On Wed, May 28, 2025 at 04:47:00PM +0200, Danilo Krummrich wrote: > On Wed, May 28, 2025 at 04:39:01PM +0200, Danilo Krummrich wrote: > > On Wed, May 28, 2025 at 09:29:30AM -0400, Alex Deucher wrote: > > > On Wed, May 28, 2025 at 8:45 AM Simona Vetter <simona.vet...@ffwll.ch> > > > wrote: > > > > I do occasionally find it useful as a record of different approaches > > > > considered, which sometimes people fail to adequately cover in their > > > > commit messages. Also useful indicator of how cursed a patch is :-) > > > > > > > > But as long as anything relevant does end up in the commit message and > > > > people don't just delete stuff I don't care how it's done at all. It's > > > > just that the cost of deleting something that should have been there can > > > > be really nasty sometimes, and storage is cheap. > > > > > > I like them for the same reasons. Also, even with links, sometimes > > > there are forks of the conversation that get missed that a changelog > > > provides some insight into. I find it useful in my own development as > > > I can note what I've changed in a patch and can retain that in the > > > commit rather than as something I need to track separately and then > > > add to the patches when I send them out. > > > > Personally, I don't think it's super useful in the commit message, it still > > remains in the patches sent to the mailing list though. And since we put > > lore > > links everywhere, it's easily accessible, *including* the context of why a > > change was made from one version to another, i.e. the full conversation. > > > > However, if we really want that, we should make it an offical thing, since > > currently the kernel's process documentation [1] clearly states otherwise: > > > > "Please put this information after the '---' line which separates the > > changelog > > from the rest of the patch. The version information is not part of the > > changelog > > which gets committed to the git tree. It is additional information for the > > reviewers. If it's placed above the commit tags, it needs manual > > interaction to > > remove it." > > > > Alternatively, it can go into the cover letter. > > One additional note: > > This is not me trying to be super bureaucratic; instead I think being > consistent > in the process across the whole kernel results in a better experience for > (new) > contributors.
Yeah I agree with this part, which is why in the past I didn't ask people to keep that part, but also won't complain if it's kept. The entire goal being "minimal amount to get to a commit message that's hopefully complete". I think with b4 this has now also become a bit easier than 10+ years ago. Also all the kernel fd.o lists are on lore and the archive on fd.o is under our control, so hopefully the archive situation shouldn't ever be an issue for us. Anyway no strong opinion from me, but we might want to document that we're a bit more relaxed here. *shrugs* Cheers, Sima > > > [1] https://docs.kernel.org/process/submitting-patches.html#commentary -- Simona Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch
