Re: [PATCH 1/3] lib/vsprintf: add snprintf_noterm

From: Jeff Layton <jlayton@kernel.org>
To: Joe Perches <joe@perches.com>, "Yan, Zheng" <ukernel@gmail.com>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	ceph-devel <ceph-devel@vger.kernel.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Ilya Dryomov <idryomov@gmail.com>, Zheng Yan <zyan@redhat.com>,
	Sage Weil <sage@redhat.com>,
	agruenba@redhat.com
Subject: Re: [PATCH 1/3] lib/vsprintf: add snprintf_noterm
Date: Sat, 15 Jun 2019 07:08:41 -0400	[thread overview]
Message-ID: <ef030ba8553a8bc81fde998df4bd927bfba17537.camel@kernel.org> (raw)
In-Reply-To: <75c8f066c3aa2e20db2e1554a4d28c20b2952724.camel@perches.com>

On Fri, 2019-06-14 at 19:58 -0700, Joe Perches wrote:
> On Sat, 2019-06-15 at 10:41 +0800, Yan, Zheng wrote:
> > On Fri, Jun 14, 2019 at 9:48 PM Jeff Layton <jlayton@kernel.org> wrote:
> > > The getxattr interface returns a length after filling out the value
> > > buffer, and the convention with xattrs is to not NULL terminate string
> > > data.
> > > 
> > > CephFS implements some virtual xattrs by using snprintf to fill the
> > > buffer, but that always NULL terminates the string. If userland sends
> > > down a buffer that is just the right length to hold the text without
> > > termination then we end up truncating the value.
> > > 
> > > Factor the formatting piece of vsnprintf into a separate helper
> > > function, and have vsnprintf call that and then do the NULL termination
> > > afterward. Then add a snprintf_noterm function that calls the new helper
> > > to populate the string but skips the termination.
> 
> Is this function really necessary enough to add
> the additional stack use to the generic case?
> 

The only alternative I saw was to allocate an extra buffer in the
callers, call snprintf to populate that and then copy the result into
the destination buffer sans termination. I really would like to avoid
that here.

Does breaking this code out into a helper add any significant stack
usage? I didn't see it that way, but I am quite concerned about not
slowing down the generic vsnprintf routine.

> Why not add have this function call vsnprintf
> and then terminate the string separately?
> 

I don't quite follow what you're suggesting here. vsnprintf is what does
the termination today, and we need a function that doesn't do that.

-- 
Jeff Layton <jlayton@kernel.org>