xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
To: Matthew Rushton <mvrushton@gmail.com>
Cc: Keir Fraser <keir@xen.org>, Matt Wilson <msw@amazon.com>,
	Matt Wilson <msw@linux.com>, Tim Deegan <tim@xen.org>,
	Jan Beulich <jbeulich@suse.com>,
	Andrew Cooper <andrew.cooper3@citrix.com>,
	xen-devel@lists.xenproject.org
Subject: Re: [RFC PATCH] page_alloc: use first half of higher order chunks when halving
Date: Wed, 26 Mar 2014 12:36:09 -0400	[thread overview]
Message-ID: <20140326163609.GD21368@phenom.dumpdata.com> (raw)
In-Reply-To: <5332F948.1020909@gmail.com>

On Wed, Mar 26, 2014 at 08:59:04AM -0700, Matthew Rushton wrote:
> On 03/26/14 08:15, Matt Wilson wrote:
> >On Wed, Mar 26, 2014 at 11:08:01AM -0400, Konrad Rzeszutek Wilk wrote:
> >>Could you elaborate a bit more on the use-case please?
> >>My understanding is that most drivers use a scatter gather list - in which
> >>case it does not matter if the underlaying MFNs in the PFNs spare are
> >>not contingous.
> >>
> >>But I presume the issue you are hitting is with drivers doing dma_map_page
> >>and the page is not 4KB but rather large (compound page). Is that the
> >>problem you have observed?
> >Drivers are using very large size arguments to dma_alloc_coherent()
> >for things like RX and TX descriptor rings.

Large size like larger than 512kB? That would also cause problems
on baremetal then when swiotlb is activated I believe.

> >
> >--msw
> 
> It's the dma streaming api I've noticed the problem with, so
> dma_map_single(). Applicable swiotlb code would be
> xen_swiotlb_map_page() and range_straddles_page_boundary(). So yes
> for larger buffers it can cause bouncing.

  reply	other threads:[~2014-03-26 16:36 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-03-25 11:22 [RFC PATCH] page_alloc: use first half of higher order chunks when halving Matt Wilson
2014-03-25 11:44 ` Andrew Cooper
2014-03-25 13:20   ` Matt Wilson
2014-03-25 20:18     ` Matthew Rushton
2014-03-25 12:19 ` Tim Deegan
2014-03-25 13:27   ` Matt Wilson
2014-03-25 20:09     ` Matthew Rushton
2014-03-26  9:55       ` Tim Deegan
2014-03-26 10:17         ` Matt Wilson
2014-03-26 10:44           ` David Vrabel
2014-03-26 10:48             ` Matt Wilson
2014-03-26 11:13               ` Ian Campbell
2014-03-26 11:41                 ` Matt Wilson
2014-03-26 11:45                   ` Andrew Cooper
2014-03-26 11:50                     ` Matt Wilson
2014-03-26 12:43               ` David Vrabel
2014-03-26 12:48                 ` Matt Wilson
2014-03-26 15:08           ` Konrad Rzeszutek Wilk
2014-03-26 15:15             ` Matt Wilson
2014-03-26 15:59               ` Matthew Rushton
2014-03-26 16:36                 ` Konrad Rzeszutek Wilk [this message]
2014-03-26 17:47                   ` Matthew Rushton
2014-03-26 17:56                     ` Konrad Rzeszutek Wilk
2014-03-26 22:15                       ` Matthew Rushton
2014-03-28 17:02                         ` Konrad Rzeszutek Wilk
2014-03-28 22:06                           ` Matthew Rushton
2014-03-31 14:15                             ` Konrad Rzeszutek Wilk
2014-04-01  3:25                               ` Matthew Rushton
2014-04-01 10:48                                 ` Konrad Rzeszutek Wilk
2014-04-01 12:22                                   ` Tim Deegan
2014-04-02  0:17                                     ` Matthew Rushton
2014-04-02  7:52                                       ` Jan Beulich
2014-04-02 10:06                                         ` Ian Campbell
2014-04-02 10:15                                           ` Jan Beulich
2014-04-02 10:20                                             ` Ian Campbell
2014-04-09 22:21                                               ` Matthew Rushton
2014-04-10  6:14                                                 ` Jan Beulich
2014-04-11 20:20                                                   ` Matthew Rushton
2014-04-11 17:05                                                 ` Konrad Rzeszutek Wilk
2014-04-11 20:28                                                   ` Matthew Rushton
2014-04-12  1:34                                                     ` Konrad Rzeszutek Wilk
2014-04-13 21:32                                                   ` Tim Deegan
2014-04-14  8:51                                                     ` Jan Beulich
2014-04-14 14:40                                                       ` Konrad Rzeszutek Wilk
2014-04-14 15:34                                                         ` Jan Beulich
2014-04-16 14:15                                                           ` Konrad Rzeszutek Wilk
2014-04-17  1:34                                                             ` Matthew Rushton
2014-05-07 23:16                                                             ` Matthew Rushton
2014-05-08 18:05                                                               ` Konrad Rzeszutek Wilk
2014-05-14 15:06                                                               ` Konrad Rzeszutek Wilk
2014-05-20 19:26                                                                 ` Matthew Rushton
2014-05-23 19:00                                                                   ` Konrad Rzeszutek Wilk
2014-06-04 22:25                                                                     ` Matthew Rushton
2014-06-05  9:32                                                                       ` David Vrabel
2014-03-26 16:34               ` Konrad Rzeszutek Wilk

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=20140326163609.GD21368@phenom.dumpdata.com \
    --to=konrad.wilk@oracle.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=jbeulich@suse.com \
    --cc=keir@xen.org \
    --cc=msw@amazon.com \
    --cc=msw@linux.com \
    --cc=mvrushton@gmail.com \
    --cc=tim@xen.org \
    --cc=xen-devel@lists.xenproject.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 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).