From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nate Lawson Subject: Re: double power button events on presario 900 - apparently a timing issue? Date: Thu, 25 Sep 2003 09:37:14 -0700 (PDT) Sender: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org Message-ID: <20030925093512.H59477@root.org> References: <200309240723.28526.fedor@karpelevitch.net> <20030924111121.F53806@root.org> <200309241621.11993.fedor@karpelevitch.net> <200309250859.46970.fedor@karpelevitch.net> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-path: In-Reply-To: <200309250859.46970.fedor-ugkMPi23kyXAHCkTjeiZhQ@public.gmane.org> Errors-To: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org List-Help: List-Post: List-Subscribe: , List-Unsubscribe: , List-Archive: To: Fedor Karpelevitch Cc: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org List-Id: linux-acpi@vger.kernel.org On Thu, 25 Sep 2003, Fedor Karpelevitch wrote: > Fedor Karpelevitch wrote: > > Nate Lawson wrote: > > > On Wed, 24 Sep 2003, Fedor Karpelevitch wrote: > > > > as I previously mentioned I am seeing an annoying problem on my > > > > presario 900 - when power button (works as FF here) is pressed > > > > it generates two events every time. I was experimentin with > > > > debugging output and noticed an interesting thing - if I do > > > > echo 0xffffffff > /proc/acpi/debug_level > > > > to get verbose debug output the problem goes away - I only get > > > > one event every time button is pressed. Looks like the problem > > > > is caused by some kind timing issue which goes away when > > > > debugging slows it down. Does this sound like a reasonable > > > > guess? Would anyone suggest how this can be debugged further? > > > > > > It means you need to debounce the event. > > > > do you mean debouncing on the receiving end? or would that be some > > way to avoid that in the kernel code? > > > by the way, I found that it is sufficient to do > > echo 0x0800000f > /proc/acpi/debug_level and problem disappears. That > bit is ACPI_DB_INTERRUPTS which is only referenced it two places... The kernel should debounce the button events after receiving them from ACPICA. Your debugging level change merely increases the latency between handling events so that the second event is missed. -Nate ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf