Peter Teuben wrote:
> However, because I'm doing my own "device driver setup" in a RT
> module, IRQ 5 - in my case - and IObase 0x304, are never registered by the
> kernel. That must be dangerous if you decide to load something after
> the PIO12 has been loaded. Is there a better solution to this?
What I am doing is registering the interrupt and IO base in the
initialisation code using standard kernel functions:
int init_module(void whatever){
int err;
if ((err=check_region(your_base_address,0x1F))<0) { / / check if
address already used
printk("<1>IO address used: reconfigure your device\n");
return err;
}
request_region(your_base_address,address_range,"YOUR_DEVICE");
}
and later in your cleanup code:
release_region(your_base_address,address_range);
The same works for interrupts, but I have forgotten the name of the
corresponding function calls, if you cannot find them, let me know, it must
be somewhere in my code.
That way, your device will be listed if you do "cat /proc/interrupts" or "cat
/proc/ioports"
Bye,
Hartwig
--- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
----
For more information on Real-Time Linux see:
http://www.rtlinux.org/~rtlinux/
