Skip to content
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

NSight Systems Collection, Analysis, and TorchScript Operator #48

Closed
wants to merge 1 commit into from

Conversation

louisfeng
Copy link
Contributor

Summary:

New features

  • -l, --log-level to control logging verbosity.
  • --cuda-l2-cache to control whether to clear cache in discrete mode.
  • --nsys to run NSight Systems and collect kernel and runtime latency metrics.
  • Use the new PyTorch NVTX range start/end to avoid dependency additional nvtx library (require PyTorch 1.12, to be released)
  • New nsys_analysis.py script to process NSight Systems sqlite report.
  • New command line options to specific --batch-cmd, --ncu-bin, --nsys-bin.

Differential Revision: D35107534

@facebook-github-bot facebook-github-bot added CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. fb-exported labels Mar 24, 2022
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D35107534

louisfeng added a commit to louisfeng/param that referenced this pull request Apr 10, 2022
…okresearch#48)

Summary:
Pull Request resolved: facebookresearch#48

# New features
* -l, --log-level to control logging verbosity.
* --cuda-l2-cache to control whether to clear cache in discrete mode.
* --nsys to run NSight Systems and collect kernel and runtime latency metrics.
* Use the new PyTorch NVTX range start/end to avoid dependency additional nvtx library (require PyTorch 1.12, to be released)
* New nsys_analysis.py script to process NSight Systems sqlite report.
* New command line options to specific --batch-cmd, --ncu-bin, --nsys-bin.

Differential Revision: D35107534

fbshipit-source-id: 8fc7ac1ad171935f613494e11222c7b7640c8498
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D35107534

louisfeng added a commit to louisfeng/param that referenced this pull request Apr 10, 2022
…okresearch#48)

Summary:
Pull Request resolved: facebookresearch#48

# New features
* -l, --log-level to control logging verbosity.
* --cuda-l2-cache to control whether to clear cache in discrete mode.
* --nsys to run NSight Systems and collect kernel and runtime latency metrics.
* Use the new PyTorch NVTX range start/end to avoid dependency additional nvtx library (require PyTorch 1.12, to be released)
* New nsys_analysis.py script to process NSight Systems sqlite report.
* New command line options to specific --batch-cmd, --ncu-bin, --nsys-bin.

Differential Revision: D35107534

fbshipit-source-id: a8b2ba135b8ad8da06b9caf48128d127cb825766
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D35107534

…okresearch#48)

Summary:
Pull Request resolved: facebookresearch#48

# New features
* -l, --log-level to control logging verbosity.
* --cuda-l2-cache to control whether to clear cache in discrete mode.
* --nsys to run NSight Systems and collect kernel and runtime latency metrics.
* Use the new PyTorch NVTX range start/end to avoid dependency additional nvtx library (require PyTorch 1.12, to be released)
* New nsys_analysis.py script to process NSight Systems sqlite report.
* New command line options to specific --batch-cmd, --ncu-bin, --nsys-bin.

Differential Revision: D35107534

fbshipit-source-id: bb1345b1b1e92a5e004cc94a63658d4420ab2900
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D35107534

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. fb-exported
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants