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
191 lines
6.6 KiB
LLVM
191 lines
6.6 KiB
LLVM
; RUN: llc -mtriple=amdgcn -mcpu=gfx908 < %s | FileCheck -check-prefixes=GCN,GFX908 %s
|
|
|
|
; GCN-LABEL: {{^}}accvgpr_write_read:
|
|
; GFX908: v_accvgpr_write [[AREG:a[0-9]+]], 1
|
|
; GFX908: v_accvgpr_read [[VREG:v[0-9]+]], [[AREG]]
|
|
; GFX908: global_store_dword v{{[0-9]+}}, [[VREG]], s{{\[[0-9]+:[0-9]+\]}}
|
|
define amdgpu_kernel void @accvgpr_write_read(ptr addrspace(1) %arg) {
|
|
bb:
|
|
%in.1 = load float, ptr addrspace(1) %arg
|
|
%init = tail call float asm "v_accvgpr_write $0, 1", "=a"()
|
|
%read = tail call float asm "v_accvgpr_read $0, $1", "=v,a"(float %init)
|
|
store float %read, ptr addrspace(1) %arg
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: {{^}}v_mfma_f32_4x4x1f32_avva
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_mfma_f32_4x4x1f32 a[{{[0-9:]+}}], v{{[0-9]+}}, v{{[0-9]+}}, a[{{[0-9:]+}}]
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
define amdgpu_kernel void @v_mfma_f32_4x4x1f32_avva(ptr addrspace(1) %arg) {
|
|
bb:
|
|
%in.1 = load <4 x float>, ptr addrspace(1) %arg
|
|
%mai.1 = tail call <4 x float> asm "v_mfma_f32_4x4x1f32 $0, $1, $2, $3", "=a,v,v,a"(float 1.0, float 2.0, <4 x float> %in.1)
|
|
store <4 x float> %mai.1, ptr addrspace(1) %arg
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: {{^}}v_mfma_f32_4x4x1f32_aaaa
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_mfma_f32_4x4x1f32 a[{{[0-9:]+}}], a{{[0-9]+}}, a{{[0-9]+}}, a[{{[0-9:]+}}]
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
define amdgpu_kernel void @v_mfma_f32_4x4x1f32_aaaa(ptr addrspace(1) %arg) {
|
|
bb:
|
|
%in.1 = load <4 x float>, ptr addrspace(1) %arg
|
|
%mai.1 = tail call <4 x float> asm "v_mfma_f32_4x4x1f32 $0, $1, $2, $3", "=a,a,a,a"(float 1.0, float 2.0, <4 x float> %in.1)
|
|
store <4 x float> %mai.1, ptr addrspace(1) %arg
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: {{^}}v_mfma_f32_4x4x4f16_aaaa
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_mfma_f32_4x4x4f16 a[{{[0-9:]+}}], a[{{[0-9]+:[0-9]+}}], a[{{[0-9]+:[0-9]+}}], a[{{[0-9:]+}}]
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
define amdgpu_kernel void @v_mfma_f32_4x4x4f16_aaaa(ptr addrspace(1) %arg) {
|
|
bb:
|
|
%in.1 = load <4 x float>, ptr addrspace(1) %arg
|
|
%mai.1 = tail call <4 x float> asm "v_mfma_f32_4x4x4f16 $0, $1, $2, $3", "=a,a,a,a"(<4 x half> <half 0xH3800, half 0xH3800, half 0xH3800, half 0xH3800>, <4 x half> <half 0xH03FF, half 0xH03FF, half 0xH03FF, half 0xH03FF>, <4 x float> %in.1)
|
|
store <4 x float> %mai.1, ptr addrspace(1) %arg
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: {{^}}v_mfma_f32_16x16x1f32_avaa
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_mfma_f32_16x16x1f32 a[{{[0-9:]+}}], v{{[0-9]+}}, a{{[0-9]+}}, a[{{[0-9:]+}}]
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
define amdgpu_kernel void @v_mfma_f32_16x16x1f32_avaa(ptr addrspace(1) %arg) {
|
|
bb:
|
|
%in.1 = load <16 x float>, ptr addrspace(1) %arg
|
|
%mai.1 = tail call <16 x float> asm "v_mfma_f32_16x16x1f32 $0, $1, $2, $3", "=a,v,a,a"(float 1.0, float 2.0, <16 x float> %in.1)
|
|
store <16 x float> %mai.1, ptr addrspace(1) %arg
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: {{^}}v_mfma_f32_32x32x1f32_avaa
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_accvgpr_write_b32
|
|
; GFX908: v_mfma_f32_32x32x1f32 a[{{[0-9:]+}}], v{{[0-9]+}}, a{{[0-9]+}}, a[{{[0-9:]+}}]
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
; GFX908: v_accvgpr_read_b32
|
|
define amdgpu_kernel void @v_mfma_f32_32x32x1f32_avaa(ptr addrspace(1) %arg) {
|
|
bb:
|
|
%in.1 = load <32 x i32>, ptr addrspace(1) %arg
|
|
%mai.1 = tail call <32 x i32> asm "v_mfma_f32_32x32x1f32 $0, $1, $2, $3", "=a,v,a,a"(float 1.0, float 2.0, <32 x i32> %in.1)
|
|
store <32 x i32> %mai.1, ptr addrspace(1) %arg
|
|
ret void
|
|
}
|