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
159 lines
5.0 KiB
YAML
159 lines
5.0 KiB
YAML
TestParameters:
|
|
marks: [skip-gfx900, skip-gfx906, skip-gfx908, skip-gfx90a, skip-gfx950, skip-gfx1010, skip-gfx1011, skip-gfx1012, skip-gfx1030, skip-gfx1100, skip-gfx1101, skip-gfx1102, skip-gfx1200, skip-gfx1201, skip-gfx1250] # not supported by arch
|
|
|
|
GlobalParameters:
|
|
SyncsPerBenchmark: 0
|
|
MinimumRequiredVersion: 5.0.0
|
|
SleepPercent: 50
|
|
NumElementsToValidate: -1
|
|
DataInitTypeBias: 0
|
|
DataInitTypeBeta: 1
|
|
DataInitTypeAlpha: 1
|
|
DataInitTypeScaleA: 2
|
|
DataInitTypeScaleB: 3
|
|
CSVExportWinner: true
|
|
CSVMergeSameProblemID: true
|
|
Device: 0
|
|
MaxWorkspaceSize: 3355443200
|
|
|
|
BenchmarkProblems:
|
|
########################################
|
|
# TN - standard
|
|
########################################
|
|
-
|
|
- # ProblemType
|
|
OperationType: GEMM
|
|
DataType: F8B8N
|
|
DestDataType: S
|
|
ComputeDataType: S
|
|
HighPrecisionAccumulate: True
|
|
TransposeA: True
|
|
TransposeB: False
|
|
UseBeta: True
|
|
Batched: True
|
|
UseBias: 1
|
|
BiasDataTypeList: [s,b]
|
|
UseScaleAB: "Scalar"
|
|
Activation: True
|
|
ActivationType: hipblaslt_all
|
|
- # BenchmarkProblemSizeGroup - Standard
|
|
InitialSolutionParameters:
|
|
BenchmarkCommonParameters:
|
|
- KernelLanguage: ["Assembly"]
|
|
ForkParameters:
|
|
- MatrixInstruction:
|
|
- [16,16,32, 1, 1, 1,1, 1,1] # 16x16
|
|
- [16,16,32, 1, 1, 1,1, 2,2] # 32x32
|
|
- [16,16,32, 1, 1, 2,2, 1,1] # 32x32
|
|
- [16,16,32, 1, 1, 1,3, 4,1] # 64x48
|
|
- [16,16,32, 1, 1, 2,2, 2,2] # 64x64
|
|
- [32,32, 16, 1, 1, 1,1, 1,1] # 32x32
|
|
- [32,32, 16, 1, 1, 1,1, 2,2] # 64x64
|
|
- [32,32, 16, 1, 1, 2,2, 1,1] # 64x64
|
|
- [32,32, 16, 1, 1, 2,2, 2,2] # 128x128
|
|
- DepthU: [ 32, 64 ]
|
|
- PrefetchGlobalRead: [2]
|
|
- PrefetchLocalRead: [1]
|
|
- ClusterLocalRead: [1]
|
|
- VectorWidthA: [1]
|
|
- VectorWidthB: [1]
|
|
- GlobalReadVectorWidthA: [-1]
|
|
- GlobalReadVectorWidthB: [1,-1]
|
|
- LocalReadVectorWidth: [8]
|
|
- ScheduleIterAlg: [3]
|
|
- InnerUnroll: [1]
|
|
- ExpandPointerSwap: [true]
|
|
- TransposeLDS: [1]
|
|
- LdsBlockSizePerPadA: [0]
|
|
- LdsBlockSizePerPadB: [0]
|
|
- LdsPadA: [0]
|
|
- LdsPadB: [0]
|
|
- WaveSeparateGlobalReadB: [1]
|
|
- 1LDSBuffer: [-1]
|
|
- GlobalSplitU: [1, 2]
|
|
- GlobalSplitUAlgorithm: ["MultipleBuffer"]
|
|
- GlobalReadPerMfma: [1.0]
|
|
- LocalWritePerMfma: [-1]
|
|
- StoreVectorWidth: [-1]
|
|
- SourceSwap: [true]
|
|
- NumElementsPerBatchStore: [0,2]
|
|
- StorePriorityOpt: [false, true]
|
|
BenchmarkJoinParameters:
|
|
BenchmarkFinalParameters:
|
|
- ProblemSizes:
|
|
- Exact: [4608, 1335, 1, 640]
|
|
- Exact: [4608, 1408, 1, 640]
|
|
- Exact: [128, 128, 1, 640]
|
|
- BiasTypeArgs: ['s', 'b']
|
|
- ActivationArgs:
|
|
- [Enum: Relu]
|
|
|
|
########################################
|
|
# NT - standard
|
|
########################################
|
|
-
|
|
- # ProblemType
|
|
OperationType: GEMM
|
|
DataType: F8B8N
|
|
DestDataType: S
|
|
ComputeDataType: S
|
|
HighPrecisionAccumulate: True
|
|
TransposeA: False
|
|
TransposeB: True
|
|
UseBeta: True
|
|
Batched: True
|
|
UseBias: 1
|
|
BiasDataTypeList: [s,b]
|
|
UseScaleAB: "Scalar"
|
|
Activation: True
|
|
- # BenchmarkProblemSizeGroup - Standard
|
|
InitialSolutionParameters:
|
|
BenchmarkCommonParameters:
|
|
- KernelLanguage: ["Assembly"]
|
|
ForkParameters:
|
|
- MatrixInstruction:
|
|
- [16,16,32, 1, 1, 1,1, 1,1] # 16x16
|
|
- [16,16,32, 1, 1, 1,1, 2,2] # 32x32
|
|
- [16,16,32, 1, 1, 2,2, 1,1] # 32x32
|
|
- [16,16,32, 1, 1, 1,3, 4,1] # 64x48
|
|
- [16,16,32, 1, 1, 2,2, 2,2] # 64x64
|
|
- [32,32, 16, 1, 1, 1,1, 1,1] # 32x32
|
|
- [32,32, 16, 1, 1, 1,1, 2,2] # 64x64
|
|
- [32,32, 16, 1, 1, 2,2, 1,1] # 64x64
|
|
- [32,32, 16, 1, 1, 2,2, 2,2] # 128x128
|
|
- DepthU: [ 32, 64 ]
|
|
- PrefetchGlobalRead: [2]
|
|
- PrefetchLocalRead: [1]
|
|
- ClusterLocalRead: [1]
|
|
- GlobalReadVectorWidthA: [-1]
|
|
- GlobalReadVectorWidthB: [1,-1]
|
|
- VectorWidthA: [1]
|
|
- VectorWidthB: [1]
|
|
- ScheduleIterAlg: [3]
|
|
- InnerUnroll: [1]
|
|
- ExpandPointerSwap: [true]
|
|
- TransposeLDS: [0]
|
|
- LdsBlockSizePerPadA: [0]
|
|
- LdsBlockSizePerPadB: [0]
|
|
- LdsPadA: [0]
|
|
- LdsPadB: [0]
|
|
- WaveSeparateGlobalReadB: [1]
|
|
- 1LDSBuffer: [-1]
|
|
- GlobalSplitU: [1, 2]
|
|
- GlobalSplitUAlgorithm: ["MultipleBuffer"]
|
|
- GlobalReadPerMfma: [1.0]
|
|
- LocalWritePerMfma: [-1]
|
|
- StoreVectorWidth: [-1]
|
|
- SourceSwap: [true]
|
|
- NumElementsPerBatchStore: [0,2]
|
|
- StorePriorityOpt: [false, true]
|
|
BenchmarkJoinParameters:
|
|
BenchmarkFinalParameters:
|
|
- ProblemSizes:
|
|
- Exact: [4608, 1335, 1, 640]
|
|
- Exact: [4608, 1408, 1, 640]
|
|
- Exact: [128, 128, 1, 640]
|
|
- BiasTypeArgs: ['s', 'b']
|
|
- ActivationArgs:
|
|
- [Enum: none]
|