Thread (25 messages) 25 messages, 4 authors, 2015-10-14

[PATCH v2 05/17] irqchip: Convert all alloc/xlate users from of_node to fwnode

From: Tomasz Nowicki <hidden>
Date: 2015-10-14 15:38:05
Also in: linux-acpi, lkml

On 14.10.2015 17:31, Marc Zyngier wrote:
On 14/10/15 16:26, Tomasz Nowicki wrote:
quoted
Hi Marc,

On 13.10.2015 13:51, Marc Zyngier wrote:
quoted
Since we now have a generic data structure to express an
interrupt specifier, convert all hierarchical irqchips that
are OF based to use a fwnode_handle as part of their alloc
and xlate (which becomes translate) callbacks.

As most of these drivers have dependencies (they exchange IRQ
specifiers), change them all in a single, massive patch...

Signed-off-by: Marc Zyngier <redacted>
---
   arch/arm/mach-exynos/suspend.c       | 55 ++++++++++++++++---------------
   arch/arm/mach-imx/gpc.c              | 55 ++++++++++++++++---------------
   arch/arm/mach-omap2/omap-wakeupgen.c | 55 ++++++++++++++++---------------
   drivers/irqchip/irq-crossbar.c       | 62 ++++++++++++++++++----------------
   drivers/irqchip/irq-gic-v2m.c        | 18 ++++++----
   drivers/irqchip/irq-gic-v3-its.c     | 20 ++++++-----
   drivers/irqchip/irq-gic-v3.c         | 49 +++++++++++++--------------
   drivers/irqchip/irq-gic.c            | 33 ++++++++++++++++---
   drivers/irqchip/irq-imx-gpcv2.c      | 64 ++++++++++++++++--------------------
   drivers/irqchip/irq-mtk-sysirq.c     | 49 ++++++++++++++-------------
   drivers/irqchip/irq-nvic.c           | 18 +++++++---
   drivers/irqchip/irq-tegra.c          | 55 ++++++++++++++++---------------
   drivers/irqchip/irq-vf610-mscm-ir.c  | 42 +++++++++++++++--------
   13 files changed, 323 insertions(+), 252 deletions(-)
[...]
quoted
   static int its_irq_domain_alloc(struct irq_domain *domain, unsigned int virq,
diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c
index 5793880..05d010b 100644
--- a/drivers/irqchip/irq-gic-v3.c
+++ b/drivers/irqchip/irq-gic-v3.c
@@ -737,32 +737,30 @@ static int gic_irq_domain_map(struct irq_domain *d, unsigned int irq,
   	return 0;
   }

-static int gic_irq_domain_xlate(struct irq_domain *d,
-				struct device_node *controller,
-				const u32 *intspec, unsigned int intsize,
-				unsigned long *out_hwirq, unsigned int *out_type)
+static int gic_irq_domain_translate(struct irq_domain *d,
+				    struct irq_fwspec *fwspec,
+				    unsigned long *hwirq,
+				    unsigned int *type)
   {
-	if (irq_domain_get_of_node(d) != controller)
-		return -EINVAL;
-	if (intsize < 3)
-		return -EINVAL;
+	if (is_of_node(fwspec->fwnode)) {
+		if (fwspec->param_count < 3)
+			return -EINVAL;

-	switch(intspec[0]) {
-	case 0:			/* SPI */
-		*out_hwirq = intspec[1] + 32;
-		break;
-	case 1:			/* PPI */
-		*out_hwirq = intspec[1] + 16;
-		break;
-	case GIC_IRQ_TYPE_LPI:	/* LPI */
-		*out_hwirq = intspec[1];
-		break;
-	default:
-		return -EINVAL;
+		/* Get the interrupt number and add 16 to skip over SGIs */
+		*hwirq = fwspec->param[1] + 16;
+
+		/*
+		 * For SPIs, we need to add 16 more to get the GIC irq
+		 * ID number
+		 */
+		if (!fwspec->param[0])
+			*hwirq += 16;
+
+		*type = fwspec->param[2] & IRQ_TYPE_SENSE_MASK;
+		return 0;
   	}

-	*out_type = intspec[2] & IRQ_TYPE_SENSE_MASK;
-	return 0;
+	return -EINVAL;
   }
What about GIC_IRQ_TYPE_LPI type? Previously we leaved interrupt line nr
and now +32.
Please see the patch I've posted earlier that address this particular issue:

https://lkml.org/lkml/2015/10/14/272
Makes sense, thanks.

Tomasz
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help