %xmm1, %xmm0
|
|
; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
|
|
; SSE2-NEXT: movdqa %xmm0, %xmm2
|
|
; SSE2-NEXT: pand %xmm1, %xmm2
|
|
; SSE2-NEXT: psrlw $2, %xmm0
|
|
; SSE2-NEXT: pand %xmm1, %xmm0
|
|
; SSE2-NEXT: paddb %xmm2, %xmm0
|
|
; SSE2-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE2-NEXT: psrlw $4, %xmm1
|
|
; SSE2-NEXT: paddb %xmm0, %xmm1
|
|
; SSE2-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE2-NEXT: movdqa %xmm1, %xmm0
|
|
; SSE2-NEXT: retq
|
|
;
|
|
; SSE3-LABEL: testv16i8:
|
|
; SSE3: # %bb.0:
|
|
; SSE3-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE3-NEXT: psrlw $1, %xmm1
|
|
; SSE3-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE3-NEXT: psubb %xmm1, %xmm0
|
|
; SSE3-NEXT: movdqa {{.*#+}} xmm1 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
|
|
; SSE3-NEXT: movdqa %xmm0, %xmm2
|
|
; SSE3-NEXT: pand %xmm1, %xmm2
|
|
; SSE3-NEXT: psrlw $2, %xmm0
|
|
; SSE3-NEXT: pand %xmm1, %xmm0
|
|
; SSE3-NEXT: paddb %xmm2, %xmm0
|
|
; SSE3-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE3-NEXT: psrlw $4, %xmm1
|
|
; SSE3-NEXT: paddb %xmm0, %xmm1
|
|
; SSE3-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE3-NEXT: movdqa %xmm1, %xmm0
|
|
; SSE3-NEXT: retq
|
|
;
|
|
; SSSE3-LABEL: testv16i8:
|
|
; SSSE3: # %bb.0:
|
|
; SSSE3-NEXT: movdqa {{.*#+}} xmm2 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; SSSE3-NEXT: movdqa %xmm0, %xmm3
|
|
; SSSE3-NEXT: pand %xmm2, %xmm3
|
|
; SSSE3-NEXT: movdqa {{.*#+}} xmm1 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; SSSE3-NEXT: movdqa %xmm1, %xmm4
|
|
; SSSE3-NEXT: pshufb %xmm3, %xmm4
|
|
; SSSE3-NEXT: psrlw $4, %xmm0
|
|
; SSSE3-NEXT: pand %xmm2, %xmm0
|
|
; SSSE3-NEXT: pshufb %xmm0, %xmm1
|
|
; SSSE3-NEXT: paddb %xmm4, %xmm1
|
|
; SSSE3-NEXT: movdqa %xmm1, %xmm0
|
|
; SSSE3-NEXT: retq
|
|
;
|
|
; SSE41-LABEL: testv16i8:
|
|
; SSE41: # %bb.0:
|
|
; SSE41-NEXT: movdqa {{.*#+}} xmm2 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; SSE41-NEXT: movdqa %xmm0, %xmm3
|
|
; SSE41-NEXT: pand %xmm2, %xmm3
|
|
; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; SSE41-NEXT: movdqa %xmm1, %xmm4
|
|
; SSE41-NEXT: pshufb %xmm3, %xmm4
|
|
; SSE41-NEXT: psrlw $4, %xmm0
|
|
; SSE41-NEXT: pand %xmm2, %xmm0
|
|
; SSE41-NEXT: pshufb %xmm0, %xmm1
|
|
; SSE41-NEXT: paddb %xmm4, %xmm1
|
|
; SSE41-NEXT: movdqa %xmm1, %xmm0
|
|
; SSE41-NEXT: retq
|
|
;
|
|
; AVX1-LABEL: testv16i8:
|
|
; AVX1: # %bb.0:
|
|
; AVX1-NEXT: vmovdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; AVX1-NEXT: vpand %xmm1, %xmm0, %xmm2
|
|
; AVX1-NEXT: vmovdqa {{.*#+}} xmm3 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; AVX1-NEXT: vpshufb %xmm2, %xmm3, %xmm2
|
|
; AVX1-NEXT: vpsrlw $4, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpand %xmm1, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpshufb %xmm0, %xmm3, %xmm0
|
|
; AVX1-NEXT: vpaddb %xmm2, %xmm0, %xmm0
|
|
; AVX1-NEXT: retq
|
|
;
|
|
; AVX2-LABEL: testv16i8:
|
|
; AVX2: # %bb.0:
|
|
; AVX2-NEXT: vmovdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
|
|
; AVX2-NEXT: vpand %xmm1, %xmm0, %xmm2
|
|
; AVX2-NEXT: vmovdqa {{.*#+}} xmm3 = [0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4]
|
|
; AVX2-NEXT: vpshufb %xmm2, %xmm3, %xmm2
|
|
; AVX2-NEXT: vpsrlw $4, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpand %xmm1, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpshufb %xmm0, %xmm3, %xmm0
|
|
; AVX2-NEXT: vpaddb %xmm2, %xmm0, %xmm0
|
|
; AVX2-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQ-LABEL: testv16i8:
|
|
; AVX512VPOPCNTDQ: # %bb.0:
|
|
; AVX512VPOPCNTDQ-NEXT: vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
|
|
; AVX512VPOPCNTDQ-NEXT: vpopcntd %zmm0, %zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpmovdb %zmm0, %xmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vzeroupper
|
|
; AVX512VPOPCNTDQ-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQVL-LABEL: testv16i8:
|
|
; AVX512VPOPCNTDQVL: # %bb.0:
|
|
; AVX512VPOPCNTDQVL-NEXT: vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
|
|
; AVX512VPOPCNTDQVL-NEXT: vpopcntd %zmm0, %zmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: vpmovdb %zmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: vzeroupper
|
|
; AVX512VPOPCNTDQVL-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: testv16i8:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
|
|
; BITALG_NOVLX-NEXT: vpopcntb %zmm0, %zmm0
|
|
; BITALG_NOVLX-NEXT: # kill: def $xmm0 killed $xmm0 killed $zmm0
|
|
; BITALG_NOVLX-NEXT: vzeroupper
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: testv16i8:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vpopcntb %xmm0, %xmm0
|
|
; BITALG-NEXT: retq
|
|
%out = call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %in)
|
|
ret <16 x i8> %out
|
|
}
|
|
|
|
define <2 x i64> @foldv2i64() nounwind {
|
|
; SSE-LABEL: foldv2i64:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: movaps {{.*#+}} xmm0 = [1,64]
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX-LABEL: foldv2i64:
|
|
; AVX: # %bb.0:
|
|
; AVX-NEXT: vmovaps {{.*#+}} xmm0 = [1,64]
|
|
; AVX-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: foldv2i64:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: vmovaps {{.*#+}} xmm0 = [1,64]
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: foldv2i64:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vmovaps {{.*#+}} xmm0 = [1,64]
|
|
; BITALG-NEXT: retq
|
|
%out = call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> <i64 256, i64 -1>)
|
|
ret <2 x i64> %out
|
|
}
|
|
|
|
define <4 x i32> @foldv4i32() nounwind {
|
|
; SSE-LABEL: foldv4i32:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: movaps {{.*#+}} xmm0 = [1,32,0,8]
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX-LABEL: foldv4i32:
|
|
; AVX: # %bb.0:
|
|
; AVX-NEXT: vmovaps {{.*#+}} xmm0 = [1,32,0,8]
|
|
; AVX-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: foldv4i32:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: vmovaps {{.*#+}} xmm0 = [1,32,0,8]
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: foldv4i32:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vmovaps {{.*#+}} xmm0 = [1,32,0,8]
|
|
; BITALG-NEXT: retq
|
|
%out = call <4 x i32> @llvm.ctpop.v4i32(<4 x i32> <i32 256, i32 -1, i32 0, i32 255>)
|
|
ret <4 x i32> %out
|
|
}
|
|
|
|
define <8 x i16> @foldv8i16() nounwind {
|
|
; SSE-LABEL: foldv8i16:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: movaps {{.*#+}} xmm0 = [1,16,0,8,0,3,2,3]
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX-LABEL: foldv8i16:
|
|
; AVX: # %bb.0:
|
|
; AVX-NEXT: vmovaps {{.*#+}} xmm0 = [1,16,0,8,0,3,2,3]
|
|
; AVX-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: foldv8i16:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: vmovaps {{.*#+}} xmm0 = [1,16,0,8,0,3,2,3]
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: foldv8i16:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vmovaps {{.*#+}} xmm0 = [1,16,0,8,0,3,2,3]
|
|
; BITALG-NEXT: retq
|
|
%out = call <8 x i16> @llvm.ctpop.v8i16(<8 x i16> <i16 256, i16 -1, i16 0, i16 255, i16 -65536, i16 7, i16 24, i16 88>)
|
|
ret <8 x i16> %out
|
|
}
|
|
|
|
define <16 x i8> @foldv16i8() nounwind {
|
|
; SSE-LABEL: foldv16i8:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: movaps {{.*#+}} xmm0 = [0,8,0,8,0,3,2,3,7,7,1,1,1,1,1,1]
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX-LABEL: foldv16i8:
|
|
; AVX: # %bb.0:
|
|
; AVX-NEXT: vmovaps {{.*#+}} xmm0 = [0,8,0,8,0,3,2,3,7,7,1,1,1,1,1,1]
|
|
; AVX-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: foldv16i8:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: vmovaps {{.*#+}} xmm0 = [0,8,0,8,0,3,2,3,7,7,1,1,1,1,1,1]
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: foldv16i8:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vmovaps {{.*#+}} xmm0 = [0,8,0,8,0,3,2,3,7,7,1,1,1,1,1,1]
|
|
; BITALG-NEXT: retq
|
|
%out = call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> <i8 256, i8 -1, i8 0, i8 255, i8 -65536, i8 7, i8 24, i8 88, i8 -2, i8 254, i8 1, i8 2, i8 4, i8 8, i8 16, i8 32>)
|
|
ret <16 x i8> %out
|
|
}
|
|
|
|
define <2 x i64> @eq_1_v2i64(<2 x i64> %0) {
|
|
; SSE2-LABEL: eq_1_v2i64:
|
|
; SSE2: # %bb.0:
|
|
; SSE2-NEXT: pxor %xmm1, %xmm1
|
|
; SSE2-NEXT: pcmpeqd %xmm2, %xmm2
|
|
; SSE2-NEXT: paddq %xmm0, %xmm2
|
|
; SSE2-NEXT: pand %xmm0, %xmm2
|
|
; SSE2-NEXT: pcmpeqd %xmm1, %xmm0
|
|
; SSE2-NEXT: pshufd {{.*#+}} xmm3 = xmm0[1,0,3,2]
|
|
; SSE2-NEXT: pand %xmm3, %xmm0
|
|
; SSE2-NEXT: pcmpeqd %xmm1, %xmm2
|
|
; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm2[1,0,3,2]
|
|
; SSE2-NEXT: pand %xmm2, %xmm1
|
|
; SSE2-NEXT: pandn %xmm1, %xmm0
|
|
; SSE2-NEXT: retq
|
|
;
|
|
; SSE3-LABEL: eq_1_v2i64:
|
|
; SSE3: # %bb.0:
|
|
; SSE3-NEXT: pxor %xmm1, %xmm1
|
|
; SSE3-NEXT: pcmpeqd %xmm2, %xmm2
|
|
; SSE3-NEXT: paddq %xmm0, %xmm2
|
|
; SSE3-NEXT: pand %xmm0, %xmm2
|
|
; SSE3-NEXT: pcmpeqd %xmm1, %xmm0
|
|
; SSE3-NEXT: pshufd {{.*#+}} xmm3 = xmm0[1,0,3,2]
|
|
; SSE3-NEXT: pand %xmm3, %xmm0
|
|
; SSE3-NEXT: pcmpeqd %xmm1, %xmm2
|
|
; SSE3-NEXT: pshufd {{.*#+}} xmm1 = xmm2[1,0,3,2]
|
|
; SSE3-NEXT: pand %xmm2, %xmm1
|
|
; SSE3-NEXT: pandn %xmm1, %xmm0
|
|
; SSE3-NEXT: retq
|
|
;
|
|
; SSSE3-LABEL: eq_1_v2i64:
|
|
; SSSE3: # %bb.0:
|
|
; SSSE3-NEXT: pxor %xmm1, %xmm1
|
|
; SSSE3-NEXT: pcmpeqd %xmm2, %xmm2
|
|
; SSSE3-NEXT: paddq %xmm0, %xmm2
|
|
; SSSE3-NEXT: pand %xmm0, %xmm2
|
|
; SSSE3-NEXT: pcmpeqd %xmm1, %xmm0
|
|
; SSSE3-NEXT: pshufd {{.*#+}} xmm3 = xmm0[1,0,3,2]
|
|
; SSSE3-NEXT: pand %xmm3, %xmm0
|
|
; SSSE3-NEXT: pcmpeqd %xmm1, %xmm2
|
|
; SSSE3-NEXT: pshufd {{.*#+}} xmm1 = xmm2[1,0,3,2]
|
|
; SSSE3-NEXT: pand %xmm2, %xmm1
|
|
; SSSE3-NEXT: pandn %xmm1, %xmm0
|
|
; SSSE3-NEXT: retq
|
|
;
|
|
; SSE41-LABEL: eq_1_v2i64:
|
|
; SSE41: # %bb.0:
|
|
; SSE41-NEXT: pxor %xmm1, %xmm1
|
|
; SSE41-NEXT: pcmpeqd %xmm2, %xmm2
|
|
; SSE41-NEXT: paddq %xmm0, %xmm2
|
|
; SSE41-NEXT: pand %xmm0, %xmm2
|
|
; SSE41-NEXT: pcmpeqq %xmm1, %xmm0
|
|
; SSE41-NEXT: pcmpeqq %xmm1, %xmm2
|
|
; SSE41-NEXT: pandn %xmm2, %xmm0
|
|
; SSE41-NEXT: retq
|
|
;
|
|
; AVX1-LABEL: eq_1_v2i64:
|
|
; AVX1: # %bb.0:
|
|
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX1-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm2
|
|
; AVX1-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX1-NEXT: vpaddq %xmm3, %xmm0, %xmm3
|
|
; AVX1-NEXT: vpand %xmm3, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpandn %xmm0, %xmm2, %xmm0
|
|
; AVX1-NEXT: retq
|
|
;
|
|
; AVX2-LABEL: eq_1_v2i64:
|
|
; AVX2: # %bb.0:
|
|
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX2-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm2
|
|
; AVX2-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX2-NEXT: vpaddq %xmm3, %xmm0, %xmm3
|
|
; AVX2-NEXT: vpand %xmm3, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpandn %xmm0, %xmm2, %xmm0
|
|
; AVX2-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQ-LABEL: eq_1_v2i64:
|
|
; AVX512VPOPCNTDQ: # %bb.0:
|
|
; AVX512VPOPCNTDQ-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpopcntq %zmm0, %zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpcmpeqq {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vzeroupper
|
|
; AVX512VPOPCNTDQ-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQVL-LABEL: eq_1_v2i64:
|
|
; AVX512VPOPCNTDQVL: # %bb.0:
|
|
; AVX512VPOPCNTDQVL-NEXT: vpopcntq %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: vpcmpeqq {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: eq_1_v2i64:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; BITALG_NOVLX-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm2
|
|
; BITALG_NOVLX-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; BITALG_NOVLX-NEXT: vpaddq %xmm3, %xmm0, %xmm3
|
|
; BITALG_NOVLX-NEXT: vpand %xmm3, %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vpandn %xmm0, %xmm2, %xmm0
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: eq_1_v2i64:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; BITALG-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm2
|
|
; BITALG-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; BITALG-NEXT: vpaddq %xmm3, %xmm0, %xmm3
|
|
; BITALG-NEXT: vpand %xmm3, %xmm0, %xmm0
|
|
; BITALG-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm0
|
|
; BITALG-NEXT: vpandn %xmm0, %xmm2, %xmm0
|
|
; BITALG-NEXT: retq
|
|
%2 = tail call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> %0)
|
|
%3 = icmp eq <2 x i64> %2, <i64 1, i64 1>
|
|
%4 = sext <2 x i1> %3 to <2 x i64>
|
|
ret <2 x i64> %4
|
|
}
|
|
|
|
define <2 x i64> @ne_1_v2i64(<2 x i64> %0) {
|
|
; SSE2-LABEL: ne_1_v2i64:
|
|
; SSE2: # %bb.0:
|
|
; SSE2-NEXT: pxor %xmm1, %xmm1
|
|
; SSE2-NEXT: pcmpeqd %xmm2, %xmm2
|
|
; SSE2-NEXT: movdqa %xmm0, %xmm3
|
|
; SSE2-NEXT: paddq %xmm2, %xmm3
|
|
; SSE2-NEXT: pand %xmm0, %xmm3
|
|
; SSE2-NEXT: pcmpeqd %xmm1, %xmm0
|
|
; SSE2-NEXT: pshufd {{.*#+}} xmm4 = xmm0[1,0,3,2]
|
|
; SSE2-NEXT: pand %xmm4, %xmm0
|
|
; SSE2-NEXT: pcmpeqd %xmm1, %xmm3
|
|
; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm3[1,0,3,2]
|
|
; SSE2-NEXT: pand %xmm3, %xmm1
|
|
; SSE2-NEXT: pxor %xmm2, %xmm1
|
|
; SSE2-NEXT: por %xmm1, %xmm0
|
|
; SSE2-NEXT: retq
|
|
;
|
|
; SSE3-LABEL: ne_1_v2i64:
|
|
; SSE3: # %bb.0:
|
|
; SSE3-NEXT: pxor %xmm1, %xmm1
|
|
; SSE3-NEXT: pcmpeqd %xmm2, %xmm2
|
|
; SSE3-NEXT: movdqa %xmm0, %xmm3
|
|
; SSE3-NEXT: paddq %xmm2, %xmm3
|
|
; SSE3-NEXT: pand %xmm0, %xmm3
|
|
; SSE3-NEXT: pcmpeqd %xmm1, %xmm0
|
|
; SSE3-NEXT: pshufd {{.*#+}} xmm4 = xmm0[1,0,3,2]
|
|
; SSE3-NEXT: pand %xmm4, %xmm0
|
|
; SSE3-NEXT: pcmpeqd %xmm1, %xmm3
|
|
; SSE3-NEXT: pshufd {{.*#+}} xmm1 = xmm3[1,0,3,2]
|
|
; SSE3-NEXT: pand %xmm3, %xmm1
|
|
; SSE3-NEXT: pxor %xmm2, %xmm1
|
|
; SSE3-NEXT: por %xmm1, %xmm0
|
|
; SSE3-NEXT: retq
|
|
;
|
|
; SSSE3-LABEL: ne_1_v2i64:
|
|
; SSSE3: # %bb.0:
|
|
; SSSE3-NEXT: pxor %xmm1, %xmm1
|
|
; SSSE3-NEXT: pcmpeqd %xmm2, %xmm2
|
|
; SSSE3-NEXT: movdqa %xmm0, %xmm3
|
|
; SSSE3-NEXT: paddq %xmm2, %xmm3
|
|
; SSSE3-NEXT: pand %xmm0, %xmm3
|
|
; SSSE3-NEXT: pcmpeqd %xmm1, %xmm0
|
|
; SSSE3-NEXT: pshufd {{.*#+}} xmm4 = xmm0[1,0,3,2]
|
|
; SSSE3-NEXT: pand %xmm4, %xmm0
|
|
; SSSE3-NEXT: pcmpeqd %xmm1, %xmm3
|
|
; SSSE3-NEXT: pshufd {{.*#+}} xmm1 = xmm3[1,0,3,2]
|
|
; SSSE3-NEXT: pand %xmm3, %xmm1
|
|
; SSSE3-NEXT: pxor %xmm2, %xmm1
|
|
; SSSE3-NEXT: por %xmm1, %xmm0
|
|
; SSSE3-NEXT: retq
|
|
;
|
|
; SSE41-LABEL: ne_1_v2i64:
|
|
; SSE41: # %bb.0:
|
|
; SSE41-NEXT: pxor %xmm2, %xmm2
|
|
; SSE41-NEXT: pcmpeqd %xmm3, %xmm3
|
|
; SSE41-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE41-NEXT: paddq %xmm3, %xmm1
|
|
; SSE41-NEXT: pand %xmm0, %xmm1
|
|
; SSE41-NEXT: pcmpeqq %xmm2, %xmm0
|
|
; SSE41-NEXT: pcmpeqq %xmm2, %xmm1
|
|
; SSE41-NEXT: pxor %xmm3, %xmm1
|
|
; SSE41-NEXT: por %xmm0, %xmm1
|
|
; SSE41-NEXT: movdqa %xmm1, %xmm0
|
|
; SSE41-NEXT: retq
|
|
;
|
|
; AVX1-LABEL: ne_1_v2i64:
|
|
; AVX1: # %bb.0:
|
|
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX1-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm2
|
|
; AVX1-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX1-NEXT: vpaddq %xmm3, %xmm0, %xmm4
|
|
; AVX1-NEXT: vpand %xmm4, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpxor %xmm3, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpor %xmm0, %xmm2, %xmm0
|
|
; AVX1-NEXT: retq
|
|
;
|
|
; AVX2-LABEL: ne_1_v2i64:
|
|
; AVX2: # %bb.0:
|
|
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX2-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm2
|
|
; AVX2-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX2-NEXT: vpaddq %xmm3, %xmm0, %xmm4
|
|
; AVX2-NEXT: vpand %xmm4, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpxor %xmm3, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpor %xmm0, %xmm2, %xmm0
|
|
; AVX2-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQ-LABEL: ne_1_v2i64:
|
|
; AVX512VPOPCNTDQ: # %bb.0:
|
|
; AVX512VPOPCNTDQ-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpopcntq %zmm0, %zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpcmpeqq {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpternlogq $15, %zmm0, %zmm0, %zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: # kill: def $xmm0 killed $xmm0 killed $zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vzeroupper
|
|
; AVX512VPOPCNTDQ-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQVL-LABEL: ne_1_v2i64:
|
|
; AVX512VPOPCNTDQVL: # %bb.0:
|
|
; AVX512VPOPCNTDQVL-NEXT: vpopcntq %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: vpcmpeqq {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: vpternlogq $15, %xmm0, %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: ne_1_v2i64:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; BITALG_NOVLX-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm2
|
|
; BITALG_NOVLX-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; BITALG_NOVLX-NEXT: vpaddq %xmm3, %xmm0, %xmm3
|
|
; BITALG_NOVLX-NEXT: vpand %xmm3, %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vpternlogq $15, %zmm0, %zmm0, %zmm0
|
|
; BITALG_NOVLX-NEXT: vpor %xmm0, %xmm2, %xmm0
|
|
; BITALG_NOVLX-NEXT: vzeroupper
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: ne_1_v2i64:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; BITALG-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm2
|
|
; BITALG-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; BITALG-NEXT: vpaddq %xmm3, %xmm0, %xmm4
|
|
; BITALG-NEXT: vpand %xmm4, %xmm0, %xmm0
|
|
; BITALG-NEXT: vpcmpeqq %xmm1, %xmm0, %xmm0
|
|
; BITALG-NEXT: vpternlogq $222, %xmm3, %xmm2, %xmm0
|
|
; BITALG-NEXT: retq
|
|
%2 = tail call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> %0)
|
|
%3 = icmp ne <2 x i64> %2, <i64 1, i64 1>
|
|
%4 = sext <2 x i1> %3 to <2 x i64>
|
|
ret <2 x i64> %4
|
|
}
|
|
|
|
define <4 x i32> @eq_1_v4i32(<4 x i32> %0) {
|
|
; SSE-LABEL: eq_1_v4i32:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: pxor %xmm1, %xmm1
|
|
; SSE-NEXT: pcmpeqd %xmm2, %xmm2
|
|
; SSE-NEXT: paddd %xmm0, %xmm2
|
|
; SSE-NEXT: pand %xmm0, %xmm2
|
|
; SSE-NEXT: pcmpeqd %xmm1, %xmm0
|
|
; SSE-NEXT: pcmpeqd %xmm1, %xmm2
|
|
; SSE-NEXT: pandn %xmm2, %xmm0
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX1-LABEL: eq_1_v4i32:
|
|
; AVX1: # %bb.0:
|
|
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX1-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm2
|
|
; AVX1-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX1-NEXT: vpaddd %xmm3, %xmm0, %xmm3
|
|
; AVX1-NEXT: vpand %xmm3, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpandn %xmm0, %xmm2, %xmm0
|
|
; AVX1-NEXT: retq
|
|
;
|
|
; AVX2-LABEL: eq_1_v4i32:
|
|
; AVX2: # %bb.0:
|
|
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX2-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm2
|
|
; AVX2-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX2-NEXT: vpaddd %xmm3, %xmm0, %xmm3
|
|
; AVX2-NEXT: vpand %xmm3, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpandn %xmm0, %xmm2, %xmm0
|
|
; AVX2-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQ-LABEL: eq_1_v4i32:
|
|
; AVX512VPOPCNTDQ: # %bb.0:
|
|
; AVX512VPOPCNTDQ-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpopcntd %zmm0, %zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpbroadcastd {{.*#+}} xmm1 = [1,1,1,1]
|
|
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vzeroupper
|
|
; AVX512VPOPCNTDQ-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQVL-LABEL: eq_1_v4i32:
|
|
; AVX512VPOPCNTDQVL: # %bb.0:
|
|
; AVX512VPOPCNTDQVL-NEXT: vpopcntd %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: vpbroadcastd {{.*#+}} xmm1 = [1,1,1,1]
|
|
; AVX512VPOPCNTDQVL-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: eq_1_v4i32:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; BITALG_NOVLX-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm2
|
|
; BITALG_NOVLX-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; BITALG_NOVLX-NEXT: vpaddd %xmm3, %xmm0, %xmm3
|
|
; BITALG_NOVLX-NEXT: vpand %xmm3, %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vpandn %xmm0, %xmm2, %xmm0
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: eq_1_v4i32:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; BITALG-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm2
|
|
; BITALG-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; BITALG-NEXT: vpaddd %xmm3, %xmm0, %xmm3
|
|
; BITALG-NEXT: vpand %xmm3, %xmm0, %xmm0
|
|
; BITALG-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm0
|
|
; BITALG-NEXT: vpandn %xmm0, %xmm2, %xmm0
|
|
; BITALG-NEXT: retq
|
|
%2 = tail call <4 x i32> @llvm.ctpop.v4i32(<4 x i32> %0)
|
|
%3 = icmp eq <4 x i32> %2, <i32 1, i32 1, i32 1, i32 1>
|
|
%4 = sext <4 x i1> %3 to <4 x i32>
|
|
ret <4 x i32> %4
|
|
}
|
|
|
|
define <4 x i32> @ne_1_v4i32(<4 x i32> %0) {
|
|
; SSE-LABEL: ne_1_v4i32:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: pxor %xmm2, %xmm2
|
|
; SSE-NEXT: pcmpeqd %xmm3, %xmm3
|
|
; SSE-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE-NEXT: paddd %xmm3, %xmm1
|
|
; SSE-NEXT: pand %xmm0, %xmm1
|
|
; SSE-NEXT: pcmpeqd %xmm2, %xmm0
|
|
; SSE-NEXT: pcmpeqd %xmm2, %xmm1
|
|
; SSE-NEXT: pxor %xmm3, %xmm1
|
|
; SSE-NEXT: por %xmm0, %xmm1
|
|
; SSE-NEXT: movdqa %xmm1, %xmm0
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX1-LABEL: ne_1_v4i32:
|
|
; AVX1: # %bb.0:
|
|
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX1-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm2
|
|
; AVX1-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX1-NEXT: vpaddd %xmm3, %xmm0, %xmm4
|
|
; AVX1-NEXT: vpand %xmm4, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpxor %xmm3, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpor %xmm0, %xmm2, %xmm0
|
|
; AVX1-NEXT: retq
|
|
;
|
|
; AVX2-LABEL: ne_1_v4i32:
|
|
; AVX2: # %bb.0:
|
|
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX2-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm2
|
|
; AVX2-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX2-NEXT: vpaddd %xmm3, %xmm0, %xmm4
|
|
; AVX2-NEXT: vpand %xmm4, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpxor %xmm3, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpor %xmm0, %xmm2, %xmm0
|
|
; AVX2-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQ-LABEL: ne_1_v4i32:
|
|
; AVX512VPOPCNTDQ: # %bb.0:
|
|
; AVX512VPOPCNTDQ-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpopcntd %zmm0, %zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpbroadcastd {{.*#+}} xmm1 = [1,1,1,1]
|
|
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpternlogq $15, %zmm0, %zmm0, %zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: # kill: def $xmm0 killed $xmm0 killed $zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vzeroupper
|
|
; AVX512VPOPCNTDQ-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQVL-LABEL: ne_1_v4i32:
|
|
; AVX512VPOPCNTDQVL: # %bb.0:
|
|
; AVX512VPOPCNTDQVL-NEXT: vpopcntd %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: vpbroadcastd {{.*#+}} xmm1 = [1,1,1,1]
|
|
; AVX512VPOPCNTDQVL-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: vpternlogq $15, %xmm0, %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: ne_1_v4i32:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; BITALG_NOVLX-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm2
|
|
; BITALG_NOVLX-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; BITALG_NOVLX-NEXT: vpaddd %xmm3, %xmm0, %xmm3
|
|
; BITALG_NOVLX-NEXT: vpand %xmm3, %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vpternlogq $15, %zmm0, %zmm0, %zmm0
|
|
; BITALG_NOVLX-NEXT: vpor %xmm0, %xmm2, %xmm0
|
|
; BITALG_NOVLX-NEXT: vzeroupper
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: ne_1_v4i32:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; BITALG-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm2
|
|
; BITALG-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; BITALG-NEXT: vpaddd %xmm3, %xmm0, %xmm4
|
|
; BITALG-NEXT: vpand %xmm4, %xmm0, %xmm0
|
|
; BITALG-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm0
|
|
; BITALG-NEXT: vpternlogd $222, %xmm3, %xmm2, %xmm0
|
|
; BITALG-NEXT: retq
|
|
%2 = tail call <4 x i32> @llvm.ctpop.v4i32(<4 x i32> %0)
|
|
%3 = icmp ne <4 x i32> %2, <i32 1, i32 1, i32 1, i32 1>
|
|
%4 = sext <4 x i1> %3 to <4 x i32>
|
|
ret <4 x i32> %4
|
|
}
|
|
|
|
define <8 x i16> @eq_1_v8i16(<8 x i16> %0) {
|
|
; SSE-LABEL: eq_1_v8i16:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: pxor %xmm1, %xmm1
|
|
; SSE-NEXT: pcmpeqd %xmm2, %xmm2
|
|
; SSE-NEXT: paddw %xmm0, %xmm2
|
|
; SSE-NEXT: pand %xmm0, %xmm2
|
|
; SSE-NEXT: pcmpeqw %xmm1, %xmm0
|
|
; SSE-NEXT: pcmpeqw %xmm1, %xmm2
|
|
; SSE-NEXT: pandn %xmm2, %xmm0
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX-LABEL: eq_1_v8i16:
|
|
; AVX: # %bb.0:
|
|
; AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX-NEXT: vpcmpeqw %xmm1, %xmm0, %xmm2
|
|
; AVX-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX-NEXT: vpaddw %xmm3, %xmm0, %xmm3
|
|
; AVX-NEXT: vpand %xmm3, %xmm0, %xmm0
|
|
; AVX-NEXT: vpcmpeqw %xmm1, %xmm0, %xmm0
|
|
; AVX-NEXT: vpandn %xmm0, %xmm2, %xmm0
|
|
; AVX-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: eq_1_v8i16:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
|
|
; BITALG_NOVLX-NEXT: vpopcntw %zmm0, %zmm0
|
|
; BITALG_NOVLX-NEXT: vpcmpeqw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vzeroupper
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: eq_1_v8i16:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vpopcntw %xmm0, %xmm0
|
|
; BITALG-NEXT: vpcmpeqw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; BITALG-NEXT: retq
|
|
%2 = tail call <8 x i16> @llvm.ctpop.v8i16(<8 x i16> %0)
|
|
%3 = icmp eq <8 x i16> %2, <i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1>
|
|
%4 = sext <8 x i1> %3 to <8 x i16>
|
|
ret <8 x i16> %4
|
|
}
|
|
|
|
define <8 x i16> @ne_1_v8i16(<8 x i16> %0) {
|
|
; SSE-LABEL: ne_1_v8i16:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: pxor %xmm2, %xmm2
|
|
; SSE-NEXT: pcmpeqd %xmm3, %xmm3
|
|
; SSE-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE-NEXT: paddw %xmm3, %xmm1
|
|
; SSE-NEXT: pand %xmm0, %xmm1
|
|
; SSE-NEXT: pcmpeqw %xmm2, %xmm0
|
|
; SSE-NEXT: pcmpeqw %xmm2, %xmm1
|
|
; SSE-NEXT: pxor %xmm3, %xmm1
|
|
; SSE-NEXT: por %xmm0, %xmm1
|
|
; SSE-NEXT: movdqa %xmm1, %xmm0
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX1-LABEL: ne_1_v8i16:
|
|
; AVX1: # %bb.0:
|
|
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX1-NEXT: vpcmpeqw %xmm1, %xmm0, %xmm2
|
|
; AVX1-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX1-NEXT: vpaddw %xmm3, %xmm0, %xmm4
|
|
; AVX1-NEXT: vpand %xmm4, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpcmpeqw %xmm1, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpxor %xmm3, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpor %xmm0, %xmm2, %xmm0
|
|
; AVX1-NEXT: retq
|
|
;
|
|
; AVX2-LABEL: ne_1_v8i16:
|
|
; AVX2: # %bb.0:
|
|
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX2-NEXT: vpcmpeqw %xmm1, %xmm0, %xmm2
|
|
; AVX2-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX2-NEXT: vpaddw %xmm3, %xmm0, %xmm4
|
|
; AVX2-NEXT: vpand %xmm4, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpcmpeqw %xmm1, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpxor %xmm3, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpor %xmm0, %xmm2, %xmm0
|
|
; AVX2-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQ-LABEL: ne_1_v8i16:
|
|
; AVX512VPOPCNTDQ: # %bb.0:
|
|
; AVX512VPOPCNTDQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX512VPOPCNTDQ-NEXT: vpcmpeqw %xmm1, %xmm0, %xmm2
|
|
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX512VPOPCNTDQ-NEXT: vpaddw %xmm3, %xmm0, %xmm3
|
|
; AVX512VPOPCNTDQ-NEXT: vpand %xmm3, %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpcmpeqw %xmm1, %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpternlogq $15, %zmm0, %zmm0, %zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpor %xmm0, %xmm2, %xmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vzeroupper
|
|
; AVX512VPOPCNTDQ-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQVL-LABEL: ne_1_v8i16:
|
|
; AVX512VPOPCNTDQVL: # %bb.0:
|
|
; AVX512VPOPCNTDQVL-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX512VPOPCNTDQVL-NEXT: vpcmpeqw %xmm1, %xmm0, %xmm2
|
|
; AVX512VPOPCNTDQVL-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX512VPOPCNTDQVL-NEXT: vpaddw %xmm3, %xmm0, %xmm4
|
|
; AVX512VPOPCNTDQVL-NEXT: vpand %xmm4, %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: vpcmpeqw %xmm1, %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: vpternlogq $222, %xmm3, %xmm2, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: ne_1_v8i16:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
|
|
; BITALG_NOVLX-NEXT: vpopcntw %zmm0, %zmm0
|
|
; BITALG_NOVLX-NEXT: vpcmpeqw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vpternlogq $15, %zmm0, %zmm0, %zmm0
|
|
; BITALG_NOVLX-NEXT: # kill: def $xmm0 killed $xmm0 killed $zmm0
|
|
; BITALG_NOVLX-NEXT: vzeroupper
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: ne_1_v8i16:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vpopcntw %xmm0, %xmm0
|
|
; BITALG-NEXT: vpcmpeqw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; BITALG-NEXT: vpternlogq $15, %xmm0, %xmm0, %xmm0
|
|
; BITALG-NEXT: retq
|
|
%2 = tail call <8 x i16> @llvm.ctpop.v8i16(<8 x i16> %0)
|
|
%3 = icmp ne <8 x i16> %2, <i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1>
|
|
%4 = sext <8 x i1> %3 to <8 x i16>
|
|
ret <8 x i16> %4
|
|
}
|
|
|
|
define <16 x i8> @eq_1_v16i8(<16 x i8> %0) {
|
|
; SSE-LABEL: eq_1_v16i8:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: pxor %xmm1, %xmm1
|
|
; SSE-NEXT: pcmpeqd %xmm2, %xmm2
|
|
; SSE-NEXT: paddb %xmm0, %xmm2
|
|
; SSE-NEXT: pand %xmm0, %xmm2
|
|
; SSE-NEXT: pcmpeqb %xmm1, %xmm0
|
|
; SSE-NEXT: pcmpeqb %xmm1, %xmm2
|
|
; SSE-NEXT: pandn %xmm2, %xmm0
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX-LABEL: eq_1_v16i8:
|
|
; AVX: # %bb.0:
|
|
; AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX-NEXT: vpcmpeqb %xmm1, %xmm0, %xmm2
|
|
; AVX-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX-NEXT: vpaddb %xmm3, %xmm0, %xmm3
|
|
; AVX-NEXT: vpand %xmm3, %xmm0, %xmm0
|
|
; AVX-NEXT: vpcmpeqb %xmm1, %xmm0, %xmm0
|
|
; AVX-NEXT: vpandn %xmm0, %xmm2, %xmm0
|
|
; AVX-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: eq_1_v16i8:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
|
|
; BITALG_NOVLX-NEXT: vpopcntb %zmm0, %zmm0
|
|
; BITALG_NOVLX-NEXT: vpcmpeqb {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vzeroupper
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: eq_1_v16i8:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vpopcntb %xmm0, %xmm0
|
|
; BITALG-NEXT: vpcmpeqb {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; BITALG-NEXT: retq
|
|
%2 = tail call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %0)
|
|
%3 = icmp eq <16 x i8> %2, <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
|
|
%4 = sext <16 x i1> %3 to <16 x i8>
|
|
ret <16 x i8> %4
|
|
}
|
|
|
|
define <16 x i8> @ne_1_v16i8(<16 x i8> %0) {
|
|
; SSE-LABEL: ne_1_v16i8:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: pxor %xmm2, %xmm2
|
|
; SSE-NEXT: pcmpeqd %xmm3, %xmm3
|
|
; SSE-NEXT: movdqa %xmm0, %xmm1
|
|
; SSE-NEXT: paddb %xmm3, %xmm1
|
|
; SSE-NEXT: pand %xmm0, %xmm1
|
|
; SSE-NEXT: pcmpeqb %xmm2, %xmm0
|
|
; SSE-NEXT: pcmpeqb %xmm2, %xmm1
|
|
; SSE-NEXT: pxor %xmm3, %xmm1
|
|
; SSE-NEXT: por %xmm0, %xmm1
|
|
; SSE-NEXT: movdqa %xmm1, %xmm0
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX1-LABEL: ne_1_v16i8:
|
|
; AVX1: # %bb.0:
|
|
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX1-NEXT: vpcmpeqb %xmm1, %xmm0, %xmm2
|
|
; AVX1-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX1-NEXT: vpaddb %xmm3, %xmm0, %xmm4
|
|
; AVX1-NEXT: vpand %xmm4, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpcmpeqb %xmm1, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpxor %xmm3, %xmm0, %xmm0
|
|
; AVX1-NEXT: vpor %xmm0, %xmm2, %xmm0
|
|
; AVX1-NEXT: retq
|
|
;
|
|
; AVX2-LABEL: ne_1_v16i8:
|
|
; AVX2: # %bb.0:
|
|
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX2-NEXT: vpcmpeqb %xmm1, %xmm0, %xmm2
|
|
; AVX2-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX2-NEXT: vpaddb %xmm3, %xmm0, %xmm4
|
|
; AVX2-NEXT: vpand %xmm4, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpcmpeqb %xmm1, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpxor %xmm3, %xmm0, %xmm0
|
|
; AVX2-NEXT: vpor %xmm0, %xmm2, %xmm0
|
|
; AVX2-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQ-LABEL: ne_1_v16i8:
|
|
; AVX512VPOPCNTDQ: # %bb.0:
|
|
; AVX512VPOPCNTDQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX512VPOPCNTDQ-NEXT: vpcmpeqb %xmm1, %xmm0, %xmm2
|
|
; AVX512VPOPCNTDQ-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX512VPOPCNTDQ-NEXT: vpaddb %xmm3, %xmm0, %xmm3
|
|
; AVX512VPOPCNTDQ-NEXT: vpand %xmm3, %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpcmpeqb %xmm1, %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpternlogq $15, %zmm0, %zmm0, %zmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vpor %xmm0, %xmm2, %xmm0
|
|
; AVX512VPOPCNTDQ-NEXT: vzeroupper
|
|
; AVX512VPOPCNTDQ-NEXT: retq
|
|
;
|
|
; AVX512VPOPCNTDQVL-LABEL: ne_1_v16i8:
|
|
; AVX512VPOPCNTDQVL: # %bb.0:
|
|
; AVX512VPOPCNTDQVL-NEXT: vpxor %xmm1, %xmm1, %xmm1
|
|
; AVX512VPOPCNTDQVL-NEXT: vpcmpeqb %xmm1, %xmm0, %xmm2
|
|
; AVX512VPOPCNTDQVL-NEXT: vpcmpeqd %xmm3, %xmm3, %xmm3
|
|
; AVX512VPOPCNTDQVL-NEXT: vpaddb %xmm3, %xmm0, %xmm4
|
|
; AVX512VPOPCNTDQVL-NEXT: vpand %xmm4, %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: vpcmpeqb %xmm1, %xmm0, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: vpternlogq $222, %xmm3, %xmm2, %xmm0
|
|
; AVX512VPOPCNTDQVL-NEXT: retq
|
|
;
|
|
; BITALG_NOVLX-LABEL: ne_1_v16i8:
|
|
; BITALG_NOVLX: # %bb.0:
|
|
; BITALG_NOVLX-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
|
|
; BITALG_NOVLX-NEXT: vpopcntb %zmm0, %zmm0
|
|
; BITALG_NOVLX-NEXT: vpcmpeqb {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; BITALG_NOVLX-NEXT: vpternlogq $15, %zmm0, %zmm0, %zmm0
|
|
; BITALG_NOVLX-NEXT: # kill: def $xmm0 killed $xmm0 killed $zmm0
|
|
; BITALG_NOVLX-NEXT: vzeroupper
|
|
; BITALG_NOVLX-NEXT: retq
|
|
;
|
|
; BITALG-LABEL: ne_1_v16i8:
|
|
; BITALG: # %bb.0:
|
|
; BITALG-NEXT: vpopcntb %xmm0, %xmm0
|
|
; BITALG-NEXT: vpcmpeqb {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; BITALG-NEXT: vpternlogq $15, %xmm0, %xmm0, %xmm0
|
|
; BITALG-NEXT: retq
|
|
%2 = tail call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %0)
|
|
%3 = icmp ne <16 x i8> %2, <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
|
|
%4 = sext <16 x i1> %3 to <16 x i8>
|
|
ret <16 x i8> %4
|
|
}
|
|
|
|
declare <2 x i64> @llvm.ctpop.v2i64(<2 x i64>)
|
|
declare <4 x i32> @llvm.ctpop.v4i32(<4 x i32>)
|
|
declare <8 x i16> @llvm.ctpop.v8i16(<8 x i16>)
|
|
declare <16 x i8> @llvm.ctpop.v16i8(<16 x i8>)
|