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 cachgrind instructions to match new default values for cache profiling #201

Open
Tropicao opened this issue May 5, 2023 · 0 comments
Labels
debugging Debugging/Profiling/Tracing training

Comments

@Tropicao
Copy link
Contributor

Tropicao commented May 5, 2023

Valgrind v3.21.0 has changed the default value of --cache-sim from yes to no. This change has been introduced because "The cache simulation is old and unlikely to match any real modern machine".
Based on the official documentation, it means that starting from this version, only instruction cache reads will be accounted when running cachegrind. Current labs are still not affected while we stay on an older version of valgrind, but at some point we will update valgrind and lab won't work anymore (it currently relies on data cache write misses). Moreover, trainees which will likely use updated version of valgrind will not be able to apply topics discovered during the labs.

We have to check for the following points:

  • how accurate is cachegrind cache simulation regarding our lab platform (STM32MP157DK1) ?
  • if it is accurate enough, we should add instructions to properly re-enable cache simulation: -cache-sim=yes
  • if it is not accurate enough, we should either:
    • change the studied metric ?
    • change the tool ?
@Tropicao Tropicao added the debugging Debugging/Profiling/Tracing training label May 5, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
debugging Debugging/Profiling/Tracing training
Projects
None yet
Development

No branches or pull requests

1 participant