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

[CI] Add benchmarks' enhancements #259

Merged
merged 11 commits into from
Nov 3, 2024
Merged

Conversation

iamazeem
Copy link
Collaborator

@iamazeem iamazeem commented Nov 3, 2024

  • Use build artifacts from other workflow runs using the run ID (optional)
  • Use the latest released build otherwise

Signed-off-by: Azeem Sajid [email protected]

@iamazeem iamazeem added documentation Improvements or additions to documentation enhancement New feature or request labels Nov 3, 2024
@iamazeem iamazeem self-assigned this Nov 3, 2024
@iamazeem
Copy link
Collaborator Author

iamazeem commented Nov 3, 2024

Steps:

@iamazeem iamazeem merged commit fabf6ce into main Nov 3, 2024
@iamazeem iamazeem deleted the ci-benchmarks-enhancements branch November 3, 2024 19:18
@iamazeem iamazeem requested a review from liquidaty November 3, 2024 19:18
@liquidaty
Copy link
Owner

@iamazeem the musl builds still seem a little off for a couple reasons. First, it seems odd that zsv would not still outperform tsv in the select benchmark just like the other zsv builds. is there any theoretical explanation for that? Second, when we build on our musl-gcc (x86_64-linux-musl-gcc (GCC) 9.4.0), the binary size is much smaller (2511640 bytes in the last run) than what is being produces in the artifacts. Could you send or post the config.musl file that is being generated by the musl build's ./configure run so that we can examine?

@iamazeem
Copy link
Collaborator Author

iamazeem commented Nov 6, 2024

@liquidaty:

First, it seems odd that zsv would not still outperform tsv in the select benchmark just like the other zsv builds. is there any theoretical explanation for that?

Not yet.
Need to look into that.
Could be due to differences in internal API implementation.

Second, when we build on our musl-gcc (x86_64-linux-musl-gcc (GCC) 9.4.0), the binary size is much smaller (2511640 bytes in the last run) than what is being produces in the artifacts.

Regarding 2511640, is that for the stripped binary?

Just downloaded zsv-0.3.9-alpha-amd64-linux-musl.tar.gz, stripped the binary and got 2416888:

$ ls -l zsv
-rwxr-xr-x 1 azeem azeem 3411440 Nov  6 10:12 zsv
$ strip -s zsv
$ ls -l zsv
-rwxr-xr-x 1 azeem azeem 2416888 Nov  6 15:03 zsv

Could you send or post the config.musl file that is being generated by the musl build's ./configure run so that we can examine?

Sure, I'll share it later today.

@iamazeem iamazeem mentioned this pull request Nov 6, 2024
@iamazeem
Copy link
Collaborator Author

iamazeem commented Nov 6, 2024

@liquidaty: Please see config.mk file here:
https://github.com/liquidaty/zsv/actions/runs/11702289131/attempts/1#summary-32590175922

We may continue this discussion there. Thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
documentation Improvements or additions to documentation enhancement New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants