From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ipmail05.adl6.internode.on.net ([150.101.137.143]:58610 "EHLO ipmail05.adl6.internode.on.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752894AbcK0XQx (ORCPT ); Sun, 27 Nov 2016 18:16:53 -0500 Date: Mon, 28 Nov 2016 10:16:39 +1100 From: Dave Chinner To: Christoph Hellwig Cc: Jens Axboe , linux-xfs@vger.kernel.org, linux-fsdevel@vger.kernel.org Subject: Re: an iomap-based direct I/O implementation V3 Message-ID: <20161127231639.GW28177@dastard> References: <1479064054-10474-1-git-send-email-hch@lst.de> <20161122225243.GN31101@dastard> <20161123000204.GO31101@dastard> <20161123083657.GA16966@lst.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20161123083657.GA16966@lst.de> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Wed, Nov 23, 2016 at 09:36:57AM +0100, Christoph Hellwig wrote: > On Tue, Nov 22, 2016 at 05:40:30PM -0700, Jens Axboe wrote: > > That's going to be somewhat harder... The REQ_IDLE used to be > > REQ_NOIDLE. bio_iov_iter_get_pages() is a single commit, so that's easy > > enough. The poll bits are 3-4 patches, so not too bad either. > > I think we'll simply need a tree that has latest master merged into > the block tree as a base. I'd prefer not to pull and entire tree into the xfs for-next branch if I can avoid it. > The only other option would be to cherry-pick > bio_iov_iter_get_pages from the block tree into the XFS tree, and I'll > resend the series so that it doesn't require the other block changes, > those are fairly mechanical merges that could be carried in linux-next. I think this is probably a better way of proceeding. Can you resend the series and in the description indicate what commit I need to cherrypick from the the block tree? Cheers, Dave. -- Dave Chinner david@fromorbit.com