From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
To: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>,
Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Thomas Gleixner <tglx@linutronix.de>,
"H. Peter Anvin" <hpa@zytor.com>,
Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com>,
Ingo Molnar <mingo@elte.hu>,
Linus Torvalds <torvalds@linux-foundation.org>,
Peter Zijlstra <peterz@infradead.org>,
the arch/x86 maintainers <x86@kernel.org>,
LKML <linux-kernel@vger.kernel.org>, Avi Kivity <avi@redhat.com>,
Marcelo Tosatti <mtosatti@redhat.com>, KVM <kvm@vger.kernel.org>,
Andi Kleen <andi@firstfloor.org>,
Xen Devel <xen-devel@lists.xensource.com>,
Virtualization <virtualization@lists.linux-foundation.org>,
Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>,
Stephan Diestelhorst <stephan.diestelhorst@amd.com>,
Stefano Stabellini <stefano.stabellini@eu.citrix.com>,
Attilio Rao <attilio.rao@citrix.com>
Subject: Re: [PATCH RFC V6 0/11] Paravirtualized ticketlocks
Date: Mon, 16 Apr 2012 11:44:29 -0400 [thread overview]
Message-ID: <20120416154429.GB4654@phenom.dumpdata.com> (raw)
In-Reply-To: <20120331040745.GC14030@linux.vnet.ibm.com>
On Sat, Mar 31, 2012 at 09:37:45AM +0530, Srivatsa Vaddagiri wrote:
> * Thomas Gleixner <tglx@linutronix.de> [2012-03-31 00:07:58]:
>
> > I know that Peter is going to go berserk on me, but if we are running
> > a paravirt guest then it's simple to provide a mechanism which allows
> > the host (aka hypervisor) to check that in the guest just by looking
> > at some global state.
> >
> > So if a guest exits due to an external event it's easy to inspect the
> > state of that guest and avoid to schedule away when it was interrupted
> > in a spinlock held section. That guest/host shared state needs to be
> > modified to indicate the guest to invoke an exit when the last nested
> > lock has been released.
>
> I had attempted something like that long back:
>
> http://lkml.org/lkml/2010/6/3/4
>
> The issue is with ticketlocks though. VCPUs could go into a spin w/o
> a lock being held by anybody. Say VCPUs 1-99 try to grab a lock in
> that order (on a host with one cpu). VCPU1 wins (after VCPU0 releases it)
> and releases the lock. VCPU1 is next eligible to take the lock. If
> that is not scheduled early enough by host, then remaining vcpus would keep
> spinning (even though lock is technically not held by anybody) w/o making
> forward progress.
>
> In that situation, what we really need is for the guest to hint to host
> scheduler to schedule VCPU1 early (via yield_to or something similar).
>
> The current pv-spinlock patches however does not track which vcpu is
> spinning at what head of the ticketlock. I suppose we can consider
> that optimization in future and see how much benefit it provides (over
> plain yield/sleep the way its done now).
Right. I think Jeremy played around with this some time?
>
> Do you see any issues if we take in what we have today and address the
> finer-grained optimization as next step?
I think that is the proper course - these patches show
that on baremetal we don't incur performance regressions and in
virtualization case we benefit greatly. Since these are the basic
building blocks of a kernel - taking it slow and just adding
this set of patches for v3.5 is a good idea - and then building on top
of that for further refinement.
>
> - vatsa
next prev parent reply other threads:[~2012-04-16 15:50 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-03-21 10:20 [PATCH RFC V6 0/11] Paravirtualized ticketlocks Raghavendra K T
2012-03-21 10:20 ` [PATCH RFC V6 1/11] x86/spinlock: replace pv spinlocks with pv ticketlocks Raghavendra K T
2012-03-21 13:04 ` Attilio Rao
2012-03-21 13:22 ` Stephan Diestelhorst
2012-03-21 13:49 ` Attilio Rao
2012-03-21 14:25 ` Stephan Diestelhorst
2012-03-21 14:33 ` Attilio Rao
2012-03-21 14:49 ` Raghavendra K T
2012-03-21 10:21 ` [PATCH RFC V6 2/11] x86/ticketlock: don't inline _spin_unlock when using paravirt spinlocks Raghavendra K T
2012-03-21 17:13 ` Linus Torvalds
2012-03-22 10:06 ` Raghavendra K T
2012-03-21 10:21 ` [PATCH RFC V6 3/11] x86/ticketlock: collapse a layer of functions Raghavendra K T
2012-03-21 10:21 ` [PATCH RFC V6 4/11] xen: defer spinlock setup until boot CPU setup Raghavendra K T
2012-03-21 10:21 ` [PATCH RFC V6 5/11] xen/pvticketlock: Xen implementation for PV ticket locks Raghavendra K T
2012-03-21 10:21 ` [PATCH RFC V6 6/11] xen/pvticketlocks: add xen_nopvspin parameter to disable xen pv ticketlocks Raghavendra K T
2012-03-21 10:21 ` [PATCH RFC V6 7/11] x86/pvticketlock: use callee-save for lock_spinning Raghavendra K T
2012-03-21 10:22 ` [PATCH RFC V6 8/11] x86/pvticketlock: when paravirtualizing ticket locks, increment by 2 Raghavendra K T
2012-03-21 10:22 ` [PATCH RFC V6 9/11] x86/ticketlock: add slowpath logic Raghavendra K T
2012-03-21 10:22 ` [PATCH RFC V6 10/11] xen/pvticketlock: allow interrupts to be enabled while blocking Raghavendra K T
2012-03-21 10:22 ` [PATCH RFC V6 11/11] xen: enable PV ticketlocks on HVM Xen Raghavendra K T
2012-03-26 14:25 ` [PATCH RFC V6 0/11] Paravirtualized ticketlocks Avi Kivity
2012-03-27 7:37 ` Raghavendra K T
2012-03-28 16:37 ` Alan Meadows
[not found] ` <CAMy5W3foop40+R1RLv_JPhnO5ZmV90uMmNERYY-e3QCeaJfqLw@mail.gmail.com>
2012-03-28 18:21 ` Raghavendra K T
2012-03-29 9:58 ` Avi Kivity
2012-03-29 18:03 ` Raghavendra K T
2012-03-30 10:07 ` Raghavendra K T
2012-04-01 13:18 ` Avi Kivity
2012-04-01 13:48 ` Raghavendra K T
2012-04-01 13:53 ` Avi Kivity
2012-04-01 13:56 ` Raghavendra K T
2012-04-02 9:51 ` Raghavendra K T
2012-04-02 12:15 ` Raghavendra K T
2012-04-05 9:01 ` Avi Kivity
2012-04-05 10:40 ` Raghavendra K T
2012-04-05 8:43 ` Raghavendra K T
2012-03-30 20:26 ` H. Peter Anvin
2012-03-30 22:07 ` Thomas Gleixner
2012-03-30 22:18 ` Andi Kleen
2012-03-30 23:04 ` Thomas Gleixner
2012-03-31 0:08 ` Andi Kleen
2012-03-31 8:11 ` Ingo Molnar
2012-03-31 4:07 ` Srivatsa Vaddagiri
2012-03-31 4:09 ` Srivatsa Vaddagiri
2012-04-16 15:44 ` Konrad Rzeszutek Wilk [this message]
2012-04-16 16:36 ` [Xen-devel] " Ian Campbell
2012-04-16 16:42 ` Jeremy Fitzhardinge
2012-04-17 2:54 ` Srivatsa Vaddagiri
2012-04-01 13:31 ` Avi Kivity
2012-04-02 9:26 ` Thomas Gleixner
2012-04-05 9:15 ` Avi Kivity
2012-04-02 4:36 ` [Xen-devel] " Juergen Gross
2012-04-02 9:42 ` Ian Campbell
2012-04-11 1:29 ` Marcelo Tosatti
2012-03-31 0:51 ` Raghavendra K T
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=20120416154429.GB4654@phenom.dumpdata.com \
--to=konrad.wilk@oracle.com \
--cc=andi@firstfloor.org \
--cc=attilio.rao@citrix.com \
--cc=avi@redhat.com \
--cc=hpa@zytor.com \
--cc=jeremy.fitzhardinge@citrix.com \
--cc=jeremy@goop.org \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=mtosatti@redhat.com \
--cc=peterz@infradead.org \
--cc=raghavendra.kt@linux.vnet.ibm.com \
--cc=stefano.stabellini@eu.citrix.com \
--cc=stephan.diestelhorst@amd.com \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=vatsa@linux.vnet.ibm.com \
--cc=virtualization@lists.linux-foundation.org \
--cc=x86@kernel.org \
--cc=xen-devel@lists.xensource.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).