Skip to content
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.

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

Add release note for parallel module creation in LLDB #116857

Merged
merged 3 commits into from
Nov 20, 2024

Conversation

JDevlieghere
Copy link
Member

Release note #110646 and #114507.

@JDevlieghere JDevlieghere force-pushed the parallel-module-release-note branch from 942deda to 4dafd39 Compare November 19, 2024 18:55
@@ -355,6 +355,8 @@ Changes to LLDB

* A new setting `target.launch-working-dir` can be used to set a persistent cwd that is used by default by `process launch` and `run`.

* LLDB now creates modules in parallel, resulting in an average 2x speedup when launching and attaching.
Copy link
Collaborator

Choose a reason for hiding this comment

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

Good idea to highlight it.

Outside of us, I don't think most people know what "creates modules" is. Maybe "reads binaries". More accurately "scans binary symbol tables", but that's pushing what people would understand again.

I do have a little concern that we've only done this for the attach use-case on Darwin, and for the "load all dependent libraries" use case on all platforms. Someone might attach to a process on linux and see no benefit at all, and be sad.

Copy link
Member Author

Choose a reason for hiding this comment

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

WDYT of the new wording?

Copy link
Collaborator

Choose a reason for hiding this comment

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

So is this does load in parallel everywhere but you only measured it on Darwin, or has the possibility to do so everywhere but only does so right now on Darwin?

The former I would write as:

  • LLDB can . On Darwin this has resulted in...

Implying that YMMV anywhere else.

The latter:

  • LLDB running on Darwin will now load libraries in parallel when attaching or.... This gives an average 2x speedup.

Copy link
Member Author

Choose a reason for hiding this comment

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

We load modules in parallel everywhere when launching. We only load modules in parallel when attaching on Darwin (because that happens in our dynamic loader plugin). Linux and other platforms could probably/likely do the same thing, but I haven't looked into it. The improvements are the same regardless of whether the attach or launch.

@DavidSpickett
Copy link
Collaborator

Assuming I remember, I will trawl the git log prior to the 20 branch and see if there's anything release note worthy.

But in the meantime glad to see more being added.

Copy link
Collaborator

@jasonmolenda jasonmolenda left a comment

Choose a reason for hiding this comment

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

New wording looks good.

@JDevlieghere JDevlieghere merged commit 4acf935 into llvm:main Nov 20, 2024
6 of 9 checks passed
@JDevlieghere JDevlieghere deleted the parallel-module-release-note branch November 20, 2024 21:25
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.

3 participants