derek.holloway
7502bf64a4
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
204 lines
8.3 KiB
LLVM
204 lines
8.3 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+mmx,+sse2 | FileCheck %s --check-prefixes=X86,X86-SSE
|
|
; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+mmx,+avx | FileCheck %s --check-prefixes=X86,X86-AVX
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+mmx,+sse2 | FileCheck %s --check-prefixes=X64,X64-SSE
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+mmx,+avx | FileCheck %s --check-prefixes=X64,X64-AVX
|
|
|
|
define i32 @f(<4 x float> %A, ptr %B, <2 x double> %C, i32 %D, <2 x i64> %E, <4 x i32> %F, <8 x i16> %G, <16 x i8> %H, i64 %I, ptr %loadptr) nounwind {
|
|
; X86-SSE-LABEL: f:
|
|
; X86-SSE: # %bb.0:
|
|
; X86-SSE-NEXT: pushl %ebp
|
|
; X86-SSE-NEXT: movl %esp, %ebp
|
|
; X86-SSE-NEXT: pushl %esi
|
|
; X86-SSE-NEXT: andl $-16, %esp
|
|
; X86-SSE-NEXT: subl $16, %esp
|
|
; X86-SSE-NEXT: movsd {{.*#+}} xmm3 = mem[0],zero
|
|
; X86-SSE-NEXT: movl 12(%ebp), %ecx
|
|
; X86-SSE-NEXT: movdqa 56(%ebp), %xmm4
|
|
; X86-SSE-NEXT: movdqa 40(%ebp), %xmm5
|
|
; X86-SSE-NEXT: movdqa 24(%ebp), %xmm6
|
|
; X86-SSE-NEXT: movl 8(%ebp), %esi
|
|
; X86-SSE-NEXT: movl 80(%ebp), %edx
|
|
; X86-SSE-NEXT: movl (%edx), %eax
|
|
; X86-SSE-NEXT: addps {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0
|
|
; X86-SSE-NEXT: movntps %xmm0, (%esi)
|
|
; X86-SSE-NEXT: paddq {{\.?LCPI[0-9]+_[0-9]+}}, %xmm2
|
|
; X86-SSE-NEXT: addl (%edx), %eax
|
|
; X86-SSE-NEXT: movntdq %xmm2, (%esi)
|
|
; X86-SSE-NEXT: addpd {{\.?LCPI[0-9]+_[0-9]+}}, %xmm1
|
|
; X86-SSE-NEXT: addl (%edx), %eax
|
|
; X86-SSE-NEXT: movntpd %xmm1, (%esi)
|
|
; X86-SSE-NEXT: paddd {{\.?LCPI[0-9]+_[0-9]+}}, %xmm6
|
|
; X86-SSE-NEXT: addl (%edx), %eax
|
|
; X86-SSE-NEXT: movntdq %xmm6, (%esi)
|
|
; X86-SSE-NEXT: paddw {{\.?LCPI[0-9]+_[0-9]+}}, %xmm5
|
|
; X86-SSE-NEXT: addl (%edx), %eax
|
|
; X86-SSE-NEXT: movntdq %xmm5, (%esi)
|
|
; X86-SSE-NEXT: paddb {{\.?LCPI[0-9]+_[0-9]+}}, %xmm4
|
|
; X86-SSE-NEXT: addl (%edx), %eax
|
|
; X86-SSE-NEXT: movntdq %xmm4, (%esi)
|
|
; X86-SSE-NEXT: addl (%edx), %eax
|
|
; X86-SSE-NEXT: movntil %ecx, (%esi)
|
|
; X86-SSE-NEXT: addl (%edx), %eax
|
|
; X86-SSE-NEXT: movsd %xmm3, (%esi)
|
|
; X86-SSE-NEXT: addl (%edx), %eax
|
|
; X86-SSE-NEXT: leal -4(%ebp), %esp
|
|
; X86-SSE-NEXT: popl %esi
|
|
; X86-SSE-NEXT: popl %ebp
|
|
; X86-SSE-NEXT: retl
|
|
;
|
|
; X86-AVX-LABEL: f:
|
|
; X86-AVX: # %bb.0:
|
|
; X86-AVX-NEXT: pushl %ebp
|
|
; X86-AVX-NEXT: movl %esp, %ebp
|
|
; X86-AVX-NEXT: pushl %esi
|
|
; X86-AVX-NEXT: andl $-16, %esp
|
|
; X86-AVX-NEXT: subl $16, %esp
|
|
; X86-AVX-NEXT: vmovsd {{.*#+}} xmm3 = mem[0],zero
|
|
; X86-AVX-NEXT: movl 12(%ebp), %ecx
|
|
; X86-AVX-NEXT: vmovdqa 56(%ebp), %xmm4
|
|
; X86-AVX-NEXT: vmovdqa 40(%ebp), %xmm5
|
|
; X86-AVX-NEXT: vmovdqa 24(%ebp), %xmm6
|
|
; X86-AVX-NEXT: movl 8(%ebp), %esi
|
|
; X86-AVX-NEXT: movl 80(%ebp), %edx
|
|
; X86-AVX-NEXT: movl (%edx), %eax
|
|
; X86-AVX-NEXT: vaddps {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0, %xmm0
|
|
; X86-AVX-NEXT: vmovntps %xmm0, (%esi)
|
|
; X86-AVX-NEXT: vpaddq {{\.?LCPI[0-9]+_[0-9]+}}, %xmm2, %xmm0
|
|
; X86-AVX-NEXT: addl (%edx), %eax
|
|
; X86-AVX-NEXT: vmovntdq %xmm0, (%esi)
|
|
; X86-AVX-NEXT: vaddpd {{\.?LCPI[0-9]+_[0-9]+}}, %xmm1, %xmm0
|
|
; X86-AVX-NEXT: addl (%edx), %eax
|
|
; X86-AVX-NEXT: vmovntpd %xmm0, (%esi)
|
|
; X86-AVX-NEXT: vpaddd {{\.?LCPI[0-9]+_[0-9]+}}, %xmm6, %xmm0
|
|
; X86-AVX-NEXT: addl (%edx), %eax
|
|
; X86-AVX-NEXT: vmovntdq %xmm0, (%esi)
|
|
; X86-AVX-NEXT: vpaddw {{\.?LCPI[0-9]+_[0-9]+}}, %xmm5, %xmm0
|
|
; X86-AVX-NEXT: addl (%edx), %eax
|
|
; X86-AVX-NEXT: vmovntdq %xmm0, (%esi)
|
|
; X86-AVX-NEXT: vpaddb {{\.?LCPI[0-9]+_[0-9]+}}, %xmm4, %xmm0
|
|
; X86-AVX-NEXT: addl (%edx), %eax
|
|
; X86-AVX-NEXT: vmovntdq %xmm0, (%esi)
|
|
; X86-AVX-NEXT: addl (%edx), %eax
|
|
; X86-AVX-NEXT: movntil %ecx, (%esi)
|
|
; X86-AVX-NEXT: addl (%edx), %eax
|
|
; X86-AVX-NEXT: vmovsd %xmm3, (%esi)
|
|
; X86-AVX-NEXT: addl (%edx), %eax
|
|
; X86-AVX-NEXT: leal -4(%ebp), %esp
|
|
; X86-AVX-NEXT: popl %esi
|
|
; X86-AVX-NEXT: popl %ebp
|
|
; X86-AVX-NEXT: retl
|
|
;
|
|
; X64-SSE-LABEL: f:
|
|
; X64-SSE: # %bb.0:
|
|
; X64-SSE-NEXT: movl (%rcx), %eax
|
|
; X64-SSE-NEXT: addps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
|
|
; X64-SSE-NEXT: movntps %xmm0, (%rdi)
|
|
; X64-SSE-NEXT: paddq {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2
|
|
; X64-SSE-NEXT: addl (%rcx), %eax
|
|
; X64-SSE-NEXT: movntdq %xmm2, (%rdi)
|
|
; X64-SSE-NEXT: addpd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; X64-SSE-NEXT: addl (%rcx), %eax
|
|
; X64-SSE-NEXT: movntpd %xmm1, (%rdi)
|
|
; X64-SSE-NEXT: paddd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm3
|
|
; X64-SSE-NEXT: addl (%rcx), %eax
|
|
; X64-SSE-NEXT: movntdq %xmm3, (%rdi)
|
|
; X64-SSE-NEXT: paddw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm4
|
|
; X64-SSE-NEXT: addl (%rcx), %eax
|
|
; X64-SSE-NEXT: movntdq %xmm4, (%rdi)
|
|
; X64-SSE-NEXT: paddb {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm5
|
|
; X64-SSE-NEXT: addl (%rcx), %eax
|
|
; X64-SSE-NEXT: movntdq %xmm5, (%rdi)
|
|
; X64-SSE-NEXT: addl (%rcx), %eax
|
|
; X64-SSE-NEXT: movntil %esi, (%rdi)
|
|
; X64-SSE-NEXT: addl (%rcx), %eax
|
|
; X64-SSE-NEXT: movntiq %rdx, (%rdi)
|
|
; X64-SSE-NEXT: addl (%rcx), %eax
|
|
; X64-SSE-NEXT: retq
|
|
;
|
|
; X64-AVX-LABEL: f:
|
|
; X64-AVX: # %bb.0:
|
|
; X64-AVX-NEXT: movl (%rcx), %eax
|
|
; X64-AVX-NEXT: vaddps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
|
|
; X64-AVX-NEXT: vmovntps %xmm0, (%rdi)
|
|
; X64-AVX-NEXT: vpaddq {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2, %xmm0
|
|
; X64-AVX-NEXT: addl (%rcx), %eax
|
|
; X64-AVX-NEXT: vmovntdq %xmm0, (%rdi)
|
|
; X64-AVX-NEXT: vaddpd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm0
|
|
; X64-AVX-NEXT: addl (%rcx), %eax
|
|
; X64-AVX-NEXT: vmovntpd %xmm0, (%rdi)
|
|
; X64-AVX-NEXT: vpaddd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm3, %xmm0
|
|
; X64-AVX-NEXT: addl (%rcx), %eax
|
|
; X64-AVX-NEXT: vmovntdq %xmm0, (%rdi)
|
|
; X64-AVX-NEXT: vpaddw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm4, %xmm0
|
|
; X64-AVX-NEXT: addl (%rcx), %eax
|
|
; X64-AVX-NEXT: vmovntdq %xmm0, (%rdi)
|
|
; X64-AVX-NEXT: vpaddb {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm5, %xmm0
|
|
; X64-AVX-NEXT: addl (%rcx), %eax
|
|
; X64-AVX-NEXT: vmovntdq %xmm0, (%rdi)
|
|
; X64-AVX-NEXT: addl (%rcx), %eax
|
|
; X64-AVX-NEXT: movntil %esi, (%rdi)
|
|
; X64-AVX-NEXT: addl (%rcx), %eax
|
|
; X64-AVX-NEXT: movntiq %rdx, (%rdi)
|
|
; X64-AVX-NEXT: addl (%rcx), %eax
|
|
; X64-AVX-NEXT: retq
|
|
%v0 = load i32, ptr %loadptr, align 1
|
|
%A2 = fadd <4 x float> %A, <float 1.0, float 2.0, float 3.0, float 4.0>
|
|
store <4 x float> %A2, ptr %B, align 16, !nontemporal !0
|
|
%v1 = load i32, ptr %loadptr, align 1
|
|
%E2 = add <2 x i64> %E, <i64 1, i64 2>
|
|
store <2 x i64> %E2, ptr %B, align 16, !nontemporal !0
|
|
%v2 = load i32, ptr %loadptr, align 1
|
|
%C2 = fadd <2 x double> %C, <double 1.0, double 2.0>
|
|
store <2 x double> %C2, ptr %B, align 16, !nontemporal !0
|
|
%v3 = load i32, ptr %loadptr, align 1
|
|
%F2 = add <4 x i32> %F, <i32 1, i32 2, i32 3, i32 4>
|
|
store <4 x i32> %F2, ptr %B, align 16, !nontemporal !0
|
|
%v4 = load i32, ptr %loadptr, align 1
|
|
%G2 = add <8 x i16> %G, <i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 8>
|
|
store <8 x i16> %G2, ptr %B, align 16, !nontemporal !0
|
|
%v5 = load i32, ptr %loadptr, align 1
|
|
%H2 = add <16 x i8> %H, <i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8>
|
|
store <16 x i8> %H2, ptr %B, align 16, !nontemporal !0
|
|
%v6 = load i32, ptr %loadptr, align 1
|
|
store i32 %D, ptr %B, align 1, !nontemporal !0
|
|
%v7 = load i32, ptr %loadptr, align 1
|
|
store i64 %I, ptr %B, align 1, !nontemporal !0
|
|
%v8 = load i32, ptr %loadptr, align 1
|
|
%sum1 = add i32 %v0, %v1
|
|
%sum2 = add i32 %sum1, %v2
|
|
%sum3 = add i32 %sum2, %v3
|
|
%sum4 = add i32 %sum3, %v4
|
|
%sum5 = add i32 %sum4, %v5
|
|
%sum6 = add i32 %sum5, %v6
|
|
%sum7 = add i32 %sum6, %v7
|
|
%sum8 = add i32 %sum7, %v8
|
|
ret i32 %sum8
|
|
}
|
|
|
|
define void @test_mmx(ptr nocapture %a0, ptr nocapture %a1) {
|
|
; X86-LABEL: test_mmx:
|
|
; X86: # %bb.0: # %entry
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
|
; X86-NEXT: movq (%ecx), %mm0
|
|
; X86-NEXT: psrlq $3, %mm0
|
|
; X86-NEXT: movntq %mm0, (%eax)
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: test_mmx:
|
|
; X64: # %bb.0: # %entry
|
|
; X64-NEXT: movq (%rdi), %mm0
|
|
; X64-NEXT: psrlq $3, %mm0
|
|
; X64-NEXT: movntq %mm0, (%rsi)
|
|
; X64-NEXT: retq
|
|
entry:
|
|
%0 = load <1 x i64>, ptr %a0
|
|
%1 = call <1 x i64> @llvm.x86.mmx.psrli.q(<1 x i64> %0, i32 3)
|
|
store <1 x i64> %1, ptr %a1, align 8, !nontemporal !0
|
|
ret void
|
|
}
|
|
declare <1 x i64> @llvm.x86.mmx.psrli.q(<1 x i64>, i32) nounwind readnone
|
|
|
|
!0 = !{i32 1}
|