Skip to content

Sync GPU stream before getting the time in TinyProfiler#3763

Merged
WeiqunZhang merged 1 commit intoAMReX-Codes:developmentfrom
AlexanderSinn:Sync_stream_before_getting_the_time_in_TinyProfiler
Feb 14, 2024
Merged

Sync GPU stream before getting the time in TinyProfiler#3763
WeiqunZhang merged 1 commit intoAMReX-Codes:developmentfrom
AlexanderSinn:Sync_stream_before_getting_the_time_in_TinyProfiler

Conversation

@AlexanderSinn
Copy link
Copy Markdown
Member

Summary

In #1505 tiny_profiler.device_synchronize_around_region = 1 was added to better measure timings on GPU using nvtx, however this would also be useful for the timings in TinyProfiler itself. Currently codes have to custom-add these syncs to get meaningful output from TinyProfiler on GPU
https://github.com/Hi-PACE/hipace/blob/development/src/utils/HipaceProfilerWrapper.H

Additional background

Checklist

The proposed changes:

  • fix a bug or incorrect behavior in AMReX
  • add new capabilities to AMReX
  • changes answers in the test suite to more than roundoff level
  • are likely to significantly affect the results of downstream AMReX users
  • include documentation in the code and/or rst files, if appropriate

@AlexanderSinn AlexanderSinn changed the title Sync stream before getting the time in TinyProfiler Sync GPU stream before getting the time in TinyProfiler Feb 14, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants