Skip to content

profiling with rocprofiler-sdk #3073

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.

Sign up for GitHub

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

Jump to bottom
Open
wants to merge 1 commit into
base: r2.19-rocm-enhanced
Choose a base branch
from
Open

profiling with rocprofiler-sdk #3073

wants to merge 1 commit into from
+1,831 −595

Conversation

cj401-amd
Copy link

This PR is based on the PR for xla 0.6.0 ROCm/xla#302 and upstream PR openxla/xla#29769

rocprofiler-sdk integration for improved profiling with rocprofiler_force_configure() and annotations, support both time-based and step-based profiling,

  • Keep roctracer(v1) for ROCm version < 6.3
  • rocprofiler-sdk and roctracer are selected at compile time based on ROCm version guard
  • Add unit tests for rocm_collector and rocm_tracer for v3 (ROCm version >= 6.3)
  • still need to figure out how to add more stats related to kernel, e.g., kernel size, occupancy, DMA copy, etc.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@i-chaochen i-chaochen Awaiting requested review from i-chaochen

@ScXfjiang ScXfjiang Awaiting requested review from ScXfjiang

@draganmladjenovic draganmladjenovic Awaiting requested review from draganmladjenovic

@mmakevic-amd mmakevic-amd Awaiting requested review from mmakevic-amd

At least 1 approving review is required to merge this pull request.

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@cj401-amd