From: David Vrabel <david.vrabel@citrix.com>
To: xen-devel@lists.xenproject.org
Cc: Keir Fraser <keir@xen.org>, Tim Deegan <tim@xen.org>,
David Vrabel <david.vrabel@citrix.com>,
Jan Beulich <jbeulich@suse.com>,
Ian Campbell <ian.campbell@citrix.com>
Subject: [PATCHv2 0/5] evtchn: Improve scalebility
Date: Mon, 15 Jun 2015 16:48:14 +0100 [thread overview]
Message-ID: <1434383299-21833-1-git-send-email-david.vrabel@citrix.com> (raw)
The per-domain event channel lock limits scalability when many VCPUs
are trying to send interdomain events. A per-channel lock is
introduced eliminating any lock contention when sending an event.
See this graph for the performance improvements:
http://xenbits.xen.org/people/dvrabel/evtchn-scalability.png
A different test (using Linux's evtchn device which masks/unmasks
event channels) showed the following lock profile improvements:
Per-domain lock:
(XEN) lock: 69267976(00000004:19830041), block: 27777407(00000002:3C7C5C96)
Per-event channel lock
(XEN) lock: 686530(00000000:076AF5F6), block: 1787(00000000:000B4D22)
Locking removed from evtchn_unmask():
(XEN) lock: 10769(00000000:00512999), block: 99(00000000:00009491)
v2:
- Use unsigned int for d->valid_evtchns.
- Compare channel pointers in double_evtchn_lock().
David
next reply other threads:[~2015-06-15 16:45 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-06-15 15:48 David Vrabel [this message]
2015-06-15 15:48 ` [PATCHv2 1/5] evtchn: factor out freeing an event channel David Vrabel
2015-06-15 15:48 ` [PATCHv2 2/5] evtchn: simplify port_is_valid() David Vrabel
2015-06-15 15:57 ` Ian Campbell
2015-06-15 15:59 ` David Vrabel
2015-06-15 16:09 ` Ian Campbell
2015-06-15 15:48 ` [PATCHv2 3/5] evtchn: use a per-event channel lock for sending events David Vrabel
2015-06-16 9:18 ` Jan Beulich
2015-06-16 9:34 ` David Vrabel
2015-06-16 9:51 ` Jan Beulich
2015-06-16 9:57 ` David Vrabel
2015-06-16 15:19 ` David Vrabel
2015-06-16 15:58 ` David Vrabel
2015-06-16 16:19 ` Jan Beulich
2015-06-16 16:39 ` David Vrabel
2015-06-17 7:05 ` Jan Beulich
2015-06-16 16:06 ` Jan Beulich
2015-06-15 15:48 ` [PATCHv2 4/5] evtchn: remove the locking when unmasking an event channel David Vrabel
2015-06-16 9:19 ` Jan Beulich
2015-06-16 9:40 ` David Vrabel
2015-06-15 15:48 ` [PATCHv2 5/5] evtchn: pad struct evtchn to 64 bytes David Vrabel
2015-06-16 9:22 ` Jan Beulich
2015-06-16 11:14 ` Julien Grall
2015-06-16 11:59 ` Jan Beulich
2015-06-16 12:57 ` Julien Grall
2015-06-16 13:13 ` David Vrabel
2015-06-16 13:27 ` Julien Grall
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=1434383299-21833-1-git-send-email-david.vrabel@citrix.com \
--to=david.vrabel@citrix.com \
--cc=ian.campbell@citrix.com \
--cc=jbeulich@suse.com \
--cc=keir@xen.org \
--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 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.