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

Enables nsys in iGPU container #621

Merged
merged 5 commits into from
Jan 9, 2025
Merged

Enables nsys in iGPU container #621

merged 5 commits into from
Jan 9, 2025

Conversation

sohamm17
Copy link
Contributor

@sohamm17 sohamm17 commented Dec 6, 2024

No description provided.

@sohamm17 sohamm17 self-assigned this Dec 6, 2024
@AndreasHeumann
Copy link
Contributor

@sohamm17 Instead of installing NSight for iGPU could we add an option which allows the user to map any local NSight installation into the dev container? Something like '--nsys_locattion=/opt/nvidia/nsight-systems/2024.2.2'. Also print a message indicating to use that option when using ./run ... --nsys_profile and nsys is not found?
This would be more flexible (can use other NSight version for dGPU as well) and would also avoid hardcoding the 2024.2.2 version. The nsys executable would also need to be mapped to /usr/local/bin to avoid having the path in the run script (or use the update-alternatives command).

@sohamm17
Copy link
Contributor Author

@sohamm17 Instead of installing NSight for iGPU could we add an option which allows the user to map any local NSight installation into the dev container? Something like '--nsys_locattion=/opt/nvidia/nsight-systems/2024.2.2'. Also print a message indicating to use that option when using ./run ... --nsys_profile and nsys is not found? This would be more flexible (can use other NSight version for dGPU as well) and would also avoid hardcoding the 2024.2.2 version. The nsys executable would also need to be mapped to /usr/local/bin to avoid having the path in the run script (or use the update-alternatives command).

Great suggestion @AndreasHeumann . Let me make the changes.

@sohamm17
Copy link
Contributor Author

@sohamm17 Instead of installing NSight for iGPU could we add an option which allows the user to map any local NSight installation into the dev container? Something like '--nsys_locattion=/opt/nvidia/nsight-systems/2024.2.2'. Also print a message indicating to use that option when using ./run ... --nsys_profile and nsys is not found? This would be more flexible (can use other NSight version for dGPU as well) and would also avoid hardcoding the 2024.2.2 version. The nsys executable would also need to be mapped to /usr/local/bin to avoid having the path in the run script (or use the update-alternatives command).

Great suggestion @AndreasHeumann . Let me make the changes.

@AndreasHeumann Kindly review.

Copy link
Contributor

@AndreasHeumann AndreasHeumann left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good to me, thanks!

Copy link
Contributor

@tbirdso tbirdso left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@tbirdso tbirdso merged commit 0983ad2 into main Jan 9, 2025
6 checks passed
@tbirdso tbirdso deleted the ss/igpu_nsys branch January 9, 2025 19:32
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants