PROBLEM: [2.4.18-3] kernel BUG at softirq.c:194!

[Clive Nicolson]

I'm (really I'm just the reporter) seeing the above kernel BUG, the code 
is thus (in tasklet_action):

	while (list) {
		struct tasklet_struct *t = list;

		list = list->next;

		if (tasklet_trylock(t)) {
			if (!atomic_read(&t->count)) {
				if (!test_and_clear_bit(TASKLET_STATE_SCHED, &t->state))
					BUG();
				t->func(t->data);
				tasklet_unlock(t);
				continue;
			}
		}

I dont see the problem on my slow PC's but it is seen at 900 Mhz!

Can someone give me a clue as to how a device driver may be inducing this!

Thanks
Clive