CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
230 lines
7.8 KiB
LLVM
230 lines
7.8 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=+avx512f,avx512bw | FileCheck %s --check-prefixes=AVX,AVX512
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw,+avx512vl | FileCheck %s --check-prefixes=AVX,AVX512
|
|
|
|
define <2 x i64> @insert_v2i64_x1(<2 x i64> %a) {
|
|
; SSE2-LABEL: insert_v2i64_x1:
|
|
; SSE2: # %bb.0:
|
|
; SSE2-NEXT: movlps {{.*#+}} xmm0 = mem[0,1],xmm0[2,3]
|
|
; SSE2-NEXT: retq
|
|
;
|
|
; SSE3-LABEL: insert_v2i64_x1:
|
|
; SSE3: # %bb.0:
|
|
; SSE3-NEXT: movlps {{.*#+}} xmm0 = mem[0,1],xmm0[2,3]
|
|
; SSE3-NEXT: retq
|
|
;
|
|
; SSSE3-LABEL: insert_v2i64_x1:
|
|
; SSSE3: # %bb.0:
|
|
; SSSE3-NEXT: movlps {{.*#+}} xmm0 = mem[0,1],xmm0[2,3]
|
|
; SSSE3-NEXT: retq
|
|
;
|
|
; SSE41-LABEL: insert_v2i64_x1:
|
|
; SSE41: # %bb.0:
|
|
; SSE41-NEXT: pcmpeqd %xmm1, %xmm1
|
|
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm1[0,1,2,3],xmm0[4,5,6,7]
|
|
; SSE41-NEXT: retq
|
|
;
|
|
; AVX1-LABEL: insert_v2i64_x1:
|
|
; AVX1: # %bb.0:
|
|
; AVX1-NEXT: vpcmpeqd %xmm1, %xmm1, %xmm1
|
|
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm1[0,1,2,3],xmm0[4,5,6,7]
|
|
; AVX1-NEXT: retq
|
|
;
|
|
; AVX2-LABEL: insert_v2i64_x1:
|
|
; AVX2: # %bb.0:
|
|
; AVX2-NEXT: vpcmpeqd %xmm1, %xmm1, %xmm1
|
|
; AVX2-NEXT: vpblendd {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
|
|
; AVX2-NEXT: retq
|
|
;
|
|
; AVX512-LABEL: insert_v2i64_x1:
|
|
; AVX512: # %bb.0:
|
|
; AVX512-NEXT: vpcmpeqd %xmm1, %xmm1, %xmm1
|
|
; AVX512-NEXT: vpblendd {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
|
|
; AVX512-NEXT: retq
|
|
%1 = insertelement <2 x i64> %a, i64 -1, i32 0
|
|
ret <2 x i64> %1
|
|
}
|
|
|
|
define <4 x i64> @insert_v4i64_01x3(<4 x i64> %a) {
|
|
; SSE2-LABEL: insert_v4i64_01x3:
|
|
; SSE2: # %bb.0:
|
|
; SSE2-NEXT: movlps {{.*#+}} xmm1 = mem[0,1],xmm1[2,3]
|
|
; SSE2-NEXT: retq
|
|
;
|
|
; SSE3-LABEL: insert_v4i64_01x3:
|
|
; SSE3: # %bb.0:
|
|
; SSE3-NEXT: movlps {{.*#+}} xmm1 = mem[0,1],xmm1[2,3]
|
|
; SSE3-NEXT: retq
|
|
;
|
|
; SSSE3-LABEL: insert_v4i64_01x3:
|
|
; SSSE3: # %bb.0:
|
|
; SSSE3-NEXT: movlps {{.*#+}} xmm1 = mem[0,1],xmm1[2,3]
|
|
; SSSE3-NEXT: retq
|
|
;
|
|
; SSE41-LABEL: insert_v4i64_01x3:
|
|
; SSE41: # %bb.0:
|
|
; SSE41-NEXT: pcmpeqd %xmm2, %xmm2
|
|
; SSE41-NEXT: pblendw {{.*#+}} xmm1 = xmm2[0,1,2,3],xmm1[4,5,6,7]
|
|
; SSE41-NEXT: retq
|
|
;
|
|
; AVX1-LABEL: insert_v4i64_01x3:
|
|
; AVX1: # %bb.0:
|
|
; AVX1-NEXT: vxorps %xmm1, %xmm1, %xmm1
|
|
; AVX1-NEXT: vcmptrueps %ymm1, %ymm1, %ymm1
|
|
; AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5],ymm0[6,7]
|
|
; AVX1-NEXT: retq
|
|
;
|
|
; AVX2-LABEL: insert_v4i64_01x3:
|
|
; AVX2: # %bb.0:
|
|
; AVX2-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
|
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5],ymm0[6,7]
|
|
; AVX2-NEXT: retq
|
|
;
|
|
; AVX512-LABEL: insert_v4i64_01x3:
|
|
; AVX512: # %bb.0:
|
|
; AVX512-NEXT: vpcmpeqd %ymm1, %ymm1, %ymm1
|
|
; AVX512-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5],ymm0[6,7]
|
|
; AVX512-NEXT: retq
|
|
%1 = insertelement <4 x i64> %a, i64 -1, i32 2
|
|
ret <4 x i64> %1
|
|
}
|
|
|
|
define <4 x i32> @insert_v4i32_01x3(<4 x i32> %a) {
|
|
; SSE2-LABEL: insert_v4i32_01x3:
|
|
; SSE2: # %bb.0:
|
|
; SSE2-NEXT: movl $-1, %eax
|
|
; SSE2-NEXT: movd %eax, %xmm1
|
|
; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[3,0]
|
|
; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,2]
|
|
; SSE2-NEXT: retq
|
|
;
|
|
; SSE3-LABEL: insert_v4i32_01x3:
|
|
; SSE3: # %bb.0:
|
|
; SSE3-NEXT: movl $-1, %eax
|
|
; SSE3-NEXT: movd %eax, %xmm1
|
|
; SSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[3,0]
|
|
; SSE3-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,2]
|
|
; SSE3-NEXT: retq
|
|
;
|
|
; SSSE3-LABEL: insert_v4i32_01x3:
|
|
; SSSE3: # %bb.0:
|
|
; SSSE3-NEXT: movl $-1, %eax
|
|
; SSSE3-NEXT: movd %eax, %xmm1
|
|
; SSSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[3,0]
|
|
; SSSE3-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,2]
|
|
; SSSE3-NEXT: retq
|
|
;
|
|
; SSE41-LABEL: insert_v4i32_01x3:
|
|
; SSE41: # %bb.0:
|
|
; SSE41-NEXT: pcmpeqd %xmm1, %xmm1
|
|
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4,5],xmm0[6,7]
|
|
; SSE41-NEXT: retq
|
|
;
|
|
; AVX1-LABEL: insert_v4i32_01x3:
|
|
; AVX1: # %bb.0:
|
|
; AVX1-NEXT: vpcmpeqd %xmm1, %xmm1, %xmm1
|
|
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4,5],xmm0[6,7]
|
|
; AVX1-NEXT: retq
|
|
;
|
|
; AVX2-LABEL: insert_v4i32_01x3:
|
|
; AVX2: # %bb.0:
|
|
; AVX2-NEXT: vpcmpeqd %xmm1, %xmm1, %xmm1
|
|
; AVX2-NEXT: vpblendd {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3]
|
|
; AVX2-NEXT: retq
|
|
;
|
|
; AVX512-LABEL: insert_v4i32_01x3:
|
|
; AVX512: # %bb.0:
|
|
; AVX512-NEXT: vpcmpeqd %xmm1, %xmm1, %xmm1
|
|
; AVX512-NEXT: vpblendd {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3]
|
|
; AVX512-NEXT: retq
|
|
%1 = insertelement <4 x i32> %a, i32 -1, i32 2
|
|
ret <4 x i32> %1
|
|
}
|
|
|
|
define <8 x i32> @insert_v8i32_x12345x7(<8 x i32> %a) {
|
|
; SSE-LABEL: insert_v8i32_x12345x7:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: orps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
|
|
; SSE-NEXT: orps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX-LABEL: insert_v8i32_x12345x7:
|
|
; AVX: # %bb.0:
|
|
; AVX-NEXT: vorps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm0, %ymm0
|
|
; AVX-NEXT: retq
|
|
%1 = insertelement <8 x i32> %a, i32 -1, i32 0
|
|
%2 = insertelement <8 x i32> %1, i32 -1, i32 6
|
|
ret <8 x i32> %2
|
|
}
|
|
|
|
define <8 x i16> @insert_v8i16_x12345x7(<8 x i16> %a) {
|
|
; SSE-LABEL: insert_v8i16_x12345x7:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: orps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX-LABEL: insert_v8i16_x12345x7:
|
|
; AVX: # %bb.0:
|
|
; AVX-NEXT: vorps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; AVX-NEXT: retq
|
|
%1 = insertelement <8 x i16> %a, i16 -1, i32 0
|
|
%2 = insertelement <8 x i16> %1, i16 -1, i32 6
|
|
ret <8 x i16> %2
|
|
}
|
|
|
|
define <16 x i16> @insert_v16i16_x12345x789ABCDEx(<16 x i16> %a) {
|
|
; SSE-LABEL: insert_v16i16_x12345x789ABCDEx:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: orps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
|
|
; SSE-NEXT: orps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX-LABEL: insert_v16i16_x12345x789ABCDEx:
|
|
; AVX: # %bb.0:
|
|
; AVX-NEXT: vorps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm0, %ymm0
|
|
; AVX-NEXT: retq
|
|
%1 = insertelement <16 x i16> %a, i16 -1, i32 0
|
|
%2 = insertelement <16 x i16> %1, i16 -1, i32 6
|
|
%3 = insertelement <16 x i16> %2, i16 -1, i32 15
|
|
ret <16 x i16> %3
|
|
}
|
|
|
|
define <16 x i8> @insert_v16i8_x123456789ABCDEx(<16 x i8> %a) {
|
|
; SSE-LABEL: insert_v16i8_x123456789ABCDEx:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: orps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX-LABEL: insert_v16i8_x123456789ABCDEx:
|
|
; AVX: # %bb.0:
|
|
; AVX-NEXT: vorps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; AVX-NEXT: retq
|
|
%1 = insertelement <16 x i8> %a, i8 -1, i32 0
|
|
%2 = insertelement <16 x i8> %1, i8 -1, i32 15
|
|
ret <16 x i8> %2
|
|
}
|
|
|
|
define <32 x i8> @insert_v32i8_x123456789ABCDEzGHIJKLMNOPQRSTxx(<32 x i8> %a) {
|
|
; SSE-LABEL: insert_v32i8_x123456789ABCDEzGHIJKLMNOPQRSTxx:
|
|
; SSE: # %bb.0:
|
|
; SSE-NEXT: orps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
|
|
; SSE-NEXT: orps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; SSE-NEXT: retq
|
|
;
|
|
; AVX-LABEL: insert_v32i8_x123456789ABCDEzGHIJKLMNOPQRSTxx:
|
|
; AVX: # %bb.0:
|
|
; AVX-NEXT: vorps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm0, %ymm0
|
|
; AVX-NEXT: retq
|
|
%1 = insertelement <32 x i8> %a, i8 -1, i32 0
|
|
%2 = insertelement <32 x i8> %1, i8 -1, i32 15
|
|
%3 = insertelement <32 x i8> %2, i8 -1, i32 30
|
|
%4 = insertelement <32 x i8> %3, i8 -1, i32 31
|
|
ret <32 x i8> %4
|
|
}
|