linux-next.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@infradead.org>
To: Kent Overstreet <kmo@daterainc.com>
Cc: Christoph Hellwig <hch@infradead.org>,
	Dave Kleikamp <dave.kleikamp@oracle.com>,
	Stephen Rothwell <sfr@canb.auug.org.au>,
	Jens Axboe <axboe@kernel.dk>,
	linux-next@vger.kernel.org, linux-kernel@vger.kernel.org,
	Zach Brown <zab@zabbo.net>, Olof Johansson <olof@lixom.net>,
	Andrew Morton <akpm@linux-foundation.org>
Subject: Re: linux-next: manual merge of the block tree with the  tree
Date: Fri, 8 Nov 2013 00:32:51 -0800	[thread overview]
Message-ID: <20131108083251.GA20121@infradead.org> (raw)
In-Reply-To: <20131108081737.GA27885@kmo-pixel>

On Fri, Nov 08, 2013 at 12:17:37AM -0800, Kent Overstreet wrote:
> The core issue isn't whether the IO is going to a block based filesystem
> (but thanks for pointing out that that's not necessarily true!) but
> whether we want to work with pinned pages or not. If pinned pages are ok
> for everything, then bios as a common interface work - likely evolving
> them a bit to be more general (it's just bi_bdev and bi_sector that's
> actually block specific) - and IMO that would be far preferable to this
> abstraction layer.
> 
> If OTOH we need a common interface that's also for places where we can't
> afford the overhead of pinning user pages - that's a different story,
> and maybe we do need all this infrastructure then. That's why I'm asking
> about the stuff you meantioned, I'm honestly not sure.

For both of them we will deal with kernel-allocated pages that are never
mapped to userspace.  This is likely to be true for all the consumers
of in-kernel aio/dio as the existing interfaces handle user pages just
fine.

> What I'm working towards though is a clean separation between buffered
> and direct code paths, so that buffered IO can continue work with iovs
> and for O_DIRECT the first thing you do is fill out a bio with pinned
> pages and send it down to filesystem code or wherever it's going to go.

I don't think pushing bios above the fs interface is a good idea. Note
that the iovecs come from userspace for the user pages cases, so there
is little we can do about that, and non-bio based direct I/O
implementations generally work directly at just that level and never
even touch the direct-io.c code.

If you want to redo the ->direct_IO address_space operation and
generic_file_direct_write and the direct I/O side of
generic_file_aio_read (both of which aren't anywhere near as generic as
the name claims) I'm all for it, but it really won't affect the consumer
of the in-kernel aio/dio code.

> That make sense? I can show you more concretely what I'm working on if
> you want. Or if I'm full of crap and this is useless for what you guys
> want I'm sure you'll let me know :)

It sounds interesting, but also a little confusing at this point, at
least from the non-block side of view.

  reply	other threads:[~2013-11-08  8:32 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-11-01  3:20 linux-next: manual merge of the block tree with the tree Stephen Rothwell
2013-11-01 15:10 ` Jens Axboe
2013-11-01 20:22   ` Stephen Rothwell
2013-11-01 20:27     ` Jens Axboe
2013-11-01 20:41       ` Dave Kleikamp
2013-11-01 20:53         ` Jens Axboe
2013-11-01 21:07           ` Dave Kleikamp
2013-11-02 20:50           ` Dave Kleikamp
2013-11-07 19:17             ` Olof Johansson
2013-11-07 19:20               ` Kent Overstreet
2013-11-07 19:20             ` Dave Kleikamp
2013-11-07 19:25               ` Kent Overstreet
2013-11-07 19:38                 ` Dave Kleikamp
2013-11-08  0:04                 ` Dave Kleikamp
2013-11-08  1:53                   ` Stephen Rothwell
2013-11-08  2:08                     ` Kent Overstreet
2013-11-08  2:32                       ` Dave Kleikamp
2013-11-08  7:33                         ` Christoph Hellwig
2013-11-08  7:39                           ` Kent Overstreet
2013-11-08  7:44                             ` Christoph Hellwig
2013-11-08  7:56                               ` Kent Overstreet
2013-11-08  8:02                                 ` Christoph Hellwig
2013-11-08  8:17                                   ` Kent Overstreet
2013-11-08  8:32                                     ` Christoph Hellwig [this message]
2013-11-08  9:21                                       ` Kent Overstreet
2013-11-08 17:56                                         ` Zach Brown
2013-11-08 15:10                           ` Dave Kleikamp
2013-11-08 15:29                           ` Jens Axboe
2013-11-08 16:15                             ` Jens Axboe
2013-11-10 21:32                               ` Stephen Rothwell
2013-11-08  2:39                     ` Jens Axboe
  -- strict thread matches above, loose matches on Subject: below --
2010-12-17  1:28 Stephen Rothwell
2010-12-17 14:53 ` James Bottomley
2010-12-18  7:15   ` Tejun Heo
2009-09-10  4:48 Stephen Rothwell
2009-09-10  7:24 ` Jens Axboe
2009-09-10  7:40   ` Stephen Rothwell
2009-09-10  7:43     ` Jens Axboe
2009-07-01  5:37 Stephen Rothwell
2009-07-01  6:59 ` Jens Axboe
2009-05-18  4:53 Stephen Rothwell
2009-05-18  6:27 ` Jens Axboe
2009-05-18 12:34   ` Rusty Russell
2009-05-18 12:42     ` Jens Axboe
2009-05-19  0:11       ` Stephen Rothwell

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=20131108083251.GA20121@infradead.org \
    --to=hch@infradead.org \
    --cc=akpm@linux-foundation.org \
    --cc=axboe@kernel.dk \
    --cc=dave.kleikamp@oracle.com \
    --cc=kmo@daterainc.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-next@vger.kernel.org \
    --cc=olof@lixom.net \
    --cc=sfr@canb.auug.org.au \
    --cc=zab@zabbo.net \
    /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;
as well as URLs for NNTP newsgroup(s).