public inbox for [email protected]
 help / color / mirror / Atom feed
From: Gao Xiang <[email protected]>
To: Ming Lei <[email protected]>
Cc: Pavel Machek <[email protected]>, Jens Axboe <[email protected]>,
	[email protected], [email protected],
	[email protected],
	Gabriel Krisman Bertazi <[email protected]>,
	ZiyangZhang <[email protected]>,
	Xiaoguang Wang <[email protected]>
Subject: Re: [RFC PATCH] ubd: add io_uring based userspace block driver
Date: Mon, 6 Jun 2022 10:15:35 +0800	[thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <YpgsTojc4mVKghZA@T590>

On Thu, Jun 02, 2022 at 11:19:42AM +0800, Ming Lei wrote:
> Hello Pavel,
> 
> On Mon, May 30, 2022 at 09:07:00AM +0200, Pavel Machek wrote:
> > Hi!
> > 
> > > This is the driver part of userspace block driver(ubd driver), the other
> > > part is userspace daemon part(ubdsrv)[1].
> > 
> > > @@ -0,0 +1,1193 @@
> > > +// SPDX-License-Identifier: GPL-2.0-or-later
> > > +/*
> > > + * Userspace block device - block device which IO is handled from userspace
> > > + *
> > > + * Take full use of io_uring passthrough command for communicating with
> > > + * ubd userspace daemon(ubdsrvd) for handling basic IO request.
> > 
> > > +
> > > +static inline unsigned int ubd_req_build_flags(struct request *req)
> > > +{
> > ...
> > > +	if (req->cmd_flags & REQ_SWAP)
> > > +		flags |= UBD_IO_F_SWAP;
> > > +
> > > +	return flags;
> > > +}
> > 
> > Does it work? How do you guarantee operation will be deadlock-free with swapping and
> > writebacks going on?
> 
> The above is just for providing command flags to user side, so that the
> user side can understand/handle the request better.
> 
> prtrl(PR_SET_IO_FLUSHER) has been merged for avoiding the deadlock.
>

I've pointed out a case before that (I think) PR_SET_IO_FLUSHER doesn't work:
https://lore.kernel.org/all/[email protected]

I don't think handling writeback in the userspace under the direct reclaim
context is _safe_ honestly. Because userspace program can call any system
call under direct reclaim, which can interconnect to another process context
and wait for it. yet I don't look into ubd implementation.

Thanks,
Gao Xiang

      reply	other threads:[~2022-06-06  2:15 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-09  9:23 [RFC PATCH] ubd: add io_uring based userspace block driver Ming Lei
2022-05-09 15:10 ` Gabriel Krisman Bertazi
2022-05-10  1:57   ` Ming Lei
2022-05-10  4:22   ` Ziyang Zhang
2022-05-09 16:00 ` Randy Dunlap
2022-05-09 18:11   ` Gabriel Krisman Bertazi
2022-05-09 18:13     ` Jens Axboe
2022-05-09 16:09 ` Jens Axboe
2022-05-10  2:58   ` Ming Lei
2022-05-10  3:29     ` Jens Axboe
2022-05-10  7:38       ` Ming Lei
2022-05-09 18:14 ` Martin Raiber
2022-05-16 19:29 ` Stefan Hajnoczi
2022-05-17  1:57   ` Ming Lei
2022-05-17  6:17     ` Stefan Hajnoczi
2022-05-30  7:07 ` Pavel Machek
2022-06-02  3:19   ` Ming Lei
2022-06-06  2:15     ` Gao Xiang [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] \
    [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