Skip to content

Conversation

@hzhou
Copy link
Contributor

@hzhou hzhou commented Aug 19, 2025

Pull Request Description

When we use malloc for pack buffer and when it is used for copying from
and to GPU, we potentially can incur expensive registration cost every
time. This is the case with ZE. Use pipeline_pool allows re-usage of pack
buffers.

[skip warnings]

Author Checklist

  • Provide Description
    Particularly focus on why, not what. Reference background, issues, test failures, xfail entries, etc.
  • Commits Follow Good Practice
    Commits are self-contained and do not do two things at once.
    Commit message is of the form: module: short description
    Commit message explains what's in the commit.
  • Passes All Tests
    Whitespace checker. Warnings test. Additional tests via comments.
  • Contribution Agreement
    For non-Argonne authors, check contribution agreement.
    If necessary, request an explicit comment from your companies PR approval manager.

@hzhou hzhou changed the title 2508 ofi needpack ch4/ofi: fix MPIDI_OFI_rndv_need_pack for reg_host Aug 19, 2025
@hzhou
Copy link
Contributor Author

hzhou commented Aug 19, 2025

@hzhou hzhou force-pushed the 2508_ofi_needpack branch 2 times, most recently from 5be040c to a19534a Compare August 21, 2025 13:45
When we use malloc for pack buffer and when it is used for copying from
and to GPU, we potentially can incur expensive registration cost every
time. This is the case with ZE. Use pipeline_pool allows reusage of pack
buffers.
@hzhou hzhou force-pushed the 2508_ofi_needpack branch from a19534a to f769f68 Compare August 21, 2025 22:00
@hzhou hzhou marked this pull request as draft August 21, 2025 22:01
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.

1 participant