Re: [RFC v1 25/25] genirq: Kill the first parameter 'irq' of irq_flow_handler_t

[Thomas Gleixner]

On Wed, 20 May 2015, Jiang Liu wrote:

> On 2015/5/20 23:40, Thomas Gleixner wrote:
> > On Wed, 20 May 2015, Jiang Liu wrote:
> >>  /* Chained IRQ handler for IPU error interrupt */
> >> -static void ipu_irq_err(unsigned int irq, struct irq_desc *desc)
> >> +static void ipu_irq_err(struct irq_desc *desc)
> >>  {
> >> +	unsigned int irq = irq_desc_to_irq(desc);
> > 
> > Why initializing? It's overwritten in the loop.
> Hi Thomas,
> 
> I'm not familiar with the hardware, but if src2map() returns 0
> on the first call, we need to initialize irq. Otherwise
> generic_handle_irq(irq) will use random value on the first call.
> ---------------------------------------------------------
>                         map = src2map(32 * i + line);
>                         if (map)
>                                 irq = map->irq;

You omitted the following code:

                        raw_spin_unlock(&bank_lock);

                        if (!map) {
                                pr_err("IPU: Interrupt on unmapped source %u bank %d\n",
                                       line, i);
                                continue;
			}

So you it does not use irq if map is 0.

Thanks,

	tglx

--
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/