Skip to contentSkip to conversation

hv: hv_fcopy: drop the obsolete message on transfer failure

PATCH In review Dexuan Cui 2 participants 2 messages Jan 14, 2015 – Jan 15, 2015 Hyper-V/Azure core and drivers

lore
Share & tools
lore

Dexuan Cui

author maintainer
#0 lore raw

PATCH v4 1/1 hv: hv_fcopy: drop the obsolete message on transfer failure

In the case the user-space daemon crashes, hangs or is killed, we need to down the semaphore, otherwise, after the daemon starts next time, the obsolete data in fcopy_transaction.message or fcopy_transaction.fcopy_msg will be used immediately.

Cc Jason Wang <[email protected]>
Cc Vitaly Kuznetsov <[email protected]>
Cc K. Y. Srinivasan <[email protected]>
Signed-off-by Dexuan Cui <[email protected]>

---

v2: I removed the "FCP" prefix as Greg asked.

I also updated the output message a little:
"FCP: failed to acquire the semaphore" -->
"can not acquire the semaphore: it is benign"

v3: I added the code in fcopy_release() as Jason Wang suggested.
I removed the pr_debug (it isn't so meaningful)and added a comment instead.

v4: this is a resend of v3, plus adding a comment before
fcopy_release().

---
 drivers/hv/hv_fcopy.c | 27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)
drivers/hv/hv_fcopy.c +27 -0
index 23b2ce2..cd453e4 100644
@@ -86,6 +86,18 @@ static void fcopy_work_func(struct work_struct *dummy)
 	 * process the pending transaction.
 	 */
 	fcopy_respond_to_host(HV_E_FAIL);
+
+	/* In the case the user-space daemon crashes, hangs or is killed, we
+	 * need to down the semaphore, otherwise, after the daemon starts next
+	 * time, the obsolete data in fcopy_transaction.message or
+	 * fcopy_transaction.fcopy_msg will be used immediately.
+	 *
+	 * NOTE: fcopy_read() happens to get the semaphore (very rare)? We're
+	 * still OK, because we've reported the failure to the host.
+	 */
+	if (down_trylock(&fcopy_transaction.read_sema))
+		;
+
 }
 
 static int fcopy_handle_handshake(u32 version)

@@ -344,6 +356,14 @@ static int fcopy_open(struct inode *inode, struct file *f)
 	return 0;
 }
 
+/* XXX: there are still some tricky corner cases, e.g.,
+ * 1) In a SMP guest, when fcopy_release() runs between
+ * schedule_delayed_work() and fcopy_send_data(), there is
+ * still a chance an obsolete message will be queued.
+ *
+ * 2) When the fcopy daemon is running, if we unload the driver,
+ * we'll notice a kernel oops when we kill the daemon later.
+ */
 static int fcopy_release(struct inode *inode, struct file *f)
 {
 	/*

@@ -351,6 +371,13 @@ static int fcopy_release(struct inode *inode, struct file *f)
 	 */
 	in_hand_shake = true;
 	opened = false;
+
+	if (cancel_delayed_work_sync(&fcopy_work)) {
+		/* We haven't up()-ed the semaphore(very rare)? */
+		if (down_trylock(&fcopy_transaction.read_sema))
+			;
+		fcopy_respond_to_host(HV_E_FAIL);
+	}
 	return 0;
 }
Signature · 1 line
1.9.1
#1 lore raw
Quoted 17 lines

-----Original Message-----
From: Dexuan Cui [mailto:[email protected]]
Sent: Wednesday, January 14, 2015 1:55 AM
To: [email protected]; [email protected]; driverdev- [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: KY Srinivasan
Subject: [PATCH v4] hv: hv_fcopy: drop the obsolete message on transfer failure

In the case the user-space daemon crashes, hangs or is killed, we need to down the semaphore, otherwise, after the daemon starts next time, the obsolete data in fcopy_transaction.message or fcopy_transaction.fcopy_msg will be used immediately.

Quoted 4 lines
Cc Jason Wang <[email protected]>
Cc Vitaly Kuznetsov <[email protected]>
Cc K. Y. Srinivasan <[email protected]>
Signed-off-by Dexuan Cui <[email protected]>
Signed-off-by K. Y. Srinivasan <[email protected]>
Quoted 76 lines

---

v2: I removed the "FCP" prefix as Greg asked.

I also updated the output message a little:
"FCP: failed to acquire the semaphore" -->
"can not acquire the semaphore: it is benign"

v3: I added the code in fcopy_release() as Jason Wang suggested.
I removed the pr_debug (it isn't so meaningful)and added a comment
instead.

v4: this is a resend of v3, plus adding a comment before
fcopy_release().

---
 drivers/hv/hv_fcopy.c | 27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)
drivers/hv/hv_fcopy.c index renamed from drivers/hv/hv_fcopy.c +0 -0

23b2ce2..cd453e4 100644

drivers/hv/hv_fcopy.c +3 -0
@@ -86,6 +86,18 @@ static void fcopy_work_func(struct work_struct
 *dummy)
 	 * process the pending transaction.
 	 */
 	fcopy_respond_to_host(HV_E_FAIL);
+
+	/* In the case the user-space daemon crashes, hangs or is killed, we
+	 * need to down the semaphore, otherwise, after the daemon starts

next

drivers/hv/hv_fcopy.c +4 -0
+	 * time, the obsolete data in fcopy_transaction.message or
+	 * fcopy_transaction.fcopy_msg will be used immediately.
+	 *
+	 * NOTE: fcopy_read() happens to get the semaphore (very rare)?

We're

drivers/hv/hv_fcopy.c +5 -0
+	 * still OK, because we've reported the failure to the host.
+	 */
+	if (down_trylock(&fcopy_transaction.read_sema))
+		;
+
 }
 
 static int fcopy_handle_handshake(u32 version) @@ -344,6 +356,14 @@

static int fcopy_open(struct inode *inode, struct file *f)

drivers/hv/hv_fcopy.c +15 -1
 	return 0;
 }
 
+/* XXX: there are still some tricky corner cases, e.g.,
+ * 1) In a SMP guest, when fcopy_release() runs between
+ * schedule_delayed_work() and fcopy_send_data(), there is
+ * still a chance an obsolete message will be queued.
+ *
+ * 2) When the fcopy daemon is running, if we unload the driver,
+ * we'll notice a kernel oops when we kill the daemon later.
+ */
 static int fcopy_release(struct inode *inode, struct file *f)  {
 	/*

@@ -351,6 +371,13 @@ static int fcopy_release(struct inode *inode, struct
 file *f)
 	 */
 	in_hand_shake = true;
 	opened = false;
+
+	if (cancel_delayed_work_sync(&fcopy_work)) {
+		/* We haven't up()-ed the semaphore(very rare)? */
+		if (down_trylock(&fcopy_transaction.read_sema))
+			;
+		fcopy_respond_to_host(HV_E_FAIL);
+	}
 	return 0;
 }
 
--

1.9.1