From: "Jesper Juhl" Subject: Re: [PATCH] NFS: nfsaclsvc_encode_getaclres() - Fix potential NULL deref and tiny optimization. Date: Tue, 31 Oct 2006 17:40:21 +0100 Message-ID: <9a8748490610310840w28d44a88xd1db60b155c03f52@mail.gmail.com> References: <200610272316.47089.jesper.juhl@gmail.com> <200610280001.49272.jesper.juhl@gmail.com> <200610311726.00411.agruen@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Cc: "David Rientjes" , linux-kernel@vger.kernel.org, "Neil Brown" , nfs@lists.sourceforge.net, "Andrew Morton" Return-path: To: "Andreas Gruenbacher" In-Reply-To: <200610311726.00411.agruen@suse.de> Sender: linux-kernel-owner@vger.kernel.org List-ID: On 31/10/06, Andreas Gruenbacher wrote: > On Saturday 28 October 2006 00:01, Jesper Juhl wrote: > > > > 3) There are two locations in the function where we may return before > > > > we use the value of the variable 'w', but we compute it at the very top > > > > of the function. So in the case where we return early we have wasted a > > > > few cycles computing a value that was never used. > > Computing w later in the function is fine. > > > > w should be an unsigned int. > > > > Makes sense. > > No, this breaks the while loop further below: with an unsigned int, the loop > counter underflows and wraps. > Whoops. OK. > Please fix this identically in fs/nfsd/nfs2acl.c and fs/nfsd/nfs3acl.c. > Sure thing, expect patches later this evening. BTW: I posted an add-on patch on top of my first one - apart from the "make w unsigned" bit, is the rest of that OK? -- Jesper Juhl Don't top-post http://www.catb.org/~esr/jargon/html/T/top-post.html Plain text mails only, please http://www.expita.com/nomime.html