From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bjorn Helgaas Subject: Re: [PATCH] [SERIAL] add TP560 data/fax/modem support Date: Mon, 07 Feb 2005 13:57:35 -0700 Message-ID: <1107809856.8074.50.camel@piglet> References: <1107805182.8074.35.camel@piglet> Reply-To: bjorn.helgaas@hp.com Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Received: from sccrmhc12.comcast.net ([204.127.202.56]:30458 "EHLO sccrmhc12.comcast.net") by vger.kernel.org with ESMTP id S261268AbVBGU5s (ORCPT ); Mon, 7 Feb 2005 15:57:48 -0500 In-Reply-To: Sender: linux-serial-owner@vger.kernel.org List-Id: linux-serial@vger.kernel.org To: linux-os@analogic.com Cc: rmk+serial@arm.linux.org.uk, linux-serial@vger.kernel.org, linux-kernel@vger.kernel.org On Mon, 2005-02-07 at 15:12 -0500, linux-os wrote: > I thought somebody promised to add a pci_route_irq(dev) or some > such so that the device didn't have to be enabled before > the IRQ was correct. > > I first reported this bad IRQ problem back in December of 2004. > Has the new function been added? That's a completely different problem. The point here is that the serial driver currently doesn't do anything with the TP560 (no pci_enable_device(), no pci_route_irq(), no nothing). Then when setserial comes along and force-feeds the driver with the IO and IRQ info, there's nothing at that point that does anything to enable the device or route its interrupt either. I did raise the idea of adding a pci_route_irq() interface, but to be honest, I was never convinced of its general usefulness. I haven't heard of any driver in the tree that requires it, so it's not clear that it would be accepted even if I (or you) wrote it. I think you mentioned a specific PCI interface chip that was susceptible to the problem; is there a public reference that would help explicate the situation?