Drivers: hv: vmbus: Implement per-CPU mapping of relid to channel
Currently the mapping of the relID to channel is done under the protection of a
single spin lock. Starting with ws2012, each channel is bound to a specific VCPU
in the guest. Use this binding to eliminate the spin lock by setting up
per-cpu state for mapping relId to the channel.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
diff --git a/drivers/hv/connection.c b/drivers/hv/connection.c
index df2363e..7f10c15 100644
--- a/drivers/hv/connection.c
+++ b/drivers/hv/connection.c
@@ -234,6 +234,28 @@
return ret;
}
+/*
+ * Map the given relid to the corresponding channel based on the
+ * per-cpu list of channels that have been affinitized to this CPU.
+ * This will be used in the channel callback path as we can do this
+ * mapping in a lock-free fashion.
+ */
+static struct vmbus_channel *pcpu_relid2channel(u32 relid)
+{
+ struct vmbus_channel *channel;
+ struct vmbus_channel *found_channel = NULL;
+ int cpu = smp_processor_id();
+ struct list_head *pcpu_head = &hv_context.percpu_list[cpu];
+
+ list_for_each_entry(channel, pcpu_head, percpu_list) {
+ if (channel->offermsg.child_relid == relid) {
+ found_channel = channel;
+ break;
+ }
+ }
+
+ return found_channel;
+}
/*
* relid2channel - Get the channel object given its
@@ -285,7 +307,7 @@
* Find the channel based on this relid and invokes the
* channel callback to process the event
*/
- channel = relid2channel(relid);
+ channel = pcpu_relid2channel(relid);
if (!channel) {
pr_err("channel not found for relid - %u\n", relid);