Received-SPF: pass (google.com: domain of linux-kernel+bounces-159402-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
Date: Thu, 25 Apr 2024 20:58:34 -0400
From: Kent Overstreet <kent.overstreet@linux.dev>
To: Kees Cook <keescook@chromium.org>
Cc: Andrew Morton <akpm@linux-foundation.org>, 
	Matthew Wilcox <willy@infradead.org>, Suren Baghdasaryan <surenb@google.com>, linux-mm@kvack.org, 
	linux-kernel@vger.kernel.org, linux-hardening@vger.kernel.org
Subject: Re: [PATCH] alloc_tag: Tighten file permissions on /proc/allocinfo
Message-ID: <qzi5ss4h4ou6yfbzadoamqocvvzviuh3eeefpv5qfkcvrrejfo@qptmu2y6u7qj>
References: <20240425200844.work.184-kees@kernel.org>
 <w6nbxvxt3itugrvtcvnayj5ducoxifwbffd7qh6vcastw77mse@2ugphwusgttz>
 <ZirCbPR1XwX2WJSX@casper.infradead.org>
 <64cngpnwyav4odustofs6hgsh7htpc5nu23tx4lb3vxaltmqf2@sxn63f2gg4gu>
 <202404251532.F8860056AE@keescook>
 <20240425164718.e8e187dd0c5b0a87371d8316@linux-foundation.org>
 <sxcyj2gif2avyx2disz62sfe3hot24w4rtszgl2dtqadegtnek@xjkozdbd6yzp>
 <202404251740.81F21E54@keescook>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <202404251740.81F21E54@keescook>

On Thu, Apr 25, 2024 at 05:43:33PM -0700, Kees Cook wrote:
> On Thu, Apr 25, 2024 at 08:27:05PM -0400, Kent Overstreet wrote:
> > On Thu, Apr 25, 2024 at 04:47:18PM -0700, Andrew Morton wrote:
> > > On Thu, 25 Apr 2024 15:42:30 -0700 Kees Cook <keescook@chromium.org> wrote:
> > > 
> > > > > The concern about leaking image layout could be addressed by sorting the
> > > > > output before returning to userspace.
> > > > 
> > > > It's trivial to change permissions from the default 0400 at boot time.
> > > > It can even have groups and ownership changed, etc. This is why we have
> > > > per-mount-namespace /proc instances:
> > > > 
> > > > # chgrp sysmonitor /proc/allocinfo
> > > > # chmod 0440 /proc/allocinfo
> > > > 
> > > > Poof, instant role-based access control. :)
> > > 
> > > Conversely, the paranoid could set it to 0400 at boot also.
> > > 
> > > > I'm just trying to make the _default_ safe.
> > > 
> > > Agree with this.
> > > 
> > > Semi-seriously, how about we set the permissions to 0000 and force
> > > distributors/users to make a decision.
> > 
> > I'm ok with 0400 for now since it's consistent with slabinfo, but I'd
> > really like to see a sysctl for debug info paranoia. We shouldn't be
> > leaving this to the distros; we're the ones with the expertise to say
> > what would be covered by that sysctl.
> 
> We've not had great luck with sysctls (see userns sysctl discussions)
> since they don't provide sufficient granularity.
> 
> All this said, I'm still not excited about any of these files living
> in /proc at all -- we were supposed to use /sys for this kind of thing,
> but its interface wasn't great for this kind of more "free-form" data,
> and debugfs isn't good for production interfaces. /proc really should
> only have pid information -- we end up exposing these top-level files to
> every mount namespace with a /proc mount. :( But that's a yet-to-be-solved
> problem...

It really wouldn't be that hard to relax the 4k file limit in sysfs.