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 80D23C433EF for ; Wed, 13 Oct 2021 17:21:20 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 60C1060F21 for ; Wed, 13 Oct 2021 17:21:20 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230232AbhJMRXW (ORCPT ); Wed, 13 Oct 2021 13:23:22 -0400 Received: from mail.itouring.de ([85.10.202.141]:46066 "EHLO mail.itouring.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230225AbhJMRXW (ORCPT ); Wed, 13 Oct 2021 13:23:22 -0400 X-Greylist: delayed 484 seconds by postgrey-1.27 at vger.kernel.org; Wed, 13 Oct 2021 13:23:22 EDT Received: from tux.applied-asynchrony.com (p5ddd741d.dip0.t-ipconnect.de [93.221.116.29]) by mail.itouring.de (Postfix) with ESMTPSA id 14CACD2992A for ; Wed, 13 Oct 2021 19:13:11 +0200 (CEST) Received: from [192.168.100.221] (hho.applied-asynchrony.com [192.168.100.221]) by tux.applied-asynchrony.com (Postfix) with ESMTP id B0993F01604 for ; Wed, 13 Oct 2021 19:13:10 +0200 (CEST) To: linux-xfs From: =?UTF-8?Q?Holger_Hoffst=c3=a4tte?= Subject: Sorting blocks in xfs_buf_delwri_submit_buffers() still necessary? Organization: Applied Asynchrony, Inc. Message-ID: <05c69404-cc05-444b-e4b0-1e358deae272@applied-asynchrony.com> Date: Wed, 13 Oct 2021 19:13:10 +0200 MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-US Content-Transfer-Encoding: 7bit Precedence: bulk List-ID: X-Mailing-List: linux-xfs@vger.kernel.org Hi, Based on what's going on in blk-mq & NVMe land I though I'd check if XFS still sorts buffers before sending them down the pipe, and sure enough that still happens in xfs_buf.c:xfs_buf_delwri_submit_buffers() (the comparson function is directly above). Before I make a fool of myself and try to remove this, do we still think this is necessary? If there's a scheduler it will do the same thing, and SSD/NVMe might do the same in HW anyway or not care. The only scenario I can think of where this might make a difference is rotational RAID without scheduler attached. Not sure. I'm looking forward to hear what a foolish idea this is. cheers Holger