From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dave Jones Subject: Re: [Fwd: Re: [PROBLEM] 2.6.3-rc1: still no suspend/resume on Centrino notebook (contains agp, lapic, swsusp)] Date: Wed, 11 Feb 2004 11:32:14 +0000 Sender: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org Message-ID: <20040211113214.GB6310@redhat.com> References: <3ACA40606221794F80A5670F0AF15F8401CBB68A@PDSMSX403.ccr.corp.intel.com> <20040211105531.GA30647@hell.org.pl> <20040211110154.GJ12634@redhat.com> <20040211112505.GE30647@hell.org.pl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20040211112505.GE30647-DETuoxkZsSqrDJvtcaxF/A@public.gmane.org> Errors-To: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , List-Archive: To: "Yu, Luming" , "Georg C. F. Greve" , sergiomb-hHo3WeeoaswVhHzd4jOs4w@public.gmane.org, acpi-devel List-Id: linux-acpi@vger.kernel.org On Wed, Feb 11, 2004 at 12:25:05PM +0100, Karol Kozimor wrote: > Thus wrote Dave Jones: > > > Simple "modprobe intel-agp" will do, the system will reboot or hang when > > > resuming from swsusp1/pmdisk even when AGP is not used. > > Hmm, some of the Intel chipset code has resume callbacks already. > > These _will_ get called even if AGP isn't in use. > > Right. > > > I'm wondering if they're doing something they shouldn't. > > It might even be possible that those callbacks should just do nothing > > if agp isn't in use. Need to think some more on this one. > > That's especially strange as Nigel's swsusp2 survives the suspend. Here's > how it looks currently on my system [1]: > > > pmdisk swsusp swsusp2 > intel-agp not loaded OK OK OK > intel-agp loaded, not in use (?) reboot OK > intel-agp in use reboot reboot OK > > I haven't used pmdisk for some time, so I might be not authoritative. > > [1] 00:01.0 PCI bridge: Intel Corp. 82845 845 (Brookdale) Chipset AGP > Bridge (rev 04) > > I'm willing to test anything you come up with. Take a look at drivers/char/agp/intel-agp.c around line 1427 or so function agp_intel_resume(). Does it still hang if you just return 0 there ? I'm guessing that will fix it. Maybe we don't quite have the initialisation routine perfect, and reinitialising the chipset twice causes it to hang.. Dave ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn