mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Viacheslav Dubeyko <slava@dubeyko.com>
To: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Cc: Wang Jianjian <wangjianjian0@foxmail.com>,
	linux-nilfs <linux-nilfs@vger.kernel.org>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 0/2] nilfs2: atime support
Date: Tue, 17 Nov 2026 10:20:44 -0800	[thread overview]
Message-ID: <0318886d45a4eb49aaaa2af5227a3d92f558712d.camel@dubeyko.com> (raw)
In-Reply-To: <20261117065122.41553-1-konishi.ryusuke@gmail.com>

On Tue, 2026-11-17 at 15:47 +0900, Ryusuke Konishi wrote:
> Hi Viacheslav,
> 
> Please queue the following two patches for persistent atime support
> in
> nilfs2 for the next merge window.
> 
> This series builds upon your original patch [1], which Wang Jianjian
> followed up on to address review comments and add compat_ro feature
> flag
> support [2][3].  I have made minor modifications to those patches,
> including filling in missing kernel-doc comments and adjusting the
> commit messages (such as fixing the Co-developed-by tags).
> 
> To use this feature, install the updated nilfs-utils from the
> 'atime-support' branch of the development repository [4] (which will
> be
> merged into the master branch once this patchset is queued), and
> either
> format a new filesystem with the atime feature enabled:
> 
>   $ sudo mkfs -t nilfs2 -O atime /dev/<device>
> 
> or enable the feature flag on an existing filesystem image using
> nilfs-tune:
> 
>   $ sudo nilfs-tune -O atime /dev/<device>
> 
> Once the atime feature flag is set, the filesystem can only be
> mounted
> read-only on currently deployed kernels that lack support for it
> (until
> the flag is cleared with "nilfs-tune -O ^atime /dev/<device>").  To
> avoid confusing users, this feature will remain optional in
> mkfs.nilfs2
> for the time being, though I plan to enable it by default in a future
> release of nilfs-utils.
> 
> Note that when enabling the feature on an existing filesystem, inodes
> written by older mkfs or kernels have their on-disk atime fields
> initialized to 0, which will expose the epoch date (1970-01-01).
> nilfs-tune warns about this when setting the flag, and image
> conversion
> (such as copying mtime to atime) is not supported.
> 
> The series consists of two patches:
> 
> - Patch 1/2 introduces the read-only compat feature flag
>   (NILFS_FEATURE_COMPAT_RO_ATIME) to manage on-disk format
> compatibility
>   for atime support.
> - Patch 2/2 reads atime from on-disk inodes when the compat_ro flag
> is
>   enabled and stops unconditionally setting SB_NOATIME on the
> superblock,
>   switching VFS behavior accordingly.
> 
> [1]
> https://lore.kernel.org/all/20260219204343.70117-1-slava@dubeyko.com
> [2]
> https://lore.kernel.org/all/tencent_E3847A19056D4E914FACD269A3C77BAF0905@qq.com
> [3]
> https://lore.kernel.org/all/tencent_F3CAF2061A8AE294D88513588FD3DE1F4806@qq.com
> [4] https://github.com/nilfs-dev/nilfs-utils.git
> 
> Thanks,
> Ryusuke Konishi
> 
> 
> Viacheslav Dubeyko (1):
>   nilfs2: introduce atime support
> 
> Wang Jianjian (1):
>   nilfs2: add atime compat_ro feature
> 
>  fs/nilfs2/Kconfig                  |  2 +-
>  fs/nilfs2/inode.c                  | 11 +++++++++--
>  fs/nilfs2/nilfs.h                  |  4 ----
>  fs/nilfs2/segment.c                |  2 --
>  fs/nilfs2/super.c                  | 11 ++++++-----
>  fs/nilfs2/the_nilfs.h              |  6 ++++++
>  include/uapi/linux/nilfs2_ondisk.h | 12 +++++++-----
>  7 files changed, 29 insertions(+), 19 deletions(-)

Applied.

Thanks,
Slava.

      parent reply	other threads:[~2026-11-17 18:20 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-17  6:47 Ryusuke Konishi
2026-11-17  6:47 ` [PATCH 1/2] nilfs2: add atime compat_ro feature Ryusuke Konishi
2026-11-17  6:47 ` [PATCH 2/2] nilfs2: introduce atime support Ryusuke Konishi
2026-11-17 18:20 ` Viacheslav Dubeyko [this message]

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=0318886d45a4eb49aaaa2af5227a3d92f558712d.camel@dubeyko.com \
    --to=slava@dubeyko.com \
    --cc=konishi.ryusuke@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-nilfs@vger.kernel.org \
    --cc=wangjianjian0@foxmail.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

all inboxes | Powered by JetHome®