-
Notifications
You must be signed in to change notification settings - Fork 29k
[SPARK-47242][BUILD] Bump ap-loader 3.0(v8) to support for async-profiler 3.0 #45351
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
Conversation
…r async-profiler 3.0
|
Thank you. |
parthchandra
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
|
To @SteNicholas , for this kind of PR, we can use simply To @parthchandra , could you confirm that this works without any issues, please? We don't have a test coverage in CI, do we? |
|
Yes there is no good way to test this in ci. Let me try it out and make sure. |
|
Thank you so much, @parthchandra ! |
|
@parthchandra, thank you try it out. Have you tried anything wrong? |
|
The original PR does not have its test together (#44021). @SteNicholas Mind describing how you tested this? |
@HyukjinKwon, I only tested this bump in production environment of Spark cluster. |
|
That's fine. My point is that please describe it in the PR description. |
HyukjinKwon
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM otherwise
I was able to try it out locally (non production) and the jfr files written were fine. I didn't see much difference in the output for my job, but others may definitely benefit. |
dongjoon-hyun
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you, @SteNicholas , @parthchandra , @HyukjinKwon .
Merged to master for Apache Spark 4.0.0.
|
I added you to the Apache Spark contributor group, @SteNicholas , and assigned SPARK-47242 to you. |
…iler 3.0 ### What changes were proposed in this pull request? Bump ap-loader from 3.0(v8) to support for async-profiler 3.0. ### Why are the changes needed? ap-loader 3.0(v8) has already been released, which supports for async-profier 3.0. The release guide refers to [Loader for 3.0 (v8): Binary launcher and AsyncGetCallTrace replacement](https://github.com/jvm-profiling-tools/ap-loader/releases/tag/3.0-8). Breaking changes with async-profiler 3.0: async-profiler 3.0 changed the meaning of the `--lib` option from `--lib path full path to libasyncProfiler.so in the container` to `-l, --lib prepend library names`, so the `AsyncProfilerLoader` will throw an `UnsupportedOperation` exception when using the --lib option with a path argument and async-profiler 3.0 or higher. ### Does this PR introduce _any_ user-facing change? No. ### How was this patch tested? Pass GA. Tested in production environment of Spark cluster. ### Was this patch authored or co-authored using generative AI tooling? No. Closes apache#45351 from SteNicholas/SPARK-47242. Authored-by: SteNicholas <[email protected]> Signed-off-by: Dongjoon Hyun <[email protected]>

What changes were proposed in this pull request?
Bump ap-loader from 3.0(v8) to support for async-profiler 3.0.
Why are the changes needed?
ap-loader 3.0(v8) has already been released, which supports for async-profier 3.0. The release guide refers to Loader for 3.0 (v8): Binary launcher and AsyncGetCallTrace replacement. Breaking changes with async-profiler 3.0:
async-profiler 3.0 changed the meaning of the
--liboption from--lib path full path to libasyncProfiler.so in the containerto-l, --lib prepend library names, so theAsyncProfilerLoaderwill throw anUnsupportedOperationexception when using the --lib option with a path argument and async-profiler 3.0 or higher.Does this PR introduce any user-facing change?
No.
How was this patch tested?
Pass GA. Tested in production environment of Spark cluster.
Was this patch authored or co-authored using generative AI tooling?
No.