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
355 lines
10 KiB
YAML
355 lines
10 KiB
YAML
GlobalParameters:
|
|
MinimumRequiredVersion: 4.4.0
|
|
PrintLevel: 1
|
|
ForceRedoBenchmarkProblems: True
|
|
ForceRedoLibraryLogic: True
|
|
ForceRedoLibraryClient: True
|
|
CMakeBuildType: Release
|
|
EnqueuesPerSync: 5
|
|
SyncsPerBenchmark: 3
|
|
LibraryPrintDebug: False
|
|
NumElementsToValidate: 128
|
|
ValidationMaxToPrint: 4
|
|
ValidationPrintValids: False
|
|
ShortNames: False
|
|
MergeFiles: False
|
|
Platform: 0
|
|
Device: 0
|
|
KernelTime: True
|
|
PinClocks: True
|
|
SleepPercent: 0
|
|
DataInitTypeBeta : 0
|
|
PrintWinnersOnly: 0
|
|
|
|
BenchmarkProblems:
|
|
########################################
|
|
# NN - standard
|
|
########################################
|
|
-
|
|
- # ProblemType
|
|
OperationType: GEMM
|
|
DataType: s
|
|
TransposeA: False
|
|
TransposeB: False
|
|
UseBeta: True
|
|
Batched: True
|
|
|
|
- # BenchmarkProblemSizeGroup - Standard
|
|
InitialSolutionParameters:
|
|
BenchmarkCommonParameters:
|
|
- ProblemSizes:
|
|
- Exact: [ 5504, 5504, 1, 3104 ]
|
|
- KernelLanguage: ["Assembly"]
|
|
- EdgeType: ["ShiftPtr"]
|
|
- LoopTail: [True]
|
|
- PrefetchLocalRead: [True]
|
|
ForkParameters:
|
|
- PrefetchGlobalRead: [False, True]
|
|
- ThreadTile:
|
|
- [ 4, 4 ]
|
|
- [ 8, 4 ]
|
|
- [ 4, 8 ]
|
|
- [ 8, 8 ]
|
|
- WorkGroup:
|
|
- [ 16, 16, 1 ]
|
|
- [ 8, 8, 4 ]
|
|
- WorkGroupMapping: [1, 8]
|
|
#- GlobalSplitU: [1, 2, 4, 8]
|
|
- GlobalSplitU: [1]
|
|
- DepthU: [ 8, 16, 32 ]
|
|
- VectorWidth: [-1, 1]
|
|
BenchmarkForkParameters:
|
|
JoinParameters:
|
|
- MacroTile
|
|
- GlobalSplitU
|
|
BenchmarkJoinParameters:
|
|
BenchmarkFinalParameters:
|
|
- ProblemSizes:
|
|
- Range: [ [64, 64, 64, 7000], [64, 64, 64, 7000], [1], [256, 512, 512, 4096] ]
|
|
|
|
########################################
|
|
# NN - VectorWidth Correctness
|
|
########################################
|
|
- # Benchmark Group
|
|
InitialSolutionParameters:
|
|
BenchmarkCommonParameters:
|
|
- EdgeType: ["ShiftPtr"]
|
|
- LoopTail: [True]
|
|
- KernelLanguage: ["Assembly"]
|
|
- PrefetchGlobalRead: [True]
|
|
- PrefetchLocalRead: [True]
|
|
- WorkGroupMapping: [1]
|
|
ForkParameters:
|
|
- ThreadTile:
|
|
- [ 2, 2 ]
|
|
- WorkGroup:
|
|
- [ 4, 4, 4 ]
|
|
#- GlobalSplitU: [2, 4, 8]
|
|
- GlobalSplitU: [1]
|
|
- DepthU: [ -1 ]
|
|
- VectorWidth: [-1, 1]
|
|
BenchmarkForkParameters:
|
|
JoinParameters:
|
|
BenchmarkJoinParameters:
|
|
BenchmarkFinalParameters:
|
|
- ProblemSizes:
|
|
- Range: [ [4], [4], [1], [256, 512, 512, 4096] ] # corner
|
|
- Range: [ [4], [64, 64, 64, 7000], [1], [256, 512, 512, 4096] ] # skinny-0
|
|
- Range: [ [64, 64, 64, 7000], [4], [1], [256, 512, 512, 4096] ] # skinny-1
|
|
|
|
########################################
|
|
# NT - standard
|
|
########################################
|
|
-
|
|
- # ProblemType
|
|
OperationType: GEMM
|
|
DataType: s
|
|
TransposeA: False
|
|
TransposeB: True
|
|
UseBeta: True
|
|
Batched: True
|
|
|
|
- # BenchmarkProblemSizeGroup - Standard
|
|
InitialSolutionParameters:
|
|
BenchmarkCommonParameters:
|
|
- ProblemSizes:
|
|
- Exact: [ 5504, 5504, 1, 3104 ]
|
|
- KernelLanguage: ["Assembly"]
|
|
- EdgeType: ["ShiftPtr"]
|
|
- LoopTail: [True]
|
|
- PrefetchLocalRead: [True]
|
|
ForkParameters:
|
|
- PrefetchGlobalRead: [False, True]
|
|
- ThreadTile:
|
|
- [ 4, 4 ]
|
|
- [ 8, 4 ]
|
|
- [ 4, 8 ]
|
|
- [ 8, 8 ]
|
|
- WorkGroup:
|
|
- [ 16, 16, 1 ]
|
|
- [ 8, 8, 4 ]
|
|
- WorkGroupMapping: [1, 8]
|
|
#- GlobalSplitU: [1, 2, 4, 8]
|
|
- GlobalSplitU: [1]
|
|
- DepthU: [ 8, 16, 32 ]
|
|
- VectorWidth: [-1, 1]
|
|
BenchmarkForkParameters:
|
|
JoinParameters:
|
|
- MacroTile
|
|
- GlobalSplitU
|
|
BenchmarkJoinParameters:
|
|
BenchmarkFinalParameters:
|
|
- ProblemSizes:
|
|
- Range: [ [64, 64, 64, 7000], [64, 64, 64, 7000], [1], [256, 512, 512, 4096] ]
|
|
|
|
########################################
|
|
# NT - VectorWidth Correctness
|
|
########################################
|
|
- # Benchmark Group
|
|
InitialSolutionParameters:
|
|
BenchmarkCommonParameters:
|
|
- EdgeType: ["ShiftPtr"]
|
|
- LoopTail: [True]
|
|
- KernelLanguage: ["Assembly"]
|
|
- PrefetchGlobalRead: [True]
|
|
- PrefetchLocalRead: [True]
|
|
- WorkGroupMapping: [1]
|
|
ForkParameters:
|
|
- ThreadTile:
|
|
- [ 2, 2 ]
|
|
- WorkGroup:
|
|
- [ 4, 4, 4 ]
|
|
#- GlobalSplitU: [2, 4, 8]
|
|
- GlobalSplitU: [1]
|
|
- DepthU: [ -1 ]
|
|
- VectorWidth: [-1, 1]
|
|
BenchmarkForkParameters:
|
|
JoinParameters:
|
|
BenchmarkJoinParameters:
|
|
BenchmarkFinalParameters:
|
|
- ProblemSizes:
|
|
- Range: [ [4], [4], [1], [256, 512, 512, 4096] ] # corner
|
|
- Range: [ [4], [64, 64, 64, 7000], [1], [256, 512, 512, 4096] ] # skinny-0
|
|
- Range: [ [64, 64, 64, 7000], [4], [1], [256, 512, 512, 4096] ] # skinny-1
|
|
|
|
########################################
|
|
# TN - standard
|
|
########################################
|
|
-
|
|
- # ProblemType
|
|
OperationType: GEMM
|
|
DataType: s
|
|
TransposeA: True
|
|
TransposeB: False
|
|
UseBeta: True
|
|
Batched: True
|
|
|
|
- # BenchmarkProblemSizeGroup - Standard
|
|
InitialSolutionParameters:
|
|
BenchmarkCommonParameters:
|
|
- ProblemSizes:
|
|
- Exact: [ 5504, 5504, 1, 3104 ]
|
|
- KernelLanguage: ["Assembly"]
|
|
- EdgeType: ["ShiftPtr"]
|
|
- LoopTail: [True]
|
|
- PrefetchLocalRead: [True]
|
|
ForkParameters:
|
|
- PrefetchGlobalRead: [False, True]
|
|
- ThreadTile:
|
|
- [ 4, 4 ]
|
|
- [ 8, 4 ]
|
|
- [ 4, 8 ]
|
|
- [ 8, 8 ]
|
|
- WorkGroup:
|
|
- [ 16, 16, 1 ]
|
|
- [ 8, 8, 4 ]
|
|
- WorkGroupMapping: [1, 8]
|
|
#- GlobalSplitU: [1, 2, 4, 8]
|
|
- GlobalSplitU: [1]
|
|
- DepthU: [ 8, 16, 32 ]
|
|
- VectorWidth: [-1, 1]
|
|
BenchmarkForkParameters:
|
|
JoinParameters:
|
|
- MacroTile
|
|
- GlobalSplitU
|
|
BenchmarkJoinParameters:
|
|
BenchmarkFinalParameters:
|
|
- ProblemSizes:
|
|
- Range: [ [64, 64, 64, 7000], [64, 64, 64, 7000], [1], [256, 512, 512, 4096] ]
|
|
|
|
########################################
|
|
# TN - VectorWidth Correctness
|
|
########################################
|
|
- # Benchmark Group
|
|
InitialSolutionParameters:
|
|
BenchmarkCommonParameters:
|
|
- EdgeType: ["ShiftPtr"]
|
|
- LoopTail: [True]
|
|
- KernelLanguage: ["Assembly"]
|
|
- PrefetchGlobalRead: [True]
|
|
- PrefetchLocalRead: [True]
|
|
- WorkGroupMapping: [1]
|
|
ForkParameters:
|
|
- ThreadTile:
|
|
- [ 2, 2 ]
|
|
- WorkGroup:
|
|
- [ 4, 4, 4 ]
|
|
#- GlobalSplitU: [2, 4, 8]
|
|
- GlobalSplitU: [1]
|
|
- DepthU: [ -1 ]
|
|
- VectorWidth: [-1, 1]
|
|
BenchmarkForkParameters:
|
|
JoinParameters:
|
|
BenchmarkJoinParameters:
|
|
BenchmarkFinalParameters:
|
|
- ProblemSizes:
|
|
- Range: [ [4], [4], [1], [256, 512, 512, 4096] ] # corner
|
|
- Range: [ [4], [64, 64, 64, 7000], [1], [256, 512, 512, 4096] ] # skinny-0
|
|
- Range: [ [64, 64, 64, 7000], [4], [1], [256, 512, 512, 4096] ] # skinny-1
|
|
|
|
########################################
|
|
# TT - standard
|
|
########################################
|
|
-
|
|
- # ProblemType
|
|
OperationType: GEMM
|
|
DataType: s
|
|
TransposeA: True
|
|
TransposeB: True
|
|
UseBeta: True
|
|
Batched: True
|
|
|
|
- # BenchmarkProblemSizeGroup - Standard
|
|
InitialSolutionParameters:
|
|
BenchmarkCommonParameters:
|
|
- ProblemSizes:
|
|
- Exact: [ 5504, 5504, 1, 3104 ]
|
|
- KernelLanguage: ["Assembly"]
|
|
- EdgeType: ["ShiftPtr"]
|
|
- LoopTail: [True]
|
|
- PrefetchLocalRead: [True]
|
|
ForkParameters:
|
|
- PrefetchGlobalRead: [False, True]
|
|
- ThreadTile:
|
|
- [ 4, 4 ]
|
|
- [ 8, 4 ]
|
|
- [ 4, 8 ]
|
|
- [ 8, 8 ]
|
|
- WorkGroup:
|
|
- [ 16, 16, 1 ]
|
|
- [ 8, 8, 4 ]
|
|
- WorkGroupMapping: [1, 4]
|
|
#- GlobalSplitU: [1, 2, 4, 8]
|
|
- GlobalSplitU: [1]
|
|
- DepthU: [ 8, 16, 32 ]
|
|
- VectorWidth: [-1, 1]
|
|
BenchmarkForkParameters:
|
|
JoinParameters:
|
|
- MacroTile
|
|
- GlobalSplitU
|
|
BenchmarkJoinParameters:
|
|
BenchmarkFinalParameters:
|
|
- ProblemSizes:
|
|
- Range: [ [64, 64, 64, 7000], [64, 64, 64, 7000], [1], [256, 512, 512, 4096] ]
|
|
|
|
########################################
|
|
# TT - VectorWidth Correctness
|
|
########################################
|
|
- # Benchmark Group
|
|
InitialSolutionParameters:
|
|
BenchmarkCommonParameters:
|
|
- EdgeType: ["ShiftPtr"]
|
|
- LoopTail: [True]
|
|
- KernelLanguage: ["Assembly"]
|
|
- PrefetchGlobalRead: [True]
|
|
- PrefetchLocalRead: [True]
|
|
- WorkGroupMapping: [1]
|
|
ForkParameters:
|
|
- ThreadTile:
|
|
- [ 2, 2 ]
|
|
- WorkGroup:
|
|
- [ 4, 4, 4 ]
|
|
#- GlobalSplitU: [2, 4, 8]
|
|
- GlobalSplitU: [1]
|
|
- DepthU: [ -1 ]
|
|
- VectorWidth: [-1, 1]
|
|
BenchmarkForkParameters:
|
|
JoinParameters:
|
|
BenchmarkJoinParameters:
|
|
BenchmarkFinalParameters:
|
|
- ProblemSizes:
|
|
- Range: [ [4], [4], [1], [256, 512, 512, 4096] ] # corner
|
|
- Range: [ [4], [64, 64, 64, 7000], [1], [256, 512, 512, 4096] ] # skinny-0
|
|
- Range: [ [64, 64, 64, 7000], [4], [1], [256, 512, 512, 4096] ] # skinny-1
|
|
|
|
|
|
LibraryLogic:
|
|
# ScheduleName: "vega20"
|
|
# DeviceNames: ["Device 66a0", "Device 66a1", "Device 66a7", "Device 66af", "Vega 20"]
|
|
# ArchitectureName: "gfx906"
|
|
|
|
ScheduleName: "navi10"
|
|
DeviceNames: ["Device 731f"]
|
|
ArchitectureName: "gfx1010"
|
|
|
|
#ScheduleName: "vega10"
|
|
#DeviceNames: ["Device 6863", "Device 6862", "Device 687f", "Device 6860",
|
|
# "Device 6861", "Vega 10 XTX [Radeon Vega Frontier Edition]",
|
|
# "Vega [Radeon RX Vega]", "Vega 10 [Radeon Instinct MI25]",
|
|
# "Vega", "Device 6864", "Device 686c"]
|
|
#ArchitectureName: "gfx900"
|
|
|
|
# ScheduleName: "mi25"
|
|
# DeviceNames: ["Device 6860"]
|
|
# ArchitectureName: "gfx900"
|
|
|
|
# ScheduleName: "r9nano"
|
|
# DeviceNames: ["Device 7300"]
|
|
# ArchitectureName: "gfx803"
|
|
|
|
# ScheduleName: "hip"
|
|
# DeviceNames: ["Device 0000"]
|
|
# ArchitectureName: "fallback"
|
|
|
|
LibraryClient:
|