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
95 lines
3.1 KiB
LLVM
95 lines
3.1 KiB
LLVM
; RUN: llc -amdgpu-scalarize-global-loads=false -mtriple=amdgcn < %s | FileCheck --check-prefixes=GCN-NOHSA,FUNC %s
|
|
; RUN: llc -amdgpu-scalarize-global-loads=false -mtriple=amdgcn-amdhsa -mcpu=kaveri < %s | FileCheck --check-prefixes=GCN-HSA,FUNC %s
|
|
; RUN: llc -amdgpu-scalarize-global-loads=false -mtriple=amdgcn -mcpu=tonga -mattr=-flat-for-global < %s | FileCheck --check-prefixes=GCN-NOHSA,FUNC %s
|
|
|
|
; FUNC-LABEL: {{^}}global_load_f64:
|
|
; GCN-NOHSA: buffer_load_dwordx2 [[VAL:v\[[0-9]+:[0-9]+\]]]
|
|
; GCN-NOHSA: buffer_store_dwordx2 [[VAL]]
|
|
|
|
; GCN-HSA: flat_load_dwordx2 [[VAL:v\[[0-9]+:[0-9]+\]]]
|
|
; GCN-HSA: flat_store_dwordx2 {{v\[[0-9]+:[0-9]+\]}}, [[VAL]]
|
|
define amdgpu_kernel void @global_load_f64(ptr addrspace(1) %out, ptr addrspace(1) %in) #0 {
|
|
%ld = load double, ptr addrspace(1) %in
|
|
store double %ld, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; FUNC-LABEL: {{^}}global_load_v2f64:
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
define amdgpu_kernel void @global_load_v2f64(ptr addrspace(1) %out, ptr addrspace(1) %in) #0 {
|
|
entry:
|
|
%ld = load <2 x double>, ptr addrspace(1) %in
|
|
store <2 x double> %ld, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; FUNC-LABEL: {{^}}global_load_v3f64:
|
|
; GCN-NOHSA-DAG: buffer_load_dwordx4
|
|
; GCN-NOHSA-DAG: buffer_load_dwordx2
|
|
; GCN-HSA-DAG: flat_load_dwordx4
|
|
; GCN-HSA-DAG: flat_load_dwordx2
|
|
define amdgpu_kernel void @global_load_v3f64(ptr addrspace(1) %out, ptr addrspace(1) %in) #0 {
|
|
entry:
|
|
%ld = load <3 x double>, ptr addrspace(1) %in
|
|
store <3 x double> %ld, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; FUNC-LABEL: {{^}}global_load_v4f64:
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
define amdgpu_kernel void @global_load_v4f64(ptr addrspace(1) %out, ptr addrspace(1) %in) #0 {
|
|
entry:
|
|
%ld = load <4 x double>, ptr addrspace(1) %in
|
|
store <4 x double> %ld, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; FUNC-LABEL: {{^}}global_load_v8f64:
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
define amdgpu_kernel void @global_load_v8f64(ptr addrspace(1) %out, ptr addrspace(1) %in) #0 {
|
|
entry:
|
|
%ld = load <8 x double>, ptr addrspace(1) %in
|
|
store <8 x double> %ld, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; FUNC-LABEL: {{^}}global_load_v16f64:
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
define amdgpu_kernel void @global_load_v16f64(ptr addrspace(1) %out, ptr addrspace(1) %in) #0 {
|
|
entry:
|
|
%ld = load <16 x double>, ptr addrspace(1) %in
|
|
store <16 x double> %ld, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
attributes #0 = { nounwind }
|