On Thu, 18 Aug 2016 14:43:16 +0100
Zubair Lutfullah Kakakhel <zubair.kakak...@imgtec.com> wrote:

> The MIPS based xilfpga platform has the following IRQ structure
> 
> Peripherals --> xilinx_intcontroller -> mips_cpu_int controller
> 
> Add support for the driver to chain the irq handler
> 
> Signed-off-by: Zubair Lutfullah Kakakhel <zubair.kakak...@imgtec.com>
> 
> ---
> V1 -> V2
> 
> No change
> ---
>  drivers/irqchip/irq-axi-intc.c | 28 +++++++++++++++++++++++++++-
>  1 file changed, 27 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/irqchip/irq-axi-intc.c b/drivers/irqchip/irq-axi-intc.c
> index 90bec7d..a0be6fa 100644
> --- a/drivers/irqchip/irq-axi-intc.c
> +++ b/drivers/irqchip/irq-axi-intc.c
> @@ -15,6 +15,7 @@
>  #include <linux/of_address.h>
>  #include <linux/io.h>
>  #include <linux/bug.h>
> +#include <linux/of_irq.h>
>  
>  static void __iomem *intc_baseaddr;
>  
> @@ -135,11 +136,26 @@ static const struct irq_domain_ops xintc_irq_domain_ops 
> = {
>       .map = xintc_map,
>  };
>  
> +static void xil_intc_irq_handler(struct irq_desc *desc)
> +{
> +     u32 pending = get_irq();
> +
> +     if (pending != -1U) {
> +             while (true) {
> +                     pending = get_irq();
> +                     generic_handle_irq(pending);
> +                     if (pending == -1U)

Erm... So even when pending is -1, you're calling generic_handle_irq?
This doesn't seem right.

You're also missing the chained_irq_enter/exit calls around this loop.

Overall, this should be rewritten in a less cumbersome way. Something
like:

        do {
                u32 pending;

                pending = get_irq();
                if (pending == ~0)
                        break;
                generic_handle_irq(pending);
        } while (true);

> +                             break;
> +             }
> +     }
> +}
> +
>  static int __init xilinx_intc_of_init(struct device_node *intc,
>                                            struct device_node *parent)
>  {
>       u32 nr_irq, intr_mask;
> -     int ret;
> +     int ret, irq;
> +     struct device_node *parent_node;
>  
>       intc_baseaddr = of_iomap(intc, 0);
>       BUG_ON(!intc_baseaddr);
> @@ -188,6 +204,16 @@ static int __init xilinx_intc_of_init(struct device_node 
> *intc,
>       root_domain = irq_domain_add_linear(intc, nr_irq, &xintc_irq_domain_ops,
>                                                       (void *)intr_mask);
>  
> +     parent_node = of_irq_find_parent(intc);

You already have the parent node as an argument to the function. Why do
you need to do that again?

> +     if (parent_node) {
> +             irq = irq_of_parse_and_map(intc, 0);
> +             if (irq)
> +                     irq_set_chained_handler_and_data(irq,
> +                                                      xil_intc_irq_handler,
> +                                                      root_domain);
> +
> +     }
> +
>       irq_set_default_host(root_domain);
>  
>       return 0;


Thanks,

        M.
-- 
Jazz is not dead. It just smells funny.

Reply via email to