From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756276AbZERHTU (ORCPT ); Mon, 18 May 2009 03:19:20 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755902AbZERHTF (ORCPT ); Mon, 18 May 2009 03:19:05 -0400 Received: from vpn.id2.novell.com ([195.33.99.129]:19269 "EHLO vpn.id2.novell.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755917AbZERHTD convert rfc822-to-8bit (ORCPT ); Mon, 18 May 2009 03:19:03 -0400 Message-Id: <4A11280402000078000014E2@vpn.id2.novell.com> X-Mailer: Novell GroupWise Internet Agent 8.0.0 Date: Mon, 18 May 2009 08:19:00 +0100 From: "Jan Beulich" To: "Jeremy Fitzhardinge" Cc: "Ingo Molnar" , "Jun Nakajima" , "Xiaohui Xin" , "Xin Li" , "Xen-devel" , "Nick Piggin" , "Linux Kernel Mailing List" , "H. Peter Anvin" Subject: Re: [Xen-devel] Performance overhead of paravirt_ops on nativeidentified References: <4A0B62F7.5030802@goop.org> <4A0BED040200007800000DB0@vpn.id2.novell.com> <4A0C58BB.3090303@goop.org> <4A0D3F8C02000078000010A7@vpn.id2.novell.com> <4A0DB988.6000009@goop.org> In-Reply-To: <4A0DB988.6000009@goop.org> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 8BIT Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org >>> Jeremy Fitzhardinge 15.05.09 20:50 >>> >Jan Beulich wrote: >> A patch for the pv-ops kernel would require some time. What I can give you >> right away - just for reference - are the sources we currently use in our kernel: >> attached. > >Hm, I see. Putting a call out to a pv-ops function in the ticket lock >slow path looks pretty straightforward. The need for an extra lock on >the contended unlock side is a bit unfortunate; have you measured to see >what hit that has? Seems to me like you could avoid the problem by >using per-cpu storage rather than stack storage (though you'd need to >copy the per-cpu data to stack when handling a nested spinlock). Not sure how you'd imagine this to work: The unlock code has to look at all cpus' data in either case, so an inner lock would still be required imo. >What's the thinking behind the xen_spin_adjust() stuff? That's the placeholder for implementing interrupt re-enabling in the irq-save lock path. The requirement is that if a nested lock attempt hits the same lock on the same cpu that it failed to get acquired on earlier (but got a ticket already), tickets for the given (lock, cpu) pair need to be circularly shifted around so that the innermost requestor gets the earliest ticket. This is what that function's job will become if I ever get to implement this. >> static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock) { >> unsigned int token, count; bool free; __ticket_spin_lock_preamble; if >> (unlikely(!free)) token = xen_spin_adjust(lock, token); do { count = 1 >> << 10; __ticket_spin_lock_body; } while (unlikely(!count) && >> !xen_spin_wait(lock, token)); } > >How does this work? Doesn't it always go into the slowpath loop even if >the preamble got the lock with no contention? It indeed always enters the slowpath loop, but only for a single pass through part of its body (the first compare in the body macro will make it exit the loop right away: 'token' is not only the ticket here, but the full lock->slock contents). But yes, I think you're right, one could avoid entering the body altogether by moving the containing loop into the if(!free) body. The logic went through a number of re-writes, so I must have overlooked that opportunity on the last round of adjustments. Jan