irqchip/gic-v3-its: Move pending doorbell after VMOVP

After moving a VPE from a redistributor to another, we're still left
with a potential pending doorbell interrupt on the old redistributor.
That interrupt should be moved to the new one to be either cleared
or take, depending on what the hypervisor wishes to do.

So let's move it right after having execited VMOVP. This doesn't
add much cost in the !DirectLPI case (we trade a DISCARD for a MOVI),
and the cost of the DIRECTLPI case should be minimal (two extra MMIO
accesses).

Signed-off-by: Marc Zyngier <[e-mail]>

drivers/irqchip/irq-gic-v3-its.c

···
 	its_send_mapti(vpe_proxy.dev, vpe->vpe_db_lpi, vpe->vpe_proxy_event);
 }
 
+static void its_vpe_db_proxy_move(struct its_vpe *vpe, int from, int to)
+{
+	unsigned long flags;
+	struct its_collection *target_col;
+
+	if (gic_rdists->has_direct_lpi) {
+		void __iomem *rdbase;
+
+		rdbase = per_cpu_ptr(gic_rdists->rdist, from)->rd_base;
+		gic_write_lpir(vpe->vpe_db_lpi, rdbase + GICR_CLRLPIR);
+		while (gic_read_lpir(rdbase + GICR_SYNCR) & 1)
+			cpu_relax();
+
+		return;
+	}
+
+	raw_spin_lock_irqsave(&vpe_proxy.lock, flags);
+
+	its_vpe_db_proxy_map_locked(vpe);
+
+	target_col = &vpe_proxy.dev->its->collections[to];
+	its_send_movi(vpe_proxy.dev, target_col, vpe->vpe_proxy_event);
+	vpe_proxy.dev->event_map.col_map[vpe->vpe_proxy_event] = to;
+
+	raw_spin_unlock_irqrestore(&vpe_proxy.lock, flags);
+}
+
 static int its_vpe_set_affinity(struct irq_data *d,
 				const struct cpumask *mask_val,
 				bool force)
···
 	/*
 	 * Changing affinity is mega expensive, so let's be as lazy as
 	 * we can and only do it if we really have to. Also, if mapped
-	 * into the proxy device, we need to nuke that mapping.
+	 * into the proxy device, we need to move the doorbell
+	 * interrupt to its new location.
 	 */
 	if (vpe->col_idx != cpu) {
-		its_vpe_db_proxy_unmap(vpe);
+		int from = vpe->col_idx;
+
 		vpe->col_idx = cpu;
 		its_send_vmovp(vpe);
+		its_vpe_db_proxy_move(vpe, from, cpu);
 	}
 
 	return IRQ_SET_MASK_OK_DONE;