This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
Home Products Support Community News


[Xen-devel] Re: [PATCH 13/14] x86/ticketlock: add slowpath logic

To: Jeremy Fitzhardinge <jeremy@xxxxxxxx>
Subject: [Xen-devel] Re: [PATCH 13/14] x86/ticketlock: add slowpath logic
From: Avi Kivity <avi@xxxxxxxxxx>
Date: Wed, 17 Nov 2010 11:10:48 +0200
Cc: Jeremy Fitzhardinge <jeremy.fitzhardinge@xxxxxxxxxx>, Mathieu Desnoyers <mathieu.desnoyers@xxxxxxxxxx>, Nick Piggin <npiggin@xxxxxxxxx>, Peter Zijlstra <peterz@xxxxxxxxxxxxx>, Linux Kernel Mailing List <linux-kernel@xxxxxxxxxxxxxxx>, Srivatsa Vaddagiri <vatsa@xxxxxxxxxxxxxxxxxx>, Jan Beulich <JBeulich@xxxxxxxxxx>, Eric Dumazet <dada1@xxxxxxxxxxxxx>, Xen-devel <xen-devel@xxxxxxxxxxxxxxxxxxx>, "H. Peter Anvin" <hpa@xxxxxxxxx>, xiyou.wangcong@xxxxxxxxx, Linux Virtualization <virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx>
Delivery-date: Wed, 17 Nov 2010 01:11:51 -0800
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <4CE39AD2.9080606@xxxxxxxx>
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
References: <cover.1289940821.git.jeremy.fitzhardinge@xxxxxxxxxx><cover.1289940821.git.jeremy.fitzhardinge@xxxxxxxxxx> <97ed99ae9160bdb6477284b333bd6708fb7a19cb.1289940821.git.jeremy.fitzhardinge@xxxxxxxxxx> <4CE3A1060200007800022B04@xxxxxxxxxxxxxxxxxx> <4CE397C8.9040901@xxxxxxxx> <4CE3992E.2070406@xxxxxxxxxx> <4CE39AD2.9080606@xxxxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv: Gecko/20101103 Fedora/1.0-0.33.b2pre.fc14 Lightning/1.0b3pre Thunderbird/3.1.6
On 11/17/2010 11:05 AM, Jeremy Fitzhardinge wrote:
On 11/17/2010 12:58 AM, Avi Kivity wrote:
>>  Actually in this case I'm pretty sure there's already a "set bit"
>>  function which will do the job.  set_bit(), I guess, though it takes a
>>  bit number rather than a mask...
>  set_bit() operates on a long, while the intel manuals recommend
>  against operating on operands of different size, especially with
>  locked operations.  I think newer processors have more relaxed
>  requirements, though.

Despite its prototype, set_bit() is pretty specifically using "orb" for
a the constant case, or bts otherwise (I don't know what size memory
operation bts is considered to generate).

Perhaps that should be fixed.

bts will take its size from the argument, so it will be a btsq on x86_64. AFAICT, the only visible difference between btsl and btsq is a page fault if the last four bytes of the operand are in an unmapped page.

error compiling committee.c: too many arguments to function

Xen-devel mailing list