From: Christoph Hellwig <hch@lst.de>
To: John Garry <john.g.garry@oracle.com>
Cc: Christoph Hellwig <hch@lst.de>,
brauner@kernel.org, djwong@kernel.org, cem@kernel.org,
dchinner@redhat.com, linux-xfs@vger.kernel.org,
linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
ojaswin@linux.ibm.com, ritesh.list@gmail.com,
martin.petersen@oracle.com, tytso@mit.edu,
linux-ext4@vger.kernel.org
Subject: Re: [PATCH v6 11/13] xfs: add xfs_file_dio_write_atomic()
Date: Tue, 18 Mar 2025 09:46:41 +0100 [thread overview]
Message-ID: <20250318084641.GC19274@lst.de> (raw)
In-Reply-To: <08992e02-9ff4-416e-bd6c-e3e016356200@oracle.com>
On Tue, Mar 18, 2025 at 08:42:36AM +0000, John Garry wrote:
>> I see that this is what's done in the current series now. But that feels
>> very wrong. Why do you want to deprive the user of this nice and useful
>> code if they don't have the right hardware?
>
> I don't think it's fair to say that we deprive the user - so far we just
> don't and nobody has asked for atomics without HW support.
You're still keeping this nice functionality from the users..
>
>> Why do we limit us to the
>> hardware supported size when we support more in software?
>
> As I see, HW offload gives fast and predictable performance.
>
> The COW method is just a (slow) fallback is when HW offload is not possible.
>
> If we want to allow the user to avail of atomics greater than the mounted
> bdev, then we should have a method to tell the user of the optimised
> threshold. They could read the bdev atomic limits and infer this, but that
> is not a good user experience.
Yes, there should be an interface to expose that. But even without
the hardware acceleration a guaranteed untorn write is a really nice
feature to have.
next prev parent reply other threads:[~2025-03-18 8:46 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-13 17:12 [PATCH v6 00/13] large atomic writes for xfs with CoW John Garry
2025-03-13 17:12 ` [PATCH v6 01/13] iomap: inline iomap_dio_bio_opflags() John Garry
2025-03-16 13:40 ` Ritesh Harjani
2025-03-17 6:07 ` Christoph Hellwig
2025-03-13 17:12 ` [PATCH v6 02/13] iomap: comment on atomic write checks in iomap_dio_bio_iter() John Garry
2025-03-17 6:08 ` Christoph Hellwig
2025-03-17 8:22 ` John Garry
2025-03-17 14:16 ` Ritesh Harjani
2025-03-13 17:13 ` [PATCH v6 03/13] iomap: rework IOMAP atomic flags John Garry
2025-03-17 6:11 ` Christoph Hellwig
2025-03-17 9:05 ` John Garry
2025-03-18 5:32 ` Christoph Hellwig
2025-03-18 8:11 ` John Garry
2025-03-17 13:44 ` Ritesh Harjani
2025-03-17 14:25 ` John Garry
2025-03-13 17:13 ` [PATCH v6 04/13] xfs: pass flags to xfs_reflink_allocate_cow() John Garry
2025-03-17 6:15 ` Christoph Hellwig
2025-03-17 9:17 ` John Garry
2025-03-18 5:33 ` Christoph Hellwig
2025-03-18 8:12 ` John Garry
2025-03-13 17:13 ` [PATCH v6 05/13] xfs: allow block allocator to take an alignment hint John Garry
2025-03-17 6:16 ` Christoph Hellwig
2025-03-13 17:13 ` [PATCH v6 06/13] xfs: switch atomic write size check in xfs_file_write_iter() John Garry
2025-03-17 6:18 ` Christoph Hellwig
2025-03-17 9:17 ` John Garry
2025-03-13 17:13 ` [PATCH v6 07/13] xfs: refactor xfs_reflink_end_cow_extent() John Garry
2025-03-17 6:19 ` Christoph Hellwig
2025-03-13 17:13 ` [PATCH v6 08/13] xfs: reflink CoW-based atomic write support John Garry
2025-03-17 6:20 ` Christoph Hellwig
2025-03-13 17:13 ` [PATCH v6 09/13] xfs: add XFS_REFLINK_ALLOC_EXTSZALIGN John Garry
2025-03-13 18:03 ` Darrick J. Wong
2025-03-17 6:23 ` Christoph Hellwig
2025-03-13 17:13 ` [PATCH v6 10/13] xfs: iomap COW-based atomic write support John Garry
2025-03-16 6:53 ` Ritesh Harjani
2025-03-17 8:54 ` John Garry
2025-03-17 14:20 ` Ritesh Harjani
2025-03-17 14:56 ` John Garry
2025-03-18 5:35 ` Christoph Hellwig
2025-03-17 7:26 ` Christoph Hellwig
2025-03-17 10:18 ` John Garry
2025-03-18 5:39 ` Christoph Hellwig
2025-03-18 8:22 ` John Garry
2025-03-18 8:32 ` Christoph Hellwig
2025-03-18 17:44 ` John Garry
2025-03-19 7:30 ` Christoph Hellwig
2025-03-19 10:24 ` John Garry
2025-03-20 5:29 ` Christoph Hellwig
2025-03-20 9:49 ` John Garry
2025-03-20 14:12 ` Christoph Hellwig
2025-03-13 17:13 ` [PATCH v6 11/13] xfs: add xfs_file_dio_write_atomic() John Garry
2025-03-17 6:41 ` Christoph Hellwig
2025-03-17 9:36 ` John Garry
2025-03-18 5:43 ` Christoph Hellwig
2025-03-18 8:42 ` John Garry
2025-03-18 8:46 ` Christoph Hellwig [this message]
2025-03-18 9:12 ` John Garry
2025-03-13 17:13 ` [PATCH v6 12/13] xfs: commit CoW-based atomic writes atomically John Garry
2025-03-17 6:56 ` Christoph Hellwig
2025-03-17 9:43 ` John Garry
2025-03-13 17:13 ` [PATCH v6 13/13] xfs: update atomic write max size John Garry
2025-03-17 7:25 ` Christoph Hellwig
2025-03-17 9:57 ` John Garry
2025-03-18 5:47 ` Christoph Hellwig
2025-03-18 5:48 ` [PATCH v6 00/13] large atomic writes for xfs with CoW Christoph Hellwig
2025-03-18 8:44 ` John Garry
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=20250318084641.GC19274@lst.de \
--to=hch@lst.de \
--cc=brauner@kernel.org \
--cc=cem@kernel.org \
--cc=dchinner@redhat.com \
--cc=djwong@kernel.org \
--cc=john.g.garry@oracle.com \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-xfs@vger.kernel.org \
--cc=martin.petersen@oracle.com \
--cc=ojaswin@linux.ibm.com \
--cc=ritesh.list@gmail.com \
--cc=tytso@mit.edu \
/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).