From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from bombadil.infradead.org ([198.137.202.133]:55670 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727781AbeKJApM (ORCPT ); Fri, 9 Nov 2018 19:45:12 -0500 Date: Fri, 9 Nov 2018 07:04:12 -0800 From: Christoph Hellwig Subject: Re: [PATCH] xfs: defer online discard submission to a workqueue Message-ID: <20181109150412.GA9153@infradead.org> References: <20181105181021.8174-1-bfoster@redhat.com> <20181105215139.GA3160@infradead.org> <41f009c2-8f17-ca67-6981-72160bdc94c7@sandeen.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <41f009c2-8f17-ca67-6981-72160bdc94c7@sandeen.net> Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Eric Sandeen Cc: Christoph Hellwig , Brian Foster , linux-xfs@vger.kernel.org On Mon, Nov 05, 2018 at 04:20:24PM -0600, Eric Sandeen wrote: > Is there some downside to Brian's proposal, in principle? > It seems like it would be an improvement for device that might cause > a discard bottleneck like this. We create another bottleneck, we lost thottling, we don't flush the workqueue when needed and probably a few more. Nevermind that I don't see any point in optimizing for out of tree code that no one even bothers to upstream. All this VDO here and there crap from Red Hat in the last year is really more than enough.