Re: tasklet_kill will always hang for recursive tasklets on a UP

[Juergen Quade]

On Wed, Aug 27, 2003 at 11:16:52PM +0530, Nagendra Singh Tomar wrote:
> Juergen,
> 	The whole tasklet_kill function is a big confusion. It is a big 
> misnomer as Werner rightly said. For non-recursive tasklets this  
> function does not do anything. Its just an expensive "nop". If you simply 
> call tasklet_schedule after tasklet_kill, it will execute as nothing had
> happened. 
> If we remove the line 
> 
> clear_bit(TASKLET_STATE_SCHED, &t->state);
> 
> from tasklet_kill then tasklet_kill will have the desired effect of 
> "killing" the tasklet, tasklet_schedule() after tasklet_kill in that case, 
> will not call __tasklet_kill and hence it will not be queued to the CPU
> queue and hence it will not run (desired effect).

Here we have it! In my opintion, the line

	clear_bit(TASKLET_STATE_SCHED, &t->state);

is just a _BUG_. The programmer _wanted_ to write

	set_bit(TASKLET_STATE_SCHED, &t->state);

In this case the function tasklet_kill _makes sense_ (beside
the problem of not working with recursive taskets)!
It will mostly be called in the cleanup function of a module 
and - yes - it would be useful.

So in my opintion
1. we should fix the bug (very easy)
2. we should find some means to make it usable for recursive tasklets.

	Juergen.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/