From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arndt Schoenewald Subject: Re: No events when pressing power button Date: Tue, 13 Aug 2002 23:01:08 +0200 Sender: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org Message-ID: <20020813210108.GA11105@digda.intern.quelltext.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Return-path: Content-Disposition: inline In-Reply-To: Errors-To: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org List-Help: List-Post: List-Subscribe: , List-Unsubscribe: , List-Archive: To: "Grover, Andrew" Cc: acpi-devel List-Id: linux-acpi@vger.kernel.org Hi Andy, nice to hear you're back from vacation! (Hum, well, not quite as nice for yourself, I guess...!) On Tue, Aug 13, 2002 at 11:33:56AM -0700, Grover, Andrew wrote: > > From: Sérgio Monteiro Basto [mailto:sergiomb-5ng0By4jB+/QUPosyRRdSA@public.gmane.org] > > > I patched my kernel (2.4.18) today with latest version of > > acpi-patch > > > (acpi-20020726-2.4.18.diff). My power button gets detected > > but I dont get > > > any events from it. I have a ABit BP6 with dual celeron processors. > > > ACPI: Power Button (FF) [PWRF] > > > What could be wrong? I tried latest acpid rpm and 'cat > > /proc/acpi/event' > > > without results. Acpid doesn't write nothing to > > /var/log/acpid when I > > > press power button. > > > > > had you try kacpid patch > > that detects lost acpi events > > Hi, > > The problem *may* be that the ACPI interrupt is not being set to active-low, > level-triggered properly (we do this for PCI irqs but not our own ;-) > > Is your acpi interrupt shared with any PCI devices? If the above hypothesis > is true, the answer should be "no". > AHA! I own one of those machines that need eisa_set_level_irq() for some builtin devices, and I have never seen any events coming out of /var/log/acpid so far. The ACPI interrupt is not shared with anything. So I really guess you are right and eisa_set_level_irq(9) might be needed to turn ACPI interrupt reporting on. I'll add this call next time I build a kernel. Arndt > I don't know about people's comfort level with the code, but you might try > adding a call to eisa_set_level_irq() and seeing if that gets > interrupts happening properly. kacpid should not be necessary but obviously > until SCIs work 100% it remains useful for some people. > > Regards -- Andy ------------------------------------------------------- This sf.net email is sponsored by: Dice - The leading online job board for high-tech professionals. Search and apply for tech jobs today! http://seeker.dice.com/seeker.epl?rel_code=31