On 10/12/18 10:55, Marc Zyngier wrote:
+void gic_enable_of_quirks(const struct device_node *np,
+                         const struct gic_quirk *quirks, void *data)
+{
+       for (; quirks->desc; quirks++) {
So you expect quirks->desc to be NULL at some point, just like for any
other quirk table we have.

+               if (!of_device_is_compatible(np, quirks->compatible))
+                       continue;
+               if (quirks->init(data))
+                       pr_info("GIC: enabling workaround for %s\n",
+                               quirks->desc);
+       }
+}
+
  void gic_enable_quirks(u32 iidr, const struct gic_quirk *quirks,
                void *data)
  {
diff --git a/drivers/irqchip/irq-gic-common.h b/drivers/irqchip/irq-gic-common.h
index 3919cd7c5285..97e58fb6c232 100644
--- a/drivers/irqchip/irq-gic-common.h
+++ b/drivers/irqchip/irq-gic-common.h
@@ -23,6 +23,7 @@
struct gic_quirk {
        const char *desc;
+       const char *compatible;
        bool (*init)(void *data);
        u32 iidr;
        u32 mask;
@@ -35,6 +36,8 @@ void gic_dist_config(void __iomem *base, int gic_irqs,
  void gic_cpu_config(void __iomem *base, void (*sync_access)(void));
  void gic_enable_quirks(u32 iidr, const struct gic_quirk *quirks,
                void *data);
+void gic_enable_of_quirks(const struct device_node *np,
+                         const struct gic_quirk *quirks, void *data);
void gic_set_kvm_info(const struct gic_kvm_info *info); diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c
index 29e9d47be97d..c95796fa4de6 100644
--- a/drivers/irqchip/irq-gic-v3.c
+++ b/drivers/irqchip/irq-gic-v3.c
@@ -1271,6 +1271,9 @@ static void __init gic_of_setup_kvm_info(struct 
device_node *node)
        gic_set_kvm_info(&gic_v3_kvm_info);
  }
+static const struct gic_quirk gic_quirks[] = {
+};
... and yet here you provide an empty table. That's not going to work
very well. You definitely need to have an empty entry at the end of the
array, always.

I guess you want to test your code on a non affected platform, and I'm
pretty sure you'll see it exploding.

Yes, Should have carefully looked at this!!
We need an empty entry at the end!
I will fix it and send a new version!

Thanks,
srini

Reply via email to