From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail.kernel.org ([198.145.29.99]:43128 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752417AbeESXvS (ORCPT ); Sat, 19 May 2018 19:51:18 -0400 Message-ID: <4cb7ba897fd4b5641be680cd599e74dd0c8e41bf.camel@kernel.org> Subject: Re: What's up with "locks: change POSIX lock ownership on execve when files_struct is displaced"? From: Jeff Layton Date: Sat, 19 May 2018 19:51:15 -0400 In-Reply-To: <20180519194711.GA21362@thunk.org> References: <20180519194711.GA21362@thunk.org> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: fstests-owner@vger.kernel.org To: "Theodore Y. Ts'o" Cc: linux-fsdevel@vger.kernel.org, fstests@vger.kernel.org, ebiederm@xmission.com List-ID: On Sat, 2018-05-19 at 15:47 -0400, Theodore Y. Ts'o wrote: > There is quite of noise currently with xfstests failures. In addition > to the fsync/EIO failures, there is also something going on with > dm-thin, etc. And then there's generic/484, which is failing for all > ext4 configs. > > According to test script for generic/484, the test failure is fixed by > the patch: > > locks: change POSIX lock ownership on execve when files_struct is displaced > > From what I can tell, the patch was last submitted on March 17th, with > comments from Eric Beiderman on April 2nd and 3rd --- but nothing > since then: > > https://lists.gt.net/linux/kernel/2941364 > > Are you still hoping to land this patch for 4.17? > > Thanks, regards, > > - Ted No, it's not suitable for inclusion. After I sent the above patch, I realized that it would break some important filesystems (like NFS). Eric sounded like he had the best approach to fix it, but it's non- trivial and somewhat outside the bounds of the locking code. I haven't had time to attempt a patch for it so far. -- Jeff Layton