Skip to contentSkip to conversation

WiMAX fixes for 2.6.30

GIT PULL Inaky Perez-Gonzalez 2 participants 4 messages May 6, 2009

Subsystems (2)
Networking Network drivers
lore
Share & tools
lore

Inaky Perez-Gonzalez

author
#0 lore raw

These fix a couple of oopsen found in the WiMAX stack. They are available as a pull and patches are included for revieweing convenience.

Thank you,

--

The following changes since commit 356d6c2d55b71303a17910ea2cce3eba8e44bd29:
David S. Miller (1):

        Merge branch 'master' of git://git.kernel.org/.../kaber/nf-2.6

are available in the git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/inaky/wimax.git linux-2.6.30.y

Inaky Perez-Gonzalez (2):
wimax: fix oops if netlink fails to add attribute
wimax: oops: wimax_dev_add() is the only one that can initialize the state

 net/wimax/op-msg.c |   11 ++++++-----
 net/wimax/stack.c  |   17 +++++++++++++++--
 2 files changed, 21 insertions(+), 7 deletions(-)
author
#1 lore raw

[pull 2.6.30 1/2] wimax: fix oops if netlink fails to add attribute

When sending a message to user space using wimax_msg(), if nla_put() fails, correctly interpret the return code from wimax_msg_alloc() as an err ptr and return the error code instead of crashing (as it is assuming than non-NULL means the pointer is ok).

Signed-off-by Inaky Perez-Gonzalez <[email protected]>
---
 net/wimax/op-msg.c |   11 ++++++-----
 1 files changed, 6 insertions(+), 5 deletions(-)
net/wimax/op-msg.c +6 -5
index 5d149c1..9ad4d89 100644
@@ -149,7 +149,8 @@ struct sk_buff *wimax_msg_alloc(struct wimax_dev *wimax_dev,
 	}
 	result = nla_put(skb, WIMAX_GNL_MSG_DATA, size, msg);
 	if (result < 0) {
-		dev_err(dev, "no memory to add payload in attribute\n");
+		dev_err(dev, "no memory to add payload (msg %p size %zu) in "
+			"attribute: %d\n", msg, size, result);
 		goto error_nla_put;
 	}
 	genlmsg_end(skb, genl_msg);

@@ -299,10 +300,10 @@ int wimax_msg(struct wimax_dev *wimax_dev, const char *pipe_name,
 	struct sk_buff *skb;
 
 	skb = wimax_msg_alloc(wimax_dev, pipe_name, buf, size, gfp_flags);
-	if (skb == NULL)
-		goto error_msg_new;
-	result = wimax_msg_send(wimax_dev, skb);
-error_msg_new:
+	if (IS_ERR(skb))
+		result = PTR_ERR(skb);
+	else
+		result = wimax_msg_send(wimax_dev, skb);
 	return result;
 }
 EXPORT_SYMBOL_GPL(wimax_msg);
Signature · 1 line
1.6.2.3
author
#2 lore raw

[pull 2.6.30 2/2] wimax: oops: wimax_dev_add() is the only one that can initialize the state

When a new wimax_dev is created, it's state has to be __WIMAX_ST_NULL until wimax_dev_add() is succesfully called. This allows calls into the stack that happen before said time to be rejected.

Until now, the state was being set (by mistake) to UNINITIALIZED, which was allowing calls such as wimax_report_rfkill_hw() to go through even when a call to wimax_dev_add() had failed; that was causing an oops when touching uninitialized data.

This situation is normal when the device starts reporting state before the whole initialization has been completed. It just has to be dealt with.

Signed-off-by Inaky Perez-Gonzalez <[email protected]>
---
 net/wimax/stack.c |   17 +++++++++++++++--
 1 files changed, 15 insertions(+), 2 deletions(-)
net/wimax/stack.c +15 -2
index a0ee76b..933e142 100644
@@ -338,8 +338,21 @@ out:
  */
 void wimax_state_change(struct wimax_dev *wimax_dev, enum wimax_st new_state)
 {
+	/*
+	 * A driver cannot take the wimax_dev out of the
+	 * __WIMAX_ST_NULL state unless by calling wimax_dev_add(). If
+	 * the wimax_dev's state is still NULL, we ignore any request
+	 * to change its state because it means it hasn't been yet
+	 * registered.
+	 *
+	 * There is no need to complain about it, as routines that
+	 * call this might be shared from different code paths that
+	 * are called before or after wimax_dev_add() has done its
+	 * job.
+	 */
 	mutex_lock(&wimax_dev->mutex);
-	__wimax_state_change(wimax_dev, new_state);
+	if (wimax_dev->state > __WIMAX_ST_NULL)
+		__wimax_state_change(wimax_dev, new_state);
 	mutex_unlock(&wimax_dev->mutex);
 	return;
 }

@@ -376,7 +389,7 @@ EXPORT_SYMBOL_GPL(wimax_state_get);
 void wimax_dev_init(struct wimax_dev *wimax_dev)
 {
 	INIT_LIST_HEAD(&wimax_dev->id_table_node);
-	__wimax_state_set(wimax_dev, WIMAX_ST_UNINITIALIZED);
+	__wimax_state_set(wimax_dev, __WIMAX_ST_NULL);
 	mutex_init(&wimax_dev->mutex);
 	mutex_init(&wimax_dev->mutex_reset);
 }
Signature · 1 line
1.6.2.3
maintainer
#3 lore raw
In reply to Inaky Perez-Gonzalez

From: Inaky Perez-Gonzalez <[email protected]>
Date: Wed, 6 May 2009 14:47:21 -0700

Quoted 3 lines

These fix a couple of oopsen found in the WiMAX stack. They are available as a pull and patches are included for revieweing convenience.

...

Quoted 4 lines

are available in the git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/inaky/wimax.git linux-2.6.30.y

Pulled, thanks a lot!