linux-kernel - Re: [PATCH] Cleanup: snprintf() always NUL-terminates: depend on it

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Sun, 4 Jan 2015 23:38:00 +0100
From:	Giel van Schijndel <me@...tis.eu>
To:	Linus Torvalds <torvalds@...ux-foundation.org>
Cc:	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Brian King <brking@...ibm.com>,
	"James E.J. Bottomley" <JBottomley@...allels.com>,
	"open list:SCSI SUBSYSTEM" <linux-scsi@...r.kernel.org>
Subject: Re: [PATCH] Cleanup: snprintf() always NUL-terminates: depend on it

On Sun, Jan 04, 2015 at 11:34:43 -0800, Linus Torvalds wrote:
> On Sun, Jan 4, 2015 at 10:05 AM, Giel van Schijndel <me@...tis.eu> wrote:
>> Especially since one very strange piece of code seems to be written in
>> such a way that a NUL needs to be placed where a NUL is present already.
> 
> Actually, it's worse than that. This:
> 
>>         len = snprintf(fname, 99, "%s", buf);
>> -       fname[len-1] = '\0';
> 
> is complete garbage, since the return value of snprintf() is not the
> length of the result, but length of what the result *would* have been.
> 
> So if the string doesn't fit in 99 bytes, it will actively corrupt
> some random memory after the string. It's not writing zero to what was
> already zero, it's corrupting memory.

Ah yes, I didn't even notice that nasty side effect. I just deleted that
"really, really" NUL-termination line because it was based on a
misunderstanding of snprintf()'s postcondition. Even if
len==sizeof(fname) this still would have given the wrong example for
others to follow.

> Anyway, from a quick glance your patches look fine, but you need to
> sign off on them. See Documentation/SubmittingPatches.

Ah yes, forgot that. Would it be sufficient if I sent a reply to all
those patch mails with next line tacked on, or would it require a
resubmission?
> Signed-off-by: Giel van Schijndel <me@...tis.eu>

-- 
Met vriendelijke groet,
With kind regards,
Giel van Schijndel
--
"Always code as if the guy who ends up maintaining your code will be a
 violent psychopath who knows where you live."
  -- Rick Osborne

Download attachment "signature.asc" of type "application/pgp-signature" (182 bytes)