From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from userp2130.oracle.com ([156.151.31.86]:57574 "EHLO userp2130.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726402AbfFYDZi (ORCPT ); Mon, 24 Jun 2019 23:25:38 -0400 Date: Mon, 24 Jun 2019 20:25:27 -0700 From: "Darrick J. Wong" Subject: Re: xfs cgroup writeback support Message-ID: <20190625032527.GF1611011@magnolia> References: <20190624134315.21307-1-hch@lst.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20190624134315.21307-1-hch@lst.de> Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Christoph Hellwig Cc: linux-xfs@vger.kernel.org, Stefan Priebe - Profihost AG On Mon, Jun 24, 2019 at 03:43:13PM +0200, Christoph Hellwig wrote: > Hi all, > > this small series adds cgroup writeback support to XFS. Unlike > the previous iteration of the support a few years ago it also > ensures that that trailing bios in an ioend inherit the right > cgroup. It has been tested with the shared/011 xfstests test > that was written to test this functionality in all file systems, > and manually by Stefan Priebe. > > This work was funded by Profihost AG. > > Note that the first patch was also in my series to move the xfs > writepage code to iomap.c and the second one will conflict with > it. We'll need to sort out which series to merge first, but given > how simple this one I would suggest to go for this one. By the way, did all the things Dave complained about in last year's attempt[1] to add cgroup writeback support get fixed? IIRC someone whose name I didn't recognise complained about log starvation due to REQ_META bios being charged to the wrong cgroup and other misbehavior. Also, I remember that in the earlier 2017 discussion[2] we talked about a fstest to test that writeback throttling actually capped bandwidth usage correctly. I haven't been following cgroupwb development since 2017 -- does it not ratelimit bandwidth now, or is there some test for that? The only test I could find was shared/011 which only tests the accounting, not bandwidth. --D [1] https://patchwork.kernel.org/comment/21658249/ [2] https://patchwork.kernel.org/comment/21042703/