From owner-freebsd-acpi@FreeBSD.ORG Mon Apr 19 12:03:45 2004 Return-Path: Delivered-To: freebsd-acpi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id A0F9716A4CE for ; Mon, 19 Apr 2004 12:03:45 -0700 (PDT) Received: from root.org (root.org [67.118.192.226]) by mx1.FreeBSD.org (Postfix) with SMTP id 3430843D54 for ; Mon, 19 Apr 2004 12:03:45 -0700 (PDT) (envelope-from nate@root.org) Received: (qmail 22630 invoked by uid 1000); 19 Apr 2004 19:03:46 -0000 Date: Mon, 19 Apr 2004 12:03:46 -0700 (PDT) From: Nate Lawson To: Peter Schultz In-Reply-To: <408421AA.7040905@bis.midco.net> Message-ID: <20040419120026.L22535@root.org> References: <4083FC0D.1060505@bis.midco.net> <20040419105533.T22015@root.org> <408421AA.7040905@bis.midco.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: freebsd-acpi@freebsd.org Subject: Re: OS Interrupted X-BeenThere: freebsd-acpi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: ACPI and power management development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 19 Apr 2004 19:03:45 -0000 On Mon, 19 Apr 2004, Peter Schultz wrote: > Nate Lawson wrote: > > On Mon, 19 Apr 2004, Peter Schultz wrote: > > > >>ACPI is working again on my machine but there's a terrible amount of > >>latency introduced when using it. The mouse cursor is really jerky and > >>realaudio streams will constantly skip where they never do otherwise. > >> > >>Are there ACPI tweaks I can try to see if this can be eliminated? > > > > set hw.acpi.force_sci_lo=1 at the loader prompt. > > This isn't interrupt storm bad, but it makes listening to > music very annoying and the mouse cursor really jumpy. So you're saying vmstat -i doesn't show anything out of the ordinary? How often does the jumpiness happen? If it's once every 30 seconds, it could be thermal polling. If it's constantly, it's like a GPE issue or interrupt storm. You could stick a printf in acpi_ec.c:EcGpeQueryHandler to see if you're getting a lot of GPEs. If that's the case, it will likely be solved on the next import. You can work around this by removing the line in acpi.c that sets Gbl*Gpe* to FALSE. -Nate