derek/TheRock
1
0 Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
Files
1a182e2dac58879f215c0fad0f5f78c76dcafe56
TheRock/rocm-libraries/shared/tensile/Tensile/Perf/DLRM/sgemm_xdlops.yaml
T
derek.holloway 7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
Details
CodeQL / Analyze (python) (push) Canceled after 0s
Details
pre-commit / pre-commit (push) Canceled after 0s
Details
Multi-Arch CI / setup (push) Canceled after 0s
Details
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Details
Setup (Multi-Arch) / setup (push) Canceled after 0s
Details
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Details
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
Details
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Details
Multi-Arch CI / CI Summary (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Details
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Details
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
Details
Flatten subprojects into a single base monorepo
2026-09-02 16:34:47 -07:00

594 lines
21 KiB
YAML
Raw Blame History

# A config to generate a fast kernel(s) for DLRM model GEMM problem sizes
# problem sizes comes from Kaggle and Medium_4 data sets
# update parameters to produce best kernel(S)
# to keep number of kernel produced minimum
# 1) add new sections if certain paramters are more conducive to certain problem sizes
# 2) create separate sections for replacement kernel(s)
GlobalParameters:
MinimumRequiredVersion: 4.14.0
PrintLevel: 1
ForceRedoBenchmarkProblems: True
ForceRedoLibraryLogic: True
ForceRedoLibraryClient: True
CMakeBuildType: Release
EnqueuesPerSync: 1
SyncsPerBenchmark: 1
LibraryPrintDebug: False
NumElementsToValidate: 1
ValidationMaxToPrint: 0
ValidationPrintValids: False
ShortNames: False
MergeFiles: True
Platform: 0
Device: 0
KernelTime: True
PinClocks: False
SleepPercent: 0
PrintSolutionRejectionReason: True
DataInitTypeA: 3
DataInitTypeB: 3
DataInitTypeBeta: 0
DataInitTypeAlpha: 1
PrintTensorA: 0
PrintTensorB: 0
PrintTensorD: 0
BenchmarkProblems:
########################################
# TN -
########################################
-
- # ProblemType
OperationType: GEMM
DataType: s
TransposeA: True
TransposeB: False
UseBeta: True
Batched: True
########################################
# TN - standard 2Blocks/Inst
# - medium_4 dataset
########################################
- # BenchmarkProblemSizeGroup - Standard
InitialSolutionParameters:
BenchmarkCommonParameters:
- KernelLanguage: ["Assembly"]
- EdgeType: ["ShiftPtr"]
- PrefetchLocalRead: [True]
- PrefetchGlobalRead: [True]
ForkParameters:
- MatrixInstruction:
- [16, 16, 4, 1]
- ThreadTile:
- [ 8, 16 ]
- [ 8, 32 ]
- WorkGroup:
- [ 16, 16, 1 ]
- TransposeLDS: [1]
- LdsBlockSizePerPad: [128,256]
- LdsPadA: [4]
- LdsPadB: [4]
- DepthU: [ 16 ]
- VectorWidth: [4,2]
- DisableVgprOverlapping: [True,False]
- WaveSeparateGlobalReadA: [True,False]
- WaveSeparateGlobalReadB: [True,False]
- GlobalReadVectorWidth: [4]
- InnerUnroll: [1,2]
- ScheduleIterAlg: [1,3]
- OptNoLoadLoop: [1,2]
- StoreVectorWidth: [-1,2,4]
BenchmarkForkParameters:
JoinParameters:
BenchmarkJoinParameters:
BenchmarkFinalParameters:
- ProblemSizes:
- Exact: [ 1024,1600,1,1024,1056,1056,1088,1088]
- Exact: [ 1024,1600,1,1024,1056,1056,1056,1088]
- Exact: [ 1024,1600,1,560,1056,1056,592,592]
- Exact: [ 1024,1600,1,560,1056,1056,592,560]
- Exact: [ 2048,512,1,100,2080,2080,132,164]
- Exact: [ 2048,512,1,100,2080,2080,132,132]
- Exact: [ 1024,200,1,560,1056,1056,592,560]
- Exact: [ 1024,200,1,560,1056,1056,592,592]
- Exact: [ 2048,512,1,74,2080,2080,106,106]
- Exact: [ 2048,512,1,74,2080,2080,74,106]
- Exact: [ 512,32768,1,13,544,544,32,64]
- Exact: [ 256,32768,1,512,288,288,544,576]
- Exact: [ 128,32768,1,512,160,160,544,576]
- Exact: [ 1024,32768,1,479,1056,1056,511,543]
- Exact: [ 1024,32768,1,1024,1056,1056,1024,1024]
- Exact: [ 512,32768,1,1024,544,544,1024,1024]
- Exact: [ 256,32768,1,512,544,544,512,512]
- # BenchmarkProblemSizeGroup - Standard
InitialSolutionParameters:
BenchmarkCommonParameters:
- KernelLanguage: ["Assembly"]
- EdgeType: ["ShiftPtr"]
- PrefetchLocalRead: [True]
- PrefetchGlobalRead: [True]
ForkParameters:
- MatrixInstruction:
- [16, 16, 4, 1]
- ThreadTile:
- [ 1, 16 ]
- [ 2, 16 ]
- WorkGroup:
- [ 32, 8, 1 ]
- TransposeLDS: [1]
- LdsBlockSizePerPad: [128,256]
- LdsPadA: [4]
- LdsPadB: [4]
- DepthU: [ 16,32 ]
- VectorWidth: [4,2]
- DisableVgprOverlapping: [True,False]
- WaveSeparateGlobalReadA: [True,False]
- WaveSeparateGlobalReadB: [True,False]
- GlobalReadVectorWidth: [2,4]
- InnerUnroll: [1,2]
- ScheduleIterAlg: [1,3]
- OptNoLoadLoop: [1,2]
- StoreVectorWidth: [-1,2,4]
BenchmarkForkParameters:
JoinParameters:
BenchmarkJoinParameters:
BenchmarkFinalParameters:
- ProblemSizes:
- Exact: [ 2048,512,1,100,2080,2080,132,164]
- Exact: [ 2048,512,1,100,2080,2080,132,132]
- Exact: [ 1024,200,1,560,1056,1056,592,560]
- Exact: [ 1024,200,1,560,1056,1056,592,592]
- Exact: [ 2048,512,1,74,2080,2080,106,106]
- Exact: [ 2048,512,1,74,2080,2080,74,106]
- Exact: [ 512,32768,1,13,544,544,32,64]
- Exact: [ 256,32768,1,512,288,288,544,576]
- Exact: [ 128,32768,1,512,160,160,544,576]
- Exact: [ 1024,32768,1,479,1056,1056,511,543]
- Exact: [ 1024,32768,1,1024,1056,1056,1024,1024]
- Exact: [ 512,32768,1,1024,544,544,1024,1024]
- Exact: [ 256,32768,1,512,544,544,512,512]
- # BenchmarkProblemSizeGroup - Standard
InitialSolutionParameters:
BenchmarkCommonParameters:
- KernelLanguage: ["Assembly"]
- EdgeType: ["ShiftPtr"]
- PrefetchLocalRead: [True]
- PrefetchGlobalRead: [True]
ForkParameters:
- MatrixInstruction:
- [16, 16, 4, 1]
- ThreadTile:
- [ 1, 16 ]
- [ 2, 16 ]
- WorkGroup:
- [ 64, 4, 1 ]
- TransposeLDS: [1]
- LdsBlockSizePerPad: [128,256]
- LdsPadA: [4]
- LdsPadB: [4]
- DepthU: [ 16,32 ]
- VectorWidth: [4,2]
- DisableVgprOverlapping: [True,False]
- WaveSeparateGlobalReadA: [True,False]
- WaveSeparateGlobalReadB: [True,False]
- GlobalReadVectorWidth: [2,4]
- InnerUnroll: [1,2]
- ScheduleIterAlg: [1,3]
- OptNoLoadLoop: [1,2]
- StoreVectorWidth: [-1,2,4]
BenchmarkForkParameters:
JoinParameters:
BenchmarkJoinParameters:
BenchmarkFinalParameters:
- ProblemSizes:
- Exact: [ 2048,512,1,100,2080,2080,132,164]
- Exact: [ 2048,512,1,100,2080,2080,132,132]
- Exact: [ 1024,200,1,560,1056,1056,592,560]
- Exact: [ 1024,200,1,560,1056,1056,592,592]
- Exact: [ 2048,512,1,74,2080,2080,106,106]
- Exact: [ 2048,512,1,74,2080,2080,74,106]
- Exact: [ 512,32768,1,13,544,544,32,64]
- Exact: [ 256,32768,1,512,288,288,544,576]
- Exact: [ 128,32768,1,512,160,160,544,576]
- Exact: [ 1024,32768,1,479,1056,1056,511,543]
- Exact: [ 1024,32768,1,1024,1056,1056,1024,1024]
- Exact: [ 512,32768,1,1024,544,544,1024,1024]
- Exact: [ 256,32768,1,512,544,544,512,512]
- # BenchmarkProblemSizeGroup - Standard
InitialSolutionParameters:
BenchmarkCommonParameters:
- KernelLanguage: ["Assembly"]
- EdgeType: ["ShiftPtr"]
- PrefetchLocalRead: [True]
- PrefetchGlobalRead: [True]
ForkParameters:
- MatrixInstruction:
- [16, 16, 1, 4]
- ThreadTile:
- [ 1, 16 ]
- [ 2, 16 ]
- [ 1, 32 ]
- [ 2, 32 ]
- WorkGroup:
- [ 64, 4, 1 ]
- TransposeLDS: [1]
- WaveSeparateGlobalReadA: [True,False]
- WaveSeparateGlobalReadB: [True,False]
- LdsBlockSizePerPad: [128,256]
- LdsPadA: [4]
- LdsPadB: [4]
- DepthU: [ 16 ]
- VectorWidth: [4,2]
- DisableVgprOverlapping: [True,False]
- GlobalReadVectorWidth: [2,4]
- InnerUnroll: [1,2]
- ScheduleIterAlg: [1,3]
- OptNoLoadLoop: [1,2]
- StoreVectorWidth: [-1,2,4]
BenchmarkForkParameters:
JoinParameters:
BenchmarkJoinParameters:
BenchmarkFinalParameters:
- ProblemSizes:
- Exact: [ 1024,1600,1,1024,1056,1056,1088,1088]
- Exact: [ 1024,1600,1,1024,1056,1056,1056,1088]
- Exact: [ 1024,1600,1,560,1056,1056,592,592]
- Exact: [ 1024,1600,1,560,1056,1056,592,560]
- Exact: [ 2048,512,1,100,2080,2080,132,164]
- Exact: [ 2048,512,1,100,2080,2080,132,132]
- Exact: [ 1024,200,1,560,1056,1056,592,560]
- Exact: [ 1024,200,1,560,1056,1056,592,592]
- Exact: [ 2048,512,1,74,2080,2080,106,106]
- Exact: [ 2048,512,1,74,2080,2080,74,106]
- Exact: [ 512,32768,1,13,544,544,32,64]
- Exact: [ 256,32768,1,512,288,288,544,576]
- Exact: [ 128,32768,1,512,160,160,544,576]
- Exact: [ 1024,32768,1,479,1056,1056,511,543]
- Exact: [ 1024,32768,1,1024,1056,1056,1024,1024]
- Exact: [ 512,32768,1,1024,544,544,1024,1024]
- Exact: [ 256,32768,1,512,544,544,512,512]
- # BenchmarkProblemSizeGroup - Standard
InitialSolutionParameters:
BenchmarkCommonParameters:
- KernelLanguage: ["Assembly"]
- EdgeType: ["ShiftPtr"]
- PrefetchLocalRead: [True]
- PrefetchGlobalRead: [True]
ForkParameters:
- MatrixInstruction:
- [16, 16, 4, 1]
- ThreadTile:
- [ 8, 16 ]
- [ 8, 32 ]
- WorkGroup:
- [ 16, 16, 1 ]
- TransposeLDS: [0]
- DepthU: [ 16 ]
- VectorWidth: [4,2]
- DisableVgprOverlapping: [True,False]
- GlobalReadVectorWidth: [2,4]
- InnerUnroll: [1,2]
- ScheduleIterAlg: [1,3]
- OptNoLoadLoop: [1,2]
- StoreVectorWidth: [-1,2,4]
BenchmarkForkParameters:
JoinParameters:
BenchmarkJoinParameters:
BenchmarkFinalParameters:
- ProblemSizes:
- Exact: [ 1024,1600,1,1024,1056,1056,1088,1088]
- Exact: [ 1024,1600,1,1024,1056,1056,1056,1088]
- Exact: [ 1024,1600,1,560,1056,1056,592,592]
- Exact: [ 1024,1600,1,560,1056,1056,592,560]
- Exact: [ 2048,512,1,100,2080,2080,132,164]
- Exact: [ 2048,512,1,100,2080,2080,132,132]
- Exact: [ 1024,200,1,560,1056,1056,592,560]
- Exact: [ 1024,200,1,560,1056,1056,592,592]
- Exact: [ 2048,512,1,74,2080,2080,106,106]
- Exact: [ 2048,512,1,74,2080,2080,74,106]
- Exact: [ 512,32768,1,13,544,544,32,64]
- Exact: [ 256,32768,1,512,288,288,544,576]
- Exact: [ 128,32768,1,512,160,160,544,576]
- Exact: [ 1024,32768,1,479,1056,1056,511,543]
- Exact: [ 1024,32768,1,1024,1056,1056,1024,1024]
- Exact: [ 512,32768,1,1024,544,544,1024,1024]
- Exact: [ 256,32768,1,512,544,544,512,512]
- # BenchmarkProblemSizeGroup - Standard
InitialSolutionParameters:
BenchmarkCommonParameters:
- KernelLanguage: ["Assembly"]
- EdgeType: ["ShiftPtr"]
- PrefetchLocalRead: [True]
- PrefetchGlobalRead: [True]
ForkParameters:
- MatrixInstruction:
- [16, 16, 1, 4]
- ThreadTile:
- [ 1, 16 ]
- [ 2, 16 ]
- [ 1, 32 ]
- [ 2, 32 ]
- WorkGroup:
- [ 64, 4, 1 ]
- TransposeLDS: [0]
- DepthU: [ 16 ]
- VectorWidth: [4,2]
- DisableVgprOverlapping: [True,False]
- GlobalReadVectorWidth: [2,4]
- InnerUnroll: [1,2]
- ScheduleIterAlg: [1,3]
- OptNoLoadLoop: [1,2]
- StoreVectorWidth: [-1,2,4]
BenchmarkForkParameters:
JoinParameters:
BenchmarkJoinParameters:
BenchmarkFinalParameters:
- ProblemSizes:
- Exact: [ 1024,1600,1,1024,1056,1056,1088,1088]
- Exact: [ 1024,1600,1,1024,1056,1056,1056,1088]
- Exact: [ 1024,1600,1,560,1056,1056,592,592]
- Exact: [ 1024,1600,1,560,1056,1056,592,560]
- Exact: [ 2048,512,1,100,2080,2080,132,164]
- Exact: [ 2048,512,1,100,2080,2080,132,132]
- Exact: [ 1024,200,1,560,1056,1056,592,560]
- Exact: [ 1024,200,1,560,1056,1056,592,592]
- Exact: [ 2048,512,1,74,2080,2080,106,106]
- Exact: [ 2048,512,1,74,2080,2080,74,106]
- Exact: [ 512,32768,1,13,544,544,32,64]
- Exact: [ 256,32768,1,512,288,288,544,576]
- Exact: [ 128,32768,1,512,160,160,544,576]
- Exact: [ 1024,32768,1,479,1056,1056,511,543]
- Exact: [ 1024,32768,1,1024,1056,1056,1024,1024]
- Exact: [ 512,32768,1,1024,544,544,1024,1024]
- Exact: [ 256,32768,1,512,544,544,512,512]
- # BenchmarkProblemSizeGroup - Standard
InitialSolutionParameters:
BenchmarkCommonParameters:
- KernelLanguage: ["Assembly"]
- EdgeType: ["ShiftPtr"]
- PrefetchLocalRead: [True]
- PrefetchGlobalRead: [True]
ForkParameters:
- MatrixInstruction:
- [32, 32, 2, 1]
- ThreadTile:
- [ 2, 32 ]
- [ 4, 32 ]
- WorkGroup:
- [ 16, 16, 1 ]
- TransposeLDS: [1]
- LdsBlockSizePerPad: [128,256]
- LdsPadA: [4]
- LdsPadB: [4]
- DepthU: [ 16 ]
- VectorWidth: [4,2]
- DisableVgprOverlapping: [True,False]
- GlobalReadVectorWidth: [2,4]
- WaveSeparateGlobalReadA: [True,False]
- InnerUnroll: [1,2]
- ScheduleIterAlg: [1,3]
- OptNoLoadLoop: [1,2]
- StoreVectorWidth: [-1,2,4]
BenchmarkForkParameters:
JoinParameters:
BenchmarkJoinParameters:
BenchmarkFinalParameters:
- ProblemSizes:
- Exact: [ 1024,1600,1,1024,1056,1056,1088,1088]
- Exact: [ 1024,1600,1,1024,1056,1056,1056,1088]
- Exact: [ 1024,1600,1,560,1056,1056,592,592]
- Exact: [ 1024,1600,1,560,1056,1056,592,560]
- Exact: [ 2048,512,1,100,2080,2080,132,164]
- Exact: [ 2048,512,1,100,2080,2080,132,132]
- Exact: [ 1024,200,1,560,1056,1056,592,560]
- Exact: [ 1024,200,1,560,1056,1056,592,592]
- Exact: [ 2048,512,1,74,2080,2080,106,106]
- Exact: [ 2048,512,1,74,2080,2080,74,106]
- Exact: [ 512,32768,1,13,544,544,32,64]
- Exact: [ 256,32768,1,512,288,288,544,576]
- Exact: [ 128,32768,1,512,160,160,544,576]
- Exact: [ 1024,32768,1,479,1056,1056,511,543]
- Exact: [ 1024,32768,1,1024,1056,1056,1024,1024]
- Exact: [ 512,32768,1,1024,544,544,1024,1024]
- Exact: [ 256,32768,1,512,544,544,512,512]
- # BenchmarkProblemSizeGroup - Standard
InitialSolutionParameters:
BenchmarkCommonParameters:
- KernelLanguage: ["Assembly"]
- EdgeType: ["ShiftPtr"]
- PrefetchLocalRead: [True]
- PrefetchGlobalRead: [True]
ForkParameters:
- MatrixInstruction:
- [32, 32, 1, 2]
- ThreadTile:
- [ 1, 32 ]
- [ 2, 32 ]
- WorkGroup:
- [ 64, 4, 1 ]
- TransposeLDS: [1]
- LdsBlockSizePerPad: [128,256]
- LdsPadA: [4]
- LdsPadB: [4]
- DepthU: [ 16 ]
- VectorWidth: [4,2]
- DisableVgprOverlapping: [True,False]
- GlobalReadVectorWidth: [2,4]
- InnerUnroll: [1,2]
- WaveSeparateGlobalReadA: [True,False]
- WaveSeparateGlobalReadB: [True,False]
- ScheduleIterAlg: [1,3]
- OptNoLoadLoop: [1,2]
- StoreVectorWidth: [-1,2,4]
BenchmarkForkParameters:
JoinParameters:
BenchmarkJoinParameters:
BenchmarkFinalParameters:
- ProblemSizes:
- Exact: [ 1024,1600,1,1024,1056,1056,1088,1088]
- Exact: [ 1024,1600,1,1024,1056,1056,1056,1088]
- Exact: [ 1024,1600,1,560,1056,1056,592,592]
- Exact: [ 1024,1600,1,560,1056,1056,592,560]
- Exact: [ 2048,512,1,100,2080,2080,132,164]
- Exact: [ 2048,512,1,100,2080,2080,132,132]
- Exact: [ 1024,200,1,560,1056,1056,592,560]
- Exact: [ 1024,200,1,560,1056,1056,592,592]
- Exact: [ 2048,512,1,74,2080,2080,106,106]
- Exact: [ 2048,512,1,74,2080,2080,74,106]
- Exact: [ 512,32768,1,13,544,544,32,64]
- Exact: [ 256,32768,1,512,288,288,544,576]
- Exact: [ 128,32768,1,512,160,160,544,576]
- Exact: [ 1024,32768,1,479,1056,1056,511,543]
- Exact: [ 1024,32768,1,1024,1056,1056,1024,1024]
- Exact: [ 512,32768,1,1024,544,544,1024,1024]
- Exact: [ 256,32768,1,512,544,544,512,512]
- # BenchmarkProblemSizeGroup - Standard
InitialSolutionParameters:
BenchmarkCommonParameters:
- KernelLanguage: ["Assembly"]
- EdgeType: ["ShiftPtr"]
- PrefetchLocalRead: [True]
- PrefetchGlobalRead: [True]
ForkParameters:
- MatrixInstruction:
- [32, 32, 2, 1]
- ThreadTile:
- [ 2, 32 ]
- [ 4, 32 ]
- WorkGroup:
- [ 16, 16, 1 ]
- TransposeLDS: [0]
- DepthU: [ 16 ]
- VectorWidth: [4,2]
- DisableVgprOverlapping: [True,False]
- GlobalReadVectorWidth: [2,4]
- InnerUnroll: [1,2]
- ScheduleIterAlg: [1,3]
- OptNoLoadLoop: [1,2]
- StoreVectorWidth: [-1,2,4]
BenchmarkForkParameters:
JoinParameters:
BenchmarkJoinParameters:
BenchmarkFinalParameters:
- ProblemSizes:
- Exact: [ 1024,1600,1,1024,1056,1056,1088,1088]
- Exact: [ 1024,1600,1,1024,1056,1056,1056,1088]
- Exact: [ 1024,1600,1,560,1056,1056,592,592]
- Exact: [ 1024,1600,1,560,1056,1056,592,560]
- Exact: [ 2048,512,1,100,2080,2080,132,164]
- Exact: [ 2048,512,1,100,2080,2080,132,132]
- Exact: [ 1024,200,1,560,1056,1056,592,560]
- Exact: [ 1024,200,1,560,1056,1056,592,592]
- Exact: [ 2048,512,1,74,2080,2080,106,106]
- Exact: [ 2048,512,1,74,2080,2080,74,106]
- Exact: [ 512,32768,1,13,544,544,32,64]
- Exact: [ 256,32768,1,512,288,288,544,576]
- Exact: [ 128,32768,1,512,160,160,544,576]
- Exact: [ 1024,32768,1,479,1056,1056,511,543]
- Exact: [ 1024,32768,1,1024,1056,1056,1024,1024]
- Exact: [ 512,32768,1,1024,544,544,1024,1024]
- Exact: [ 256,32768,1,512,544,544,512,512]
- # BenchmarkProblemSizeGroup - Standard
InitialSolutionParameters:
BenchmarkCommonParameters:
- KernelLanguage: ["Assembly"]
- EdgeType: ["ShiftPtr"]
- PrefetchLocalRead: [True]
- PrefetchGlobalRead: [True]
ForkParameters:
- MatrixInstruction:
- [32, 32, 1, 2]
- ThreadTile:
- [ 1, 32 ]
- [ 2, 32 ]
- WorkGroup:
- [ 64, 4, 1 ]
- TransposeLDS: [0]
- DepthU: [ 16 ]
- VectorWidth: [4,2]
- DisableVgprOverlapping: [True,False]
- GlobalReadVectorWidth: [2,4]
- InnerUnroll: [1,2]
- ScheduleIterAlg: [1,3]
- OptNoLoadLoop: [1,2]
- StoreVectorWidth: [-1,2,4]
BenchmarkForkParameters:
JoinParameters:
BenchmarkJoinParameters:
BenchmarkFinalParameters:
- ProblemSizes:
- Exact: [ 1024,1600,1,1024,1056,1056,1088,1088]
- Exact: [ 1024,1600,1,1024,1056,1056,1056,1088]
- Exact: [ 1024,1600,1,560,1056,1056,592,592]
- Exact: [ 1024,1600,1,560,1056,1056,592,560]
- Exact: [ 2048,512,1,100,2080,2080,132,164]
- Exact: [ 2048,512,1,100,2080,2080,132,132]
- Exact: [ 1024,200,1,560,1056,1056,592,560]
- Exact: [ 1024,200,1,560,1056,1056,592,592]
- Exact: [ 2048,512,1,74,2080,2080,106,106]
- Exact: [ 2048,512,1,74,2080,2080,74,106]
- Exact: [ 512,32768,1,13,544,544,32,64]
- Exact: [ 256,32768,1,512,288,288,544,576]
- Exact: [ 128,32768,1,512,160,160,544,576]
- Exact: [ 1024,32768,1,479,1056,1056,511,543]
- Exact: [ 1024,32768,1,1024,1056,1056,1024,1024]
- Exact: [ 512,32768,1,1024,544,544,1024,1024]
- Exact: [ 256,32768,1,512,544,544,512,512]
- # BenchmarkProblemSizeGroup - Standard
InitialSolutionParameters:
BenchmarkCommonParameters:
- KernelLanguage: ["Assembly"]
- EdgeType: ["ShiftPtr"]
- PrefetchLocalRead: [True]
- PrefetchGlobalRead: [True]
ForkParameters:
- MatrixInstruction:
- [16, 16, 4, 1]
- ThreadTile:
- [ 1, 32 ]
- [ 2, 32 ]
- WorkGroup:
- [ 64, 4, 1 ]
- TransposeLDS: [1]
- WaveSeparateGlobalReadA: [True,False]
- WaveSeparateGlobalReadB: [True,False]
- LdsBlockSizePerPad: [128,256]
- LdsPadA: [4]
- LdsPadB: [4]
- DepthU: [ 16 ]
- VectorWidth: [4,2]
- DisableVgprOverlapping: [True,False]
- GlobalReadVectorWidth: [2,4]
- InnerUnroll: [1,2]
- ScheduleIterAlg: [1,3]
- OptNoLoadLoop: [1,2]
- StoreVectorWidth: [-1,2,4]
BenchmarkForkParameters:
JoinParameters:
BenchmarkJoinParameters:
BenchmarkFinalParameters:
- ProblemSizes:
- Exact: [ 1024,200,1,560,1056,1056,592,560]
- Exact: [ 1024,200,1,560,1056,1056,592,592]
########################################
LibraryLogic:
ScheduleName: "arcturus"
DeviceNames: ["Device 7380", "Device 7388", "Device 738c", "Device 7390"]
ArchitectureName: "gfx908"
Reference in new issue View Git Blame Copy Permalink