[PATCH V2 1/2] Drivers: hv: vmbus: Fix a Host signaling bug

K. Y. Srinivasan kys at microsoft.com
Fri Nov 20 01:50:12 UTC 2015


Currently we have two policies for deciding when to signal the host:
One based on the ring buffer state and the other based on what the
VMBUS client driver wants to do. Consider the case when the client
wants to explicitly control when to signal the host. In this case,
if the client were to defer signaling, we will not be able to signal
the host subsequently when the client does want to signal since the
ring buffer state will prevent the signaling. Implement logic to
have only one signaling policy in force for a given channel.

Signed-off-by: K. Y. Srinivasan <kys at microsoft.com>
Reviewed-by: Haiyang Zhang <haiyangz at microsoft.com>
Tested-by: Haiyang Zhang <haiyangz at microsoft.com>
Cc: <stable at vger.kernel.org> # v4.2+
---
	V2: Made signaling policy an enum - Vitaly

 drivers/hv/channel.c   |   18 ++++++++++++++++++
 include/linux/hyperv.h |   18 ++++++++++++++++++
 2 files changed, 36 insertions(+), 0 deletions(-)

diff --git a/drivers/hv/channel.c b/drivers/hv/channel.c
index 77d2579..2889d97 100644
--- a/drivers/hv/channel.c
+++ b/drivers/hv/channel.c
@@ -653,10 +653,19 @@ int vmbus_sendpacket_ctl(struct vmbus_channel *channel, void *buffer,
 	 *    on the ring. We will not signal if more data is
 	 *    to be placed.
 	 *
+	 * Based on the channel signal state, we will decide
+	 * which signaling policy will be applied.
+	 *
 	 * If we cannot write to the ring-buffer; signal the host
 	 * even if we may not have written anything. This is a rare
 	 * enough condition that it should not matter.
 	 */
+
+	if (channel->signal_policy)
+		signal = true;
+	else
+		kick_q = true;
+
 	if (((ret == 0) && kick_q && signal) || (ret))
 		vmbus_setevent(channel);
 
@@ -756,10 +765,19 @@ int vmbus_sendpacket_pagebuffer_ctl(struct vmbus_channel *channel,
 	 *    on the ring. We will not signal if more data is
 	 *    to be placed.
 	 *
+	 * Based on the channel signal state, we will decide
+	 * which signaling policy will be applied.
+	 *
 	 * If we cannot write to the ring-buffer; signal the host
 	 * even if we may not have written anything. This is a rare
 	 * enough condition that it should not matter.
 	 */
+
+	if (channel->signal_policy)
+		signal = true;
+	else
+		kick_q = true;
+
 	if (((ret == 0) && kick_q && signal) || (ret))
 		vmbus_setevent(channel);
 
diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h
index 437c9c8..227b08d 100644
--- a/include/linux/hyperv.h
+++ b/include/linux/hyperv.h
@@ -629,6 +629,11 @@ struct hv_input_signal_event_buffer {
 	struct hv_input_signal_event event;
 };
 
+enum hv_signal_policy {
+	HV_SIGNAL_POLICY_DEFAULT = 0,
+	HV_SIGNAL_POLICY_EXPLICIT,
+};
+
 struct vmbus_channel {
 	/* Unique channel id */
 	int id;
@@ -756,8 +761,21 @@ struct vmbus_channel {
 	 * link up channels based on their CPU affinity.
 	 */
 	struct list_head percpu_list;
+	/*
+	 * Host signaling policy: The default policy will be
+	 * based on the ring buffer state. We will also support
+	 * a policy where the client driver can have explicit
+	 * signaling control.
+	 */
+	enum hv_signal_policy  signal_policy;
 };
 
+static inline void set_channel_signal_state(struct vmbus_channel *c,
+					    enum hv_signal_policy policy)
+{
+	c->signal_policy = policy;
+}
+
 static inline void set_channel_read_state(struct vmbus_channel *c, bool state)
 {
 	c->batched_reading = state;
-- 
1.7.4.1



More information about the devel mailing list