Re: [PATCH v3 2/2] vfs: avoid duplicating creds in faccessat if possible

From: Christian Brauner <brauner@kernel.org>
To: Al Viro <viro@zeniv.linux.org.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
	Mateusz Guzik <mjguzik@gmail.com>,
	serge@hallyn.com, paul@paul-moore.com,
	linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-security-module@vger.kernel.org
Subject: Re: [PATCH v3 2/2] vfs: avoid duplicating creds in faccessat if possible
Date: Fri, 3 Mar 2023 15:27:32 +0100	[thread overview]
Message-ID: <20230303142732.2xsfnkhphlivwjb3@wittgenstein> (raw)
In-Reply-To: <ZADmuzDEr6yznutq@ZenIV>

On Thu, Mar 02, 2023 at 06:11:07PM +0000, Al Viro wrote:
> On Thu, Mar 02, 2023 at 09:30:25AM +0100, Christian Brauner wrote:
> > On Mon, Feb 27, 2023 at 04:44:06PM -0800, Linus Torvalds wrote:
> > > On Wed, Jan 25, 2023 at 7:56 AM Mateusz Guzik <mjguzik@gmail.com> wrote:
> > > >
> > > > Turns out for typical consumers the resulting creds would be identical
> > > > and this can be checked upfront, avoiding the hard work.
> > > 
> > > I've applied this v3 of the two patches.
> > > 
> > > Normally it would go through Al, but he's clearly been under the
> > > weather and is drowning in email. Besides, I'm comfortable with this
> > > particular set of patches anyway as I was involved in the previous
> > > round of access() overhead avoidance with the whole RCU grace period
> > > thing.
> > > 
> > > So I think we're all better off having Al look at any iov_iter issues.
> > > 
> > > Anybody holler if there are issues,
> > 
> > Fwiw, as long as you, Al, and others are fine with it and I'm aware of
> > it I'm happy to pick up more stuff like this. I've done it before and
> > have worked in this area so I'm happy to help with some of the load.
> 
> TBH, I've missed that series; patches look sane, so consider them
> belatedly ACKed.
> 
> And I've no problem with sharing the load - you have been doing that with
> idmapping stuff anyway.  As far as I'm concerned, I generally trust your
> taste; it doesn't matter that I won't disagree with specific decisions,
> of course, but...

Thanks, I appreciate that!

Sure, you won't agree with everything. That's kinda expected for reasons
of preference alone but also simply because there's a lot of stuff that
probably only you know atm. In general, I never pick up any stuff I
can't review myself unless it's in-your-face obvious or it already has
acks from the subject matter experts.

But it is much easier to help maintain vfs stuff now that you made it
clear that you're ok with this. So I'm happy to share more of the vfs
maintenance workload.