Re: [PATCH] Documentation/submitting-patches: Add blurb about backtraces in commit messages

From: Borislav Petkov
Date: Tue Jan 05 2021 - 05:48:52 EST


On Mon, Jan 04, 2021 at 04:19:11PM -0700, Jonathan Corbet wrote:
> So I have some questions, I guess... How often is a backtrace *in a commit
> message* really helpful at all? The value in problem reports is clear, but
> I'm not sure how often having a backtrace in a commit message will really
> help the reader understand why the patch was written. But perhaps I'm
> wrong?

Does the subthread here with Sean shed some light on the matter or... ?

> If we do want this advice in our already-too-long submitting-patches
> document,

Thought the same thing when looking at that doc - it is a *lot* and I
guess we should put only very globally relevant info in there...

> we should perhaps give some advice as to what is "relevant
> information" and what is not?

Right, in that subthread, the gist of what we wanna say is to almost
always put the splat in the commit message - except for the example I
gave there and other early boot cases - but leave it to the committer to
do the final decision whether to keep or ditch the splat.

Something like that. Yah, I know, it is fuzzy :-\

--
Regards/Gruss,
Boris.

https://people.kernel.org/tglx/notes-about-netiquette