From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from moutng.kundenserver.de (moutng.kundenserver.de [212.227.126.186]) by ozlabs.org (Postfix) with ESMTP id 3F0B9DDEC5 for ; Thu, 26 Apr 2007 08:03:01 +1000 (EST) From: Arnd Bergmann To: Segher Boessenkool Subject: Re: [PATCH] generic check_legacy_ioport Date: Thu, 26 Apr 2007 00:02:54 +0200 References: <20070417210745.GA3567@aepfle.de> <200704250949.17795.arnd@arndb.de> <3f8a4cde6b06ee98203d62dc33c1630f@kernel.crashing.org> In-Reply-To: <3f8a4cde6b06ee98203d62dc33c1630f@kernel.crashing.org> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Message-Id: <200704260002.54627.arnd@arndb.de> Cc: Christian Krafft , Olaf Hering , linuxppc-dev@ozlabs.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Wednesday 25 April 2007, Segher Boessenkool wrote: > > It won'd be found by the of_platform_driver part of ipmi, > > since devices on the ISA bus do not get added to the linux > > device tree. >=20 > Drat. =A0Why not? We have a number of buses on which devices get automatically added as of_devices into the Linux driver infrastructures. Which buses are probed depends on the platform, it may include e.g. "soc", "plb5" and "siliconbackplane", but no platform currently scans the ISA bus. > > We might still want to have the driver use a proper of_device, > > which would require the maple platform code to add this device > > during probing. >=20 > That sounds like the only option then. =A0Does > this have to be done per platform though? =A0It > sounds perfectly safe to do it in more generic > code. I think the right solution would be to add automatic probing of all ISA devices into the platform independent code. Besides making IPMI work on JS21, it would result in the legacy serial ports to show up in the right place in sysfs. I am however worried about what other side-effects this change could have on platforms that have a broken ISA device in their device trees. Arnd <><