public inbox for [email protected]
 help / color / mirror / Atom feed
From: Jan Kara <[email protected]>
To: Stefan Roesch <[email protected]>
Cc: [email protected], [email protected], [email protected],
	[email protected], [email protected],
	[email protected], [email protected], [email protected],
	[email protected]
Subject: Re: [PATCH v7 11/15] fs: Optimization for concurrent file time updates.
Date: Thu, 2 Jun 2022 10:59:58 +0200	[thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>

On Wed 01-06-22 14:01:37, Stefan Roesch wrote:
> This introduces the S_PENDING_TIME flag. If an async buffered write
> needs to update the time, it cannot be processed in the fast path of
> io-uring. When a time update is pending this flag is set for async
> buffered writes. Other concurrent async buffered writes for the same
> file do not need to wait while this time update is pending.
> 
> This reduces the number of async buffered writes that need to get punted
> to the io-workers in io-uring.
> 
> Signed-off-by: Stefan Roesch <[email protected]>

Thinking about this, there is a snag with this S_PENDING_TIME scheme. It
can happen that we report write as completed to userspace before timestamps
are actually updated. So following stat(2) can still return old time stamp
which might confuse some userspace application. It might be even nastier
with i_version which is used by NFS and can thus cause data consistency
issues for NFS.

								Honza

> ---
>  fs/inode.c         | 11 +++++++++--
>  include/linux/fs.h |  3 +++
>  2 files changed, 12 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/inode.c b/fs/inode.c
> index 4503bed063e7..7185d860d423 100644
> --- a/fs/inode.c
> +++ b/fs/inode.c
> @@ -2150,10 +2150,17 @@ static int file_modified_flags(struct file *file, int flags)
>  	ret = inode_needs_update_time(inode, &now);
>  	if (ret <= 0)
>  		return ret;
> -	if (flags & IOCB_NOWAIT)
> +	if (flags & IOCB_NOWAIT) {
> +		if (IS_PENDING_TIME(inode))
> +			return 0;
> +
> +		inode_set_flags(inode, S_PENDING_TIME, S_PENDING_TIME);
>  		return -EAGAIN;
> +	}
>  
> -	return __file_update_time(file, &now, ret);
> +	ret = __file_update_time(file, &now, ret);
> +	inode_set_flags(inode, 0, S_PENDING_TIME);
> +	return ret;
>  }
>  
>  /**
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 553e57ec3efa..15f9a7beba55 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -2151,6 +2151,8 @@ struct super_operations {
>  #define S_CASEFOLD	(1 << 15) /* Casefolded file */
>  #define S_VERITY	(1 << 16) /* Verity file (using fs/verity/) */
>  #define S_KERNEL_FILE	(1 << 17) /* File is in use by the kernel (eg. fs/cachefiles) */
> +#define S_PENDING_TIME (1 << 18) /* File update time is pending */
> +
>  
>  /*
>   * Note that nosuid etc flags are inode-specific: setting some file-system
> @@ -2193,6 +2195,7 @@ static inline bool sb_rdonly(const struct super_block *sb) { return sb->s_flags
>  #define IS_ENCRYPTED(inode)	((inode)->i_flags & S_ENCRYPTED)
>  #define IS_CASEFOLDED(inode)	((inode)->i_flags & S_CASEFOLD)
>  #define IS_VERITY(inode)	((inode)->i_flags & S_VERITY)
> +#define IS_PENDING_TIME(inode) ((inode)->i_flags & S_PENDING_TIME)
>  
>  #define IS_WHITEOUT(inode)	(S_ISCHR(inode->i_mode) && \
>  				 (inode)->i_rdev == WHITEOUT_DEV)
> -- 
> 2.30.2
> 
-- 
Jan Kara <[email protected]>
SUSE Labs, CR

  reply	other threads:[~2022-06-02  9:00 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-01 21:01 [PATCH v7 00/15] io-uring/xfs: support async buffered writes Stefan Roesch
2022-06-01 21:01 ` [PATCH v7 01/15] mm: Move starting of background writeback into the main balancing loop Stefan Roesch
2022-06-01 21:01 ` [PATCH v7 02/15] mm: Move updates of dirty_exceeded into one place Stefan Roesch
2022-06-01 21:01 ` [PATCH v7 03/15] mm: Add balance_dirty_pages_ratelimited_flags() function Stefan Roesch
2022-06-01 21:01 ` [PATCH v7 04/15] iomap: Add flags parameter to iomap_page_create() Stefan Roesch
2022-06-02 16:26   ` Darrick J. Wong
2022-06-01 21:01 ` [PATCH v7 05/15] iomap: Add async buffered write support Stefan Roesch
2022-06-01 21:01 ` [PATCH v7 06/15] iomap: Return error code from iomap_write_iter() Stefan Roesch
2022-06-02 12:38   ` Matthew Wilcox
2022-06-02 17:08     ` Stefan Roesch
2022-06-06 16:39     ` Stefan Roesch
2022-06-06 19:18       ` Matthew Wilcox
2022-06-06 19:21         ` Stefan Roesch
2022-06-06 19:25           ` Matthew Wilcox
2022-06-06 19:28             ` Stefan Roesch
2022-06-06 20:30               ` Matthew Wilcox
2022-06-06 20:34                 ` Stefan Roesch
2022-06-01 21:01 ` [PATCH v7 07/15] fs: Add check for async buffered writes to generic_write_checks Stefan Roesch
2022-06-01 21:01 ` [PATCH v7 08/15] fs: add __remove_file_privs() with flags parameter Stefan Roesch
2022-06-02  9:04   ` Jan Kara
2022-06-01 21:01 ` [PATCH v7 09/15] fs: Split off inode_needs_update_time and __file_update_time Stefan Roesch
2022-06-02  8:44   ` Jan Kara
2022-06-02 12:57   ` Matthew Wilcox
2022-06-02 21:04     ` Stefan Roesch
2022-06-01 21:01 ` [PATCH v7 10/15] fs: Add async write file modification handling Stefan Roesch
2022-06-02  8:44   ` Jan Kara
2022-06-02  9:06   ` Jan Kara
2022-06-02 21:00     ` Stefan Roesch
2022-06-03 10:12       ` Jan Kara
2022-06-06 16:35         ` Stefan Roesch
2022-06-01 21:01 ` [PATCH v7 11/15] fs: Optimization for concurrent file time updates Stefan Roesch
2022-06-02  8:59   ` Jan Kara [this message]
2022-06-02 21:06     ` Stefan Roesch
2022-06-01 21:01 ` [PATCH v7 12/15] io_uring: Add support for async buffered writes Stefan Roesch
2022-06-01 21:01 ` [PATCH v7 13/15] io_uring: Add tracepoint for short writes Stefan Roesch
2022-06-01 21:01 ` [PATCH v7 14/15] xfs: Specify lockmode when calling xfs_ilock_for_iomap() Stefan Roesch
2022-06-02 16:25   ` Darrick J. Wong
2022-06-01 21:01 ` [PATCH v7 15/15] xfs: Add async buffered write support Stefan Roesch
2022-07-01  4:39   ` Al Viro
2022-07-01 14:19     ` Jens Axboe
2022-07-01 14:30       ` Jens Axboe
2022-07-01 14:38         ` Jens Axboe
2022-07-01 18:05           ` Darrick J. Wong
2022-07-01 18:14             ` Jens Axboe
2022-07-05 13:47               ` Josef Bacik
2022-07-05 14:23                 ` Goldwyn Rodrigues
2022-07-05 16:11                 ` Christoph Hellwig
2022-06-02  8:09 ` [PATCH v7 00/15] io-uring/xfs: support async buffered writes Jens Axboe
2022-06-03  2:43   ` Dave Chinner
2022-06-03 13:04     ` Jens Axboe
2022-06-07 16:41       ` Stefan Roesch

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 \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    /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