linux-kernel - Re: [PATCH v2] nfsd: prevent integer overflow in decode_cb

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] [day] [month] [year] [list]

Message-ID: <183154d2-cb65-4161-a737-394aba9c1d43@suswa.mountain>
Date: Fri, 20 Sep 2024 14:10:02 +0300
From: Dan Carpenter <dan.carpenter@...aro.org>
To: Chuck Lever III <chuck.lever@...cle.com>
Cc: Jeff Layton <jlayton@...nel.org>, Neil Brown <neilb@...e.de>,
	Olga Kornievskaia <okorniev@...hat.com>,
	Dai Ngo <dai.ngo@...cle.com>, Tom Talpey <tom@...pey.com>,
	Linux NFS Mailing List <linux-nfs@...r.kernel.org>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	"kernel-janitors@...r.kernel.org" <kernel-janitors@...r.kernel.org>
Subject: Re: [PATCH v2] nfsd: prevent integer overflow in
 decode_cb_compound4res()

On Thu, Sep 19, 2024 at 02:02:19PM +0000, Chuck Lever III wrote:
> 
> 
> > On Sep 19, 2024, at 1:12 AM, Dan Carpenter <dan.carpenter@...aro.org> wrote:
> > 
> > If "length" is >= U32_MAX - 3 then the "length + 4" addition can result
> > in an integer overflow.  The impact of this bug is not totally clear to
> > me, but it's safer to not allow the integer overflow.
> > 
> > Check that "length" is valid right away before doing any math.
> > 
> > Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
> > Signed-off-by: Dan Carpenter <dan.carpenter@...aro.org>
> > ---
> > v2: Check that "len" is valid instead of just checking for integer
> >    overflows.
> > 
> > fs/nfsd/nfs4callback.c | 2 ++
> > 1 file changed, 2 insertions(+)
> > 
> > diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
> > index 43b8320c8255..0f5b7b6fba74 100644
> > --- a/fs/nfsd/nfs4callback.c
> > +++ b/fs/nfsd/nfs4callback.c
> > @@ -317,6 +317,8 @@ static int decode_cb_compound4res(struct xdr_stream *xdr,
> > hdr->status = be32_to_cpup(p++);
> > /* Ignore the tag */
> > length = be32_to_cpup(p++);
> > + if (unlikely(length > xdr->buf->len))
> > + goto out_overflow;
> > p = xdr_inline_decode(xdr, length + 4);
> > if (unlikely(p == NULL))
> > goto out_overflow;
> > -- 
> > 2.34.1
> > 
> 
> Hi Dan, I've already gone with
> 
> https://lore.kernel.org/linux-nfs/172658972371.2454.15715383792386404543.stgit@oracle-102.chuck.lever.oracle.com.nfsv4.dev/T/#u
> 
> Let me know if that doesn't make sense.

Oh, sorry, I got mixed up which things were patched.  That looks good.
The truth is I always prefer when people give me a reported by tag so I
don't have to write a v2 patch.  It just saves a back and forth.
Thanks!

regards,
dan carpenter