From: "Theodore Ts'o" <tytso@mit.edu>
To: Eric Biggers <ebiggers@kernel.org>
Cc: youling257 <youling257@gmail.com>,
jack@suse.cz, hch@infradead.org, hch@lst.de,
linux-ext4@vger.kernel.org, ritesh.list@gmail.com,
keescook@chromium.org
Subject: Re: [PATCH v4 12/13] ext4: Stop providing .writepage hook
Date: Tue, 9 May 2023 14:36:49 -0400 [thread overview]
Message-ID: <ZFqSwegsnsqi3vAu@mit.edu> (raw)
In-Reply-To: <20230509050227.GA1180@quark.localdomain>
On Mon, May 08, 2023 at 10:02:27PM -0700, Eric Biggers wrote:
> On Tue, May 09, 2023 at 01:51:08AM +0800, youling257 wrote:
> > I using linux mainline kernel on android. https://github.com/youling257/android-mainline/commits/6.4 https://github.com/youling257/android-mainline/commits/6.3
> > "ext4: Stop providing .writepage hook" cause some android app unable to read storage/emulated/0 files, i need to say android esdfs file system storage/emulated is ext4 data/media bind mount.
> > I want to ask, why android storage/emulated need .writepage hook?
>
> "esdfs" doesn't exist upstream, so linux-ext4 can't provide support for it.
>
> Also, it doesn't exist in the Android Common Kernels either, so the Android team
> cannot help you either.
The problem with esdfs is that it's based on the old stackable file
system paradigm which is filled with races and is inherently
unreliable (just for fun, try running fsstress on the upper and lower
file systems of a stackable file system simultaneously, and watch the
kernel crash and burn). For that reason, some number of us have been
working for a while to eliminate the need for stacking file systems,
such as sdcardfs. esdfs, etc. from the Android kernel.
The other thing I would add is that upstream has been working[1] on
getting rid of writepage function. So out-of-tree file systems are
going to need to adapt --- or die.
[1] https://lore.kernel.org/all/20221202102644.770505-1-hch@lst.de/
It looks like esdfs is coming from the Chromium kernel? The latest
Chromium kernel I can find is 5.15 based, and it has esdfs in it. I'm
sad to see that esdfs hasn't been removed from the Chromium kernel
yet, and replaced with something more stable and reliable, but maybe
we can find someone who is more familiar with the Chromium kernel to
comment.
Cheers,
- Ted
next prev parent reply other threads:[~2023-05-09 18:37 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-07 11:27 [PATCH v4 0/13] ext4: Stop using ext4_writepage() for writeout of ordered data Jan Kara
2022-12-07 11:27 ` [PATCH v4 01/13] ext4: Handle redirtying in ext4_bio_write_page() Jan Kara
2022-12-07 11:27 ` [PATCH v4 02/13] ext4: Move keep_towrite handling to ext4_bio_write_page() Jan Kara
2022-12-07 11:27 ` [PATCH v4 03/13] ext4: Remove nr_submitted from ext4_bio_write_page() Jan Kara
2022-12-07 11:27 ` [PATCH v4 04/13] ext4: Drop pointless IO submission " Jan Kara
2022-12-07 11:27 ` [PATCH v4 05/13] ext4: Add support for writepages calls that cannot map blocks Jan Kara
2022-12-07 11:27 ` [PATCH v4 06/13] ext4: Provide ext4_do_writepages() Jan Kara
2022-12-07 11:27 ` [PATCH v4 07/13] ext4: Move percpu_rwsem protection into ext4_writepages() Jan Kara
2022-12-07 11:27 ` [PATCH v4 08/13] ext4: Switch to using ext4_do_writepages() for ordered data writeout Jan Kara
2022-12-07 11:27 ` [PATCH v4 09/13] jbd2: Switch jbd2_submit_inode_data() to use fs-provided hook for " Jan Kara
2022-12-08 15:28 ` Ritesh Harjani (IBM)
2022-12-07 11:27 ` [PATCH v4 10/13] ext4: Switch to using write_cache_pages() for data=journal writeout Jan Kara
2022-12-08 15:40 ` Ritesh Harjani (IBM)
2022-12-08 16:33 ` Jan Kara
2022-12-07 11:27 ` [PATCH v4 11/13] mm: Export buffer_migrate_folio_norefs() Jan Kara
2022-12-07 11:59 ` Christoph Hellwig
2022-12-07 11:27 ` [PATCH v4 12/13] ext4: Stop providing .writepage hook Jan Kara
2023-05-08 17:51 ` youling257
2023-05-09 0:25 ` Jan Kara
2023-05-09 5:02 ` Eric Biggers
2023-05-09 18:36 ` Theodore Ts'o [this message]
2023-05-10 5:17 ` youling 257
2023-05-10 5:47 ` youling 257
2023-05-10 6:50 ` Eric Biggers
2023-05-10 22:00 ` Theodore Ts'o
2022-12-07 11:27 ` [PATCH v4 13/13] ext4: Remove ordered data support from ext4_writepage() Jan Kara
2022-12-08 14:23 ` Theodore Ts'o
2022-12-08 15:40 ` Jan Kara
2022-12-08 15:57 ` Theodore Ts'o
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ZFqSwegsnsqi3vAu@mit.edu \
--to=tytso@mit.edu \
--cc=ebiggers@kernel.org \
--cc=hch@infradead.org \
--cc=hch@lst.de \
--cc=jack@suse.cz \
--cc=keescook@chromium.org \
--cc=linux-ext4@vger.kernel.org \
--cc=ritesh.list@gmail.com \
--cc=youling257@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).