From: Rickard Strandqvist Subject: Re: [PATCH] crypto: rng.c: Cleaning up missing null-terminate in conjunction with strncpy Date: Sun, 27 Jul 2014 14:14:00 +0200 Message-ID: References: <1406384308-2269-1-git-send-email-rickard_strandqvist@spectrumdigital.se> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8BIT Cc: Neil Horman , Herbert Xu , "David S. Miller" , linux-crypto@vger.kernel.org, "linux-kernel@vger.kernel.org" To: Mark D Rustad Return-path: Received: from mail-oi0-f43.google.com ([209.85.218.43]:48660 "EHLO mail-oi0-f43.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750885AbaG0MOV convert rfc822-to-8bit (ORCPT ); Sun, 27 Jul 2014 08:14:21 -0400 Received: by mail-oi0-f43.google.com with SMTP id u20so5024329oif.16 for ; Sun, 27 Jul 2014 05:14:21 -0700 (PDT) In-Reply-To: Sender: linux-crypto-owner@vger.kernel.org List-ID: 2014-07-27 4:35 GMT+02:00 Mark D Rustad : > Rickard, > > On Jul 26, 2014, at 7:18 AM, Rickard Strandqvist wrote: > >> Replacing strncpy with strlcpy to avoid strings that lacks null terminate. >> >> Signed-off-by: Rickard Strandqvist >> --- >> crypto/rng.c | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/crypto/rng.c b/crypto/rng.c >> index e0a25c2..c3d4fb3 100644 >> --- a/crypto/rng.c >> +++ b/crypto/rng.c >> @@ -65,7 +65,7 @@ static int crypto_rng_report(struct sk_buff *skb, struct crypto_alg *alg) >> { >> struct crypto_report_rng rrng; >> >> - strncpy(rrng.type, "rng", sizeof(rrng.type)); >> + strlcpy(rrng.type, "rng", sizeof(rrng.type)); >> >> rrng.seedsize = alg->cra_rng.seedsize; > > Not to pick on this patch in particular, but you need to be careful about changing strncpy to strlcpy. Although strlcpy ensures termination, it does not prevent information leakage - strncpy ensures that the entire destination buffer is written. When leakage is a concern, it is better to use strncpy and then to store a zero in the last location of the buffer to ensure termination. > > These "simple" transformations can be risky - and many of these do not represent any sort of problem when the source is smaller than the destination. I hope information leakage is being considered. > Hi Mark You clearly have a point there! Information leakage is not something I think about very often otherwise. It is really unbelievable that C messed up when trying to make a better solution to strcpy. strncpy doing something completely new with zeroes whole size but still does not guarantee a terminating null character. And strlcpy that works great as a replacement for strcpy, but really can not replace strncpy in all cases. Incredibly stupid! :-( But ok, I'll submit a [Patch V2] with foo[sizeof(foo) - 1] = '\ 0' on all under crypto/ then ? Kind regards Rickard Strandqvist