From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from userp1040.oracle.com ([156.151.31.81]:51874 "EHLO userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750852AbdIFWcE (ORCPT ); Wed, 6 Sep 2017 18:32:04 -0400 Date: Wed, 6 Sep 2017 15:28:31 -0700 From: Liu Bo To: dsterba@suse.cz, linux-btrfs@vger.kernel.org Subject: Re: [PATCH] Btrfs: use the new helper wbc_to_write_flags Message-ID: <20170906222831.GA16924@lim.localdomain> Reply-To: bo.li.liu@oracle.com References: <20170825001948.7759-1-bo.li.liu@oracle.com> <20170906143806.GQ31874@twin.jikos.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <20170906143806.GQ31874@twin.jikos.cz> Sender: linux-btrfs-owner@vger.kernel.org List-ID: On Wed, Sep 06, 2017 at 04:38:06PM +0200, David Sterba wrote: > On Thu, Aug 24, 2017 at 06:19:48PM -0600, Liu Bo wrote: > > This updates btrfs to use the helper wbc_to_write_flags which has been > > applied in ext4/xfs/f2fs/block. > > Added in commit 7637241e651ec36e4094 in 11/2016, I wonder why btrfs > wasn't been changed as well as it uses the same code patterns as the > other filesystems. > /me is curious, too. > > Please note that, with this, btrfs's dirty pages written by a > > writeback job will carry the flag REQ_BACKGROUND, which is currently > > used by writeback-throttle to determine whether it should go to get a > > request or wait. > > Which in my understanding of the WBT behaviour is what we want, right? > Yes, it gives a hint to WBT whether it should wait or not. > > Signed-off-by: Liu Bo > > Reviewed-by: David Sterba > > I'm still expecting some surprises (== performance drop) from the recent > WBT changes, as we still may miss some request flags, but more subtle > than eg. the superblock REQ_ tags. Looks like this is the only missing flag I could find, the missing REQ_META (in another patch) doesn't have an impact on WBT. thanks, -liubo