From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263442AbTH0P1o (ORCPT ); Wed, 27 Aug 2003 11:27:44 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263441AbTH0P1o (ORCPT ); Wed, 27 Aug 2003 11:27:44 -0400 Received: from fw.osdl.org ([65.172.181.6]:39383 "EHLO mail.osdl.org") by vger.kernel.org with ESMTP id S263442AbTH0P1l (ORCPT ); Wed, 27 Aug 2003 11:27:41 -0400 Date: Wed, 27 Aug 2003 08:22:38 -0700 From: "Randy.Dunlap" To: "Thomas S. Iversen" Cc: linux-kernel@vger.kernel.org Subject: Re: Help with debugging of a framebuffer driver on a legacyfree system. Message-Id: <20030827082238.0508aa9e.rddunlap@osdl.org> In-Reply-To: <20030827142035.GA25937@zensonic.dk> References: <20030827142035.GA25937@zensonic.dk> Organization: OSDL X-Mailer: Sylpheed version 0.9.4 (GTK+ 1.2.10; i686-pc-linux-gnu) X-Face: +5V?h'hZQPB9kW Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 27 Aug 2003 16:20:35 +0200 "Thomas S. Iversen" wrote: | Hi there | | As part of a project I am trying to write a framebuffer device | driver for a graphic chip. Up until now I have compiled my | code as a module and been doing insmod, rmmod and that have worked | nicely. | | I then tried to compile the driver into the kernel but that | makes the kernel hang. As my development system are a legacy | free laptop and my driver initializes the screen, the kernel | hangs without me being able to figure out where and why that | happend. First choice would be to develop on a system where you can add (PCI?) bus probes or other adapters/gadgets, like a port-80 card. | So I seek advice on how to debug the driver! As said, the | laptop are legacy free, so I have not got a serial port. | I have tried a usb->serial adapter, but that requires a driver. | Can I do USB->USB on another computer? Or? | | Or am I stuck with coding my own printk variant and writing to the | screen or is there any other option I have not thought of? There are already early_printk patches and an x86-specific "poke a character into video RAM" patch. The latter one is at: http://kernelnewbies.org/documents/videochar.txt -- ~Randy