From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ric Wheeler Subject: Re: [Bug 13909] md5sum on 3GB file is constantly hanging Date: Mon, 17 Aug 2009 10:31:21 -0400 Message-ID: <4A8969B9.7000905@redhat.com> References: <200908151354.n7FDsmWp001652@demeter.kernel.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: linux-ext4@vger.kernel.org To: bugzilla-daemon@bugzilla.kernel.org Return-path: Received: from mx2.redhat.com ([66.187.237.31]:48195 "EHLO mx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752601AbZHQO3Q (ORCPT ); Mon, 17 Aug 2009 10:29:16 -0400 In-Reply-To: <200908151354.n7FDsmWp001652@demeter.kernel.org> Sender: linux-ext4-owner@vger.kernel.org List-ID: On 08/15/2009 09:54 AM, bugzilla-daemon@bugzilla.kernel.org wrote: > http://bugzilla.kernel.org/show_bug.cgi?id=13909 > > > > > > --- Comment #28 from Roland Kletzing 2009-08-15 13:54:43 --- > rus,any chance to try the attached patch and confirm if it is the real fix for > the problem ? > > > > We're now able to duplicate this bug. Not sure why it only affects ext4. > > ima: kernel_read parm integer overflow bugfix > > Reading large files>2G causes process to loop. > (http://bugzilla.kernel.org/show_bug.cgi?id=13909) > > Signed-off-by: Mimi Zohar > > Index: security-testing-2.6/security/integrity/ima/ima_crypto.c > =================================================================== > --- security-testing-2.6.orig/security/integrity/ima/ima_crypto.c > +++ security-testing-2.6/security/integrity/ima/ima_crypto.c > @@ -47,7 +47,8 @@ int ima_calc_hash(struct file *file, cha > struct scatterlist sg[1]; > loff_t i_size; > char *rbuf; > - int rc, offset = 0; > + unsigned long offset = 0; > + int rc; > > rc = init_desc(&desc); > if (rc != 0) > > I suspect that you will need to use the official offset type to avoid overflow at 4GB - loff_t should be more consistent, Ric