public inbox for [email protected]
 help / color / mirror / Atom feed
From: "[email protected]" <[email protected]>
To: Jens Axboe <[email protected]>
Cc: "Matias Bjørling" <[email protected]>,
	"Damien Le Moal" <[email protected]>,
	"Kanchan Joshi" <[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]>
Subject: Re: [PATCH 3/3] io_uring: add support for zone-append
Date: Sun, 21 Jun 2020 20:55:05 +0200	[thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>

On 19.06.2020 09:44, Jens Axboe wrote:
>On 6/19/20 9:40 AM, Matias Bjørling wrote:
>> On 19/06/2020 17.20, Jens Axboe wrote:
>>> On 6/19/20 9:14 AM, Matias Bjørling wrote:
>>>> On 19/06/2020 16.18, Jens Axboe wrote:
>>>>> On 6/19/20 5:15 AM, Matias Bjørling wrote:
>>>>>> On 19/06/2020 11.41, [email protected] wrote:
>>>>>>> Jens,
>>>>>>>
>>>>>>> Would you have time to answer a question below in this thread?
>>>>>>>
>>>>>>> On 18.06.2020 11:11, [email protected] wrote:
>>>>>>>> On 18.06.2020 08:47, Damien Le Moal wrote:
>>>>>>>>> On 2020/06/18 17:35, [email protected] wrote:
>>>>>>>>>> On 18.06.2020 07:39, Damien Le Moal wrote:
>>>>>>>>>>> On 2020/06/18 2:27, Kanchan Joshi wrote:
>>>>>>>>>>>> From: Selvakumar S <[email protected]>
>>>>>>>>>>>>
>>>>>>>>>>>> Introduce three new opcodes for zone-append -
>>>>>>>>>>>>
>>>>>>>>>>>>     IORING_OP_ZONE_APPEND     : non-vectord, similiar to
>>>>>>>>>>>> IORING_OP_WRITE
>>>>>>>>>>>>     IORING_OP_ZONE_APPENDV    : vectored, similar to IORING_OP_WRITEV
>>>>>>>>>>>>     IORING_OP_ZONE_APPEND_FIXED : append using fixed-buffers
>>>>>>>>>>>>
>>>>>>>>>>>> Repurpose cqe->flags to return zone-relative offset.
>>>>>>>>>>>>
>>>>>>>>>>>> Signed-off-by: SelvaKumar S <[email protected]>
>>>>>>>>>>>> Signed-off-by: Kanchan Joshi <[email protected]>
>>>>>>>>>>>> Signed-off-by: Nitesh Shetty <[email protected]>
>>>>>>>>>>>> Signed-off-by: Javier Gonzalez <[email protected]>
>>>>>>>>>>>> ---
>>>>>>>>>>>> fs/io_uring.c                 | 72
>>>>>>>>>>>> +++++++++++++++++++++++++++++++++++++++++--
>>>>>>>>>>>> include/uapi/linux/io_uring.h |  8 ++++-
>>>>>>>>>>>> 2 files changed, 77 insertions(+), 3 deletions(-)
>>>>>>>>>>>>
>>>>>>>>>>>> diff --git a/fs/io_uring.c b/fs/io_uring.c
>>>>>>>>>>>> index 155f3d8..c14c873 100644
>>>>>>>>>>>> --- a/fs/io_uring.c
>>>>>>>>>>>> +++ b/fs/io_uring.c
>>>>>>>>>>>> @@ -649,6 +649,10 @@ struct io_kiocb {
>>>>>>>>>>>>       unsigned long        fsize;
>>>>>>>>>>>>       u64            user_data;
>>>>>>>>>>>>       u32            result;
>>>>>>>>>>>> +#ifdef CONFIG_BLK_DEV_ZONED
>>>>>>>>>>>> +    /* zone-relative offset for append, in bytes */
>>>>>>>>>>>> +    u32            append_offset;
>>>>>>>>>>> this can overflow. u64 is needed.
>>>>>>>>>> We chose to do it this way to start with because struct io_uring_cqe
>>>>>>>>>> only has space for u32 when we reuse the flags.
>>>>>>>>>>
>>>>>>>>>> We can of course create a new cqe structure, but that will come with
>>>>>>>>>> larger changes to io_uring for supporting append.
>>>>>>>>>>
>>>>>>>>>> Do you believe this is a better approach?
>>>>>>>>> The problem is that zone size are 32 bits in the kernel, as a number
>>>>>>>>> of sectors.
>>>>>>>>> So any device that has a zone size smaller or equal to 2^31 512B
>>>>>>>>> sectors can be
>>>>>>>>> accepted. Using a zone relative offset in bytes for returning zone
>>>>>>>>> append result
>>>>>>>>> is OK-ish, but to match the kernel supported range of possible zone
>>>>>>>>> size, you
>>>>>>>>> need 31+9 bits... 32 does not cut it.
>>>>>>>> Agree. Our initial assumption was that u32 would cover current zone size
>>>>>>>> requirements, but if this is a no-go, we will take the longer path.
>>>>>>> Converting to u64 will require a new version of io_uring_cqe, where we
>>>>>>> extend at least 32 bits. I believe this will need a whole new allocation
>>>>>>> and probably ioctl().
>>>>>>>
>>>>>>> Is this an acceptable change for you? We will of course add support for
>>>>>>> liburing when we agree on the right way to do this.
>>>>>> I took a quick look at the code. No expert, but why not use the existing
>>>>>> userdata variable? use the lowest bits (40 bits) for the Zone Starting
>>>>>> LBA, and use the highest (24 bits) as index into the completion data
>>>>>> structure?
>>>>>>
>>>>>> If you want to pass the memory address (same as what fio does) for the
>>>>>> data structure used for completion, one may also play some tricks by
>>>>>> using a relative memory address to the data structure. For example, the
>>>>>> x86_64 architecture uses 48 address bits for its memory addresses. With
>>>>>> 24 bit, one can allocate the completion entries in a 32MB memory range,
>>>>>> and then use base_address + index to get back to the completion data
>>>>>> structure specified in the sqe.
>>>>> For any current request, sqe->user_data is just provided back as
>>>>> cqe->user_data. This would make these requests behave differently
>>>>> from everything else in that sense, which seems very confusing to me
>>>>> if I was an application writer.
>>>>>
>>>>> But generally I do agree with you, there are lots of ways to make
>>>>> < 64-bit work as a tag without losing anything or having to jump
>>>>> through hoops to do so. The lack of consistency introduced by having
>>>>> zone append work differently is ugly, though.
>>>>>
>>>> Yep, agree, and extending to three cachelines is big no-go. We could add
>>>> a flag that said the kernel has changes the userdata variable. That'll
>>>> make it very explicit.
>>> Don't like that either, as it doesn't really change the fact that you're
>>> now doing something very different with the user_data field, which is
>>> just supposed to be passed in/out directly. Adding a random flag to
>>> signal this behavior isn't very explicit either, imho. It's still some
>>> out-of-band (ish) notification of behavior that is different from any
>>> other command. This is very different from having a flag that says
>>> "there's extra information in this other field", which is much cleaner.
>>>
>> Ok. Then it's pulling in the bits from cqe->res and cqe->flags that you
>> mention in the other mail. Sounds good.
>
>I think that's the best approach, if we need > 32-bits. Maybe we can get
>by just using ->res, if we switch to multiples of 512b instead for the
>result like Pavel suggested. That'd provide enough room in ->res, and
>would be preferable imho. But if we do need > 32-bits, then we can use
>this approach.

Sounds good.

Thanks Matias too for chipping in with more ideas. We have enough for a
v2.

Javier

  reply	other threads:[~2020-06-21 18:55 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20200617172653epcas5p488de50090415eb802e62acc0e23d8812@epcas5p4.samsung.com>
2020-06-17 17:23 ` [PATCH 0/3] zone-append support in aio and io-uring Kanchan Joshi
     [not found]   ` <CGME20200617172702epcas5p4dbf4729d31d9a85ab1d261d04f238e61@epcas5p4.samsung.com>
2020-06-17 17:23     ` [PATCH 1/3] fs,block: Introduce IOCB_ZONE_APPEND and direct-io handling Kanchan Joshi
2020-06-17 19:02       ` Pavel Begunkov
2020-06-18  7:16       ` Damien Le Moal
2020-06-18 18:35         ` Kanchan Joshi
     [not found]   ` <CGME20200617172706epcas5p4dcbc164063f58bad95b211b9d6dfbfa9@epcas5p4.samsung.com>
2020-06-17 17:23     ` [PATCH 2/3] aio: add support for zone-append Kanchan Joshi
2020-06-18  7:33       ` Damien Le Moal
     [not found]   ` <CGME20200617172713epcas5p352f2907a12bd4ee3c97be1c7d8e1569e@epcas5p3.samsung.com>
2020-06-17 17:23     ` [PATCH 3/3] io_uring: " Kanchan Joshi
2020-06-17 18:55       ` Pavel Begunkov
2020-06-18  7:39       ` Damien Le Moal
2020-06-18  8:35         ` [email protected]
2020-06-18  8:47           ` Damien Le Moal
2020-06-18  9:11             ` [email protected]
2020-06-19  9:41               ` [email protected]
2020-06-19 11:15                 ` Matias Bjørling
2020-06-19 14:18                   ` Jens Axboe
2020-06-19 15:14                     ` Matias Bjørling
2020-06-19 15:20                       ` Jens Axboe
2020-06-19 15:40                         ` Matias Bjørling
2020-06-19 15:44                           ` Jens Axboe
2020-06-21 18:55                             ` [email protected] [this message]
2020-06-19 14:15                 ` Jens Axboe
2020-06-19 14:59                   ` Pavel Begunkov
2020-06-19 15:02                     ` Jens Axboe
2020-06-21 18:52                       ` [email protected]
2020-06-17 17:42   ` [PATCH 0/3] zone-append support in aio and io-uring Matthew Wilcox
2020-06-18  6:56   ` Christoph Hellwig
2020-06-18  8:29     ` Javier González
2020-06-18 17:52     ` Kanchan Joshi
2020-06-19  3:08       ` Damien Le Moal
2020-06-19  7:56       ` Christoph Hellwig
2020-06-18  8:04   ` Matias Bjørling
2020-06-18  8:27     ` Javier González
2020-06-18  8:32       ` Matias Bjørling
2020-06-18  8:39         ` Javier González
2020-06-18  8:46           ` Matias Bjørling
2020-06-18 14:16     ` Christoph Hellwig
2020-06-18 19:21     ` Kanchan Joshi
2020-06-18 20:04       ` Matias Bjørling
2020-06-19  1:03         ` Damien Le Moal

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=20200621185505.4i46y4kndzvqlzdm@MacBook-Pro.localdomain \
    [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