linux-arch.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jonathan Corbet <corbet@lwn.net>
To: Jens Axboe <axboe@kernel.dk>
Cc: linux-fsdevel@vger.kernel.org, linux-aio@kvack.org,
	linux-block@vger.kernel.org, linux-arch@vger.kernel.org,
	hch@lst.de, jmoyer@redhat.com, avi@scylladb.com
Subject: Re: [PATCH 05/16] Add io_uring IO interface
Date: Tue, 15 Jan 2019 09:51:34 -0700	[thread overview]
Message-ID: <20190115095134.6286b7d6@lwn.net> (raw)
In-Reply-To: <20190115025531.13985-6-axboe@kernel.dk>

On Mon, 14 Jan 2019 19:55:20 -0700
Jens Axboe <axboe@kernel.dk> wrote:

So the [0/16] cover letter seems to have gone astray this time?

Anyway, a couple of minor notes/questions:

> The submission queue (SQ) and completion queue (CQ) rings are shared
> between the application and the kernel. This eliminates the need to
> copy data back and forth to submit and complete IO.
> 
> IO submissions use the io_uring_sqe data structure, and completions
> are generated in the form of io_uring_sqe data structures. The SQ
> ring is an index into the io_uring_sqe array, which makes it possible
> to submit a batch of IOs without them being contiguous in the ring.
> The CQ ring is always contiguous, as completion events are inherently
> unordered and can point to any io_uring_iocb.
> 
> Two new system calls are added for this:
> 
> io_uring_setup(entries, iovecs, params)
> 	Sets up a context for doing async IO. On success, returns a file
> 	descriptor that the application can mmap to gain access to the
> 	SQ ring, CQ ring, and io_uring_iocbs.

Looking at the code, it would appear that the "iovecs" parameter doesn't
actually exist.

> io_uring_enter(fd, to_submit, min_complete, flags)
> 	Initiates IO against the rings mapped to this fd, or waits for
> 	them to complete, or both The behavior is controlled by the
> 	parameters passed in. If 'min_complete' is non-zero, then we'll
> 	try and submit new IO. If IORING_ENTER_GETEVENTS is set, the
> 	kernel will wait for 'min_complete' events, if they aren't
> 	already available.

I feel like I'm missing something here.  Rather than have the
IORING_ENTER_GETEVENTS flag, why not just wait if min_complete > 0 ?

Thanks,

jon

--
To unsubscribe, send a message with 'unsubscribe linux-aio' in
the body to majordomo@kvack.org.  For more info on Linux AIO,
see: http://www.kvack.org/aio/
Don't email: <a href=mailto:"aart@kvack.org">aart@kvack.org</a>

WARNING: multiple messages have this Message-ID (diff)
From: Jonathan Corbet <corbet@lwn.net>
To: Jens Axboe <axboe@kernel.dk>
Cc: linux-fsdevel@vger.kernel.org, linux-aio@kvack.org,
	linux-block@vger.kernel.org, linux-arch@vger.kernel.org,
	hch@lst.de, jmoyer@redhat.com, avi@scylladb.com
Subject: Re: [PATCH 05/16] Add io_uring IO interface
Date: Tue, 15 Jan 2019 09:51:34 -0700	[thread overview]
Message-ID: <20190115095134.6286b7d6@lwn.net> (raw)
Message-ID: <20190115165134.ztvfV7uhQPlE2vPkXm8rJ3OPf5K46GRUgcHkhmZ1hqk@z> (raw)
In-Reply-To: <20190115025531.13985-6-axboe@kernel.dk>

On Mon, 14 Jan 2019 19:55:20 -0700
Jens Axboe <axboe@kernel.dk> wrote:

So the [0/16] cover letter seems to have gone astray this time?

Anyway, a couple of minor notes/questions:

> The submission queue (SQ) and completion queue (CQ) rings are shared
> between the application and the kernel. This eliminates the need to
> copy data back and forth to submit and complete IO.
> 
> IO submissions use the io_uring_sqe data structure, and completions
> are generated in the form of io_uring_sqe data structures. The SQ
> ring is an index into the io_uring_sqe array, which makes it possible
> to submit a batch of IOs without them being contiguous in the ring.
> The CQ ring is always contiguous, as completion events are inherently
> unordered and can point to any io_uring_iocb.
> 
> Two new system calls are added for this:
> 
> io_uring_setup(entries, iovecs, params)
> 	Sets up a context for doing async IO. On success, returns a file
> 	descriptor that the application can mmap to gain access to the
> 	SQ ring, CQ ring, and io_uring_iocbs.

Looking at the code, it would appear that the "iovecs" parameter doesn't
actually exist.

> io_uring_enter(fd, to_submit, min_complete, flags)
> 	Initiates IO against the rings mapped to this fd, or waits for
> 	them to complete, or both The behavior is controlled by the
> 	parameters passed in. If 'min_complete' is non-zero, then we'll
> 	try and submit new IO. If IORING_ENTER_GETEVENTS is set, the
> 	kernel will wait for 'min_complete' events, if they aren't
> 	already available.

I feel like I'm missing something here.  Rather than have the
IORING_ENTER_GETEVENTS flag, why not just wait if min_complete > 0 ?

Thanks,

jon

  parent reply	other threads:[~2019-01-15 16:51 UTC|newest]

Thread overview: 73+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-01-15  2:55 (unknown), Jens Axboe
2019-01-15  2:55 ` Jens Axboe
2019-01-15  2:55 ` [PATCH 01/16] fs: add an iopoll method to struct file_operations Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-15  2:55 ` [PATCH 02/16] block: wire up block device iopoll method Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-15  2:55 ` [PATCH 03/16] block: add bio_set_polled() helper Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-15  2:55 ` [PATCH 04/16] iomap: wire up the iopoll method Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-15  2:55 ` [PATCH 05/16] Add io_uring IO interface Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-15 16:51   ` Jonathan Corbet [this message]
2019-01-15 16:51     ` Jonathan Corbet
2019-01-15 16:55     ` Jens Axboe
2019-01-15 16:55       ` Jens Axboe
2019-01-15 17:26       ` Jens Axboe
2019-01-15 17:26         ` Jens Axboe
2019-01-16 10:41   ` Arnd Bergmann
2019-01-16 10:41     ` Arnd Bergmann
2019-01-16 11:00     ` Arnd Bergmann
2019-01-16 11:00       ` Arnd Bergmann
2019-01-16 15:12     ` Jens Axboe
2019-01-16 15:12       ` Jens Axboe
2019-01-16 15:16       ` Arnd Bergmann
2019-01-16 15:16         ` Arnd Bergmann
2019-01-16 15:25         ` Jens Axboe
2019-01-16 15:25           ` Jens Axboe
2019-01-15  2:55 ` [PATCH 06/16] io_uring: add fsync support Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-15  2:55 ` [PATCH 07/16] io_uring: support for IO polling Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-15  2:55 ` [PATCH 08/16] io_uring: add submission side request cache Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-15  2:55 ` [PATCH 09/16] fs: add fget_many() and fput_many() Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-15  2:55 ` [PATCH 10/16] io_uring: use fget/fput_many() for file references Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-15  2:55 ` [PATCH 11/16] io_uring: batch io_kiocb allocation Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-15  2:55 ` [PATCH 12/16] block: implement bio helper to add iter bvec pages to bio Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-15  2:55 ` [PATCH 13/16] io_uring: add support for pre-mapped user IO buffers Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-16 10:53   ` Arnd Bergmann
2019-01-16 15:14     ` Jens Axboe
2019-01-16 15:14       ` Jens Axboe
2019-01-16 15:32       ` Jens Axboe
2019-01-16 15:32         ` Jens Axboe
2019-01-16 15:41         ` Arnd Bergmann
2019-01-16 15:47           ` Jens Axboe
2019-01-16 15:47             ` Jens Axboe
2019-01-15  2:55 ` [PATCH 14/16] io_uring: add submission polling Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-15  2:55 ` [PATCH 15/16] io_uring: add file registration Jens Axboe
2019-01-15  2:55   ` Jens Axboe
2019-01-16 10:45   ` Arnd Bergmann
2019-01-16 10:45     ` Arnd Bergmann
2019-01-16 15:15     ` Jens Axboe
2019-01-16 15:15       ` Jens Axboe
2019-01-15  2:55 ` [PATCH 16/16] io_uring: add io_uring_event cache hit information Jens Axboe
2019-01-15  2:55   ` Jens Axboe
  -- strict thread matches above, loose matches on Subject: below --
2019-01-12 21:29 [PATCHSET v3] io_uring IO interface Jens Axboe
2019-01-12 21:30 ` [PATCH 05/16] Add " Jens Axboe
2019-01-12 21:30   ` Jens Axboe
2019-01-08 16:56 [PATCHSET v1] " Jens Axboe
2019-01-08 16:56 ` [PATCH 05/16] Add " Jens Axboe
2019-01-08 16:56   ` Jens Axboe
2019-01-09 12:10   ` Christoph Hellwig
2019-01-09 15:53     ` Jens Axboe
2019-01-09 15:53       ` Jens Axboe
2019-01-09 18:30       ` Christoph Hellwig
2019-01-09 18:30         ` Christoph Hellwig
2019-01-09 20:07         ` Jens Axboe
2019-01-09 20:07           ` Jens Axboe

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=20190115095134.6286b7d6@lwn.net \
    --to=corbet@lwn.net \
    --cc=avi@scylladb.com \
    --cc=axboe@kernel.dk \
    --cc=hch@lst.de \
    --cc=jmoyer@redhat.com \
    --cc=linux-aio@kvack.org \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    /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;
as well as URLs for NNTP newsgroup(s).