Skip to contentSkip to conversation

hidp: Make hidp_get_raw_report abort if the session is terminating.

PATCH In review Karl Relton 2 participants 3 messages Feb 20, 2013 – Feb 26, 2013

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 hidp: Make hidp_get_raw_report abort if the session is terminating.

From: Karl Relton <[email protected]>

After linux 3.2 the hid_destroy_device call in hidp_session cleaning up invokes a hook to the power_supply code which in turn tries to read the battery capacity. This read will trigger a call to hidp_get_raw_report which is bound to fail because the device is being taken away - so rather than wait for the 5 second timeout failure this changes enables it to fail straight away.

---
 net/bluetooth/hidp/core.c |    4 ++++
 1 file changed, 4 insertions(+)
net/bluetooth/hidp/core.c +4 -0
index b2bcbe2..4b26ea7 100644
@@ -311,6 +311,9 @@ static int hidp_get_raw_report(struct hid_device *hid,
 	int numbered_reports = hid->report_enum[report_type].numbered;
 	int ret;
 
+	if (atomic_read(&session->terminate))
+		return -EIO;
+
 	switch (report_type) {
 	case HID_FEATURE_REPORT:
 		report_type = HIDP_TRANS_GET_REPORT | HIDP_DATA_RTYPE_FEATURE;
@@ -722,6 +725,7 @@ static int hidp_session(void *arg)
 		set_current_state(TASK_INTERRUPTIBLE);
 	}
 	set_current_state(TASK_RUNNING);
+	atomic_inc(&session->terminate);
 	remove_wait_queue(sk_sleep(intr_sk), &intr_wait);
 	remove_wait_queue(sk_sleep(ctrl_sk), &ctrl_wait);
 
Signature · 1 line
1.7.9.5
#1 lore raw
Quoted 21 lines · Karl Relton

From: Karl Relton <[email protected]>

After linux 3.2 the hid_destroy_device call in hidp_session cleaning up invokes a hook to the power_supply code which in turn tries to read the battery capacity. This read will trigger a call to hidp_get_raw_report which is bound to fail because the device is being taken away - so rather than wait for the 5 second timeout failure this changes enables it to fail straight away.

---
 net/bluetooth/hidp/core.c |    4 ++++
 1 file changed, 4 insertions(+)
net/bluetooth/hidp/core.c +0 -0
index b2bcbe2..4b26ea7 100644

net/bluetooth/hidp/core.c static int hidp_get_raw_report(struct hid_device *hid, · line 317

Quoted 7 lines
net/bluetooth/hidp/core.c +3 -0
@@ -311,6 +311,9 @@ static int hidp_get_raw_report(struct hid_device *hid,
 	int numbered_reports = hid->report_enum[report_type].numbered;
 	int ret;
 
+	if (atomic_read(&session->terminate))
+		return -EIO;
+
 	switch (report_type) {

Actually I've just realised that hidp_get_raw_report calls __hidp_send_ctrl_message which already has the session->terminate test, also returning -EIO on failure. So adding it again here is redundant.

net/bluetooth/hidp/core.c static int hidp_session(void *arg) · line 731

Quoted 8 lines
net/bluetooth/hidp/core.c +1 -0
@@ -722,6 +725,7 @@ static int hidp_session(void *arg)
 		set_current_state(TASK_INTERRUPTIBLE);
 	}
 	set_current_state(TASK_RUNNING);
+	atomic_inc(&session->terminate);
 	remove_wait_queue(sk_sleep(intr_sk), &intr_wait);
 	remove_wait_queue(sk_sleep(ctrl_sk), &ctrl_wait);
 

So its just this one single line we need - to ensure session->terminate is non-zero, which will ensure calls to other things (like hidp_get_raw_report) abort swiftly.

I'll change the patch and send v3 this evening (need to get back to the day job!).

Karl

#2 lore raw
In reply to Karl Relton

Hi Karl,

Quoted 10 lines · Karl Relton

From: Karl Relton <[email protected]>

After linux 3.2 the hid_destroy_device call in hidp_session cleaning up invokes a hook to the power_supply code which in turn tries to read the battery capacity. This read will trigger a call to hidp_get_raw_report which is bound to fail because the device is being taken away - so rather than wait for the 5 second timeout failure this changes enables it to fail straight away.

Quoted 5 lines
---
 net/bluetooth/hidp/core.c |    4 ++++
 1 file changed, 4 insertions(+)

Patch has been applied to bluetooth-next. Thanks.

Gustavo