forked from pytorch/pytorch
-
Notifications
You must be signed in to change notification settings - Fork 0
137 lines (126 loc) · 7.15 KB
/
inductor-cu124.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
name: inductor-cu124
on:
push:
tags:
- ciflow/inductor-cu124/*
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
- cron: 29 8 * * * # about 1:29am PDT, for mem leak check and rerun disabled tests
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: ${{ github.ref_type }}
check_experiments: "awsa100"
linux-focal-cuda12_4-py3_10-gcc9-inductor-build:
# Should be synced with the one in inductor.yml, but this doesn't run inductor_timm
name: cuda12.4-py3.10-gcc9-sm86
uses: ./.github/workflows/_linux-build.yml
needs: get-default-label-prefix
with:
runner_prefix: "${{ 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: "inductor", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "inductor", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "inductor_distributed", shard: 1, num_shards: 1, runner: "linux.g5.12xlarge.nvidia.gpu" },
{ config: "inductor_huggingface", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "inductor_torchbench", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "inductor_torchbench", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ 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" },
{ config: "inductor_cpp_wrapper", shard: 1, num_shards: 1, runner: "linux.g5.4xlarge.nvidia.gpu" },
]}
secrets:
HUGGING_FACE_HUB_TOKEN: ${{ secrets.HUGGING_FACE_HUB_TOKEN }}
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:
HUGGING_FACE_HUB_TOKEN: ${{ secrets.HUGGING_FACE_HUB_TOKEN }}
linux-focal-cuda12_4-py3_10-gcc9-inductor-build-gcp:
if: github.repository_owner == 'pytorch'
name: cuda12.4-py3.10-gcc9-sm80
uses: ./.github/workflows/_linux-build.yml
needs: get-a100-test-label-type
with:
build-environment: linux-focal-cuda12.4-py3.10-gcc9-sm80
docker-image-name: pytorch-linux-focal-cuda12.4-cudnn9-py3-gcc9-inductor-benchmarks
cuda-arch-list: '8.0'
test-matrix: |
{ include: [
{ config: "inductor_torchbench_smoketest_perf", shard: 1, num_shards: 1, runner: "${{ needs.get-a100-test-label-type.outputs.label-type }}linux.gcp.a100" },
]}
secrets:
HUGGING_FACE_HUB_TOKEN: ${{ secrets.HUGGING_FACE_HUB_TOKEN }}
linux-focal-cuda12_4-py3_10-gcc9-inductor-test-gcp:
name: cuda12.4-py3.10-gcc9-sm80
uses: ./.github/workflows/_linux-test.yml
needs: linux-focal-cuda12_4-py3_10-gcc9-inductor-build-gcp
with:
build-environment: linux-focal-cuda12.4-py3.10-gcc9-sm80
docker-image: ${{ needs.linux-focal-cuda12_4-py3_10-gcc9-inductor-build-gcp.outputs.docker-image }}
test-matrix: ${{ needs.linux-focal-cuda12_4-py3_10-gcc9-inductor-build-gcp.outputs.test-matrix }}
use-gha: anything-non-empty-to-use-gha
secrets:
HUGGING_FACE_HUB_TOKEN: ${{ secrets.HUGGING_FACE_HUB_TOKEN }}
linux-focal-cuda12_4-py3_12-gcc9-inductor-build:
if: github.repository_owner == 'pytorch'
name: cuda12.4-py3.12-gcc9-sm86
uses: ./.github/workflows/_linux-build.yml
with:
build-environment: linux-focal-cuda12.4-py3.12-gcc9-sm86
docker-image-name: pytorch-linux-focal-cuda12.4-cudnn9-py3.12-gcc9-inductor-benchmarks
cuda-arch-list: '8.6'
test-matrix: |
{ include: [
{ config: "inductor", shard: 1, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
{ config: "inductor", shard: 2, num_shards: 2, runner: "linux.g5.4xlarge.nvidia.gpu" },
]}
linux-focal-cuda12_4-py3_12-gcc9-inductor-test:
name: cuda12.4-py3.12-gcc9-sm86
uses: ./.github/workflows/_linux-test.yml
needs: linux-focal-cuda12_4-py3_12-gcc9-inductor-build
with:
build-environment: linux-focal-cuda12.4-py3.12-gcc9-sm86
docker-image: ${{ needs.linux-focal-cuda12_4-py3_12-gcc9-inductor-build.outputs.docker-image }}
test-matrix: ${{ needs.linux-focal-cuda12_4-py3_12-gcc9-inductor-build.outputs.test-matrix }}