(copying qemu-devel)
Xu, Anthony wrote:
Subject: [PATCH] Irq assignment
1. use bimodal _PRT
2. pci device can use irq > 15, reduce interrupt sharing
3. test by running linux guest in kvm-ia64, kvm-i32(w/ wo/ -no-kvm)
+
+static int ioapic_irq_count[IOAPIC_NUM_PINS];
+
void ioapic_set_irq(void *opaque, int vector, int level)
{
IOAPICState *s = opaque;
-
+ if( vector >= 16 ){
+ if( level )
+ ioapic_irq_count[vector] += 1;
+ else
+ ioapic_irq_count[vector] -= 1;
+ level = (ioapic_irq_count[vector] != 0);
+ }
+#ifdef KVM_CAP_IRQCHIP
+ if (kvm_enabled())
+ if (kvm_set_irq(vector, ioapic_irq_count[vector] == 0))
+ return;
+#endif
It's legal to call ioapic_set_irq(vector, 1) twice, which will screw up
the level calculation.
We need qemu_irq_or(), similar to qemu_irq_invert():
qemu_irq qemu_irq_or(qemu_irq irqs[], int nr);
Also, this is not the place for doing the or. The ioapic does not know
which interrupts are level connected and which are not. This belongs on
the pci level (or the mainboard level).
--
error compiling committee.c: too many arguments to function
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
