derek/TheRock
1
0 Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
Files
main
TheRock/rocm-libraries/projects/hipsparselt/clients/test/spmm_gtest.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

604 lines
12 KiB
YAML
Raw Permalink Blame History

---
include: hipsparselt_common.yaml
include: known_bugs.yaml
include: spmm_common.yaml
Definitions:
- &alpha_beta_range
- { alpha: 5, beta: 0 }
- { alpha: 0, beta: 3 }
- { alpha: 1, beta: 3 }
- { alpha: 1, beta: 1 }
- &alpha_beta_range_small
- { alpha: 2, alphai: 2, beta: -1.0, betai: 2.0 }
- &transA_transB_range
- { transA: N, transB: N }
- { transA: N, transB: T }
- { transA: T, transB: N }
- { transA: T, transB: T }
- &deepbench_alpha_beta_range
- { alpha: 1, beta: 0 }
- { alpha: 1, beta: 1 }
- &deepbench_transA_transB_range
- { transA: N, transB: N }
- { transA: N, transB: T }
- { transA: T, transB: N }
- &ldd_transA_transB_range
- { transA: N, transB: N }
- { transA: N, transB: T }
- { transA: T, transB: N }
- { transA: T, transB: T }
Tests:
- name: spmm_bad_arg
category: smoke
function:
- spmm_bad_arg: *real_precisions_2b
transA: N
transB: N
# Tests confirm no NaN propagation when alpha = 0, 2 and beta = 0. Value .NaN is converted into zero
- {name: alpha_beta_zero_NaN, category: pre_checkin, precision: *real_precisions_2b,
function: spmm, transA: N, transB: N, M: 256, N: 128, K: 64, alpha: [ .NaN, 2 ], beta: [ .NaN, 2 ] }
# Split *real_precisions_2b into *int8 and *nonint8_real_precisions. Since int8 has flags 0,1
- name: spmm_small
category: quick
function:
spmm: *real_precisions_2b
matrix_size: *small_matrix_size_range
transA_transB: *transA_transB_range
alpha_beta: *alpha_beta_range
sparse_b: [true, false]
alpha_vector_scaling: [true, false]
- name: spmm_small
category: quick
function:
spmm: *real_precisions_2b
matrix_size: *small_matrix_size_range
transA_transB: *transA_transB_range
alpha_beta: *alpha_beta_range
bias_vector: [true]
bias_stride: [0, -1, 256]
bias_type: [f32_r, f16_r, bf16_r]
sparse_b: [true, false]
alpha_vector_scaling: [true, false]
- name: spmm_small
category: quick
function:
spmm: *hpa_half_precision
matrix_size: *small_matrix_size_range
transA_transB: *transA_transB_range
alpha_beta: *alpha_beta_range
bias_vector: [true, false]
sparse_b: [true, false]
alpha_vector_scaling: [true, false]
gate_residual: [true]
gpu_arch: '942'
- name: spmm_medium
category: pre_checkin
function:
spmm: *real_precisions_2b
matrix_size: *medium_matrix_size_range
transA_transB: *transA_transB_range
alpha_beta: *alpha_beta_range
sparse_b: [true, false]
bias_vector: [true]
bias_stride: [0, -1, 256]
bias_type: [f32_r, f16_r, bf16_r]
sparse_b: [true, false]
alpha_vector_scaling: [true, false]
search: [true, false]
- name: spmm_medium_alt
category: pre_checkin
function:
spmm: *hpa_half_precision
matrix_size: *medium_matrix_size_range
transA_transB: *transA_transB_range
alpha: 1
beta: 0
initialization: special
sparse_b: [true, false]
- name: spmm_medium_HMM
category: HMM
function:
spmm: *real_precisions_2b
matrix_size: *medium_matrix_size_range
transA: [ N ]
transB: [ N ]
alpha: 1
beta: 1
HMM: true
sparse_b: [true, false]
- name: spmm_chunk
category: pre_checkin
function:
spmm: *real_precisions_2b
matrix_size: *chunk_matrix_size_range
transA_transB: *transA_transB_range
alpha: 2
beta: 3
sparse_b: [true, false]
# Split *int8_half_single_precisions into *int8 and *half_single_precisions. Since int8 has flags 0,1
- name: spmm_deepbench
category: nightly
function:
spmm: *real_precisions_2b
matrix_size: *deepbench_sizes
alpha_beta: *deepbench_alpha_beta_range
transA_transB: *deepbench_transA_transB_range
sparse_b: [true, false]
- name: spmm_deepbench_alt
category: nightly
function:
spmm: *hpa_half_precision
matrix_size: *deepbench_sizes
alpha: 1
beta: 0
flags: 4
initialization: special
transA_transB: *deepbench_transA_transB_range
sparse_b: [true, false]
- name: resnet50_fwd
category: nightly
function:
spmm: *real_precisions_2b
transA: N
transB: N
matrix_size: *resnet50_fwd_sizes
alpha: 1
beta: 0
sparse_b: [true, false]
orderA: [C, R]
orderB: [C, R]
- name: resnet50_fwd_alt
category: nightly
function:
spmm: *hpa_half_precision
transA: N
transB: N
matrix_size: *resnet50_fwd_sizes
alpha: 1
beta: 0
initialization: special
sparse_b: [true, false]
- name: resnet50_bwdwrw
category: nightly
function:
spmm: *real_precisions_2b
transA: T
transB: N
matrix_size: *resnet50_bwdwrw_sizes
alpha: 1
beta: 1
sparse_b: [true, false]
- name: resnet50_bwddata
category: nightly
function:
spmm: *real_precisions_2b
transA: N
transB: T
matrix_size: *resnet50_bwddata_sizes
alpha: 1
beta: 0
sparse_b: [true, false]
- name: inception4_fwd
category: nightly
function:
spmm: *real_precisions_2b
transA: N
transB: N
matrix_size: *inception4_fwd_sizes
alpha: 1
beta: 0
sparse_b: [true, false]
- name: inception4_bwdwrw
category: nightly
function:
spmm: *real_precisions_2b
transA: T
transB: N
matrix_size: *inception4_bwdwrw_sizes
alpha: 1
beta: 1
sparse_b: [true, false]
- name: inception4_bwddata
category: nightly
function:
spmm: *real_precisions_2b
transA: N
transB: T
matrix_size: *inception4_bwddata_sizes
alpha: 1
beta: 0
sparse_b: [true, false]
- name: ctest_bwdwrw
category: nightly
function:
spmm: *real_precisions_2b
transA: T
transB: N
alpha: 1
beta: 1
matrix_size: *ctest_bwdwrw_sizes
sparse_b: [true, false]
- name: ctest_fwd
category: nightly
function:
spmm: *real_precisions_2b
transA: N
transB: N
alpha: 1
beta: 0
matrix_size: *ctest_fwd_sizes
sparse_b: [true, false]
- name: spmm_8
category: quick
function:
spmm: *real_precisions_2b
M: 8
N: 8
K: 8
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_16
category: pre_checkin
function:
spmm: *real_precisions_2b
M: 16
K: 16
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_24
category: pre_checkin
function:
spmm: *real_precisions_2b
M: 24
N: 24
K: 24
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_algo
category: pre_checkin
function:
spmm: *real_precisions_2b
M: 32
N: 32
K: 32
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
spmm_algo: [ 0, 1 ]
sparse_b: [true, false]
- name: spmm_32_8_128
category: nightly
function:
spmm: *real_precisions_2b
M: [8, 16, 24, 32, 40, 48, 56, 64, 72, 80, 88, 96, 104, 112, 120, 128]
N: 32
K: 32
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_48_8_128
category: nightly
function:
spmm: *real_precisions_2b
M: 48
N: [8, 16, 24, 32, 40, 48, 56, 64, 72, 80, 88, 96, 104, 112, 120, 128]
K: 48
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_64_8_128
category: nightly
function:
spmm: *real_precisions_2b
M: 64
N: 64
K: [16, 24, 32, 40, 48, 56, 64, 72, 80, 88, 96, 104, 112, 120, 128]
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_64_8
category: quick
function:
spmm: *real_precisions_2b
M: 64
N: 8
K: 8
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_8_64
category: quick
function:
spmm: *real_precisions_2b
M: 8
N: 8
K: 64
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_96
category: pre_checkin
function:
spmm: *real_precisions_2b
M: 96
N: 96
K: 96
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_128
category: pre_checkin
function:
spmm: *real_precisions_2b
M: 128
N: 128
K: 128
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_256
category: pre_checkin
function:
spmm: *real_precisions_2b
M: 256
N: 256
K: 256
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_256_8_16
category: pre_checkin
function:
spmm: *real_precisions_2b
M: 256
N: 8
K: 16
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_16_256_8
category: pre_checkin
function:
spmm: *real_precisions_2b
M: 16
N: 256
K: 8
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_8_16_256
category: pre_checkin
function:
spmm: *real_precisions_2b
M: 8
N: 16
K: 256
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_512
category: pre_checkin
function:
spmm: *real_precisions_2b
M: 512
N: 512
K: 512
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_1024
category: nightly
function:
spmm: *real_precisions_2b
M: 1024
N: 1024
K: 1024
alpha_beta: *alpha_beta_range
transA_transB: *transA_transB_range
sparse_b: [true, false]
- name: spmm_bf16
category: pre_checkin
function:
spmm: *hpa_bf16_precision
transA: T
transB: N
alpha: 1
beta: 0
matrix_size:
- { M: 512, N: 512, K: 512 }
- { M: 1024, N: 1024, K: 1024 }
- { M: 2048, N: 2048, K: 2048 }
- { M: 4096, N: 4096, K: 4096 }
- { M: 960, N: 1024, K: 1024 }
- { M: 3840, N: 4096, K: 4096 }
sparse_b: [true, false]
- name: spmm_small2
category: pre_checkin
transA: N
transB: N
function:
spmm: *real_precisions_2b
matrix_size:
- { M: 512, N: 512, K: 512 }
- { M: 960, N: 1024, K: 1024 }
- { M: 1024, N: 1024, K: 1024 }
alpha: [ 0.0, 0.5, 1.0 ]
beta: [ 0.0, 0.5, 1.0 ]
sparse_b: [true, false]
- name: spmm_128_activation
category: pre_checkin
function:
spmm: *real_precisions_2b
M: 128
N: 128
K: 128
transA: N
transB: N
alpha: 1
beta: 0
activation_type: [ none, relu, abs]
sparse_b: [true, false]
- name: spmm_128_activation_sigmoid
category: pre_checkin
function:
spmm: *activation_sigmoid_tanh_precisions
M: 128
N: 128
K: 128
transA: N
transB: N
alpha: 1
beta: 0
activation_type: [sigmoid]
sparse_b: [true, false]
- name: spmm_128_activation_tanh
category: pre_checkin
function:
spmm: *activation_sigmoid_tanh_precisions
M: 128
N: 128
K: 128
transA: N
transB: N
alpha: 1
beta: 0
activation_type: [tanh]
activation_arg1 : [-1.0, 0.0, 0.5, 1.0]
activation_arg2 : [-1.0, 0.0, 0.5, 1.0, 3.0]
sparse_b: [true, false]
- name: spmm_128_activation_leakyrelu
category: pre_checkin
function:
spmm: *real_precisions_2b
M: 128
N: 128
K: 128
transA: N
transB: N
alpha: 1
beta: 0
activation_type: [leakyrelu]
activation_arg1 : [-1.0, 0.0, 0.5, 1.0]
sparse_b: [true, false]
- name: spmm_128_activation_gelu
category: pre_checkin
function:
spmm: *real_precisions_2b
M: 128
N: 128
K: 128
transA: N
transB: N
alpha: 1
beta: 0
activation_type: [gelu]
activation_arg1 : [-1.0, 0.0, 0.5, 1.0]
sparse_b: [true, false]
unit_check: 1
norm_check: 0
- name: spmm_128_activation_clippedrelu
category: pre_checkin
function:
spmm: *real_precisions_2b
M: 128
N: 128
K: 128
transA: N
transB: N
alpha: 1
beta: 0
activation_type: clippedrelu
activation_arg1 : [-1.0, 0.0, 0.5]
activation_arg2 : [-1.0, 0.0, 0.5, 1.0, 3.0]
sparse_b: [true, false]
- name: aux_plan_assign
category: smoke
function:
aux_plan_assign: *hpa_int8_precision
M: 128
N: 128
K: 128
transA: T
transB: N
alpha: 1
beta: 0
sparse_b: [false]
- name: spmm_logging
category: pre_checkin
function:
spmm_logging: *hpa_half_precision
M: 128
N: 128
K: 128
transA: N
transB: N
alpha: 1
beta: 0
activation_type: clippedrelu
activation_arg1 : [0.5]
activation_arg2 : [1.0]
sparse_b: [false]
logging: [5]
...
Reference in new issue View Git Blame Copy Permalink