Skip to content

Conversation

@rymurr
Copy link
Contributor

@rymurr rymurr commented Jul 8, 2020

As per #7619 (comment) the vector tests should be run for both netty and unsafe allocators. The default tests are run with the Netty allocator, and run-unsafe tests are done with the Unsafe Allocator.

@github-actions
Copy link

github-actions bot commented Jul 8, 2020

@rymurr rymurr marked this pull request as ready for review July 9, 2020 19:43
Copy link
Contributor

Choose a reason for hiding this comment

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

Can we simply remove this section?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Hey @liyafan82, we can skip this section but we will run the tests 3 times then. One of the three will be with an unknown Allocator (it will be chosen at runtime depending on classpath). I thought it was a bit of a waste.

Copy link
Contributor

Choose a reason for hiding this comment

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

Thanks for your clarification.

Copy link
Contributor

Choose a reason for hiding this comment

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

When we run tests from IDE, it is also with an unknown allocator, right?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

yes, thats correct. We would have to set an env variable in the IDE to choose one or the other.

Copy link
Member

Choose a reason for hiding this comment

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

would it work to remove the skip=true here and add <classpathDependencyExclude>org.apache.arrow:arrow-memory-unsafe</classpathDependencyExclude>, so that the default test runs netty, then leave the additional run-unsafe section only?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

yes, that is quite a bit simpler. Apologies for misunderstanding originally.

I have made that change

Copy link
Contributor

Choose a reason for hiding this comment

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

I do not find the code for closing the allocator

Copy link
Contributor

Choose a reason for hiding this comment

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

It would be nice to implement the code by try-with-resource

Copy link
Contributor Author

Choose a reason for hiding this comment

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

thanks, agreed. Fixed!

@rymurr rymurr force-pushed the ARROW-9371 branch 2 times, most recently from e993dc0 to 7cf3ecc Compare July 23, 2020 11:33
Copy link
Contributor

@liyafan82 liyafan82 left a comment

Choose a reason for hiding this comment

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

LGTM, thanks.

@rymurr
Copy link
Contributor Author

rymurr commented Jul 23, 2020

Build fail could be a random network error or related to the parallel build. Have rebased & retriggered.

Copy link
Member

@BryanCutler BryanCutler left a comment

Choose a reason for hiding this comment

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

LGTM, just a couple minor things

Copy link
Member

Choose a reason for hiding this comment

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

I think you can remove the comment now

Copy link
Contributor Author

Choose a reason for hiding this comment

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

👍

Copy link
Member

Choose a reason for hiding this comment

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

rename to run-unsafe please

Copy link
Contributor Author

Choose a reason for hiding this comment

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

👍

@rymurr
Copy link
Contributor Author

rymurr commented Jul 24, 2020

LGTM, just a couple minor things

Thanks for the review @BryanCutler

I won't rush a small change next time and save you the trouble ;-)

@BryanCutler
Copy link
Member

Did a quick timing test for this, as expected it doubles the test time for arrow-vector, which is ~10s on my laptop. That seems ok to me.

Before
[INFO] Arrow Vectors ...................................... SUCCESS [ 10.986 s]

After
[INFO] Arrow Vectors ...................................... SUCCESS [ 20.730 s]

@BryanCutler
Copy link
Member

@rymurr looks like this needs a rebase

As per apache#7619 (comment) the vector tests should be run for
 both netty and unsafe allocators
@rymurr
Copy link
Contributor Author

rymurr commented Jul 24, 2020

rebase is done, thanks @BryanCutler

I agree re timing, I saw the same numbers. On my laptop the vector compilation time dwarfs the test time so doubling the test time didn't seem too bad.

Copy link
Member

@BryanCutler BryanCutler left a comment

Choose a reason for hiding this comment

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

LGTM

@BryanCutler
Copy link
Member

merged to master, thanks @rymurr !

pribor pushed a commit to GlobalWebIndex/arrow that referenced this pull request Oct 24, 2025
As per apache#7619 (comment) the vector tests should be run for both netty and unsafe allocators. The default tests are run with the Netty allocator, and `run-unsafe` tests are done with the Unsafe Allocator.

Closes apache#7676 from rymurr/ARROW-9371

Authored-by: Ryan Murray <rymurr@dremio.com>
Signed-off-by: Bryan Cutler <cutlerb@gmail.com>
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