Re: Interrupt not getting re-enabled after handler is killed

[anish singh]

adding Linux Kernel in loop.

On Wed, Aug 10, 2011 at 5:16 PM, anish singh
<anish198519851985@xxxxxxxxx> wrote:
> Intention:
> Wanted to modify the behavior of threaded interrupt handler
> not being called after the threaded_handler was killed.
>
> Reason:
> Well there is no reason as such but just want to see what all
> needs to be done to get this right.I know that what i am doing
> is completely wrong as i would get a strom of irq's and device
> will be unstable[1].
>
> Modification:
> I thought if i remove the below line from exit_irq_thread function
> set_bit(IRQTF_DIED, &tsk->irqaction->flags);
> it would do.But nothing happened then i inspected closely i found that
> in-case we need to do that we have to again create kthread
> which i am not doing in-case we again receive the interrupt of the killed
> threaded_handler.
>
> So i came to this conclusion:
> Right now we need more that just setting the flag bit to get this right.
> Is the above right ?Yes/no?
>
> Or
>
> if my understanding is not right then probably the interrupt controller
> itself is not sending the interrupt after the thread is killed.If this is right
> where is the code for that?
> In my limited search i couldn't find where we are disabling the interrupt
> once the threaded_handler is killed.
>
> Hope i am clear about what i want to know.
>
> [1] http://thread.gmane.org/gmane.linux.kernel/811061
>
--
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/