From owner-cvs-all  Mon Sep  3 20:45:56 2001
Delivered-To: cvs-all@freebsd.org
Received: from mail.speakeasy.net (mail5.speakeasy.net [216.254.0.205])
	by hub.freebsd.org (Postfix) with ESMTP id 5AF1A37B40F
	for <cvs-all@FreeBSD.org>; Mon,  3 Sep 2001 20:45:05 -0700 (PDT)
Received: (qmail 61525 invoked from network); 4 Sep 2001 03:45:04 -0000
Received: from unknown (HELO laptop.baldwin.cx) ([64.81.54.73]) (envelope-sender <jhb@FreeBSD.org>)
          by mail5.speakeasy.net (qmail-ldap-1.03) with SMTP
          for <bright@mu.org>; 4 Sep 2001 03:45:04 -0000
Message-ID: <XFMail.010903204500.jhb@FreeBSD.org>
X-Mailer: XFMail 1.4.0 on FreeBSD
X-Priority: 3 (Normal)
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 8bit
MIME-Version: 1.0
In-Reply-To: <20010903195424.T81307@elvis.mu.org>
Date: Mon, 03 Sep 2001 20:45:00 -0700 (PDT)
From: John Baldwin <jhb@FreeBSD.org>
To: Alfred Perlstein <bright@mu.org>
Subject: Re: cvs commit: src/sys/sparc64/include atomic.h
Cc: cvs-all@FreeBSD.org, cvs-committers@FreeBSD.org,
	Jake Burkholder <jake@FreeBSD.org>
Sender: owner-cvs-all@FreeBSD.ORG
Precedence: bulk
List-ID: <cvs-all.FreeBSD.ORG>
List-Archive: <http://docs.freebsd.org/mail/> (Web Archive)
List-Help: <mailto:majordomo@FreeBSD.ORG?subject=help> (List Instructions)
List-Subscribe: <mailto:majordomo@FreeBSD.ORG?subject=subscribe%20cvs-all>
List-Unsubscribe: <mailto:majordomo@FreeBSD.ORG?subject=unsubscribe%20cvs-all>
X-Loop: FreeBSD.ORG


On 04-Sep-01 Alfred Perlstein wrote:
> * Jake Burkholder <jake@FreeBSD.org> [010903 17:03] wrote:
>> jake        2001/09/03 15:03:25 PDT
>> 
>>   Modified files:
>>     sys/sparc64/include  atomic.h 
>>   Log:
>>   Add atomic_load and store functions without membars, fwiw.
>>   
>>   Revision  Changes    Path
>>   1.4       +25 -6     src/sys/sparc64/include/atomic.h
> 
> My apologies, but these look wrong, my sparc v9 manual says
> that although the atomic ops will work atomically without
> membar... that if you don't use membar you don't know the
> sequencing of the stores and loads surrounding the atomic
> op, so if it is an atomic op to perform a lock, your memory
> sequencing may cause accesses to a locked object before
> the lock is aquired or at least allow memory accesses to
> bleed through after the lock is released.
> 
> Of course I could be smoking the good crack today, so take
> this with a grain of salt. :-)

They are still atomic, they just aren't very useful. :)

-- 

John Baldwin <jhb@FreeBSD.org> -- http://www.FreeBSD.org/~jhb/
PGP Key: http://www.baldwin.cx/~john/pgpkey.asc
"Power Users Use the Power to Serve!"  -  http://www.FreeBSD.org/

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message