[PATCH] irqchip: vic: Support cascaded VIC in device tree

[PATCH] irqchip: vic: Support cascaded VIC in device tree

[Linus Walleij]

When transitioning some elder platforms to device tree it
becomes necessary to cascade VIC IRQ chips off another
interrupt controller.

Tested with the cascaded VIC on the Integrator/AP attached
logic module IM-PD1.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
---
 drivers/irqchip/irq-vic.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/irqchip/irq-vic.c b/drivers/irqchip/irq-vic.c
index f3f20a3cff50..caff21b4bac6 100644
--- a/drivers/irqchip/irq-vic.c
+++ b/drivers/irqchip/irq-vic.c
@@ -509,9 +509,7 @@ static int __init vic_of_init(struct device_node *node,
 	void __iomem *regs;
 	u32 interrupt_mask = ~0;
 	u32 wakeup_mask = ~0;
-
-	if (WARN(parent, "non-root VICs are not supported"))
-		return -EINVAL;
+	int parent_irq = 0;
 
 	regs = of_iomap(node, 0);
 	if (WARN_ON(!regs))
@@ -519,11 +517,14 @@ static int __init vic_of_init(struct device_node *node,
 
 	of_property_read_u32(node, "valid-mask", &interrupt_mask);
 	of_property_read_u32(node, "valid-wakeup-mask", &wakeup_mask);
+	parent_irq = of_irq_get(node, 0);
+	if (parent_irq < 0)
+		parent_irq = 0;
 
 	/*
 	 * Passing 0 as first IRQ makes the simple domain allocate descriptors
 	 */
-	__vic_init(regs, 0, 0, interrupt_mask, wakeup_mask, node);
+	__vic_init(regs, parent_irq, 0, interrupt_mask, wakeup_mask, node);
 
 	return 0;
 }
-- 
2.23.0

Re: [PATCH] irqchip: vic: Support cascaded VIC in device tree

[Marc Zyngier]

Hi Linus,

Please use my kernel.org email address, I just rescued this email
from the ML...

On 2020-02-13 12:47, Linus Walleij wrote:
> When transitioning some elder platforms to device tree it
> becomes necessary to cascade VIC IRQ chips off another
> interrupt controller.
> 
> Tested with the cascaded VIC on the Integrator/AP attached
> logic module IM-PD1.
> 
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
> ---
>  drivers/irqchip/irq-vic.c | 9 +++++----
>  1 file changed, 5 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/irqchip/irq-vic.c b/drivers/irqchip/irq-vic.c
> index f3f20a3cff50..caff21b4bac6 100644
> --- a/drivers/irqchip/irq-vic.c
> +++ b/drivers/irqchip/irq-vic.c
> @@ -509,9 +509,7 @@ static int __init vic_of_init(struct device_node 
> *node,
>  	void __iomem *regs;
>  	u32 interrupt_mask = ~0;
>  	u32 wakeup_mask = ~0;
> -
> -	if (WARN(parent, "non-root VICs are not supported"))
> -		return -EINVAL;
> +	int parent_irq = 0;

nit: Do you need this to be initialized to zero?

> 
>  	regs = of_iomap(node, 0);
>  	if (WARN_ON(!regs))
> @@ -519,11 +517,14 @@ static int __init vic_of_init(struct device_node 
> *node,
> 
>  	of_property_read_u32(node, "valid-mask", &interrupt_mask);
>  	of_property_read_u32(node, "valid-wakeup-mask", &wakeup_mask);
> +	parent_irq = of_irq_get(node, 0);
> +	if (parent_irq < 0)
> +		parent_irq = 0;
> 
>  	/*
>  	 * Passing 0 as first IRQ makes the simple domain allocate 
> descriptors
>  	 */
> -	__vic_init(regs, 0, 0, interrupt_mask, wakeup_mask, node);
> +	__vic_init(regs, parent_irq, 0, interrupt_mask, wakeup_mask, node);
> 
>  	return 0;
>  }

Do you want this as a fix?

Thanks,

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