Skip to content

Commit 7242da9

Browse files
committed
Merge branch 'main' of https://github.com/tenstorrent/tt-metal into gpt-hang-debug-base
2 parents 153a27b + 20c1d1f commit 7242da9

File tree

378 files changed

+58594
-5578
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

378 files changed

+58594
-5578
lines changed

.github/CODEOWNERS

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -108,7 +108,7 @@ tt_metal/tools/profiler/noc_event_profiler.hpp @bgrady-tt @sohaibnadeemTT @mo-te
108108
tt_metal/tools/profiler/noc_event_profiler_utils.hpp @bgrady-tt @sohaibnadeemTT @mo-tenstorrent @tenstorrent/codeowner-bypass
109109
tt_metal/experimental/udm/ @yugaoTT @SeanNijjar @tenstorrent/codeowner-bypass
110110
tt_metal/hw/inc/experimental/udm/ @yugaoTT @SeanNijjar @tenstorrent/codeowner-bypass
111-
tt_metal/experimental/dataflow_buffer/ @abhullar-tt @tenstorrent/codeowner-bypass
111+
tt_metal/experimental/dataflow_buffer/ @abhullar-tt @arikTT @nhuang-tt @tenstorrent/codeowner-bypass
112112

113113
# metal misc
114114
tt_metal/sfpi-info.sh @nathan-TT @tenstorrent/metalium-developers-infra @tenstorrent/codeowner-bypass

.github/actions/setup-job/action.yml

Lines changed: 1 addition & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -84,7 +84,7 @@ runs:
8484
working-directory: ${{ inputs.path }}
8585
run: |
8686
echo "Enabling TT Metal Watcher"
87-
echo "TT_METAL_WATCHER=120" >> $GITHUB_ENV
87+
echo "TT_METAL_WATCHER=2" >> $GITHUB_ENV
8888
echo "TT_METAL_WATCHER_APPEND=1" >> $GITHUB_ENV
8989
echo "TT_METAL_WATCHER_NOINLINE=1" >> $GITHUB_ENV
9090
shell: bash
@@ -112,5 +112,4 @@ runs:
112112
echo "Enabling detection of timeout on fast dispatch and automatically call tt-triage"
113113
echo "TT_METAL_DISPATCH_TIMEOUT_COMMAND_TO_EXECUTE=/opt/venv/bin/python3 $(pwd)/tools/tt-triage.py --disable-progress 1>&2" >> $GITHUB_ENV
114114
echo "TT_METAL_OPERATION_TIMEOUT_SECONDS=${{ inputs.hang-detection-timeout }}" >> $GITHUB_ENV
115-
echo "TT_RUN_DISABLED_TRIAGE_SCRIPTS_IN_CI=1" >> $GITHUB_ENV
116115
shell: bash
Lines changed: 76 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,76 @@
1+
name: 'Sweep Run Analysis'
2+
description: 'Analyze sweep run results and send Slack notification with regression detection'
3+
4+
inputs:
5+
github-run-id:
6+
description: 'GitHub Actions run ID to analyze'
7+
required: true
8+
run-type:
9+
description: 'Type of sweep run (nightly, comprehensive, model_traced, lead_models)'
10+
required: true
11+
conclusion:
12+
description: 'Workflow conclusion (success, failure, cancelled)'
13+
required: true
14+
slack-webhook-url:
15+
description: 'Slack webhook URL for notifications'
16+
required: true
17+
database-url:
18+
description: 'PostgreSQL connection string for sweep database'
19+
required: true
20+
21+
runs:
22+
using: 'composite'
23+
steps:
24+
- name: Set up Python
25+
uses: actions/setup-python@v5
26+
with:
27+
python-version: '3.10'
28+
29+
- name: Install dependencies
30+
shell: bash
31+
run: pip install psycopg2-binary requests
32+
33+
- name: Query database and detect regressions
34+
id: query-db
35+
shell: bash
36+
env:
37+
DATABASE_URL: ${{ inputs.database-url }}
38+
GITHUB_RUN_ID: ${{ inputs.github-run-id }}
39+
RUN_TYPE: ${{ inputs.run-type }}
40+
RESULTS_FILE: sweep_results.json
41+
run: |
42+
echo "Querying database for run $GITHUB_RUN_ID..."
43+
python ${{ github.action_path }}/scripts/extract_sweep_results.py
44+
continue-on-error: true
45+
46+
- name: Fallback - Download run result artifact
47+
if: steps.query-db.outcome == 'failure'
48+
uses: actions/download-artifact@v4
49+
with:
50+
name: sweeps-run-result
51+
path: ./sweep-results
52+
github-token: ${{ github.token }}
53+
run-id: ${{ inputs.github-run-id }}
54+
continue-on-error: true
55+
56+
- name: Fallback - Parse artifact data
57+
if: steps.query-db.outcome == 'failure'
58+
shell: bash
59+
env:
60+
ARTIFACTS_DIR: ./sweep-results
61+
RESULTS_FILE: sweep_results.json
62+
ARCH_NAME: ${{ inputs.run-type }}
63+
run: |
64+
echo "Database unavailable, parsing artifacts instead..."
65+
python ${{ github.action_path }}/scripts/parse_sweep_artifacts.py
66+
67+
- name: Send Slack notification
68+
shell: bash
69+
env:
70+
SLACK_WEBHOOK_URL: ${{ inputs.slack-webhook-url }}
71+
CONCLUSION: ${{ inputs.conclusion }}
72+
GITHUB_RUN_ID: ${{ inputs.github-run-id }}
73+
RESULTS_FILE: sweep_results.json
74+
run: |
75+
echo "Sending Slack notification..."
76+
python ${{ github.action_path }}/scripts/send_slack_notification.py

0 commit comments

Comments
 (0)