xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Dario Faggioli <dario.faggioli@citrix.com>,
	Meng Xu <xumengpanda@gmail.com>
Cc: Ian Campbell <ian.campbell@citrix.com>,
	Sisu Xi <xisisu@gmail.com>,
	Stefano Stabellini <stefano.stabellini@eu.citrix.com>,
	George Dunlap <george.dunlap@eu.citrix.com>,
	Ian Jackson <ian.jackson@eu.citrix.com>,
	"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>,
	Meng Xu <mengxu@cis.upenn.edu>, Chong Li <lichong659@gmail.com>,
	Dagaen Golomb <dgolomb@seas.upenn.edu>
Subject: Re: [PATCH RFC v1 4/4] libxc for rt scheduler
Date: Fri, 11 Jul 2014 17:49:14 +0100	[thread overview]
Message-ID: <53C0158A.1050704@citrix.com> (raw)
In-Reply-To: <1405096513.29306.543.camel@Solace>

On 11/07/14 17:35, Dario Faggioli wrote:
>
>> My concern is: Will the former one has worse performance than the
>> later one? ​
>>
> Well, Xen has its tricks, but yes, I think performing e.g., 8 or 16
> hypercalls in a loop is worse than issueing one returning an array.

As far as hypercalls go, the path to Xen is a long one. From userspace,
a hypercall involves a context switch into the kernel (which is a double
context switch through Xen) and a context switch from the kernel into
Xen, and back again. The return from guest kernel to guest userspace
requires a TLB flush.

So, 6 context switches and a TLB flushes per hypercall (assuming a 64bit
dom0).

Use batches wherever possible, even for 2 items. It will be about twice
as fast as not batching.

~Andrew


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

  reply	other threads:[~2014-07-11 16:49 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-11  4:49 Introduce rt real-time scheduler for Xen Meng Xu
2014-07-11  4:49 ` [PATCH RFC v1 1/4] rt: Add rt scheduler to hypervisor Meng Xu
2014-07-11 14:27   ` Dario Faggioli
2014-07-11 14:37   ` Andrew Cooper
2014-07-11 15:21     ` Dario Faggioli
2014-07-11 15:40       ` Andrew Cooper
2014-07-11 15:48         ` Dario Faggioli
2014-07-16 17:05           ` Konrad Rzeszutek Wilk
2014-07-17 10:12             ` Meng Xu
2014-07-17 15:12               ` Dario Faggioli
2014-07-18  5:46                 ` Meng Xu
2014-07-18 18:40               ` Konrad Rzeszutek Wilk
2014-07-11  4:49 ` [PATCH RFC v1 2/4] xl for rt scheduler Meng Xu
2014-07-11 11:02   ` Wei Liu
2014-07-11 14:59     ` Meng Xu
2014-07-11 15:07       ` Dario Faggioli
2014-07-11 16:25         ` Meng Xu
2014-07-13 12:58         ` Meng Xu
2014-07-14  7:40           ` Dario Faggioli
2014-07-14  9:31           ` Wei Liu
2014-07-17 15:39           ` Ian Campbell
2014-07-11  4:49 ` [PATCH RFC v1 3/4] libxl " Meng Xu
2014-07-11 11:05   ` Wei Liu
2014-07-11 15:08   ` Dario Faggioli
2014-07-12 18:16     ` Meng Xu
2014-07-14 10:38       ` Dario Faggioli
2014-07-17 15:34     ` Ian Campbell
2014-07-17 15:36   ` Ian Campbell
2014-07-18 11:05     ` Meng Xu
2014-07-11  4:49 ` [PATCH RFC v1 4/4] libxc " Meng Xu
2014-07-11 14:49   ` Dario Faggioli
2014-07-11 16:23     ` Meng Xu
2014-07-11 16:35       ` Dario Faggioli
2014-07-11 16:49         ` Andrew Cooper [this message]
2014-07-12 19:46         ` Meng Xu
2014-07-17 15:29     ` Ian Campbell
2014-07-17 15:34       ` George Dunlap
2014-07-17 22:16         ` Meng Xu
2014-07-18  9:49           ` Dario Faggioli
2014-07-18  9:51           ` Ian Campbell
2014-07-18 12:11             ` Meng Xu
2014-07-18  9:47         ` Ian Campbell
2014-07-18 10:00           ` Dario Faggioli
2014-07-11 10:50 ` Introduce rt real-time scheduler for Xen Wei Liu
2014-07-11 11:06   ` Dario Faggioli
2014-07-11 16:14     ` Meng Xu
2014-07-11 16:19 ` Dario Faggioli

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=53C0158A.1050704@citrix.com \
    --to=andrew.cooper3@citrix.com \
    --cc=dario.faggioli@citrix.com \
    --cc=dgolomb@seas.upenn.edu \
    --cc=george.dunlap@eu.citrix.com \
    --cc=ian.campbell@citrix.com \
    --cc=ian.jackson@eu.citrix.com \
    --cc=lichong659@gmail.com \
    --cc=mengxu@cis.upenn.edu \
    --cc=stefano.stabellini@eu.citrix.com \
    --cc=xen-devel@lists.xen.org \
    --cc=xisisu@gmail.com \
    --cc=xumengpanda@gmail.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).