public inbox for [email protected]
 help / color / mirror / Atom feed
From: Kent Overstreet <[email protected]>
To: Matthew Wilcox <[email protected]>
Cc: John Garry <[email protected]>,
	[email protected],  [email protected], [email protected],
	[email protected], [email protected],
	 [email protected], [email protected],
	[email protected], [email protected],
	 [email protected], [email protected], [email protected],
	 [email protected], [email protected],
	[email protected],  [email protected],
	[email protected], [email protected],
	 [email protected], [email protected],
	[email protected],  [email protected],
	[email protected], [email protected]
Subject: Re: [PATCH v6 00/10] block atomic writes
Date: Fri, 5 Apr 2024 02:14:03 -0400	[thread overview]
Message-ID: <zn6kcqq24f7ltg5fpbsnuxyr7qqzqmydz4mnmaqlmsntq4pekz@h3jkkp5xps73> (raw)
In-Reply-To: <[email protected]>

On Wed, Mar 27, 2024 at 03:50:07AM +0000, Matthew Wilcox wrote:
> On Tue, Mar 26, 2024 at 01:38:03PM +0000, John Garry wrote:
> > The goal here is to provide an interface that allows applications use
> > application-specific block sizes larger than logical block size
> > reported by the storage device or larger than filesystem block size as
> > reported by stat().
> > 
> > With this new interface, application blocks will never be torn or
> > fractured when written. For a power fail, for each individual application
> > block, all or none of the data to be written. A racing atomic write and
> > read will mean that the read sees all the old data or all the new data,
> > but never a mix of old and new.
> > 
> > Three new fields are added to struct statx - atomic_write_unit_min,
> > atomic_write_unit_max, and atomic_write_segments_max. For each atomic
> > individual write, the total length of a write must be a between
> > atomic_write_unit_min and atomic_write_unit_max, inclusive, and a
> > power-of-2. The write must also be at a natural offset in the file
> > wrt the write length. For pwritev2, iovcnt is limited by
> > atomic_write_segments_max.
> > 
> > There has been some discussion on supporting buffered IO and whether the
> > API is suitable, like:
> > https://lore.kernel.org/linux-nvme/[email protected]/
> > 
> > Specifically the concern is that supporting a range of sizes of atomic IO
> > in the pagecache is complex to support. For this, my idea is that FSes can
> > fix atomic_write_unit_min and atomic_write_unit_max at the same size, the
> > extent alignment size, which should be easier to support. We may need to
> > implement O_ATOMIC to avoid mixing atomic and non-atomic IOs for this. I
> > have no proposed solution for atomic write buffered IO for bdev file
> > operations, but I know of no requirement for this.
> 
> The thing is that there's no requirement for an interface as complex as
> the one you're proposing here.  I've talked to a few database people
> and all they want is to increase the untorn write boundary from "one
> disc block" to one database block, typically 8kB or 16kB.
> 
> So they would be quite happy with a much simpler interface where they
> set the inode block size at inode creation time, and then all writes to
> that inode were guaranteed to be untorn.  This would also be simpler to
> implement for buffered writes.
> 
> Who's asking for this more complex interface?

I get the impression the atomic writes stuff has suffered from too
/much/ review and too many maintainers asking for and demanding all
their different must-haves.

      parent reply	other threads:[~2024-04-05  6:14 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-03-26 13:38 [PATCH v6 00/10] block atomic writes John Garry
2024-03-26 13:38 ` [PATCH v6 01/10] block: Pass blk_queue_get_max_sectors() a request pointer John Garry
2024-04-10 22:58   ` Luis Chamberlain
2024-03-26 13:38 ` [PATCH v6 02/10] block: Call blkdev_dio_unaligned() from blkdev_direct_IO() John Garry
2024-04-10 22:53   ` Luis Chamberlain
2024-04-11  8:06     ` John Garry
2024-03-26 13:38 ` [PATCH v6 03/10] fs: Initial atomic write support John Garry
2024-03-26 13:38 ` [PATCH v6 04/10] fs: Add initial atomic write support info to statx John Garry
2024-03-26 13:38 ` [PATCH v6 05/10] block: Add core atomic write support John Garry
2024-03-26 17:11   ` Randy Dunlap
2024-04-10 23:34     ` Luis Chamberlain
2024-04-11  8:15       ` John Garry
2024-03-26 13:38 ` [PATCH v6 06/10] block: Add atomic write support for statx John Garry
2024-03-26 13:38 ` [PATCH v6 07/10] block: Add fops atomic write support John Garry
2024-03-26 13:38 ` [PATCH v6 08/10] scsi: sd: Atomic " John Garry
2024-03-26 13:38 ` [PATCH v6 09/10] scsi: scsi_debug: " John Garry
2024-03-26 13:38 ` [PATCH v6 10/10] nvme: " John Garry
2024-04-11  0:29   ` Luis Chamberlain
2024-04-11  8:59     ` John Garry
2024-04-11 16:22       ` Luis Chamberlain
2024-04-11 23:32         ` Dan Helmick
2024-03-27  3:50 ` [PATCH v6 00/10] block atomic writes Matthew Wilcox
2024-03-27 13:37   ` John Garry
2024-04-04 16:48     ` Matthew Wilcox
2024-04-05 10:06       ` John Garry
2024-04-08 17:50         ` Luis Chamberlain
2024-04-10  4:05           ` Matthew Wilcox
2024-04-10  6:20             ` Hannes Reinecke
2024-04-11  0:38               ` Luis Chamberlain
2024-04-14 20:50             ` Luis Chamberlain
2024-04-15 21:18               ` Matthew Wilcox
2024-04-16 21:11                 ` Luis Chamberlain
2024-04-10  8:34           ` John Garry
2024-04-11 19:07             ` Luis Chamberlain
2024-04-12  8:15               ` John Garry
2024-04-12 18:28                 ` Luis Chamberlain
2024-03-27 20:31   ` Dave Chinner
2024-04-05 10:20     ` Kent Overstreet
2024-04-05 10:55       ` John Garry
2024-04-05  6:14   ` Kent Overstreet [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=zn6kcqq24f7ltg5fpbsnuxyr7qqzqmydz4mnmaqlmsntq4pekz@h3jkkp5xps73 \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [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