Merge tag 'hyperv-fixes-signed-20210722' of git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux

Pull hyperv fixes from Wei Liu:

 - bug fix from Haiyang for vmbus CPU assignment

 - revert of a bogus patch that went into 5.14-rc1

* tag 'hyperv-fixes-signed-20210722' of git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux:
  Revert "x86/hyperv: fix logical processor creation"
  Drivers: hv: vmbus: Fix duplicate CPU assignments within a device

2 files changed, 65 insertions, 33 deletions

diff --git a/arch/x86/kernel/cpu/mshyperv.c b/arch/x86/kernel/cpu/mshyperv.c
index cc8f1773deca..c890d67a64ad 100644
--- a/arch/x86/kernel/cpu/mshyperv.c
+++ b/arch/x86/kernel/cpu/mshyperv.c
@@ -237,7 +237,7 @@ static void __init hv_smp_prepare_cpus(unsigned int max_cpus)
 	for_each_present_cpu(i) {
 		if (i == 0)
 			continue;
-		ret = hv_call_add_logical_proc(numa_cpu_node(i), i, i);
+		ret = hv_call_add_logical_proc(numa_cpu_node(i), i, cpu_physical_id(i));
 		BUG_ON(ret);
 	}
 
diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c
index caf6d0c4bc1b..142308526ec6 100644
--- a/drivers/hv/channel_mgmt.c
+++ b/drivers/hv/channel_mgmt.c
@@ -605,6 +605,17 @@ static void vmbus_process_offer(struct vmbus_channel *newchannel)
 	 */
 	mutex_lock(&vmbus_connection.channel_mutex);
 
+	list_for_each_entry(channel, &vmbus_connection.chn_list, listentry) {
+		if (guid_equal(&channel->offermsg.offer.if_type,
+			       &newchannel->offermsg.offer.if_type) &&
+		    guid_equal(&channel->offermsg.offer.if_instance,
+			       &newchannel->offermsg.offer.if_instance)) {
+			fnew = false;
+			newchannel->primary_channel = channel;
+			break;
+		}
+	}
+
 	init_vp_index(newchannel);
 
 	/* Remember the channels that should be cleaned up upon suspend. */
@@ -617,16 +628,6 @@ static void vmbus_process_offer(struct vmbus_channel *newchannel)
 	 */
 	atomic_dec(&vmbus_connection.offer_in_progress);
 
-	list_for_each_entry(channel, &vmbus_connection.chn_list, listentry) {
-		if (guid_equal(&channel->offermsg.offer.if_type,
-			       &newchannel->offermsg.offer.if_type) &&
-		    guid_equal(&channel->offermsg.offer.if_instance,
-			       &newchannel->offermsg.offer.if_instance)) {
-			fnew = false;
-			break;
-		}
-	}
-
 	if (fnew) {
 		list_add_tail(&newchannel->listentry,
 			      &vmbus_connection.chn_list);
@@ -647,7 +648,6 @@ static void vmbus_process_offer(struct vmbus_channel *newchannel)
 		/*
 		 * Process the sub-channel.
 		 */
-		newchannel->primary_channel = channel;
 		list_add_tail(&newchannel->sc_list, &channel->sc_list);
 	}
 
@@ -684,6 +684,30 @@ static void vmbus_process_offer(struct vmbus_channel *newchannel)
 }
 
 /*
+ * Check if CPUs used by other channels of the same device.
+ * It should only be called by init_vp_index().
+ */
+static bool hv_cpuself_used(u32 cpu, struct vmbus_channel *chn)
+{
+	struct vmbus_channel *primary = chn->primary_channel;
+	struct vmbus_channel *sc;
+
+	lockdep_assert_held(&vmbus_connection.channel_mutex);
+
+	if (!primary)
+		return false;
+
+	if (primary->target_cpu == cpu)
+		return true;
+
+	list_for_each_entry(sc, &primary->sc_list, sc_list)
+		if (sc != chn && sc->target_cpu == cpu)
+			return true;
+
+	return false;
+}
+
+/*
  * We use this state to statically distribute the channel interrupt load.
  */
 static int next_numa_node_id;
@@ -702,6 +726,7 @@ static int next_numa_node_id;
 static void init_vp_index(struct vmbus_channel *channel)
 {
 	bool perf_chn = hv_is_perf_channel(channel);
+	u32 i, ncpu = num_online_cpus();
 	cpumask_var_t available_mask;
 	struct cpumask *alloced_mask;
 	u32 target_cpu;
@@ -724,31 +749,38 @@ static void init_vp_index(struct vmbus_channel *channel)
 		return;
 	}
 
-	while (true) {
-		numa_node = next_numa_node_id++;
-		if (numa_node == nr_node_ids) {
-			next_numa_node_id = 0;
-			continue;
+	for (i = 1; i <= ncpu + 1; i++) {
+		while (true) {
+			numa_node = next_numa_node_id++;
+			if (numa_node == nr_node_ids) {
+				next_numa_node_id = 0;
+				continue;
+			}
+			if (cpumask_empty(cpumask_of_node(numa_node)))
+				continue;
+			break;
+		}
+		alloced_mask = &hv_context.hv_numa_map[numa_node];
+
+		if (cpumask_weight(alloced_mask) ==
+		    cpumask_weight(cpumask_of_node(numa_node))) {
+			/*
+			 * We have cycled through all the CPUs in the node;
+			 * reset the alloced map.
+			 */
+			cpumask_clear(alloced_mask);
 		}
-		if (cpumask_empty(cpumask_of_node(numa_node)))
-			continue;
-		break;
-	}
-	alloced_mask = &hv_context.hv_numa_map[numa_node];
 
-	if (cpumask_weight(alloced_mask) ==
-	    cpumask_weight(cpumask_of_node(numa_node))) {
-		/*
-		 * We have cycled through all the CPUs in the node;
-		 * reset the alloced map.
-		 */
-		cpumask_clear(alloced_mask);
-	}
+		cpumask_xor(available_mask, alloced_mask,
+			    cpumask_of_node(numa_node));
 
-	cpumask_xor(available_mask, alloced_mask, cpumask_of_node(numa_node));
+		target_cpu = cpumask_first(available_mask);
+		cpumask_set_cpu(target_cpu, alloced_mask);
 
-	target_cpu = cpumask_first(available_mask);
-	cpumask_set_cpu(target_cpu, alloced_mask);
+		if (channel->offermsg.offer.sub_channel_index >= ncpu ||
+		    i > ncpu || !hv_cpuself_used(target_cpu, channel))
+			break;
+	}
 
 	channel->target_cpu = target_cpu;