mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Jiachen Zhang <zhangjiachen.jaycee@bytedance.com>
To: Bernd Schubert <bernd.schubert@fastmail.fm>,
	Miklos Szeredi <miklos@szeredi.hu>
Cc: Jonathan Corbet <corbet@lwn.net>,
	linux-fsdevel@vger.kernel.org, linux-doc@vger.kernel.org,
	linux-kernel@vger.kernel.org, me@jcix.top
Subject: Re: [PATCH 4/5] fuse: writeback_cache consistency enhancement (writeback_cache_v2)
Date: Wed, 23 Aug 2023 18:59:49 +0800	[thread overview]
Message-ID: <029cb695-9b8e-8fb3-ef0f-b223f34e7639@bytedance.com> (raw)
In-Reply-To: <699673a6-ff82-8968-6310-9a0b1c429be3@fastmail.fm>

On 2023/8/23 18:35, Bernd Schubert wrote:
> On 8/23/23 11:07, Miklos Szeredi wrote:
>> On Tue, 11 Jul 2023 at 06:36, Jiachen Zhang
>> <zhangjiachen.jaycee@bytedance.com> wrote:
>>>
>>> Some users may want both the high performance of the writeback_cahe mode
>>> and a little bit more consistency among FUSE mounts. Current
>>> writeback_cache mode never updates attributes from server, so can never
>>> see the file attributes changed by other FUSE mounts, which means
>>> 'zero-consisteny'.
>>>
>>> This commit introduces writeback_cache_v2 mode, which allows the 
>>> attributes
>>> to be updated from server to kernel when the inode is clean and no
>>> writeback is in-progressing. FUSE daemons can select this mode by the
>>> FUSE_WRITEBACK_CACHE_V2 init flag.
>>>
>>> In writeback_cache_v2 mode, the server generates official attributes.
>>> Therefore,
>>>
>>>      1. For the cmtime, the cmtime generated by kernel are just 
>>> temporary
>>>      values that are never flushed to server by fuse_write_inode(), 
>>> and they
>>>      could be eventually updated by the official server cmtime. The
>>>      mtime-based revalidation of the fc->auto_inval_data mode is also
>>>      skipped, as the kernel-generated temporary cmtime are likely not 
>>> equal
>>>      to the offical server cmtime.
>>>
>>>      2. For the file size, we expect server updates its file size on
>>>      FUSE_WRITEs. So we increase fi->attr_version in 
>>> fuse_writepage_end() to
>>>      check the staleness of the returning file size.
>>>
>>> Together with FOPEN_INVAL_ATTR, a FUSE daemon is able to implement
>>> close-to-open (CTO) consistency like NFS client implementations.
>>
>> What I'd prefer is mode similar to NFS: getattr flushes pending writes
>> so that server ctime/mtime are always in sync with client.  FUSE
>> probably should have done that from the beginning, but at that time I
>> wasn't aware of the NFS solution.
> 
> 
> I think it would be good to have flush-on-getattr configurable - systems 
> with a distributed lock manager (DLM) and notifications from 
> server/daemon to kernel should not need it.
> 
> 
> Thanks,
> Bernd

Hi Miklos and Bernd,

I agree that flush-on-getattr is a good solution to keep the c/mtime
consistency for the view of userspace applications.

Maybe in the next version, we can add the flush-on-getattr just for the
writeback_cache_v2 mode, as daemons replying on reverse notifications
are likely not need the writeback_cache_v2 mode. What do you think?

Thanks,
Jiachen



  reply	other threads:[~2023-08-23 10:59 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-07-11  4:34 [PATCH 0/5] FUSE consistency improvements Jiachen Zhang
2023-07-11  4:34 ` [PATCH 1/5] fuse: check attributes staleness on fuse_iget() Jiachen Zhang
2023-08-23  8:57   ` Miklos Szeredi
2023-07-11  4:34 ` [PATCH 2/5] fuse: invalidate dentry on EEXIST creates or ENOENT deletes Jiachen Zhang
2023-08-16 12:27   ` Miklos Szeredi
2023-07-11  4:34 ` [PATCH 3/5] fuse: add FOPEN_INVAL_ATTR Jiachen Zhang
2023-08-23  9:01   ` Miklos Szeredi
2023-08-23 11:12     ` Jiachen Zhang
2023-07-11  4:34 ` [PATCH 4/5] fuse: writeback_cache consistency enhancement (writeback_cache_v2) Jiachen Zhang
2023-08-23  9:07   ` Miklos Szeredi
2023-08-23 10:35     ` Bernd Schubert
2023-08-23 10:59       ` Jiachen Zhang [this message]
2023-08-23 22:55         ` Bernd Schubert
2023-07-11  4:34 ` [PATCH 5/5] docs: fuse: improve FUSE consistency explanation Jiachen Zhang
2023-07-11  4:42   ` Randy Dunlap
2023-07-11  7:15     ` Jiachen Zhang
2023-07-14  5:50 ` [PATCH 0/5] FUSE consistency improvements Jingbo Xu

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=029cb695-9b8e-8fb3-ef0f-b223f34e7639@bytedance.com \
    --to=zhangjiachen.jaycee@bytedance.com \
    --cc=bernd.schubert@fastmail.fm \
    --cc=corbet@lwn.net \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=me@jcix.top \
    --cc=miklos@szeredi.hu \
    /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®