From: Jens Axboe <[email protected]>
To: [email protected]
Subject: Re: [PATCHSET] Cleanup io_uring sqe handling
Date: Mon, 23 Dec 2019 21:04:48 -0700 [thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>
On 12/20/19 10:47 AM, Jens Axboe wrote:
> This series contains some prep work, then two patches that once and for
> all cleanup the sqe handling. After this patchset, the prep and issue
> handling is fully split, so each opcode has a prep handler that is called
> in the same way, and an issue handler that doesn't call the prep handler.
>
> The sqe pointer is removed from io_kiocb, so there cannot be any
> accidental dereference after we've done prep. Prep is always done in the
> original context, so we can have no reuse issues either.
>
> I've rebased for-5.6/io_uring to have this series first, so this series
> applies on top of io_uring-5.5. Ideally we'd put this into 5.5, but...
>
> In any case, please take a look, I think this is a massive improvement
> in terms of verifying that we're doing the right thing.
>
> fs/io_uring.c | 690 ++++++++++++++++++++++++++------------------------
> 1 file changed, 355 insertions(+), 335 deletions(-)
I've done various testing and everything seems to work just fine, no
issues found.
I'm tempted to push this to 5.5-rc4. It's a bit big, but I do think the
end result is much cleaner and we'll be better of for it.
--
Jens Axboe
prev parent reply other threads:[~2019-12-24 4:04 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-20 17:47 [PATCHSET] Cleanup io_uring sqe handling Jens Axboe
2019-12-20 17:47 ` [PATCH 1/7] io_uring: use u64_to_user_ptr() consistently Jens Axboe
2019-12-20 17:47 ` [PATCH 2/7] io_uring: add and use struct io_rw for read/writes Jens Axboe
2019-12-20 17:47 ` [PATCH 3/7] io_uring: move all prep state for IORING_OP_CONNECT to prep handler Jens Axboe
2019-12-20 17:47 ` [PATCH 4/7] io_uring: move all prep state for IORING_OP_{SEND,RECV}_MGS " Jens Axboe
2019-12-20 17:47 ` [PATCH 5/7] io_uring: read 'count' for IORING_OP_TIMEOUT in " Jens Axboe
2019-12-20 17:47 ` [PATCH 6/7] io_uring: standardize the prep methods Jens Axboe
2019-12-20 17:47 ` [PATCH 7/7] io_uring: pass in 'sqe' to the prep handlers Jens Axboe
2019-12-24 4:04 ` Jens Axboe [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 \
[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