From: Christoph Hellwig Subject: Re: [PATCH 06/14] block: lift setting the readahead size into the block layer Date: Tue, 15 Sep 2020 09:05:22 +0200 Message-ID: <20200915070522.GA19974@lst.de> References: <20200726150333.305527-1-hch@lst.de> <20200726150333.305527-7-hch@lst.de> <20200826220737.GA25613@redhat.com> <20200902151144.GA1738@lst.de> <20200902162007.GB5513@redhat.com> <20200910092813.GA27229@lst.de> <20200910171541.GB21919@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20200910171541.GB21919@redhat.com> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: dm-devel-bounces@redhat.com Errors-To: dm-devel-bounces@redhat.com Content-Disposition: inline To: Mike Snitzer Cc: Jens Axboe , linux-raid@vger.kernel.org, martin.petersen@oracle.com, Hans de Goede , Richard Weinberger , Minchan Kim , drbd-dev@tron.linbit.com, linux-kernel@vger.kernel.org, linux-block@vger.kernel.org, Song Liu , dm-devel@redhat.com, linux-mtd@lists.infradead.org, linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, cgroups@vger.kernel.org, Christoph Hellwig On Thu, Sep 10, 2020 at 01:15:41PM -0400, Mike Snitzer wrote: > > I'll move it to blk_register_queue, which should work just fine. > > That'll work for initial DM table load as part of DM device creation > (dm_setup_md_queue). But it won't account for DM table reloads that > might change underlying devices on a live DM device (done using > __bind). > > Both dm_setup_md_queue() and __bind() call dm_table_set_restrictions() > to set/update queue_limits. It feels like __bind() will need to call a > new block helper to set/update parts of queue_limits (e.g. ra_pages and > io_pages). > > Any chance you're open to factoring out that block function as an > exported symbol for use by blk_register_queue() and code like DM's > __bind()? I agree with the problem statement. OTOH adding an exported helper for two trivial assignments seems a little silly.. For now I'll just keep the open coded ->io_pages assignment in dm. Note that dm doesn't currently update the ->ra_pages value based on the underlying devices, so an incremental patch to do that might be useful as well.