Re: The state of ext4 tree merging (was Re: Linux 6.3-rc1)

From: Jan Kara <jack@suse.cz>
To: Bagas Sanjaya <bagasdotme@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Theodore Ts'o <tytso@mit.edu>,
	Andrew Morton <akpm@linux-foundation.org>,
	Matthew Wilcox <willy@infradead.org>, Jan Kara <jack@suse.cz>,
	Linux Next Mailing List <linux-next@vger.kernel.org>,
	ext4 Development <linux-ext4@vger.kernel.org>
Subject: Re: The state of ext4 tree merging (was Re: Linux 6.3-rc1)
Date: Mon, 6 Mar 2023 13:41:34 +0100	[thread overview]
Message-ID: <20230306124134.hmeuvjhihs4ubpmz@quack3> (raw)
In-Reply-To: <ZAVbZJSyOdF0BxAJ@debian.me>

On Mon 06-03-23 10:17:56, Bagas Sanjaya wrote:
> On Sun, Mar 05, 2023 at 03:24:41PM -0800, Linus Torvalds wrote:
> > In fact, it was quite nice in a couple of ways: not only didn't I have
> > a hugely compressed merge window where I felt I had to cram as much as
> > possible into the first few days, but the fact that we _have_ had a
> > couple of merge windows where I really asked for people to have
> > everything ready when the merge window opened seems to have set a
> > pattern: the bulk of everything really did come in early.
> > 
> 
> Not so for me watching updates to ext4 merging hell...
> 
> In this merge window, Ted only submitted the first part of ext4 updates
> [1] as noted in the resolution message [2]. The second part didn't make
> through the merge window (PR not sent). As such, the data=writepage
> cleanups have to wait for 6.4 merge window, and it is IMO inconvenient
> for linux-next to contain ext4 tree from next-20230217 for about
> seven weeks, as any enhancements and fixes applied to the tree are
> holding back from testing in linux-next until this hell can be sorted
> out.
> 
> In the long term, I'd like to see a co-maintainer step in to help
> maintaining the tree in case Ted is busy. Of couse I'm not eligible
> for that role (I played as documentation janitor instead), but
> any developer with deep knowledge and experience for the fs and its
> internals should fit the role.

To be fair, the data=journal cleanups got held back only partially due to
the merge issues. Another problem is that they somehow make problems with
filesystem freezing in data=journal mode more frequent and we wanted to
understand (and hopefully fix) that. Of course if Ted could look into this
earlier or I could earlier debug these issues, we could have merged the
cleanups but that's always the case that you have to prioritize and these
cleanups don't have that high priority...

								Honza
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR