Skip to content

Track all commit waits, including no wait - #2694

Merged
iwiznia merged 3 commits into
mainfrom
tyler-count-commit-wait-time
Jul 29, 2026
Merged

Track all commit waits, including no wait#2694
iwiznia merged 3 commits into
mainfrom
tyler-count-commit-wait-time

Conversation

@tylerkaraszewski

Copy link
Copy Markdown
Contributor

Details

For diagnostic purposes, we want to see how much time we can remove from committing with HC-Tree not requiring waiting on a commit lock.

Fixed Issues

Fixes GH_LINK

Tests


Internal Testing Reminder: when changing bedrock, please compile auth against your new changes

@tylerkaraszewski tylerkaraszewski self-assigned this Jul 29, 2026
@tylerkaraszewski

Copy link
Copy Markdown
Contributor Author

@codex review

@chatgpt-codex-connector

Copy link
Copy Markdown

Codex Review: Didn't find any major issues. 🎉

Reviewed commit: 265bd5627d

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

@tylerkaraszewski
tylerkaraszewski requested a review from iwiznia July 29, 2026 17:11
Comment thread sqlitecluster/SQLite.cpp Outdated
Comment thread sqlitecluster/SQLite.cpp
if (waited) {
SINFO("Waited " << chrono::duration_cast<chrono::microseconds>(elapsed) << " for commit lock.");
} else {
SINFO("Acquired commit lock immediately.");

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hmmm maybe we should remove this and log the elapsed every time? That way graph is still correct as it would track 0?

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think it may still track 0 though. I'm not sure how likely it is we can get 0us as a time when we waited, but I wouldn't doubt that it's possible, and that's a different case from successfully getting the lock on the first try. It might not be enough of a difference to matter, but I feel this provides more info. I could change it though, I'm not sure if the increase in accuracy makes a real difference.

Co-authored-by: Ionatan Wiznia <ionatan@expensify.com>
@iwiznia
iwiznia merged commit 03d78e0 into main Jul 29, 2026
8 checks passed
@iwiznia
iwiznia deleted the tyler-count-commit-wait-time branch July 29, 2026 17:25
@tylerkaraszewski

Copy link
Copy Markdown
Contributor Author

@MelvinBot - create an issue, make it weekly, assign it to me. It should be to undo this change after we've gathered the required data to see how long we spend waiting on this lock.

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.

2 participants