From owner-freebsd-hackers Mon Nov 3 23:10:50 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id XAA18112 for hackers-outgoing; Mon, 3 Nov 1997 23:10:50 -0800 (PST) (envelope-from owner-freebsd-hackers) Received: from Campino.Informatik.RWTH-Aachen.DE (campino.Informatik.RWTH-Aachen.DE [137.226.116.240]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA18103 for ; Mon, 3 Nov 1997 23:10:47 -0800 (PST) (envelope-from kuku@gilberto.physik.RWTH-Aachen.DE) Received: from gil.physik.rwth-aachen.de (gilberto.physik.rwth-aachen.de [137.226.30.2]) by Campino.Informatik.RWTH-Aachen.DE (8.8.7/RBI-Z14) with ESMTP id IAA25591; Tue, 4 Nov 1997 08:10:42 +0100 (MET) Received: (from kuku@localhost) by gil.physik.rwth-aachen.de (8.8.5/8.6.9) id IAA12745; Tue, 4 Nov 1997 08:26:11 +0100 (MET) Message-ID: <19971104082611.29224@gil.physik.rwth-aachen.de> Date: Tue, 4 Nov 1997 08:26:11 +0100 From: Christoph Kukulies To: Mike Smith Cc: Christoph Kukulies , freebsd-hackers@freefall.FreeBSD.org Subject: Re: still mmap and my driver References: <19971102090752.08503@gil.physik.rwth-aachen.de> <199711030108.LAA00713@word.smith.net.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.81e In-Reply-To: <199711030108.LAA00713@word.smith.net.au>; from Mike Smith on Mon, Nov 03, 1997 at 11:38:21AM +1030 Sender: owner-freebsd-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Mon, Nov 03, 1997 at 11:38:21AM +1030, Mike Smith wrote: > > On Sun, Nov 02, 1997 at 03:39:24PM +1030, Mike Smith wrote: > > > > > > > > Things are still weird with my driver ISA memory mapping. > > > > I noticed that the space I need to map is 16386 bytes > > > > from C8000 on. I changed isa_device id->id_msize to have 16386. > > > > And I have an iosiz 0x4000 in my config file now. > > > > > > > > I still get a kernel panic, when I access 0xca000 in the driver. > > > > > > > > Did I still overlook anything? > > > > > > The details of the "panic". Was it really a panic? Do you perhaps > > > mean that it was a page fault (probably good ol' trap 12)? > > > > Sure. I mean page fault because it cannot access 0xca000. > > It happens during the driver probe phase in the portion of code > > I wrote. It's quite obvious that it's because the page is not > > available. The kernel trap 12 was perfectly right: I was using the wrong address (not the kva, but the physical in one place). I got it now. Accessing the ISA memory through the kernel still uses a virtual address, mapped to 0xf0000000. > > This doesn't make a great deal of sense, unfortunately. If the correct > window is being mapped for your driver, it should work. It sounds like > the mapping isn't being done right. Source for your probe/attach > available? > > mike > -- Chris Christoph P. U. Kukulies kuku@gil.physik.rwth-aachen.de