From owner-freebsd-stable@FreeBSD.ORG Tue Feb 14 13:00:48 2006 Return-Path: X-Original-To: freebsd-stable@freebsd.org Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 684F216A422 for ; Tue, 14 Feb 2006 13:00:48 +0000 (GMT) (envelope-from miguel@anjos.strangled.net) Received: from mailrly05.isp.novis.pt (mailrly05.isp.novis.pt [195.23.133.215]) by mx1.FreeBSD.org (Postfix) with ESMTP id 8541543D8A for ; Tue, 14 Feb 2006 13:00:44 +0000 (GMT) (envelope-from miguel@anjos.strangled.net) Received: (qmail 17434 invoked from network); 14 Feb 2006 13:00:42 -0000 Received: from unknown (HELO mailfrt04.isp.novis.pt) ([195.23.133.196]) (envelope-sender ) by mailrly05.isp.novis.pt with compressed SMTP; 14 Feb 2006 13:00:42 -0000 Received: (qmail 14101 invoked from network); 14 Feb 2006 13:00:42 -0000 Received: from unknown (HELO compaq.anjos.strangled.net) ([87.196.141.32]) (envelope-sender ) by mailfrt04.isp.novis.pt with SMTP; 14 Feb 2006 13:00:42 -0000 Received: from dual.anjos.strangled.net (dual.anjos.strangled.net [192.168.0.3]) by compaq.anjos.strangled.net (8.13.4/8.13.4) with ESMTP id k1ED0fp8002011 for ; Tue, 14 Feb 2006 13:00:41 GMT (envelope-from miguel@anjos.strangled.net) From: Miguel Ramos To: freebsd-stable@freebsd.org Content-Type: text/plain Date: Tue, 14 Feb 2006 13:00:40 +0000 Message-Id: <1139922040.1758.22.camel@dual.anjos.strangled.net> Mime-Version: 1.0 X-Mailer: Evolution 2.4.2.1 FreeBSD GNOME Team Port Content-Transfer-Encoding: 7bit Subject: AMD Mobile Athlon 64 VERY slow on 6-STABLE and 7-CURRENT X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 14 Feb 2006 13:00:48 -0000 This particular laptop, with a AMD Mobile Athlon 64 3000+ processor and a MSI-made motherboard runs VERY SLOWLY with FreeBSD versions on the 6-STABLE branch or 7-CURRENT branch. - This happens when running either the i386 architecture version or the amd64 architecture version. It is not specific of the amd64 version. - This happens with acpi either enabled or disabled, although results aren't exactly the same. - This happens with apic either enabled or disabled (on 5.4-RELEASE, amd64 architecture, apic had to be disabled on this computer to avoid crash). - This machine is fine on 5-STABLE, including 5.5-BETA1. This is about CPU speed, it is most noted when generating a RSA key, for instance. I bothered use the dhrystone 2 benchmark (from unixbench) and the result is about 13000 dhrystones/sec with acpi enabled, and 38000 dhrystones/sec with acpi disabled. It should be above 5 million dhrystones/sec, the system is totally unusable, about as fast as a 386. This is not related to acpi sleep states or acpi throttle since it happens with acpi disabled. The brand of the laptop is Targa Traveller 826 (motherboard is MSI). I've tried everything I could, I couldn't yet discover some sysctl value that could make any difference, or even in what section of the kernel may the problem be. Has anyone ever had a similar problem? Any sugestions? I was hopping to narrow the circumstances of this problem before sending a pr. ---- Miguel Ramos, miguel@anjos.strangled.net