-
-
Notifications
You must be signed in to change notification settings - Fork 281
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 meeting notes 2024-11-27 #2388
Conversation
✅ Deploy Preview for conda-forge-previews ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
- [x] (WV) multi-output cache for v1 recipes | ||
- [x] Already available with `--experimental` | ||
- [x] Wrote up a (proto)-CEP https://github.com/conda/ceps/pull/102 | ||
- [x] (UK) `-fno-omit-frame-pointer` |
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.
What was the outcome of this discussion? I remember that @jjerphan and colleagues were interested in this. AFAIU it's an ABI break though, so we'd probably have to queue that up for whenever we rebuild the world?
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.
For such a change, I would be more than prudent and assume that ABI-compatibility depends on platforms.
If they were to be activated again, I guess the best would be to have a conda-forge reboot or at least a batch rebuild for an entire set of packages.
See this part of Fedora Project Wiki for more information.
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.
Is it actually an ABI break to activate frame pointers? That sounds unexpected to me: you can compile any code with or without frame pointers, and still link to the same existing third-party or system libs.
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.
@xhochy was going to make a PR and see what people said
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.
Yes, but could it in practice alter some calling conventions?
I am not knowledgeable or experienced enough to be able to tell; I think it might be worth contacting people who re-activated them on RPM-based or DEB-based distributions.
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.
For such a change, I would be super prudent and assume that ABI-compatibility depends on platforms.
Actually, I'm confused about whether this would change the ABI. My understanding is that a specific register is always filled, independent of whether the function uses it or not. I don't think this should be a world-rebuilt.
Let's move this discussion to the PR over here: conda-forge/clang-compiler-activation-feedstock#151
New meeting notes available at https://hackmd.io/@conda-forge/SkSj0ONmJx.
Once done with the meeting, sync the note back to the repository by adding the
sync-hackmd-notes
label.