-
Notifications
You must be signed in to change notification settings - Fork 654
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Build and use GPU-enabled FFmpeg in doc CI #3045
Conversation
ca59f4c
to
9e08117
Compare
805cdb5
to
8372c60
Compare
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/audio/3045
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New FailureAs of commit bce5345: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
e820d55
to
56a2a3e
Compare
This commit set Nvidia GPU driver capabilities to `all`. Currently, attempting to use GPU video decoder/encoder with `linux_job` fails due to the corresponding driver libraries not exposed to Docker container (though they exist in host) Example failure: https://github.com/pytorch/audio/actions/runs/4899270270/jobs/8749113374#step:10:2022 This is because, GPU is enabled with `--gpus all` flag, which set default drive capability to `utility` and `compute`. To run GPU video decoder/encoder, `video` needs to be added or we can specify `all`. https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/user-guide.html#driver-capabilities According to NVIDIA/nvidia-container-runtime#112 (comment), nowadays it should be fine to set `all`. This change is being tested on pytorch/audio#3045, the job is still failing for different reason, but In https://github.com/pytorch/audio/actions/runs/4930570795/jobs/8811634409#step:10:1339, the failure mentioned above is no longer happening.
94e3495
to
8993920
Compare
@mthrok has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
Hey @mthrok. Some guidance:Use 'module: ops' for operations under 'torchaudio/{transforms, functional}', and ML-related components under 'torchaudio/csrc' (e.g. RNN-T loss). Things in "examples" directory:
Please use 'other' tag only when you’re sure the changes are not much relevant to users, or when all other tags are not applicable. Try not to use it often, in order to minimize efforts required when we prepare release notes. When preparing release notes, please make sure 'documentation' and 'tutorials' occur as the last sub-categories under each primary category like 'new feature', 'improvements' or 'prototype'. Things related to build are by default excluded from the release note, except when it impacts users. For example: |
Follow-up pytorch#3045 - Revert the removal of HW acceleration doc - comment out FFmpeg CLI test run
This commit add the step to build FFmpeg with GPU decoder in build_doc job so that we can use GPU decoder/encoder in documentations.