From owner-freebsd-current@FreeBSD.ORG Thu Apr 7 18:59:25 2011 Return-Path: Delivered-To: freebsd-current@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id B918A106566B; Thu, 7 Apr 2011 18:59:25 +0000 (UTC) (envelope-from cswiger@mac.com) Received: from asmtpout028.mac.com (asmtpout028.mac.com [17.148.16.103]) by mx1.freebsd.org (Postfix) with ESMTP id A379E8FC0C; Thu, 7 Apr 2011 18:59:25 +0000 (UTC) MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; CHARSET=US-ASCII Received: from cswiger1.apple.com ([17.209.4.71]) by asmtp028.mac.com (Oracle Communications Messaging Exchange Server 7u4-20.01 64bit (built Nov 21 2010)) with ESMTPSA id <0LJA003RUN94ZG70@asmtp028.mac.com>; Thu, 07 Apr 2011 10:58:17 -0700 (PDT) X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.2.15,1.0.148,0.0.0000 definitions=2011-04-07_06:2011-04-07, 2011-04-07, 1970-01-01 signatures=0 X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 ipscore=0 suspectscore=0 phishscore=0 bulkscore=0 adultscore=0 classifier=spam adjust=0 reason=mlx engine=6.0.2-1012030000 definitions=main-1104070093 From: Chuck Swiger In-reply-to: <4D9DF086.9020906@FreeBSD.org> Date: Thu, 07 Apr 2011 10:58:16 -0700 Message-id: <709EE35F-B124-4670-81BD-FBFE6121928B@mac.com> References: <4D9DF086.9020906@FreeBSD.org> To: Andriy Gapon X-Mailer: Apple Mail (2.1084) Cc: FreeBSD current Subject: Re: prefer tsc timecounter when it's good X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 07 Apr 2011 18:59:25 -0000 Hi-- On Apr 7, 2011, at 10:12 AM, Andriy Gapon wrote: > what do you think about the following change? > The idea is mark TSC as the best timecounter when it's invariant and synchronized > between cores. > Unfortunately I don't have code to auto-detect the synchronization and keep > relying on the corresponding tunable. I thought about auto-setting it for > single-package configurations, but even that information is currently not trivial > to get out of our mp (i386/amd64) machdep code. In theory, most machines with P-state invariant TSCs should have their counters completely synchronized, even if there are multiple packages: http://software.intel.com/en-us/forums/showthread.php?t=74798 http://www.intel.com/Assets/PDF/manual/253668.pdf ...the exception would be very large machines with multiple mainboards. Regards, -- -Chuck