Skip to content
Cuda - Signed integer tests on H100

fix(gpu): general fixes on indexes used in multi-gpu context #1401

Sign in to view logs

fix(gpu): general fixes on indexes used in multi-gpu context

fix(gpu): general fixes on indexes used in multi-gpu context #1401

Triggered via pull request January 30, 2025 11:32
@pdroalvespdroalves
labeled #1997
Status Failure
Total duration 52s
Artifacts –

gpu_signed_integer_h100_tests.yml

on: pull_request_target
should-run
4s
should-run
check-ci-files  /  check-changes
5s
check-ci-files / check-changes
check-user-permission  /  check-actor-permission
2s
check-user-permission / check-actor-permission
Setup instance (cuda-h100-tests)
19s
Setup instance (cuda-h100-tests)
Matrix: CUDA H100 signed integer tests
Slack Notification
0s
Slack Notification
Teardown instance (cuda-h100-tests)
0s
Teardown instance (cuda-h100-tests)
Fit to window
Zoom out
Zoom in

Annotations

4 errors
Setup instance (cuda-h100-tests)
Instance task failed (details: api call failed (reason: bad_request, message: Not Enough Stock of H100-80G-PCIe. Unable to launch virtual-machines.))
Setup instance (cuda-h100-tests)
Failure occurred while waiting for instance.
Setup instance (cuda-h100-tests)
Instance stop request has failed (HTTP status code: 500, body: Failed to stop backend instance (reason: backend provider not found with instance key: InstanceHashMapKey("298640a33e4eca06067b25fd13f23967adc2d346", "acd0c7bc-b6ba-42b4-b39b-25559358f4cb")))
Setup instance (cuda-h100-tests)
instance stop request failed