From: Gleb Natapov <gleb@redhat.com>
To: Andi Kleen <andi@firstfloor.org>
Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
avi@redhat.com, hpa@zytor.com, mingo@elte.hu, npiggin@suse.de,
tglx@linutronix.de, mtosatti@redhat.com
Subject: Re: [PATCH] use unfair spinlock when running on hypervisor.
Date: Tue, 1 Jun 2010 19:24:14 +0300 [thread overview]
Message-ID: <20100601162414.GA6191@redhat.com> (raw)
In-Reply-To: <87sk56ycka.fsf@basil.nowhere.org>
On Tue, Jun 01, 2010 at 05:53:09PM +0200, Andi Kleen wrote:
> Gleb Natapov <gleb@redhat.com> writes:
> >
> > The patch below allows to patch ticket spinlock code to behave similar to
> > old unfair spinlock when hypervisor is detected. After patching unlocked
>
> The question is what happens when you have a system with unfair
> memory and you run the hypervisor on that. There it could be much worse.
>
How much worse performance hit could be?
> Your new code would starve again, right?
>
Yes, of course it may starve with unfair spinlock. Since vcpus are not
always running there is much smaller chance then vcpu on remote memory
node will starve forever. Old kernels with unfair spinlocks are running
fine in VMs on NUMA machines with various loads.
> There's a reason the ticket spinlocks were added in the first place.
>
I understand that reason and do not propose to get back to old spinlock
on physical HW! But with virtualization performance hit is unbearable.
--
Gleb.
next prev parent reply other threads:[~2010-06-01 16:24 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-06-01 9:35 [PATCH] use unfair spinlock when running on hypervisor Gleb Natapov
2010-06-01 15:53 ` Andi Kleen
2010-06-01 16:24 ` Gleb Natapov [this message]
2010-06-01 16:38 ` Andi Kleen
2010-06-01 16:52 ` Avi Kivity
2010-06-01 17:27 ` Andi Kleen
2010-06-02 2:51 ` Avi Kivity
2010-06-02 5:26 ` Srivatsa Vaddagiri
2010-06-02 8:50 ` Andi Kleen
2010-06-02 9:00 ` Avi Kivity
2010-06-03 4:20 ` Srivatsa Vaddagiri
2010-06-03 4:51 ` Eric Dumazet
2010-06-03 5:38 ` Srivatsa Vaddagiri
2010-06-03 8:52 ` Andi Kleen
2010-06-03 9:26 ` Srivatsa Vaddagiri
2010-06-03 10:22 ` Nick Piggin
2010-06-03 10:38 ` Nick Piggin
2010-06-03 12:04 ` Srivatsa Vaddagiri
2010-06-03 12:38 ` Nick Piggin
2010-06-03 12:58 ` Srivatsa Vaddagiri
2010-06-03 13:04 ` Srivatsa Vaddagiri
2010-06-03 13:45 ` Nick Piggin
2010-06-03 14:48 ` Srivatsa Vaddagiri
2010-06-03 15:17 ` Andi Kleen
2010-06-03 15:35 ` Nick Piggin
2010-06-03 17:25 ` Andi Kleen
2010-06-01 17:39 ` Valdis.Kletnieks
2010-06-02 2:46 ` Avi Kivity
2010-06-02 7:39 ` H. Peter Anvin
2010-06-01 17:54 ` john cooper
2010-06-01 19:36 ` Andi Kleen
2010-06-03 11:06 ` David Woodhouse
2010-06-03 15:15 ` Andi Kleen
2010-06-01 21:39 ` Eric Dumazet
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=20100601162414.GA6191@redhat.com \
--to=gleb@redhat.com \
--cc=andi@firstfloor.org \
--cc=avi@redhat.com \
--cc=hpa@zytor.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=mtosatti@redhat.com \
--cc=npiggin@suse.de \
--cc=tglx@linutronix.de \
/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).