All of lore.kernel.org
 help / color / mirror / Atom feed
From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
To: Tobias Diedrich <ranma+xen@tdiedrich.de>,
	xen-users@lists.xensource.com, linux-kernel@vger.kernel.org
Subject: Re: 3.0.0-rc2: Xen: High amount of kernel "reserved" memory, about 33% in 256MB DOMU [workaround included]
Date: Tue, 14 Jun 2011 15:48:11 -0400	[thread overview]
Message-ID: <20110614194810.GA26728@dumpdata.com> (raw)
In-Reply-To: <20110614001055.GB7417@yumi.tdiedrich.de>

On Tue, Jun 14, 2011 at 02:10:55AM +0200, Tobias Diedrich wrote:
> Konrad Rzeszutek Wilk wrote:
> > On Mon, Jun 13, 2011 at 10:50:03PM +0200, Tobias Diedrich wrote:
> > > Hi,
> > > 
> > > another issue I'm seeing with 3.0-rc2 and Xen is that there is an
> > > unexpectedly high amount of kernel reserved memory.
> > 
> > > 
> > > I suspect that Linux allocates page table entries and corresponding
> > > data structures for the whole 6GB areas of the provided 'physical
> > > RAM map' even though it has rather big unusable holes in it.
> > 
> > Can you run it with 'memblock=debug debug loglevel=8 initcall_debug'?
> > It should tell you where it tries (and for much space) the pagetables.

Ugh. In the meantime I would suggest you do the combination of:

Xen hypervisor line: "dom0_mem=max:512M" and on the Linux line: "mem=512M" to
cut down on the extra pagetable creation..

.. snip ..
> [    0.000000] init_memory_mapping: 0000000000000000-0000000010000000
> [    0.000000]  0000000000 - 0010000000 page 4k
> [    0.000000] kernel direct mapping tables up to 10000000 @ ff7e000-10000000
> [    0.000000]     memblock_x86_reserve_range: [0x0ff7e000-0x0ffe9fff]          PGTABLE
> [    0.000000] xen: setting RW the range ffea000 - 10000000

So ff7e->ffea pages, or 432kBytes
> [    0.000000] init_memory_mapping: 0000000100000000-000000016fef0000
> [    0.000000]  0100000000 - 016fef0000 page 4k
> [    0.000000] kernel direct mapping tables up to 16fef0000 @ f3f7000-ff7e000
> [    0.000000]     memblock_x86_reserve_range: [0x0f3f7000-0x0f778fff]          PGTABLE
> [    0.000000] xen: setting RW the range f779000 - ff7e000

And here from f3f7 through f779, so 3592kB or 3.5MB for un-used potential balloon memory.

  reply	other threads:[~2011-06-14 19:48 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-13 20:50 [Xen-users] 3.0.0-rc2: Xen: High amount of kernel "reserved" memory, about 33% in 256MB DOMU Tobias Diedrich
2011-06-13 21:59 ` Konrad Rzeszutek Wilk
2011-06-14  0:10   ` Tobias Diedrich
2011-06-14 19:48     ` Konrad Rzeszutek Wilk [this message]
2011-06-14 21:37       ` 3.0.0-rc2: Xen: High amount of kernel "reserved" memory, about 33% in 256MB DOMU [workaround included] Tobias Diedrich
2011-06-14 21:14   ` [Xen-users] 3.0.0-rc2: Xen: High amount of kernel "reserved" memory, about 33% in 256MB DOMU Tobias Diedrich

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=20110614194810.GA26728@dumpdata.com \
    --to=konrad.wilk@oracle.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=ranma+xen@tdiedrich.de \
    --cc=xen-users@lists.xensource.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 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.