From: Jens Axboe <[email protected]>
To: Christoph Hellwig <[email protected]>
Cc: Bob Liu <[email protected]>,
[email protected], [email protected],
[email protected], [email protected],
[email protected]
Subject: Re: [PATCH 1/4] io_uring: add IORING_OP_READ{WRITE}V_PI cmd
Date: Wed, 26 Feb 2020 08:58:46 -0700 [thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>
On 2/26/20 8:57 AM, Christoph Hellwig wrote:
> On Wed, Feb 26, 2020 at 07:24:06AM -0700, Jens Axboe wrote:
>> On 2/26/20 1:37 AM, Bob Liu wrote:
>>> diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h
>>> index a3300e1..98fa3f1 100644
>>> --- a/include/uapi/linux/io_uring.h
>>> +++ b/include/uapi/linux/io_uring.h
>>> @@ -62,6 +62,8 @@ enum {
>>> IORING_OP_NOP,
>>> IORING_OP_READV,
>>> IORING_OP_WRITEV,
>>> + IORING_OP_READV_PI,
>>> + IORING_OP_WRITEV_PI,
>>> IORING_OP_FSYNC,
>>> IORING_OP_READ_FIXED,
>>> IORING_OP_WRITE_FIXED,
>>
>> So this one renumbers everything past IORING_OP_WRITEV, breaking the
>> ABI in a very bad way. I'm guessing that was entirely unintentional?
>> Any new command must go at the end of the list.
>>
>> You're also not updating io_op_defs[] with the two new commands,
>> which means it won't compile at all. I'm guessing you tested this on
>> an older version of the kernel which didn't have io_op_defs[]?
>
> And the real question is why we need ops insted of just a flag and
> using previously reserved fields for the PI pointer.
Yeah, should probably be a RWF_ flag instead, and a 64-bit SQE field
for the PI data. The 'last iovec is PI' is kind of icky.
--
Jens Axboe
next prev parent reply other threads:[~2020-02-26 15:58 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-26 8:37 [RFC PATCH 0/4] userspace PI passthrough via io_uring Bob Liu
2020-02-26 8:37 ` [PATCH 1/4] io_uring: add IORING_OP_READ{WRITE}V_PI cmd Bob Liu
2020-02-26 14:24 ` Jens Axboe
2020-02-26 15:57 ` Christoph Hellwig
2020-02-26 15:58 ` Jens Axboe [this message]
2020-02-26 16:03 ` Darrick J. Wong
2020-02-26 16:53 ` Christoph Hellwig
2020-02-27 9:19 ` Bob Liu
2020-02-27 9:05 ` Bob Liu
2020-02-26 8:37 ` [PATCH 2/4] bio-integrity: introduce two funcs handle protect information Bob Liu
2020-02-26 16:03 ` Darrick J. Wong
2020-02-27 9:23 ` Bob Liu
2020-02-26 8:37 ` [PATCH 3/4] block_dev: support protect information passthrough Bob Liu
2020-02-26 16:04 ` Darrick J. Wong
2020-02-26 8:37 ` [PATCH 4/4] liburing/test: add testcase for " Bob Liu
2020-02-26 14:25 ` [RFC PATCH 0/4] userspace PI passthrough via io_uring 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 \
[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