Re: [PATCH AUTOSEL 6.1 12/21] fs/super.c: stop calling fscrypt_destroy_keyring() from __put_super()

[Slade Watkins]

On Sun, Feb 26, 2023 at 2:24 PM Eric Biggers <ebiggers@xxxxxxxxxx> wrote:
>
> On Sat, Feb 25, 2023 at 09:30:37PM -0800, Eric Biggers wrote:
> > On Sat, Feb 25, 2023 at 08:07:55PM -0800, Eric Biggers wrote:
> > > On Sat, Feb 25, 2023 at 10:42:47PM -0500, Sasha Levin wrote:
> > > > From: Eric Biggers <ebiggers@xxxxxxxxxx>
> > > >
> > > > [ Upstream commit ec64036e68634231f5891faa2b7a81cdc5dcd001 ]
> > > >
> > > > Now that the key associated with the "test_dummy_operation" mount option
> > > > is added on-demand when it's needed, rather than immediately when the
> > > > filesystem is mounted, fscrypt_destroy_keyring() no longer needs to be
> > > > called from __put_super() to avoid a memory leak on mount failure.
> > > >
> > > > Remove this call, which was causing confusion because it appeared to be
> > > > a sleep-in-atomic bug (though it wasn't, for a somewhat-subtle reason).
> > > >
> > > > Signed-off-by: Eric Biggers <ebiggers@xxxxxxxxxx>
> > > > Link: https://lore.kernel.org/r/20230208062107.199831-5-ebiggers@xxxxxxxxxx
> > > > Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
> > >
> > > Why is this being backported?
> > >
> > > - Eric
> >
> > BTW, can you please permanently exclude all commits authored by me from AUTOSEL
> > so that I don't have to repeatedly complain about every commit individually?
> > Especially when these mails often come on weekends and holidays.
> >
> > I know how to use Cc stable, and how to ask explicitly for a stable backport if
> > I find out after the fact that one is needed.  (And other real people can always
> > ask too... not counting AUTOSEL, even though you are sending the AUTOSEL emails,
> > since clearly they go through no or very little human review.)
> >
> > Of course, it's not just me that AUTOSEL isn't working for.  So, you'll still
> > continue backporting random commits that I have to spend hours bisecting, e.g.
> > https://lore.kernel.org/stable/20220921155332.234913-7-sashal@xxxxxxxxxx.
> >
> > But at least I won't have to deal with this garbage for my own commits.
> >
> > Now, I'm not sure I'll get a response to this --- I received no response to my
> > last AUTOSEL question at
> > https://lore.kernel.org/stable/Y1DTFiP12ws04eOM@sol.localdomain.  So to
> > hopefully entice you to actually do something, I'm also letting you know that I
> > won't be reviewing any AUTOSEL mails for my commits anymore.
> >
>
> The really annoying thing is that someone even replied to your AUTOSEL email for
> that broken patch and told you it is broken
> (https://lore.kernel.org/stable/d91aaff1-470f-cfdf-41cf-031eea9d6aca@xxxxxxxxxxx),
> and ***you ignored it and applied the patch anyway***.
>
> Why are you even sending these emails if you are ignoring feedback anyway?
>
> How do I even get you to not apply a patch?  Is it even possible?
>
> I guess I might as well just add an email filter that auto-deletes all AUTOSEL
> emails, as apparently there's no point in responding anyway?

I test this branch for Greg but don't pay attention to these emails
Sasha sends out (because there's just waaaaay too many of them to look
through unless they get a reply; I find them quite annoying
otherwise.) But if these commits automatically get applied to stable
trees, even with objections from the committers, then I personally
question the methodology for having AUTOSEL in the first place.
Commits should be tested and backported with explicit purpose by their
developers, IMO.

-- Slade