public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* debugging "irq 7: nobody cared" in my module
@ 2006-06-05 12:51 Thomas Andrews
  2006-06-05 16:11 ` Thomas Andrews
  0 siblings, 1 reply; 3+ messages in thread
From: Thomas Andrews @ 2006-06-05 12:51 UTC (permalink / raw)
  To: linux-kernel

Hi,

What can cause "irq x: nobody cared" ? On exit from my IRQ handler, I am
checking to make sure that there are no residual/pending unhandled
events on exit. Here is part of my handler:

irqreturn_t scx200_acb_irq(int irq_no, void *dev_id, struct pt_regs *regs)
{
    struct scx200_acb_iface *iface = dev_id;
    unsigned long flags;
    spin_lock_irqsave(lock, flags);
    if (inb(ACBST)) {
//      tasklet_schedule(&iface->tasklet);
        scx200_acb_machine(iface, inb(ACBST));
        if (iface->state == state_idle)     /* Finished */
            wake_up_interruptible(&iface->acb_queue);
    } else {
        /* Should never get here */
        printk("causeless IRQ!\n");
        /* Reset the ACB to clear any pending IRQ */
        outb((inb(ACBCTL2) & 0xfe), ACBCTL2);
        outb((inb(ACBCTL2) | 0x01), ACBCTL2);
    }
    /* Check to see if some event was not handled */
    if (inb(ACBST) & 0xBC)
        printk("============== ACBST=%#x ===============\n",inb(ACBST));
    spin_unlock_irqrestore(lock, flags);
    return IRQ_RETVAL(0);
}

I've also had the problem when I move everything into a tasklet, and
just have a very short IRQ handler.

Many thanks,
Thomas

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: debugging "irq 7: nobody cared" in my module
  2006-06-05 12:51 debugging "irq 7: nobody cared" in my module Thomas Andrews
@ 2006-06-05 16:11 ` Thomas Andrews
  2006-06-05 16:47   ` Michael Buesch
  0 siblings, 1 reply; 3+ messages in thread
From: Thomas Andrews @ 2006-06-05 16:11 UTC (permalink / raw)
  To: linux-kernel

On Mon, Jun 05, 2006 at 02:51:32PM +0200, Thomas Andrews wrote:

> What can cause "irq x: nobody cared" ? On exit from my IRQ handler, I am
> checking to make sure that there are no residual/pending unhandled
> events on exit. Here is part of my handler:
> 
> irqreturn_t scx200_acb_irq(int irq_no, void *dev_id, struct pt_regs *regs)
> {
>     struct scx200_acb_iface *iface = dev_id;
>     unsigned long flags;
>     spin_lock_irqsave(lock, flags);
>     if (inb(ACBST)) {
> //      tasklet_schedule(&iface->tasklet);
>         scx200_acb_machine(iface, inb(ACBST));
>         if (iface->state == state_idle)     /* Finished */
>             wake_up_interruptible(&iface->acb_queue);
>     } else {
>         /* Should never get here */
>         printk("causeless IRQ!\n");
>         /* Reset the ACB to clear any pending IRQ */
>         outb((inb(ACBCTL2) & 0xfe), ACBCTL2);
>         outb((inb(ACBCTL2) | 0x01), ACBCTL2);
>     }
>     /* Check to see if some event was not handled */
>     if (inb(ACBST) & 0xBC)
>         printk("============== ACBST=%#x ===============\n",inb(ACBST));
>     spin_unlock_irqrestore(lock, flags);
>     return IRQ_RETVAL(0);
> }

I've found the problem now. The last line is wrong - it should read:

    return IRQ_RETVAL(IRQ_HANDLED);



^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: debugging "irq 7: nobody cared" in my module
  2006-06-05 16:11 ` Thomas Andrews
@ 2006-06-05 16:47   ` Michael Buesch
  0 siblings, 0 replies; 3+ messages in thread
From: Michael Buesch @ 2006-06-05 16:47 UTC (permalink / raw)
  To: Thomas Andrews; +Cc: linux-kernel

On Monday 05 June 2006 18:11, Thomas Andrews wrote:
> On Mon, Jun 05, 2006 at 02:51:32PM +0200, Thomas Andrews wrote:
> 
> > What can cause "irq x: nobody cared" ? On exit from my IRQ handler, I am
> > checking to make sure that there are no residual/pending unhandled
> > events on exit. Here is part of my handler:
> > 
> > irqreturn_t scx200_acb_irq(int irq_no, void *dev_id, struct pt_regs *regs)
> > {
> >     struct scx200_acb_iface *iface = dev_id;
> >     unsigned long flags;
> >     spin_lock_irqsave(lock, flags);
> >     if (inb(ACBST)) {
> > //      tasklet_schedule(&iface->tasklet);
> >         scx200_acb_machine(iface, inb(ACBST));
> >         if (iface->state == state_idle)     /* Finished */
> >             wake_up_interruptible(&iface->acb_queue);
> >     } else {
> >         /* Should never get here */
> >         printk("causeless IRQ!\n");
> >         /* Reset the ACB to clear any pending IRQ */
> >         outb((inb(ACBCTL2) & 0xfe), ACBCTL2);
> >         outb((inb(ACBCTL2) | 0x01), ACBCTL2);
> >     }
> >     /* Check to see if some event was not handled */
> >     if (inb(ACBST) & 0xBC)
> >         printk("============== ACBST=%#x ===============\n",inb(ACBST));
> >     spin_unlock_irqrestore(lock, flags);
> >     return IRQ_RETVAL(0);
> > }
> 
> I've found the problem now. The last line is wrong - it should read:
> 
>     return IRQ_RETVAL(IRQ_HANDLED);

No, either
return IRQ_HANDLED;
or
return IRQ_RETVAL(boolean);

You returned the boolean 0, which is "false".

-- 
Greetings Michael.

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2006-06-05 16:48 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-06-05 12:51 debugging "irq 7: nobody cared" in my module Thomas Andrews
2006-06-05 16:11 ` Thomas Andrews
2006-06-05 16:47   ` Michael Buesch

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox