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

Update logging and reporting #305

Open
wants to merge 8 commits into
base: master
Choose a base branch
from

Conversation

denisalevi
Copy link
Member

@denisalevi denisalevi commented Oct 31, 2022

This PR

  1. Removes most of the INFO prints during cuda_standalone simulations. They are now debug prints.
  2. Adds a simple logging system for the generated code, with the same log levels as the brian logger. All the information that was printed before is now also debug information and only shown when the brian log level is set to debug. The generated code does not print to the log file (even though this could easily be implemented).
  3. Implements a ComputationTimeSummary class, which is very similar to the ProfilingSummary class, but reporting the times for compilation, initialization, simulation loop and finalization. In the current implementation, this information is printed via set_device('cuda_standalone', report_timers=True). This implementation belongs into brian2 though, but I want it in the brian2cuda release I'm preparing this week, hence it will be in brian2cuda for now.

TODO after:

  • Give user access to the macro settings in cuda_utils.h for debugging
  • Decide on what is printed out by default (I want by default some information to be shown so users understand what is going on, but also a preference to disable all information)
  • Write standalone logs into the Brian2 log file based on the brian2 file log level.
  • Make Brian2 PR for ComputationTimeSummary

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.

None yet

1 participant