From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id EB40DC433F5 for ; Mon, 11 Oct 2021 12:45:11 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id CD06760EFE for ; Mon, 11 Oct 2021 12:45:11 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235062AbhJKMrH (ORCPT ); Mon, 11 Oct 2021 08:47:07 -0400 Received: from outgoing-auth-1.mit.edu ([18.9.28.11]:45138 "EHLO outgoing.mit.edu" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S234921AbhJKMrG (ORCPT ); Mon, 11 Oct 2021 08:47:06 -0400 Received: from cwcc.thunk.org (pool-72-74-133-215.bstnma.fios.verizon.net [72.74.133.215]) (authenticated bits=0) (User authenticated as tytso@ATHENA.MIT.EDU) by outgoing.mit.edu (8.14.7/8.12.4) with ESMTP id 19BCj1nH002647 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Mon, 11 Oct 2021 08:45:01 -0400 Received: by cwcc.thunk.org (Postfix, from userid 15806) id B7FB015C00CA; Mon, 11 Oct 2021 08:44:54 -0400 (EDT) Date: Mon, 11 Oct 2021 08:44:54 -0400 From: "Theodore Ts'o" To: =?iso-8859-1?Q?Lu=EDs?= Henriques Cc: Ming Lei , fstests@vger.kernel.org, fio@vger.kernel.org Subject: Re: generic/095 failing in ext4 and xfs Message-ID: References: <20211010083125.dne6pcjjboixdllo@fedora> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: Precedence: bulk List-ID: X-Mailing-List: fstests@vger.kernel.org On Mon, Oct 11, 2021 at 11:16:09AM +0100, Luís Henriques wrote: > > Ah, I see what you mean. Ok, I've used the patch bellow and the test now > passes. The patch changes the block size only for the DIO jobs, there are > other jobs where bs=1k is still used. Do you think other jobs should be > changed as well? When you try to submit this to fstests, what I'd suggest is to use _min_dio_alignment (defined in common/rc) to set the block size for DIO in the fio jobs. Cheers, - Ted