From: Mike Latimer <mlatimer@suse.com>
To: xen-devel@lists.xen.org
Subject: freemem-slack and large memory environments
Date: Mon, 09 Feb 2015 18:27:54 -0700 [thread overview]
Message-ID: <4321015.nah3j6dvJq@mlatimer1.dnsdhcp.provo.novell.com> (raw)
Hi,
While testing commit 2563bca1, I found that libxl_get_free_memory returns 0
until there is more free memory than required for freemem-slack. This means
that during the domain creation process, freed memory is first set aside for
freemem-slack, then marked as truly free for consumption.
On machines with large amounts of memory, freemem-slack can be very high (26GB
on a 2TB test machine). If freeing this memory takes more time than allowed
during domain startup, domain creation fails with ERROR_NOMEM. (Commit
2563bca1 doesn't help here, as free_memkb remains 0 until freemem-slack is
satisfied.)
There is already a 15% limit on the size of freemem-slack (commit a39b5bc6),
but this does not take into consideration very large memory environments.
(26GB is only 1.2% of 2TB), where this limit is not hit.
It seems that there are two approaches to resolve this:
- Introduce a hard limit on freemem-slack to avoid unnecessarily large
reservations
- Increase the retry count during domain creation to ensure enough time is
set aside for any cycles spent freeing memory for freemem-slack (on the test
machine, doubling the retry count to 6 is the minimum required)
Which is the best approach (or did I miss something)?
Thanks!
Mike
next reply other threads:[~2015-02-10 1:27 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-02-10 1:27 Mike Latimer [this message]
2015-02-10 21:34 ` freemem-slack and large memory environments Mike Latimer
2015-02-13 11:13 ` Wei Liu
2015-02-13 23:16 ` Mike Latimer
2015-02-18 14:10 ` Ian Campbell
2015-02-24 16:06 ` Stefano Stabellini
2015-02-24 16:54 ` Ian Campbell
2015-02-25 11:39 ` Stefano Stabellini
2015-02-25 12:00 ` Ian Campbell
2015-02-25 14:03 ` Ian Campbell
2015-02-25 14:09 ` Stefano Stabellini
2015-02-26 15:36 ` Mike Latimer
2015-02-26 15:57 ` Ian Campbell
2015-02-26 17:38 ` Mike Latimer
2015-02-26 17:47 ` Ian Campbell
2015-02-26 20:38 ` Mike Latimer
2015-02-27 10:17 ` Ian Campbell
2015-02-27 11:05 ` Stefano Stabellini
2015-02-26 17:53 ` Stefano Stabellini
2015-02-26 20:45 ` Mike Latimer
2015-02-26 23:30 ` Mike Latimer
2015-02-27 10:21 ` Ian Campbell
2015-02-27 10:52 ` Stefano Stabellini
2015-02-27 15:28 ` Mike Latimer
2015-02-27 18:29 ` Mike Latimer
2015-02-28 0:31 ` Mike Latimer
2015-03-02 10:12 ` Stefano Stabellini
2015-03-02 10:44 ` Jan Beulich
2015-03-02 12:13 ` Stefano Stabellini
2015-03-02 13:04 ` Jan Beulich
[not found] ` <54F46DDB020000780006505B@suse.com>
2015-03-02 22:49 ` Mike Latimer
2015-03-02 11:42 ` Ian Campbell
2015-03-02 15:19 ` Stefano Stabellini
2015-03-02 16:04 ` Ian Campbell
2015-03-02 16:15 ` Stefano Stabellini
2015-03-02 22:49 ` Mike Latimer
2015-03-03 10:02 ` Ian Campbell
2015-03-03 10:32 ` Stefano Stabellini
2015-03-03 10:40 ` Stefano Stabellini
2015-02-27 8:24 ` Jan Beulich
2015-02-27 10:52 ` Stefano Stabellini
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=4321015.nah3j6dvJq@mlatimer1.dnsdhcp.provo.novell.com \
--to=mlatimer@suse.com \
--cc=xen-devel@lists.xen.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.