xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Dario Faggioli <dfaggioli@suse.com>
To: Andrew Cooper <andrew.cooper3@citrix.com>,
	Jan Beulich <JBeulich@suse.com>
Cc: Dario Faggioli <dario.faggioli@citrix.com>,
	David Woodhouse <dwmw@amazon.co.uk>,
	Xen-devel <xen-devel@lists.xen.org>
Subject: Re: [PATCH v10 09/11] x86/ctxt: Issue a speculation barrier between vcpu contexts
Date: Thu, 25 Jan 2018 19:49:09 +0100	[thread overview]
Message-ID: <1516906149.15341.66.camel@suse.com> (raw)
In-Reply-To: <345bd2ef-1319-d9a6-522c-31e42bcd2c2e@citrix.com>


[-- Attachment #1.1: Type: text/plain, Size: 1594 bytes --]

On Thu, 2018-01-25 at 16:09 +0000, Andrew Cooper wrote:
> On 25/01/18 15:57, Jan Beulich wrote:
> > > > > 
> > For the record, the overwhelming majority of calls to
> > __sync_local_execstate() being responsible for the behavior
> > come from invalidate_interrupt(), which suggests to me that
> > there's a meaningful number of cases where a vCPU is migrated
> > to another CPU and then back, without another vCPU having
> > run on the original CPU in between. If I'm not wrong with this,
> > I have to question why the vCPU is migrated then in the first
> > place.
> 
> This is a known (mis)feature of the Credit scheduler.  When a PCPU
> goes
> idle, it aggressively steals work from the adjacent cpus, even if the
> adjacent cpu only has a single vcpu to run and is running it.
> 
> XenServer has this gross hack which makes aggregate networking
> measurements far far better
> 
> https://github.com/xenserver/xen.pg/blob/XS-7.1.x/master/sched-credit
> 1-use-per-pcpu-runqueue-count.patch
> 
> Dario made a different fix to Credit1 upstream which was supposed to
> resolve this behaviour (although I can't locate the patch by a list
> of
> titles), but based on these observations, I'd say the misfeature
> hasn't
> been fixed.
> 
Yes, it's 341450eaf753 ("xen: credit1: increase efficiency and
scalability of load balancing."), and that commit and the XenServer
patch are functionally equivalent.

So, I'm not convinced about that being the actual cause of the
described behaviour. Tracing would be the way to tell (hopefully) for
sure.

Regards,
Dario

[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

[-- Attachment #2: Type: text/plain, Size: 157 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

  parent reply	other threads:[~2018-01-25 18:49 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-01-24 13:12 [PATCH v10 00/11] x86: Mitigations for SP2/CVE-2017-5715/Branch Target Injection Andrew Cooper
2018-01-24 13:12 ` [PATCH v10 01/11] x86/cpuid: Handling of IBRS/IBPB, STIBP and IBRS for guests Andrew Cooper
2018-02-01  9:06   ` Jan Beulich
2018-02-01 13:53     ` Andrew Cooper
2018-01-24 13:12 ` [PATCH v10 02/11] x86/msr: Emulation of MSR_{SPEC_CTRL, PRED_CMD} " Andrew Cooper
2018-01-25 12:25   ` Jan Beulich
2018-01-24 13:12 ` [PATCH v10 03/11] x86/migrate: Move MSR_SPEC_CTRL on migrate Andrew Cooper
2018-01-24 13:12 ` [PATCH v10 04/11] x86/hvm: Permit guests direct access to MSR_{SPEC_CTRL, PRED_CMD} Andrew Cooper
2018-01-24 13:12 ` [PATCH v10 05/11] x86/entry: Organise the use of MSR_SPEC_CTRL at each entry/exit point Andrew Cooper
2018-01-25 13:08   ` Jan Beulich
2018-01-25 14:12     ` Andrew Cooper
2018-01-25 14:36       ` Jan Beulich
2018-01-25 14:46         ` Andrew Cooper
2018-01-25 15:08           ` Jan Beulich
2018-01-25 15:10             ` Andrew Cooper
2018-01-25 16:52   ` [PATCH v11 5/11] " Andrew Cooper
2018-01-24 13:12 ` [PATCH v10 06/11] x86/entry: Organise the clobbering of the RSB/RAS on entry to Xen Andrew Cooper
2018-01-25 13:19   ` Jan Beulich
2018-01-25 14:17     ` Andrew Cooper
2018-01-25 14:40       ` Jan Beulich
2018-01-25 14:44         ` Andrew Cooper
2018-01-25 14:48           ` Jan Beulich
2018-01-25 16:54   ` [PATCH v11 6/11] " Andrew Cooper
2018-01-26 12:17     ` Jan Beulich
2018-01-24 13:12 ` [PATCH v10 07/11] x86/entry: Avoid using alternatives in NMI/#MC paths Andrew Cooper
2018-01-25 13:43   ` Jan Beulich
2018-01-25 15:04     ` Andrew Cooper
2018-01-25 15:14       ` Jan Beulich
2018-01-25 15:19         ` Andrew Cooper
2018-01-25 16:17           ` Jan Beulich
2018-01-25 17:21   ` [PATCH v11 7/11] " Andrew Cooper
2018-01-26 12:23     ` Jan Beulich
2018-01-26 12:28       ` Andrew Cooper
2018-01-24 13:12 ` [PATCH v10 08/11] x86/boot: Calculate the most appropriate BTI mitigation to use Andrew Cooper
2018-01-25 13:52   ` Jan Beulich
2018-02-01  8:41   ` Jan Beulich
2018-02-01 13:58     ` Andrew Cooper
2018-01-24 13:12 ` [PATCH v10 09/11] x86/ctxt: Issue a speculation barrier between vcpu contexts Andrew Cooper
2018-01-24 13:34   ` Woodhouse, David
2018-01-24 13:49     ` Andrew Cooper
2018-01-24 14:31       ` David Woodhouse
2018-01-25 14:46         ` Konrad Rzeszutek Wilk
2018-01-25 15:57   ` Jan Beulich
2018-01-25 16:09     ` Andrew Cooper
2018-01-25 16:15       ` Andrew Cooper
2018-01-27  1:27         ` Dario Faggioli
2018-01-29  9:28           ` Jan Beulich
2018-02-05 11:37             ` George Dunlap
2018-01-25 16:31       ` Jan Beulich
2018-01-25 16:48         ` Andrew Cooper
2018-01-25 18:49       ` Dario Faggioli [this message]
2018-01-26  1:08         ` Dario Faggioli
2018-01-26  9:43           ` Jan Beulich
2018-01-26 11:13             ` Dario Faggioli
2018-01-26 11:38               ` Jan Beulich
2018-01-24 13:12 ` [PATCH v10 10/11] x86/cpuid: Offer Indirect Branch Controls to guests Andrew Cooper
2018-01-24 13:12 ` [PATCH v10 11/11] x86/idle: Clear SPEC_CTRL while idle Andrew Cooper

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=1516906149.15341.66.camel@suse.com \
    --to=dfaggioli@suse.com \
    --cc=JBeulich@suse.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=dario.faggioli@citrix.com \
    --cc=dwmw@amazon.co.uk \
    --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 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).