public inbox for [email protected]
 help / color / mirror / Atom feed
From: Bernd Schubert <[email protected]>
To: Miklos Szeredi <[email protected]>
Cc: "[email protected]" <[email protected]>,
	"[email protected]" <[email protected]>,
	"[email protected]" <[email protected]>,
	"[email protected]" <[email protected]>,
	"[email protected]" <[email protected]>,
	Dharmendra Singh <[email protected]>
Subject: Re: [PATCH 1/2] fs: add FMODE_DIO_PARALLEL_WRITE flag
Date: Thu, 13 Apr 2023 09:25:54 +0000	[thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <CAJfpegt_ZCVodOhQCzF9OqKnCr65mKax0Gu4OTN8M51zP+8TcA@mail.gmail.com>

On 4/13/23 09:40, Miklos Szeredi wrote:
> On Wed, 12 Apr 2023 at 15:42, Bernd Schubert <[email protected]> wrote:
>>
>> Miklos, Jens,
>>
>> could we please also set this flag for fuse?
>>
>>
>> Thanks,
>> Bernd
>>
>>
>> fuse: Set FMODE_DIO_PARALLEL_WRITE flag
>>
>> From: Bernd Schubert <[email protected]>
>>
>> Fuse can also do parallel DIO writes, if userspace has enabled it.
>>
>> Signed-off-by: Bernd Schubert <[email protected]>
>> ---
>>   fs/fuse/file.c |    3 +++
>>   1 file changed, 3 insertions(+)
>>
>> diff --git a/fs/fuse/file.c b/fs/fuse/file.c
>> index 875314ee6f59..46e7f1196fd1 100644
>> --- a/fs/fuse/file.c
>> +++ b/fs/fuse/file.c
>> @@ -215,6 +215,9 @@ void fuse_finish_open(struct inode *inode, struct file *file)
>>          }
>>          if ((file->f_mode & FMODE_WRITE) && fc->writeback_cache)
>>                  fuse_link_write_file(file);
>> +
>> +       if (ff->open_flags & FOPEN_PARALLEL_DIRECT_WRITES)
>> +               file->f_mode |= FMODE_DIO_PARALLEL_WRITE;
> 
> fuse_direct_write_iter():
> 
> bool exclusive_lock =
>      !(ff->open_flags & FOPEN_PARALLEL_DIRECT_WRITES) ||
>      iocb->ki_flags & IOCB_APPEND ||
>      fuse_direct_write_extending_i_size(iocb, from);
> 
> If the write is size extending, then it will take the lock exclusive.
> OTOH, I guess that it would be unusual for lots of  size extending
> writes to be done in parallel.
> 
> What would be the effect of giving the  FMODE_DIO_PARALLEL_WRITE hint
> and then still serializing the writes?

It used here

https://lore.kernel.org/io-uring/20230403-wound-roundworm-c1660e059b8c@brauner/T/#m5f86985d6c67dd1d01a977475dab542c338372dd


fuse_finish_open has its own lock, so letting uring handle requests in 
parallel should not hurt? Is this going like

application -> uring does parallel requests -> fuse.ko -> fuse-deamon

So when fuse-deamon signals that it can handle parallel DIO, it is just 
fuse.ko that might need its own lock to extend the file?


Thanks,
Bernd

  reply	other threads:[~2023-04-13  9:26 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-07 17:20 [PATCHSET for-next 0/2] Flag file systems as supporting parallel dio writes Jens Axboe
2023-03-07 17:20 ` [PATCH 1/2] fs: add FMODE_DIO_PARALLEL_WRITE flag Jens Axboe
2023-04-12 13:40   ` Bernd Schubert
2023-04-12 13:43     ` Bernd Schubert
2023-04-13  7:40     ` Miklos Szeredi
2023-04-13  9:25       ` Bernd Schubert [this message]
2023-04-14  5:11       ` Christoph Hellwig
2023-04-14 15:36         ` Darrick J. Wong
2023-04-15 13:15           ` Jens Axboe
2023-04-18 12:42             ` Miklos Szeredi
2023-04-18 12:55               ` Bernd Schubert
2023-04-18 22:13                 ` Dave Chinner
2023-04-19  1:28                   ` Jens Axboe
2023-04-16  5:54           ` Christoph Hellwig
2023-04-19  1:29             ` Jens Axboe
2023-03-07 17:20 ` [PATCH 2/2] io_uring: avoid hashing O_DIRECT writes if the filesystem doesn't need it Jens Axboe
2023-03-15 17:40 ` [PATCHSET for-next 0/2] Flag file systems as supporting parallel dio writes Jens Axboe
2023-03-16  4:29   ` Darrick J. Wong
2023-03-17  2:53     ` Jens Axboe
2023-04-03 12:24 ` Christian Brauner

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] \
    /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