From: Konrad Rzeszutek Wilk <konrad@darnok.org>
To: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: "lersek@redhat.com" <lersek@redhat.com>,
xen-devel <xen-devel@lists.xensource.com>,
Carsten Schiers <carsten@schiers.de>,
Ian Campbell <Ian.Campbell@citrix.com>,
"zhenzhong.duan@oracle.com" <zhenzhong.duan@oracle.com>
Subject: Re: Load increase after memory upgrade (part2)
Date: Wed, 14 Dec 2011 16:23:51 -0400 [thread overview]
Message-ID: <20111214202351.GA25896@andromeda.dapyr.net> (raw)
In-Reply-To: <20111206032621.GA6568@phenom.dumpdata.com>
On Mon, Dec 05, 2011 at 10:26:21PM -0500, Konrad Rzeszutek Wilk wrote:
> On Sun, Dec 04, 2011 at 01:09:28PM +0100, Carsten Schiers wrote:
> > Here with two cards enabled and creating a bit "work" by watching TV with one oft hem:
> >
> > [ 23.842720] Starting SWIOTLB debug thread.
> > [ 23.842750] swiotlb_start_thread: Go!
> > [ 23.842838] xen_swiotlb_start_thread: Go!
> > [ 28.841451] 0 [budget_av 0000:00:01.0] bounce: from:435596(slow:0)to:0 map:658 unmap:0 sync:435596
> > [ 28.841592] SWIOTLB is 4% full
> > [ 33.840147] 0 [budget_av 0000:00:01.0] bounce: from:127652(slow:0)to:0 map:0 unmap:0 sync:127652
> > [ 33.840283] SWIOTLB is 4% full
> > [ 33.844222] 0 budget_av 0000:00:01.0 alloc coherent: 8, free: 0
> > [ 38.840227] 0 [budget_av 0000:00:01.0] bounce: from:128310(slow:0)to:0 map:0 unmap:0 sync:128310
>
> Whoa. Yes. You are definitly using the bounce buffer :-)
>
> Now it is time to look at why the drive is not using those coherent ones - it
> looks to allocate just eight of them but does not use them.. Unless it is
> using them _and_ bouncing them (which would be odd).
>
> And BTW, you can lower your 'swiotlb=XX' value. The 4% is how much you
> are using of the default size.
So I able to see this with an atl1c ethernet driver on my SandyBridge i3
box. It looks as if the card is truly 32-bit so on a box with 8GB it
bounces the data. If I booted the Xen hypervisor with 'mem=4GB' I get no
bounces (no surprise there).
In other words - I see the same behavior you are seeing. Now off to:
>
> I should find out_why_ the old Xen kernels do not use the bounce buffer
> so much...
which will require some fiddling around.
next prev parent reply other threads:[~2011-12-14 20:23 UTC|newest]
Thread overview: 66+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-11-24 12:28 Load increase after memory upgrade (part2) Carsten Schiers
2011-11-25 18:42 ` Konrad Rzeszutek Wilk
2011-11-25 22:11 ` Carsten Schiers
2011-11-28 15:28 ` Konrad Rzeszutek Wilk
2011-11-28 15:40 ` Ian Campbell
2011-11-28 16:45 ` Konrad Rzeszutek Wilk
2011-11-29 8:31 ` Jan Beulich
2011-11-29 9:31 ` Carsten Schiers
2011-11-29 9:46 ` Carsten Schiers
2011-11-29 10:23 ` Ian Campbell
2011-11-29 15:33 ` Konrad Rzeszutek Wilk
2011-12-02 15:23 ` Konrad Rzeszutek Wilk
2011-12-04 11:59 ` Carsten Schiers
2011-12-04 12:09 ` Carsten Schiers
2011-12-06 3:26 ` Konrad Rzeszutek Wilk
2011-12-14 20:23 ` Konrad Rzeszutek Wilk [this message]
2011-12-14 22:07 ` Konrad Rzeszutek Wilk
2011-12-15 14:52 ` Carsten Schiers
2011-12-16 14:56 ` Carsten Schiers
2011-12-16 15:04 ` Konrad Rzeszutek Wilk
2011-12-16 15:51 ` Carsten Schiers
2011-12-16 16:19 ` Konrad Rzeszutek Wilk
2011-12-17 22:12 ` Carsten Schiers
2011-12-18 0:19 ` Sander Eikelenboom
2011-12-19 14:56 ` Konrad Rzeszutek Wilk
2012-01-10 21:55 ` Konrad Rzeszutek Wilk
2012-01-12 22:06 ` Sander Eikelenboom
2012-01-13 8:12 ` Jan Beulich
2012-01-13 15:13 ` Konrad Rzeszutek Wilk
2012-01-15 11:32 ` Sander Eikelenboom
2012-01-17 21:02 ` Konrad Rzeszutek Wilk
2012-01-18 11:28 ` Pasi Kärkkäinen
2012-01-18 11:39 ` Jan Beulich
2012-01-18 11:35 ` Jan Beulich
2012-01-18 14:29 ` Konrad Rzeszutek Wilk
2012-01-23 22:32 ` Konrad Rzeszutek Wilk
2012-01-24 8:58 ` Jan Beulich
2012-01-24 14:17 ` Konrad Rzeszutek Wilk
2012-01-24 21:32 ` Carsten Schiers
2012-01-25 12:02 ` Carsten Schiers
2012-01-25 19:06 ` Carsten Schiers
2012-01-25 21:02 ` Konrad Rzeszutek Wilk
2012-02-15 19:28 ` Konrad Rzeszutek Wilk
2012-02-16 8:56 ` Jan Beulich
2012-02-17 15:07 ` Konrad Rzeszutek Wilk
2012-02-28 14:35 ` Carsten Schiers
2012-02-29 12:10 ` Carsten Schiers
2012-02-29 12:56 ` Carsten Schiers
2012-05-11 9:39 ` Carsten Schiers
2012-05-11 19:41 ` Konrad Rzeszutek Wilk
2012-06-13 16:55 ` Konrad Rzeszutek Wilk
2012-06-14 7:07 ` Jan Beulich
2012-06-14 18:33 ` Konrad Rzeszutek Wilk
2012-06-14 18:43 ` Carsten Schiers
2012-06-14 8:38 ` David Vrabel
2012-06-14 18:31 ` Konrad Rzeszutek Wilk
2012-06-14 18:40 ` Carsten Schiers
2012-06-14 19:16 ` Carsten Schiers
2011-12-19 14:54 ` Konrad Rzeszutek Wilk
2011-12-04 12:18 ` Carsten Schiers
2011-11-28 16:58 ` Laszlo Ersek
2011-11-29 9:37 ` Carsten Schiers
2011-11-28 15:52 ` Carsten Schiers
2011-11-26 9:14 ` Carsten Schiers
2011-11-28 15:30 ` Konrad Rzeszutek Wilk
2011-11-29 9:42 ` Carsten Schiers
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=20111214202351.GA25896@andromeda.dapyr.net \
--to=konrad@darnok.org \
--cc=Ian.Campbell@citrix.com \
--cc=carsten@schiers.de \
--cc=konrad.wilk@oracle.com \
--cc=lersek@redhat.com \
--cc=xen-devel@lists.xensource.com \
--cc=zhenzhong.duan@oracle.com \
/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).