xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Dario Faggioli <dario.faggioli@citrix.com>
To: Andrew Cooper <andrew.cooper3@citrix.com>,
	George Dunlap <george.dunlap@citrix.com>,
	xen-devel@lists.xenproject.org
Cc: George Dunlap <george.dunlap@eu.citrix.com>,
	Meng Xu <mengxu@cis.upenn.edu>
Subject: Re: [PATCH 3/9] xen: sched: make locking for {insert, remove}_vcpu consistent
Date: Thu, 8 Oct 2015 22:44:37 +0200	[thread overview]
Message-ID: <1444337077.22254.78.camel@citrix.com> (raw)
In-Reply-To: <5616A6A2.2090102@citrix.com>


[-- Attachment #1.1: Type: text/plain, Size: 1675 bytes --]

On Thu, 2015-10-08 at 18:23 +0100, Andrew Cooper wrote:
> On 08/10/15 17:46, George Dunlap wrote:
> > On 08/10/15 16:20, Andrew Cooper wrote:

> > > There are a lot of codepaths, but I cant see one which is
> > > definitely
> > > called with interrupts disables.  (OTOH, I can see several where
> > > interrupts are definitely enabled).
> > Oh, I think I misunderstood you.  You meant, "No codepaths *calling
> > into* generic scheduling code should need interrupts disabled at
> > all".
> > I can certainly believe that to be true in most cases; there's no
> > sense
> > in saving the flags if we don't need to.
> 
> My original statement came from the observation that schedule() runs
> with interrupts disabled, and takes between 2.2 and 4 microseconds to
> run (as measured during the ticketlock performance analysis).
> 
Ok, this part, I got, I think.

BTW, do you still have the scripts/harness/whatever to perform such
analysis? I'm asking because I'm about to give it a try at reducing the
amount of time during which the scheduler keeps interrupt disabled.

If I manage, it'd be nice to retry it, an get updated numbers.

> I am going to go out on a limb and say that the majority of that time
> does not need to be spent with interrupts disabled.  I might easily
> be
> wrong, but I suspect I am not.
> 
I'll have a look and let you know what I find.

Regards,
Dario
-- 
<<This happens because I choose it to happen!>> (Raistlin Majere)
-----------------------------------------------------------------
Dario Faggioli, Ph.D, http://about.me/dario.faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)


[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

[-- Attachment #2: Type: text/plain, Size: 126 bytes --]

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

  reply	other threads:[~2015-10-08 20:44 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-09-29 16:55 [PATCH 0/9] xen: sched: improve (a lot! :-D) Credit2 runqueue handling Dario Faggioli
2015-09-29 16:55 ` [PATCH 1/9] xen: sched: fix an 'off by one \t' in credit2 debug dump Dario Faggioli
2015-10-01  5:22   ` Juergen Gross
2015-10-08 14:09   ` George Dunlap
2015-09-29 16:55 ` [PATCH 2/9] xen: sched: improve scope and placement of credit2 boot parameters Dario Faggioli
2015-10-01  5:23   ` Juergen Gross
2015-10-01  7:51   ` Jan Beulich
2015-10-01  8:17     ` Dario Faggioli
2015-09-29 16:55 ` [PATCH 3/9] xen: sched: make locking for {insert, remove}_vcpu consistent Dario Faggioli
2015-09-29 17:31   ` Andrew Cooper
2015-09-29 21:40     ` Dario Faggioli
2015-09-29 21:56       ` Dario Faggioli
2015-09-30  9:00       ` Andrew Cooper
2015-10-08 14:58     ` George Dunlap
2015-10-08 15:20       ` Andrew Cooper
2015-10-08 16:46         ` George Dunlap
2015-10-08 17:23           ` Andrew Cooper
2015-10-08 20:44             ` Dario Faggioli [this message]
2015-10-12  9:44             ` George Dunlap
2015-10-08 20:39         ` Dario Faggioli
2015-10-09 13:05           ` Andrew Cooper
2015-10-09 16:56             ` Dario Faggioli
2015-10-01  8:03   ` Jan Beulich
2015-10-01 11:59     ` Dario Faggioli
2015-09-29 16:55 ` [PATCH 4/9] xen: sched: add .init_pdata hook to the scheduler interface Dario Faggioli
2015-10-01  5:21   ` Juergen Gross
2015-10-01  6:33     ` Dario Faggioli
2015-10-01  7:43       ` Juergen Gross
2015-10-01  9:32         ` Andrew Cooper
2015-10-01  9:40           ` Dario Faggioli
2015-10-01  8:17   ` Jan Beulich
2015-10-01  9:26     ` Dario Faggioli
2015-10-01 10:12       ` Jan Beulich
2015-10-01 10:35         ` Dario Faggioli
2015-10-01 10:47           ` Jan Beulich
2015-09-29 16:56 ` [PATCH 5/9] xen: sched: make implementing .alloc_pdata optional Dario Faggioli
2015-10-01  5:28   ` Juergen Gross
2015-10-01  6:35     ` Dario Faggioli
2015-10-01  7:49   ` Jan Beulich
2015-10-01  8:13     ` Dario Faggioli
2015-09-29 16:56 ` [PATCH 6/9] xen: sched: implement .init_pdata in all schedulers Dario Faggioli
2015-09-29 16:56 ` [PATCH 7/9] xen: sched: fix per-socket runqueue creation in credit2 Dario Faggioli
2015-09-29 16:56 ` [PATCH 8/9] xen: sched: allow for choosing credit2 runqueues configuration at boot Dario Faggioli
2015-10-01  5:48   ` Juergen Gross
2015-10-01  7:23     ` Dario Faggioli
2015-10-01  7:46       ` Juergen Gross
2015-09-29 16:56 ` [PATCH 9/9] xen: sched: per-core runqueues as default in credit2 Dario Faggioli
2015-10-01  5:48   ` Juergen Gross

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=1444337077.22254.78.camel@citrix.com \
    --to=dario.faggioli@citrix.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=george.dunlap@citrix.com \
    --cc=george.dunlap@eu.citrix.com \
    --cc=mengxu@cis.upenn.edu \
    --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 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).