[patch 1/2] HID: sensors: use GFP_ATOMIC under spinlock

Subsystems: hid core layer, hid sensor hub drivers, the rest

STALE5091d

3 messages, 3 authors, 2012-09-17 · open the first message on its own page

[patch 1/2] HID: sensors: use GFP_ATOMIC under spinlock

From: Dan Carpenter <hidden>
Date: 2012-09-14 06:53:36

We're holding a spinlock here so we can't call kmalloc() with
GFP_KERNEL.

Signed-off-by: Dan Carpenter <redacted>

diff --git a/drivers/hid/hid-sensor-hub.c b/drivers/hid/hid-sensor-hub.c
index 4ac759c..0c93b10 100644
--- a/drivers/hid/hid-sensor-hub.c
+++ b/drivers/hid/hid-sensor-hub.c

@@ -173,7 +173,7 @@ int sensor_hub_register_callback(struct hid_sensor_hub_device *hsdev,
 			spin_unlock(&pdata->dyn_callback_lock);
 			return -EINVAL;
 		}
-	callback = kzalloc(sizeof(*callback), GFP_KERNEL);
+	callback = kzalloc(sizeof(*callback), GFP_ATOMIC);
 	if (!callback) {
 		spin_unlock(&pdata->dyn_callback_lock);
 		return -ENOMEM;

@@ -462,7 +462,7 @@ static int sensor_hub_raw_event(struct hid_device *hdev,
 		if (pdata->pending.status && pdata->pending.attr_usage_id ==
 				report->field[i]->usage->hid) {
 			hid_dbg(hdev, "data was pending ...\n");
-			pdata->pending.raw_data = kmalloc(sz, GFP_KERNEL);
+			pdata->pending.raw_data = kmalloc(sz, GFP_ATOMIC);
 			if (pdata->pending.raw_data) {
 				memcpy(pdata->pending.raw_data, ptr, sz);
 				pdata->pending.raw_size  = sz;

Re: [patch 1/2] HID: sensors: use GFP_ATOMIC under spinlock

From: Jiri Kosina <hidden>
Date: 2012-09-17 10:59:24

On Fri, 14 Sep 2012, Dan Carpenter wrote:

We're holding a spinlock here so we can't call kmalloc() with
GFP_KERNEL.

Signed-off-by: Dan Carpenter <redacted>

Signed-off-by: Jiri Kosina <redacted>

Jonathan, could you please take this one as well, as it depends on the 
series in your tree?

Thanks.

quoted hunk

diff --git a/drivers/hid/hid-sensor-hub.c b/drivers/hid/hid-sensor-hub.c
index 4ac759c..0c93b10 100644
--- a/drivers/hid/hid-sensor-hub.c
+++ b/drivers/hid/hid-sensor-hub.c

@@ -173,7 +173,7 @@ int sensor_hub_register_callback(struct hid_sensor_hub_device *hsdev,
 			spin_unlock(&pdata->dyn_callback_lock);
 			return -EINVAL;
 		}
-	callback = kzalloc(sizeof(*callback), GFP_KERNEL);
+	callback = kzalloc(sizeof(*callback), GFP_ATOMIC);
 	if (!callback) {
 		spin_unlock(&pdata->dyn_callback_lock);
 		return -ENOMEM;

@@ -462,7 +462,7 @@ static int sensor_hub_raw_event(struct hid_device *hdev,
 		if (pdata->pending.status && pdata->pending.attr_usage_id ==
 				report->field[i]->usage->hid) {
 			hid_dbg(hdev, "data was pending ...\n");
-			pdata->pending.raw_data = kmalloc(sz, GFP_KERNEL);
+			pdata->pending.raw_data = kmalloc(sz, GFP_ATOMIC);
 			if (pdata->pending.raw_data) {
 				memcpy(pdata->pending.raw_data, ptr, sz);
 				pdata->pending.raw_size  = sz;

-- 
Jiri Kosina
SUSE Labs

RE: [patch 1/2] HID: sensors: use GFP_ATOMIC under spinlock

From: Pandruvada, Srinivas <hidden>
Date: 2012-09-17 15:21:34

Tested this and works. 

-----Original Message-----
From: Jiri Kosina [mailto:jkosina@suse.cz] 
Sent: Monday, September 17, 2012 3:59 AM
To: Dan Carpenter
Cc: Pandruvada, Srinivas; linux-input@vger.kernel.org; kernel-janitors@vger.kernel.org; Jonathan Cameron
Subject: Re: [patch 1/2] HID: sensors: use GFP_ATOMIC under spinlock

On Fri, 14 Sep 2012, Dan Carpenter wrote:

We're holding a spinlock here so we can't call kmalloc() with 
GFP_KERNEL.

Signed-off-by: Dan Carpenter <redacted>

Signed-off-by: Jiri Kosina <redacted>

Jonathan, could you please take this one as well, as it depends on the series in your tree?

Thanks.

quoted hunk

diff --git a/drivers/hid/hid-sensor-hub.c

b/drivers/hid/hid-sensor-hub.c index 4ac759c..0c93b10 100644

--- a/drivers/hid/hid-sensor-hub.c
+++ b/drivers/hid/hid-sensor-hub.c

@@ -173,7 +173,7 @@ int sensor_hub_register_callback(struct hid_sensor_hub_device *hsdev,
 			spin_unlock(&pdata->dyn_callback_lock);
 			return -EINVAL;
 		}
-	callback = kzalloc(sizeof(*callback), GFP_KERNEL);
+	callback = kzalloc(sizeof(*callback), GFP_ATOMIC);
 	if (!callback) {
 		spin_unlock(&pdata->dyn_callback_lock);
 		return -ENOMEM;

@@ -462,7 +462,7 @@ static int sensor_hub_raw_event(struct hid_device *hdev,
 		if (pdata->pending.status && pdata->pending.attr_usage_id ==
 				report->field[i]->usage->hid) {
 			hid_dbg(hdev, "data was pending ...\n");
-			pdata->pending.raw_data = kmalloc(sz, GFP_KERNEL);
+			pdata->pending.raw_data = kmalloc(sz, GFP_ATOMIC);
 			if (pdata->pending.raw_data) {
 				memcpy(pdata->pending.raw_data, ptr, sz);
 				pdata->pending.raw_size  = sz;

--
Jiri Kosina
SUSE Labs

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help