On Aug 22, 2008, at 6:11 AM, Martyn Welch wrote:
+static void __init
+gef_sbc610_init_irq(void)
+{
+ struct mpic *mpic1;
+ struct device_node *np;
+ struct resource res;
+
+ /* Determine PIC address. */
+ np = of_find_node_by_type(NULL, "open-pic");
+ if (np == NULL)
+ return;
+ of_address_to_resource(np, 0, &res);
+
+ mpic1 = mpic_alloc(np, res.start,
+ MPIC_PRIMARY | MPIC_WANTS_RESET | MPIC_BIG_ENDIAN,
+ 0, 256, "mpic");
+ of_node_put(np);
+ BUG_ON(mpic1 == NULL);
+
+ mpic_init(mpic1);
+}
any reason not to use mpc86xx_init_irq() in pic.c?
Yes, the afore-mentioned chained interrupt controller that is rather
specific to our boards that I will/have added to this function in a
patch that I shall post soon.
Hmm, does that chained interrupt-controller really effect this bit of
code any? I'd expect you'd be able to still call mpc86xx_init_irq()
and than in your board file hook up the cascade (like how we hookup
the i8259 in pic.c).
If not here are a few comments:
* Use *mpic instead mpic1
* Add MPIC_BROKEN_FRR_NIRQS to the flags
- k
_______________________________________________
Linuxppc-dev mailing list
Linuxppc-dev@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-dev