Em Sun, 23 May 2021 09:36:36 +0100
Sean Young <sean@xxxxxxxx> escreveu:
> Hi Mauro,
>
> On Sun, May 23, 2021 at 12:49:53PM +1000, Stephen Rothwell wrote:
> > Hi all,
> >
> > In commit
> >
> > 7c8a36e1fb30 ("media: rc: i2c: Fix an error message")
> >
> > Fixes tag
> >
> > Fixes: acaa34bf06e9 ('media: rc: implement zilog transmitter")
> >
> > has these problem(s):
> >
> > - Subject has leading but no trailing quotes
> > - Subject does not match target commit subject
> > Just use
> > git log -1 --format='Fixes: %h ("%s")'
>
> Now that we have media-staging, can the commit message be ammended?
If the patch didn't reach media_tree, yes, but this patch can't
be fixed anymore, as it is too old:
commit acaa34bf06e963f0b9481a3c16bfd6867e2369a0
Author: Sean Young <sean@xxxxxxxx>
AuthorDate: Sat Oct 21 08:16:47 2017 -0400
Commit: Mauro Carvalho Chehab <mchehab@xxxxxxxxxx>
CommitDate: Thu Dec 14 09:58:20 2017 -0500
media: rc: implement zilog transmitter
We need to double-check why linux-next is detecting it as a new one.
-
Linux-next should be pulling media work from those tree branches:
- v4l-dvb-fixes/fixes, e. g.:
git://linuxtv.org/media_tree.git fixes
- v4l-dvb/master, e. g.:
git://linuxtv.org/media_tree.git master
- v4l-dvb-next/master, e. g.:
git://linuxtv.org/mchehab/media-next.git master
In the past, I was using media-next (a.k.a. v4l-dvb-next at linux-next
nomenclature) to solve some special cases like when I need to deal with
patches that depends on other trees, or when I want to submit a separate
PR for some patches.
-
After the last merge window, I modified my workflow in order to:
1. be able of correcting problems on media patches, if pointed after
being merged at linux-next;
2. keep picking patches during the merge window.
The change is simple: I'm now pushing non-fixes stuff initially at
media-next master branch. Outside the merge window, the contents of
this branch is identical to the contents of the media stage tree:
git://linuxtv.org/media_stage.git
But I'll freeze changes there at the end of -rc6.
A couple of days after media-stage is merged at linux-next
(via media-next/linux-next tree), if no issues, I merge the
patches back at media_tree.
If something gets wrong in the mean time, I simply rebase
media-stage, fixing the issue again and push it back to the
media-next/linux-next tree.
-
Now, I can't understand why this specific patch had a trouble
detected today, as it was this patch is already upstream:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=acaa34bf06e9
Apparently since 4.15-rc1:
$ git describe acaa34bf06e9
v4.15-rc1-204-gacaa34bf06e9
Thanks,
Mauro
