Hi!
> > it is pretty hard to compile into kernel. (When you compile it into
> > kernel, you get oops and machine locks at boottime.)
>
> Does it help if you remove irda_device_init() from irda_init() in
> net/irda/irmod.c and call it from lets say ethif_probe() in
> linux/drivers/net/Space.c instead? That should make sure it's called at a
> later time, and after all the protocol stuff is finished initializing.
>
> I don't know where the "right" place for irda_device_init() in Space.c
> would be, but lets try to call it from some place in that file. We only
> need a little hock ;-)
Ok, this did trick for me:
--- clean/init/main.c Thu Jul 1 22:39:38 1999
+++ linux/init/main.c Fri Jul 9 10:21:15 1999
@@ -1331,6 +1374,10 @@
/* .. filesystems .. */
filesystem_setup();
+#ifdef CONFIG_IRDA
+ irda_device_init();
+#endif
+
/* Mount the root filesystem.. */
mount_root();
--- clean/net/irda/irmod.c Tue Jun 15 11:25:48 1999
+++ linux/net/irda/irmod.c Thu Jul 8 23:09:31 1999
@@ -193,13 +193,12 @@
EXPORT_SYMBOL(irtty_unregister_dongle);
#endif
__initfunc(int irda_init(void))
{
printk(KERN_INFO "IrDA (tm) Protocols for Linux-2.2 (Dag Brattli)\n");
- irlmp_init();
+ irlmp_init();
irlap_init();
- irda_device_init();
iriap_init();
irttp_init();
--
I'm really [EMAIL PROTECTED] Look at http://195.113.31.123/~pavel. Pavel
Hi! I'm a .signature virus! Copy me into your ~/.signature, please!
