- From: Ian Abbott <abbotti_at_mev.co.uk>
- Date: Tue, 28 Mar 2006 11:44:12 +0100
On 28/03/2006 10:51, Siebe Ytsma wrote: > That is exactly where we are stuck. We have put printk calls in the > driver but nothing is output to the messages file. Apparantly the > comedi_config has some problems with my driver and consequently crashes > the kernel. Some tips for debugging: 1. Run it from a text-mode virtual console, not from an xterm. 2. run 'dmesg -n 7' to output all kernel log messages to the console. Hopefully you should see the printk's (and hopefully an "oops" message) on the console before it crashes. > > > > > 2006/3/28, Ian Abbott <abbotti_at_mev.co.uk <mailto:abbotti_at_mev.co.uk>>: > > On 28/03/2006 09:04, Siebe Ytsma wrote: > > I am currently doing my internship and work on a project where we > need > > to use a National Instruments PCI-6519 card. National Instruments > > advises the Comedi framework on their website but unfortunately the > > PCI-6519 card isn't supported yet. Does anyone know if someone is > > working on such a Comedi driver? We tried starting to build one > > ourselves with the PCI-6527 as an example, until now without > success. It > > loads fine but when we comedi_config the device it crashes the > kernel. > > From the var/log/messages no debug traces can be found so we are > a bit > > lost on what the problem is. The additional info from the skel.c file > > hasn't solved this problem either. > > > > All help is welcome. > > Well you need to find out which bit is causing the driver to crash. One > way to narrow it down is to #ifdef out bits of code until it no longer > crashes. Another way (which can be combined with the first) is to > intersperse the code with a few printk calls so you can see how far it > gets before crashing. > > -- > -=( Ian Abbott _at_ MEV Ltd. E-mail: <abbotti_at_mev.co.uk > <mailto:abbotti_at_mev.co.uk>> )=- > -=( Tel: +44 (0)161 477 1898 FAX: +44 (0)161 718 3587 )=- > >
Received on 2006-03-28Z09:44:12