From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mga06.intel.com ([134.134.136.31]:8199 "EHLO mga06.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752508AbdDKWpo (ORCPT ); Tue, 11 Apr 2017 18:45:44 -0400 Date: Tue, 11 Apr 2017 16:45:42 -0600 From: Ross Zwisler Subject: Re: [PATCH v2 3/3] DAX: mmap write readonly file Message-ID: <20170411224542.GA12746@linux.intel.com> References: <20170407171600.GA29489@linux.intel.com> <1491804353-1326-1-git-send-email-xzhou@redhat.com> <1491804353-1326-3-git-send-email-xzhou@redhat.com> <20170411114619.GK22845@eguan.usersys.redhat.com> <20170411185235.GA32027@linux.intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170411185235.GA32027@linux.intel.com> Sender: fstests-owner@vger.kernel.org To: Ross Zwisler Cc: Jeff Moyer , Eryu Guan , Xiong Zhou , fstests@vger.kernel.org, dan.j.williams@intel.com List-ID: On Tue, Apr 11, 2017 at 12:52:35PM -0600, Ross Zwisler wrote: > On Tue, Apr 11, 2017 at 09:56:14AM -0400, Jeff Moyer wrote: > > Eryu Guan writes: > > > > > On Mon, Apr 10, 2017 at 02:05:53PM +0800, Xiong Zhou wrote: > > >> Regression case that one can write to read-only > > >> file in a DAX mountpoint. > > >> > > >> Signed-off-by: Xiong Zhou > > >> --- > > >> > > >> v2: > > >> compile test programme manually in this test because default > > >> cc option -O2 prevents this issue reproduction; > > > > > > Hmm, this looks.. ugly to me :) Better to find out the exact reason that > > > prevents the bug from reproducing and update the c program accordingly. > > > > The compiler probably optimizes this bit out: > > > > + /* fault in the page */ > > + foo = *buf; > > Yep, verified this with objdump. You can prevent the compiler from > optimizing out this bit by making 'foo' volatile. Patch at the end of this > mail, and I also verified that this works with objdump. > > In my setup at least this test passes both with v4.10.0 (which does not > contain the fix we are testing for, and should fail) and with v4.10.3 (which > does contain the kernel fix, and should pass). > > So, I think the test still needs a little love. :) Umm...never mind, my testing was broken. I was accidentally testing with the upstream generic/422. :( Will retest tomorrow.