[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20110120143430.GA17815@linux.vnet.ibm.com>
Date: Thu, 20 Jan 2011 20:04:30 +0530
From: Srivatsa Vaddagiri <vatsa@...ux.vnet.ibm.com>
To: Peter Zijlstra <peterz@...radead.org>
Cc: Jeremy Fitzhardinge <jeremy@...p.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
Nick Piggin <npiggin@...nel.dk>,
Mathieu Desnoyers <mathieu.desnoyers@...ymtl.ca>,
Américo Wang <xiyou.wangcong@...il.com>,
Eric Dumazet <dada1@...mosbay.com>,
Jan Beulich <JBeulich@...ell.com>, Avi Kivity <avi@...hat.com>,
Xen-devel <xen-devel@...ts.xensource.com>,
"H. Peter Anvin" <hpa@...or.com>,
Linux Virtualization <virtualization@...ts.linux-foundation.org>,
Jeremy Fitzhardinge <jeremy.fitzhardinge@...rix.com>,
kvm@...r.kernel.org, suzuki@...ibm.com
Subject: Re: [PATCH 2/3] kvm hypervisor : Add hypercalls to support
pv-ticketlock
On Thu, Jan 20, 2011 at 02:41:46PM +0100, Peter Zijlstra wrote:
> On Thu, 2011-01-20 at 17:29 +0530, Srivatsa Vaddagiri wrote:
> >
> > If we had a yield-to [1] sort of interface _and_ information on which vcpu
> > owns a lock, then lock-spinners can yield-to the owning vcpu,
>
> and then I'd nak it for being stupid ;-)
>
> really, yield*() is retarded, never even consider using it. If you've
> got the actual owner you can do full blown PI, which is tons better than
> a 'do-something-random' call.
Yes definitely that would be much better than yield-to.
> The only reason the whole non-virt pause loop filtering muck uses it is
> because it really doesn't know anything, and do-something is pretty much
> all it can do. Its a broken interface.
- vatsa
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists