All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jens Axboe <axboe@kernel.dk>
To: Bijan Mottahedeh <bijan.mottahedeh@oracle.com>
Cc: io-uring@vger.kernel.org
Subject: Re: [RFC 1/2] io_uring: don't use kiocb.private to store buf_index
Date: Tue, 19 May 2020 16:07:52 -0600	[thread overview]
Message-ID: <6ce9f56d-d4eb-0db1-6ea3-166aed29807f@kernel.dk> (raw)
In-Reply-To: <1589925170-48687-2-git-send-email-bijan.mottahedeh@oracle.com>

On 5/19/20 3:52 PM, Bijan Mottahedeh wrote:
> kiocb.private is used in iomap_dio_rw() so store buf_index separately.

Hmm, that's no good, the owner of the iocb really should own ->private
as well.

The downside of this patch is that io_rw now spills into the next
cacheline, which propagates to io_kiocb as well. iocb has 4 bytes
of padding, but probably cleaner if we can stuff it into io_kiocb
instead. How about adding a u16 after opcode? There's a 2 byte
hole there, so it would not impact the size of io_kiocb.

-- 
Jens Axboe


  reply	other threads:[~2020-05-19 22:07 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-19 21:52 [RFC 0/2] io_uring: don't use kiocb.private to store buf_index Bijan Mottahedeh
2020-05-19 21:52 ` [RFC 1/2] " Bijan Mottahedeh
2020-05-19 22:07   ` Jens Axboe [this message]
2020-05-19 22:20     ` Jens Axboe
2020-05-19 22:48       ` Bijan Mottahedeh
2020-05-19 21:52 ` [RFC 2/2] io_uring: mark REQ_NOWAIT for a non-mq queue as unspported Bijan Mottahedeh
2020-05-19 22:16   ` Fwd: " Jens Axboe
2020-05-20 17:50     ` Christoph Hellwig
2020-05-28 18:35   ` Jeff Moyer
2020-05-28 19:01     ` Jens Axboe
2020-05-28 19:22       ` Jens Axboe
2020-05-28 19:31         ` Jeff Moyer
2020-05-28 22:12         ` Jeff Moyer
2020-05-28 23:03           ` Jens Axboe
2020-05-29 15:02             ` Jeff Moyer
2020-05-29 16:20               ` 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=6ce9f56d-d4eb-0db1-6ea3-166aed29807f@kernel.dk \
    --to=axboe@kernel.dk \
    --cc=bijan.mottahedeh@oracle.com \
    --cc=io-uring@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.