From: Christoph Hellwig <hch@lst.de>
To: Jens Axboe <axboe@kernel.dk>
Cc: Damien Le Moal <dlemoal@kernel.org>,
Christoph Hellwig <hch@lst.de>,
linux-block@vger.kernel.org
Subject: Re: [PATCH 0/2] New zoned loop block device driver
Date: Wed, 8 Jan 2025 06:44:36 +0100 [thread overview]
Message-ID: <20250108054436.GA20178@lst.de> (raw)
In-Reply-To: <3a1314db-ed44-4c22-8fc1-0cf672003026@kernel.dk>
On Tue, Jan 07, 2025 at 02:08:20PM -0700, Jens Axboe wrote:
> > ublk backend driver to do the same as zloop in userspace would need a
> > lot more code to be efficient. And even then, as Christoph already
> > mentioned, we would still have performance suffer from the context
> > switches. But that performance point was not the primary stopper
>
> I don't buy this context switch argument at all.
The zloop write goes straight from kblockd into the the filesystem.
ublk switches to userspace, which goes back to the kernel when the
file system writes. Similar double context switch on the completion
side.
> Why would it mean more
> sleeping?
?
> There's absolutely zero reason why a ublk solution would be at
> least as performant as the kernel one.
Well, prove it. From haing worked on similar schemes in the past
I highly doubt it.
> And why would it need "a lot more code to be efficient"?
Because we don't have all the nice locking and even infrastructure
in userspace that we have in the kernel.
next prev parent reply other threads:[~2025-01-08 5:44 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-01-06 14:24 [PATCH 0/2] New zoned loop block device driver Damien Le Moal
2025-01-06 14:24 ` [PATCH 1/2] block: new " Damien Le Moal
2025-01-06 14:24 ` [PATCH 2/2] Documentation: Document the " Damien Le Moal
2025-01-06 14:54 ` [PATCH 0/2] New " Jens Axboe
2025-01-06 15:21 ` Christoph Hellwig
2025-01-06 15:24 ` Jens Axboe
2025-01-06 15:32 ` Christoph Hellwig
2025-01-06 15:38 ` Jens Axboe
2025-01-06 15:44 ` Christoph Hellwig
2025-01-06 17:38 ` Jens Axboe
2025-01-06 18:05 ` Christoph Hellwig
2025-01-07 21:10 ` Jens Axboe
2025-01-08 5:49 ` Christoph Hellwig
2025-01-07 1:08 ` Damien Le Moal
2025-01-07 21:08 ` Jens Axboe
2025-01-08 5:11 ` Damien Le Moal
2025-01-08 5:44 ` Christoph Hellwig [this message]
2025-01-08 2:47 ` Ming Lei
2025-01-08 14:10 ` Theodore Ts'o
2025-01-08 2:29 ` Ming Lei
2025-01-08 5:06 ` Damien Le Moal
2025-01-08 8:13 ` Ming Lei
2025-01-08 9:09 ` Christoph Hellwig
2025-01-08 9:39 ` Ming Lei
2025-01-10 12:34 ` Ming Lei
2025-01-24 9:30 ` Damien Le Moal
2025-01-24 12:30 ` Ming Lei
2025-01-24 14:20 ` Johannes Thumshirn
2025-01-29 8:10 ` Damien Le Moal
2025-01-31 3:54 ` Ming Lei
2025-02-04 3:22 ` Damien Le Moal
2025-02-05 3:43 ` Ming Lei
2025-02-05 6:07 ` Damien Le Moal
2025-02-06 3:24 ` Ming Lei
2025-01-08 5:47 ` Christoph Hellwig
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=20250108054436.GA20178@lst.de \
--to=hch@lst.de \
--cc=axboe@kernel.dk \
--cc=dlemoal@kernel.org \
--cc=linux-block@vger.kernel.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.