Skip to content

(Single-card) Model perf tests #6463

(Single-card) Model perf tests

(Single-card) Model perf tests #6463

Manually triggered November 25, 2024 19:43
Status Failure
Total duration 14h 44m 37s
Artifacts 5

perf-models.yaml

on: workflow_dispatch
build-artifact  /  ...  /  build-docker-image
19s
build-artifact / build-docker-image / build-docker-image
Matrix: build-artifact / build-artifact
Matrix: models-perf / models-perf
Fit to window
Zoom out
Zoom in

Annotations

8 errors and 43 notices
models-perf / other GS
Process completed with exit code 2.
models-perf / other GS
Process completed with exit code 1.
models-perf / llm_javelin GS
Process completed with exit code 2.
models-perf / llm_javelin GS
Process completed with exit code 1.
models-perf / llm_javelin N300 WH B0
Process completed with exit code 2.
pcie-cards-are-being-used-cleanup
Tenstorrent cards seem to be in use. Killing PIDs and exiting unsuccessfully. This can happen if a test hung and is normally an issue with the test, rather than infra.
models-perf / llm_javelin N300 WH B0
Process completed with exit code 1.
models-perf / llm_javelin N300 WH B0
The action 'Run performance regressions' has timed out after 70 minutes.
printing-out-smi-info-cleanup
Touching and printing out SMI info
successful-reset-cleanup
tt-smi reset was successful
reset-successful-cleanup
tt-smi reset was successful
disk-usage-before-startup
Disk usage is 34 %
disk-usage-after-startup
Disk usage is 34 %
printing-smi-info-startup
Touching and printing out SMI info
reset-successful-startup
tt-smi reset was successful
disk-usage-before-startup
Disk usage is 34 %
disk-usage-after-startup
Disk usage is 34 %
printing-smi-info-startup
Touching and printing out SMI info
reset-successful-startup
tt-smi reset was successful
printing-out-smi-info-cleanup
Touching and printing out SMI info
successful-reset-cleanup
tt-smi reset was successful
reset-successful-cleanup
tt-smi reset was successful
printing-out-smi-info-cleanup
Touching and printing out SMI info
successful-reset-cleanup
tt-smi reset was successful
reset-successful-cleanup
tt-smi reset was successful
disk-usage-before-startup
Disk usage is 34 %
disk-usage-after-startup
Disk usage is 34 %
printing-smi-info-startup
Touching and printing out SMI info
reset-successful-startup
tt-smi reset was successful
printing-out-smi-info-cleanup
Touching and printing out SMI info
successful-reset-cleanup
tt-smi reset was successful
reset-successful-cleanup
tt-smi reset was successful
disk-usage-before-startup
Disk usage is 91 %
disk-usage-after-startup
Disk usage is 91 %
printing-smi-info-startup
Touching and printing out SMI info
reset-successful-startup
tt-smi reset was successful
printing-out-smi-info-cleanup
Touching and printing out SMI info
successful-reset-cleanup
tt-smi reset was successful
reset-successful-cleanup
tt-smi reset was successful
disk-usage-before-startup
Disk usage is 90 %
disk-usage-after-startup
Disk usage is 90 %
printing-smi-info-startup
Touching and printing out SMI info
reset-successful-startup
tt-smi reset was successful
disk-usage-before-startup
Disk usage is 90 %
disk-usage-after-startup
Disk usage is 90 %
printing-smi-info-startup
Touching and printing out SMI info
reset-successful-startup
tt-smi reset was successful
printing-out-smi-info-cleanup
Touching and printing out SMI info
attempting-reset-cleanup
Attempting to reset card(s).
successful-reset-cleanup
tt-smi reset was successful
reset-successful-cleanup
tt-smi reset was successful

Artifacts

Produced during runtime
Name Size
TTMetal_build_grayskull
299 MB
TTMetal_build_wormhole_b0
299 MB
perf-report-csv-cnn_javelin-grayskull-bare_metal
345 Bytes
perf-report-csv-cnn_javelin-wormhole_b0-bare_metal
505 Bytes
perf-report-csv-other-wormhole_b0-bare_metal
716 Bytes