Skip to content

Conversation

@roypat
Copy link
Contributor

@roypat roypat commented May 21, 2025

By not killing the uffd handler, we are leaking the resources associated
with the uffd handler, which can be undesirable and cose gradual
slowdown in in snapshot performance tests that restore snapshots using
uffd in a loop.

License Acceptance

By submitting this pull request, I confirm that my contribution is made under
the terms of the Apache 2.0 license. For more information on following Developer
Certificate of Origin and signing off your commits, please check
CONTRIBUTING.md.

PR Checklist

  • I have read and understand CONTRIBUTING.md.
  • I have run tools/devtool checkstyle to verify that the PR passes the
    automated style checks.
  • I have described what is done in these changes, why they are needed, and
    how they are solving the problem in a clear and encompassing way.
  • I have updated any relevant documentation (both in code and in the docs)
    in the PR.
  • I have mentioned all user-facing changes in CHANGELOG.md.
  • If a specific issue led to this PR, this PR closes the issue.
  • When making API changes, I have followed the
    Runbook for Firecracker API changes.
  • I have tested all new and changed functionalities in unit tests and/or
    integration tests.
  • I have linked an issue to every new TODO.

  • This functionality cannot be added in rust-vmm.

roypat added 3 commits May 21, 2025 10:35
Drop the requirement of having to call spawn_pf_handler() before the
call to microvm.restore_from_snapshot() in integration tests, and
instead just have restore_from_snapshot() take the name of the uffd
handler as an optional keyword argument, and do the spawning. This
removes the weird calling convention of calling restore_from_snapshot()
without a snapshot parameter (because it was inferred from uffd).

Signed-off-by: Patrick Roy <[email protected]>
It's a bit weird to have this attribute set from outside the microvm
class.

Signed-off-by: Patrick Roy <[email protected]>
By not killing the uffd handler, we are leaking the resources associated
with the uffd handler, which can be undesirable and cose gradual
slowdown in in snapshot performance tests that restore snapshots using
uffd in a loop.

Reported-by: Nikita Kalyazin <[email protected]>
Signed-off-by: Patrick Roy <[email protected]>
@codecov
Copy link

codecov bot commented May 21, 2025

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 82.93%. Comparing base (81f81fc) to head (6d06f01).
Report is 4 commits behind head on main.

Additional details and impacted files
@@            Coverage Diff             @@
##             main    #5219      +/-   ##
==========================================
+ Coverage   82.88%   82.93%   +0.05%     
==========================================
  Files         250      250              
  Lines       26936    26936              
==========================================
+ Hits        22325    22339      +14     
+ Misses       4611     4597      -14     
Flag Coverage Δ
5.10-c5n.metal 83.37% <ø> (ø)
5.10-m5n.metal 83.36% <ø> (-0.01%) ⬇️
5.10-m6a.metal 82.59% <ø> (ø)
5.10-m6g.metal 79.19% <ø> (ø)
5.10-m6i.metal 83.36% <ø> (ø)
5.10-m7a.metal-48xl 82.57% <ø> (?)
5.10-m7g.metal 79.19% <ø> (ø)
5.10-m7i.metal-24xl 83.33% <ø> (?)
5.10-m7i.metal-48xl 83.32% <ø> (?)
5.10-m8g.metal-24xl 79.19% <ø> (?)
5.10-m8g.metal-48xl 79.19% <ø> (?)
6.1-c5n.metal 83.42% <ø> (ø)
6.1-m5n.metal 83.42% <ø> (ø)
6.1-m6a.metal 82.64% <ø> (ø)
6.1-m6g.metal 79.19% <ø> (+<0.01%) ⬆️
6.1-m6i.metal 83.41% <ø> (+<0.01%) ⬆️
6.1-m7a.metal-48xl 82.62% <ø> (?)
6.1-m7g.metal 79.19% <ø> (-0.01%) ⬇️
6.1-m7i.metal-24xl 83.42% <ø> (?)
6.1-m7i.metal-48xl 83.43% <ø> (?)
6.1-m8g.metal-24xl 79.19% <ø> (?)
6.1-m8g.metal-48xl 79.19% <ø> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

If the FlushMetrics fails during teardown, then this will hide any
potential previous failures that might've resulted in Firecracker
dying.

Signed-off-by: Patrick Roy <[email protected]>
@roypat roypat added the Status: Awaiting review Indicates that a pull request is ready to be reviewed label May 21, 2025
@roypat roypat requested a review from Manciukic May 21, 2025 10:55
@roypat roypat merged commit 184dc54 into firecracker-microvm:main May 21, 2025
7 of 8 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Status: Awaiting review Indicates that a pull request is ready to be reviewed

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants