Re: [PATCH] btrfs: reloc: unlink orphan reloc roots before dropping them
From: Qu Wenruo
Date: Tue Mar 10 2026 - 05:44:56 EST
在 2026/3/10 19:11, ZhengYuan Huang 写道:
On Tue, Mar 10, 2026 at 4:13 PM Qu Wenruo <wqu@xxxxxxxx> wrote:
[...]
Put this important info into changelog, and this is not the first time I
or other reviewing asking you to do it.
Thanks a lot for the detailed review and for being patient with my patch
submissions. I'm still learning the kernel patch submission style,
especially how much detail should go into the changelog above the "---" line
versus what should stay below it.
I think part of my confusion comes from seeing many patches with very concise
changelogs, so I have been trying to keep that part short, but I may
have overdone it and moved too much useful information below the separator.
If you know the bug/cause/fix well, and pretty sure it will not affect most users, sure short changelogs are good as long as you explained it well.
But if you already hit a KASAN/crash, paste the info itself will help explaing the situation well enough, and such calltrace will help a lot in the future for the following cases:
- The end user who hits a crash with similar call trace
Who want to know if it's already fix or some one else hits the same
problem.
If the call trace is included, one can determine if it's the same thus
if it's already fixed in the latest kernel.
- The engineer who is responsible for backporting
Such call trace will help him/her to determine if it's needed for
backport.
A KASAN report/crash with call trace will definitely be more obvious.
From your feedback, my understanding is that the changelog should include
the essential root cause, the fix rationale, and the key crash symptom
(for example
a concise KASAN summary), while the material below "---" should be limited to
supplementary information such as full reproduction details or longer
logs. Is that
the right interpretation?
Yes.
If there is a patch or changelog example that you think is a good reference
for this style, I would really appreciate it. I'd like to study it carefully
and improve how I write future submissions.
I just grabbed one from Johannes:
https://lore.kernel.org/linux-btrfs/20260224125113.14831-1-johannes.thumshirn@xxxxxxx/
And from Filipe:
https://lore.kernel.org/linux-btrfs/b99cee6ce652b926463a080ef052a2e8e37bff33.1772105193.git.fdmanana@xxxxxxxx/
And myself, which is more aligned to your style:
https://lore.kernel.org/linux-btrfs/4170e39bac4a2559ad0535f9bd74a89bc44a36d4.1771488629.git.wqu@xxxxxxxx/
Thanks,
Qu
Thanks again for the guidance.
Thanks,
ZhengYuan Huang