public inbox for [email protected]
 help / color / mirror / Atom feed
From: Pavel Begunkov <[email protected]>
To: Anuj Gupta <[email protected]>,
	[email protected], [email protected], [email protected],
	[email protected], [email protected],
	[email protected], [email protected], [email protected]
Cc: [email protected], [email protected],
	[email protected], [email protected],
	[email protected], [email protected],
	[email protected],
	Kanchan Joshi <[email protected]>
Subject: Re: [PATCH v9 06/11] io_uring: introduce attributes for read/write and PI support
Date: Sat, 16 Nov 2024 00:32:25 +0000	[thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>

On 11/16/24 00:00, Pavel Begunkov wrote:
> On 11/14/24 10:45, Anuj Gupta wrote:
>> Add the ability to pass additional attributes along with read/write.
>> Application can populate an array of 'struct io_uring_attr_vec' and pass
>> its address using the SQE field:
>>     __u64    attr_vec_addr;
>>
>> Along with number of attributes using:
>>     __u8    nr_attr_indirect;
>>
>> Overall 16 attributes are allowed and currently one attribute
>> 'ATTR_TYPE_PI' is supported.
> 
> Why only 16? It's possible that might need more, 256 would
> be a safer choice and fits into u8. I don't think you even
> need to commit to a number, it should be ok to add more as
> long as it fits into the given types (u8 above). It can also
> be u16 as well.
> 
>> With PI attribute, userspace can pass following information:
>> - flags: integrity check flags IO_INTEGRITY_CHK_{GUARD/APPTAG/REFTAG}
>> - len: length of PI/metadata buffer
>> - addr: address of metadata buffer
>> - seed: seed value for reftag remapping
>> - app_tag: application defined 16b value
> 
> In terms of flexibility I like it apart from small nits,
> but the double indirection could be a bit inefficient,
> this thing:
> 
> struct pi_attr pi = {};
> attr_array = { &pi, ... };
> sqe->attr_addr = attr_array;

We can also reuse your idea from your previous iterations and
use the bitmap to list all attributes. Then preamble and
the explicit attr_type field are not needed, type checking
in the loop is removed and packing is better. And just
by looking at the map we can calculate the size of the
array and remove all size checks in the loop.

-- 
Pavel Begunkov

  reply	other threads:[~2024-11-16  0:31 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20241114105326epcas5p103b2c996293fa680092b97c747fdbd59@epcas5p1.samsung.com>
2024-11-14 10:45 ` [PATCH v9 00/11] Read/Write with meta/integrity Anuj Gupta
     [not found]   ` <CGME20241114105352epcas5p109c1742fa8a6552296b9c104f2271308@epcas5p1.samsung.com>
2024-11-14 10:45     ` [PATCH v9 01/11] block: define set of integrity flags to be inherited by cloned bip Anuj Gupta
     [not found]   ` <CGME20241114105354epcas5p49a73947c3d37be4189023f66fb7ba413@epcas5p4.samsung.com>
2024-11-14 10:45     ` [PATCH v9 02/11] block: copy back bounce buffer to user-space correctly in case of split Anuj Gupta
     [not found]   ` <CGME20241114105357epcas5p41fd14282d4abfe564e858b37babe708a@epcas5p4.samsung.com>
2024-11-14 10:45     ` [PATCH v9 03/11] block: modify bio_integrity_map_user to accept iov_iter as argument Anuj Gupta
     [not found]   ` <CGME20241114105400epcas5p270b8062a0c4f26833a5b497f057d65a7@epcas5p2.samsung.com>
2024-11-14 10:45     ` [PATCH v9 04/11] fs, iov_iter: define meta io descriptor Anuj Gupta
     [not found]   ` <CGME20241114105402epcas5p41b1f6054a557f1bda2cfddfdfb9a9477@epcas5p4.samsung.com>
2024-11-14 10:45     ` [PATCH v9 05/11] fs: introduce IOCB_HAS_METADATA for metadata Anuj Gupta
     [not found]   ` <CGME20241114105405epcas5p24ca2fb9017276ff8a50ef447638fd739@epcas5p2.samsung.com>
2024-11-14 10:45     ` [PATCH v9 06/11] io_uring: introduce attributes for read/write and PI support Anuj Gupta
2024-11-14 12:16       ` Christoph Hellwig
2024-11-14 13:09         ` Pavel Begunkov
2024-11-14 15:19           ` Christoph Hellwig
2024-11-15 16:40             ` Pavel Begunkov
2024-11-15 17:12               ` Christoph Hellwig
2024-11-15 17:44                 ` Jens Axboe
2024-11-15 18:00                   ` Christoph Hellwig
2024-11-15 19:03                 ` Pavel Begunkov
2024-11-18 12:49                   ` Christoph Hellwig
2024-11-15 18:04           ` Matthew Wilcox
2024-11-20 17:35             ` Darrick J. Wong
2024-11-21  6:54               ` Christoph Hellwig
2024-11-15 13:29         ` Anuj gupta
2024-11-16  0:00       ` Pavel Begunkov
2024-11-16  0:32         ` Pavel Begunkov [this message]
2024-11-18 12:50           ` Christoph Hellwig
2024-11-18 16:59             ` Pavel Begunkov
2024-11-18 17:03               ` Christoph Hellwig
2024-11-18 17:45                 ` Pavel Begunkov
2024-11-19 12:49                   ` Christoph Hellwig
2024-11-16 23:09       ` kernel test robot
     [not found]   ` <CGME20241114105408epcas5p3c77cda2faf7ccb37abbfd8e95b4ad1f5@epcas5p3.samsung.com>
2024-11-14 10:45     ` [PATCH v9 07/11] io_uring: inline read/write attributes and PI Anuj Gupta
     [not found]   ` <CGME20241114105410epcas5p1c6a4e6141b073ccfd6277288f7d5e28b@epcas5p1.samsung.com>
2024-11-14 10:45     ` [PATCH v9 08/11] block: introduce BIP_CHECK_GUARD/REFTAG/APPTAG bip_flags Anuj Gupta
     [not found]   ` <CGME20241114105413epcas5p2d7da8675df2de0d1efba3057144e691d@epcas5p2.samsung.com>
2024-11-14 10:45     ` [PATCH v9 09/11] nvme: add support for passing on the application tag Anuj Gupta
     [not found]   ` <CGME20241114105416epcas5p3a7aa552775cfe50f60ef89f7d982ea12@epcas5p3.samsung.com>
2024-11-14 10:45     ` [PATCH v9 10/11] scsi: add support for user-meta interface Anuj Gupta
     [not found]   ` <CGME20241114105418epcas5p1537d72b9016d10670cf97751704e2cc8@epcas5p1.samsung.com>
2024-11-14 10:45     ` [PATCH v9 11/11] block: add support to pass user meta buffer Anuj Gupta

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] \
    [email protected] \
    [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