forked from pytorch/pytorch
-
Notifications
You must be signed in to change notification settings - Fork 0
260 lines (240 loc) · 15.6 KB
/
inductor-periodic.yml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
name: inductor-periodic
on:
push:
tags:
- ciflow/inductor-periodic/*
workflow_dispatch:
schedule:
# Run every 4 hours during the week and every 12 hours on the weekend
- cron: 45 0,4,8,12,16,20 * * 1-5
- cron: 45 4,12 * * 0,6
concurrency:
group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref_name }}-${{ github.ref_type == 'branch' && github.sha }}-${{ github.event_name == 'workflow_dispatch' }}
cancel-in-progress: true
permissions: read-all
jobs:
get-default-label-prefix:
name: get-default-label-prefix
uses: pytorch/pytorch/.github/workflows/_runner-determinator.yml@main
if: ${{ (github.event_name != 'schedule' || github.repository == 'pytorch/pytorch') && github.repository_owner == 'pytorch' }}
with:
triggering_actor: ${{ github.triggering_actor }}
issue_owner: ${{ github.event.pull_request.user.login || github.event.issue.user.login }}
curr_branch: ${{ github.head_ref || github.ref_name }}
curr_ref_type: ${{ github.ref_type }}
get-a100-test-label-type:
name: get-a100-test-label-type
uses: pytorch/pytorch/.github/workflows/_runner-determinator.yml@main
if: ${{ (github.event_name != 'schedule' || github.repository == 'pytorch/pytorch') && github.repository_owner == 'pytorch' }}
with:
triggering_actor: ${{ github.triggering_actor }}
issue_owner: ${{ github.event.pull_request.user.login || github.event.issue.user.login }}
curr_branch: ${{ github.head_ref || github.ref_name }}
curr_ref_type: ${{ github.ref_type }}
check_experiments: "awsa100"
linux-focal-cuda12_1-py3_10-gcc9-periodic-dynamo-benchmarks-build:
name: cuda12.1-py3.10-gcc9-sm86-periodic-dynamo-benchmarks
uses: ./.github/workflows/_linux-build.yml
needs: get-default-label-prefix
with:
runner_prefix: "${{ needs.get-default-label-prefix.outputs.label-type }}"
build-environment: linux-focal-cuda12.1-py3.10-gcc9-sm86
docker-image-name: pytorch-linux-focal-cuda12.1-cudnn9-py3-gcc9-inductor-benchmarks
cuda-arch-list: '8.6'
test-matrix: |
{ include: [
{ config: "dynamo_eager_torchbench", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamo_eager_torchbench", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamo_eager_huggingface", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamo_eager_timm", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamo_eager_timm", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_eager_torchbench", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_eager_torchbench", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_eager_huggingface", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_eager_timm", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_eager_timm", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamic_aot_eager_torchbench", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamic_aot_eager_torchbench", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamic_aot_eager_huggingface", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamic_aot_eager_timm", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamic_aot_eager_timm", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
]}
secrets: inherit
linux-focal-cuda12_1-py3_10-gcc9-periodic-dynamo-benchmarks-test:
name: cuda12.1-py3.10-gcc9-sm86-periodic-dynamo-benchmarks
uses: ./.github/workflows/_linux-test.yml
needs: linux-focal-cuda12_1-py3_10-gcc9-periodic-dynamo-benchmarks-build
with:
build-environment: linux-focal-cuda12.1-py3.10-gcc9-sm86
docker-image: ${{ needs.linux-focal-cuda12_1-py3_10-gcc9-periodic-dynamo-benchmarks-build.outputs.docker-image }}
test-matrix: ${{ needs.linux-focal-cuda12_1-py3_10-gcc9-periodic-dynamo-benchmarks-build.outputs.test-matrix }}
secrets: inherit
linux-focal-cuda12_1-py3_10-gcc9-inductor-build-gcp:
name: cuda12.1-py3.10-gcc9-sm80
uses: ./.github/workflows/_linux-build.yml
needs:
- get-default-label-prefix
- get-a100-test-label-type
with:
runner_prefix: "${{ needs.get-default-label-prefix.outputs.label-type }}"
build-environment: linux-focal-cuda12.1-py3.10-gcc9-sm80
docker-image-name: pytorch-linux-focal-cuda12.1-cudnn9-py3-gcc9-inductor-benchmarks
cuda-arch-list: '8.0'
test-matrix: |
{ include: [
{ config: "inductor_torchbench_smoketest_perf", shard: 1, num_shards: 1, runner: "${{ needs.get-a100-test-label-type.outputs.label-type }}linux.gcp.a100" },
]}
secrets: inherit
linux-focal-cuda12_1-py3_10-gcc9-inductor-test-gcp:
name: cuda12.1-py3.10-gcc9-sm80
uses: ./.github/workflows/_linux-test.yml
needs: linux-focal-cuda12_1-py3_10-gcc9-inductor-build-gcp
with:
build-environment: linux-focal-cuda12.1-py3.10-gcc9-sm80
docker-image: ${{ needs.linux-focal-cuda12_1-py3_10-gcc9-inductor-build-gcp.outputs.docker-image }}
test-matrix: ${{ needs.linux-focal-cuda12_1-py3_10-gcc9-inductor-build-gcp.outputs.test-matrix }}
use-gha: anything-non-empty-to-use-gha
secrets: inherit
linux-jammy-cpu-py3_9-gcc11-periodic-dynamo-benchmarks-build:
name: linux-jammy-cpu-py3.9-gcc11-periodic-dynamo-benchmarks
uses: ./.github/workflows/_linux-build.yml
needs: get-default-label-prefix
with:
build-environment: linux-jammy-py3.9-gcc11-build
docker-image-name: pytorch-linux-jammy-py3.9-gcc11-inductor-benchmarks
runner_prefix: "${{ needs.get-default-label-prefix.outputs.label-type }}"
test-matrix: |
{ include: [
{ config: "cpu_inductor_huggingface", shard: 1, num_shards: 1, runner: "linux.8xlarge.amx" },
{ config: "cpu_inductor_timm", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_inductor_timm", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "dynamic_cpu_inductor_huggingface", shard: 1, num_shards: 1, runner: "linux.8xlarge.amx" },
{ config: "dynamic_cpu_inductor_timm", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "dynamic_cpu_inductor_timm", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_inductor_freezing_avx2_huggingface", shard: 1, num_shards: 1, runner: "linux.10xlarge.avx2" },
{ config: "cpu_inductor_freezing_avx2_torchbench", shard: 1, num_shards: 2, runner: "linux.10xlarge.avx2" },
{ config: "cpu_inductor_freezing_avx2_torchbench", shard: 2, num_shards: 2, runner: "linux.10xlarge.avx2" },
{ config: "cpu_inductor_freezing_avx2_timm", shard: 1, num_shards: 2, runner: "linux.10xlarge.avx2" },
{ config: "cpu_inductor_freezing_avx2_timm", shard: 2, num_shards: 2, runner: "linux.10xlarge.avx2" },
]}
secrets: inherit
linux-jammy-cpu-py3_9-gcc11-periodic-dynamo-benchmarks-test:
name: linux-jammy-cpu-py3.9-gcc11-periodic-dynamo-benchmarks
uses: ./.github/workflows/_linux-test.yml
needs: linux-jammy-cpu-py3_9-gcc11-periodic-dynamo-benchmarks-build
with:
build-environment: linux-jammy-py3.9-gcc11-build
docker-image: ${{ needs.linux-jammy-cpu-py3_9-gcc11-periodic-dynamo-benchmarks-build.outputs.docker-image }}
test-matrix: ${{ needs.linux-jammy-cpu-py3_9-gcc11-periodic-dynamo-benchmarks-build.outputs.test-matrix }}
secrets: inherit
linux-focal-cuda12_1-py3_10-gcc9-inductor-build:
name: cuda12.1-py3.10-gcc9-sm86
uses: ./.github/workflows/_linux-build.yml
needs: get-default-label-prefix
with:
build-environment: linux-focal-cuda12.1-py3.10-gcc9-sm86
docker-image-name: pytorch-linux-focal-cuda12.1-cudnn9-py3-gcc9-inductor-benchmarks
cuda-arch-list: '8.6'
runner_prefix: "${{ needs.get-default-label-prefix.outputs.label-type }}"
sync-tag: linux-focal-cuda12_1-py3_10-gcc9-inductor-build
test-matrix: |
{ include: [
{ config: "dynamic_inductor_huggingface", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamic_inductor_timm", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamic_inductor_timm", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamic_inductor_torchbench", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamic_inductor_torchbench", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_inductor_huggingface", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_inductor_timm", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_inductor_timm", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_inductor_torchbench", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_inductor_torchbench", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
]}
secrets: inherit
linux-focal-cuda12_1-py3_10-gcc9-inductor-test:
name: cuda12.1-py3.10-gcc9-sm86
uses: ./.github/workflows/_linux-test.yml
needs: linux-focal-cuda12_1-py3_10-gcc9-inductor-build
with:
build-environment: linux-focal-cuda12.1-py3.10-gcc9-sm86
docker-image: ${{ needs.linux-focal-cuda12_1-py3_10-gcc9-inductor-build.outputs.docker-image }}
test-matrix: ${{ needs.linux-focal-cuda12_1-py3_10-gcc9-inductor-build.outputs.test-matrix }}
secrets: inherit
linux-focal-cuda12_4-py3_10-gcc9-inductor-build:
# Should be synced with the benchmark tests in inductor.yml, but this doesn't run inductor_timm
name: cuda12.4-py3.10-gcc9-sm86
uses: ./.github/workflows/_linux-build.yml
needs: get-default-label-prefix
with:
runner_prefix: "${{ needs.get-default-label-prefix.outputs.label-type }}"
sync-tag: linux-focal-cuda12_4-py3_10-gcc9-inductor-build
build-environment: linux-focal-cuda12.4-py3.10-gcc9-sm86
docker-image-name: pytorch-linux-focal-cuda12.4-cudnn9-py3-gcc9-inductor-benchmarks
cuda-arch-list: '8.6'
test-matrix: |
{ include: [
{ config: "dynamic_inductor_huggingface", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamic_inductor_timm", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamic_inductor_timm", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamic_inductor_torchbench", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "dynamic_inductor_torchbench", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_inductor_huggingface", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_inductor_timm", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_inductor_timm", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_inductor_torchbench", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "aot_inductor_torchbench", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
]}
secrets: inherit
linux-focal-cuda12_4-py3_10-gcc9-inductor-test:
name: cuda12.4-py3.10-gcc9-sm86
uses: ./.github/workflows/_linux-test.yml
needs: linux-focal-cuda12_4-py3_10-gcc9-inductor-build
with:
sync-tag: linux-focal-cuda12_4-py3_10-gcc9-inductor-test
build-environment: linux-focal-cuda12.4-py3.10-gcc9-sm86
docker-image: ${{ needs.linux-focal-cuda12_4-py3_10-gcc9-inductor-build.outputs.docker-image }}
test-matrix: ${{ needs.linux-focal-cuda12_4-py3_10-gcc9-inductor-build.outputs.test-matrix }}
secrets: inherit
linux-jammy-cpu-py3_9-gcc11-inductor-build:
name: linux-jammy-cpu-py3.9-gcc11-inductor
uses: ./.github/workflows/_linux-build.yml
needs: get-default-label-prefix
with:
build-environment: linux-jammy-py3.9-gcc11-build
docker-image-name: pytorch-linux-jammy-py3.9-gcc11-inductor-benchmarks
runner_prefix: "${{ needs.get-default-label-prefix.outputs.label-type }}"
sync-tag: linux-jammy-cpu-py3_9-gcc11-inductor-build
test-matrix: |
{ include: [
{ config: "cpu_inductor_freezing_huggingface", shard: 1, num_shards: 1, runner: "linux.8xlarge.amx" },
{ config: "cpu_inductor_freezing_timm", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_inductor_freezing_timm", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_inductor_freezing_torchbench", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_inductor_freezing_torchbench", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_inductor_amp_freezing_huggingface", shard: 1, num_shards: 1, runner: "linux.8xlarge.amx" },
{ config: "cpu_inductor_amp_freezing_timm", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_inductor_amp_freezing_timm", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_inductor_amp_freezing_torchbench", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_inductor_amp_freezing_torchbench", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_aot_inductor_freezing_huggingface", shard: 1, num_shards: 1, runner: "linux.8xlarge.amx" },
{ config: "cpu_aot_inductor_freezing_timm", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_aot_inductor_freezing_timm", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_aot_inductor_freezing_torchbench", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_aot_inductor_freezing_torchbench", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_aot_inductor_amp_freezing_torchbench", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "cpu_aot_inductor_amp_freezing_torchbench", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "dynamic_cpu_aot_inductor_freezing_torchbench", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "dynamic_cpu_aot_inductor_freezing_torchbench", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "dynamic_cpu_aot_inductor_amp_freezing_torchbench", shard: 1, num_shards: 2, runner: "linux.8xlarge.amx" },
{ config: "dynamic_cpu_aot_inductor_amp_freezing_torchbench", shard: 2, num_shards: 2, runner: "linux.8xlarge.amx" },
]}
secrets: inherit
linux-jammy-cpu-py3_9-gcc11-inductor-test:
name: linux-jammy-cpu-py3.9-gcc11-inductor
uses: ./.github/workflows/_linux-test.yml
needs: linux-jammy-cpu-py3_9-gcc11-inductor-build
with:
build-environment: linux-jammy-py3.9-gcc11-build
docker-image: ${{ needs.linux-jammy-cpu-py3_9-gcc11-inductor-build.outputs.docker-image }}
test-matrix: ${{ needs.linux-jammy-cpu-py3_9-gcc11-inductor-build.outputs.test-matrix }}
secrets: inherit