From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Luck, Tony" Date: Fri, 29 Sep 2006 18:01:10 +0000 Subject: Re: KDB blindly reads keyboard port Message-Id: <20060929180110.GA4021@intel.com> List-Id: References: <14425.1159496284@kao2.melbourne.sgi.com> <200609291057.41529.bjorn.helgaas@hp.com> In-Reply-To: <200609291057.41529.bjorn.helgaas@hp.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Bjorn Helgaas Cc: Keith Owens , linux-kernel@vger.kernel.org, linux-ia64@vger.kernel.org On Fri, Sep 29, 2006 at 10:57:41AM -0600, Bjorn Helgaas wrote: > acpi_parse_fadt: acpi_kbd_controller_present 0 The logic in the kernel seems backwards here though. We start by assuming there is a keyboard, then when parsing the FADT we reset this assumption if the BAF_8042_KEYBOARD_CONTROLLER bit isn't set. Which in turn forced SGI to include some workaround code for their older PROM (which doesn't provide the FADT table). There's also a risk that if some code might get added that runs before we parse FADT that could be confused into thinking that the keyboard is present. Wouldn't it be simpler/better to assume there is no keyboard until we find positive evidence that there is one? -Tony