linux-wireless.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] at76c50x-usb: fix Oops on disconnect
@ 2009-02-17  1:39 Jason Andryuk
  2009-02-18 20:45 ` Kalle Valo
  0 siblings, 1 reply; 4+ messages in thread
From: Jason Andryuk @ 2009-02-17  1:39 UTC (permalink / raw)
  To: linux-wireless@vger.kernel.org, Kalle Valo

flush_workqueue needs to be called instead of the generic one and the
associated functions need to be modified to prevent re-adding
themselves to the workqueue.

The rx_tasklet is also killed in the small (?) chance it is scheduled.

Signed-off-by: Jason Andryuk <jandryuk@gmail.com>
--

diff --git a/drivers/net/wireless/at76c50x-usb.c
b/drivers/net/wireless/at76c50x-usb.c
index 8d508ca..7a03251 100644
--- a/drivers/net/wireless/at76c50x-usb.c
+++ b/drivers/net/wireless/at76c50x-usb.c
@@ -1860,6 +1860,9 @@ static void at76_dwork_hw_scan(struct work_struct *work)
                           dwork_hw_scan.work);
     int ret;

+    if (priv->device_unplugged)
+        return;
+
     mutex_lock(&priv->mtx);

     ret = at76_get_cmd_status(priv->udev, CMD_SCAN);
@@ -1897,6 +1900,9 @@ static int at76_hw_scan(struct ieee80211_hw *hw,

     at76_dbg(DBG_MAC80211, "%s():", __func__);

+    if (priv->device_unplugged)
+        return 0;
+
     mutex_lock(&priv->mtx);

     ieee80211_stop_queues(hw);
@@ -2003,6 +2009,10 @@ static void at76_configure_filter(struct
ieee80211_hw *hw,
     flags = changed_flags & AT76_SUPPORTED_FILTERS;
     *total_flags = AT76_SUPPORTED_FILTERS;

+    /* Bail out after updating flags to prevent a WARN_ON in mac80211. */
+          if (priv->device_unplugged)
+        return;
+
     /* FIXME: access to priv->promisc should be protected with
      * priv->mtx, but it's impossible because this function needs to be
      * atomic */
@@ -2103,8 +2113,7 @@ static struct at76_priv
*at76_alloc_new_device(struct usb_device *udev)
     INIT_WORK(&priv->work_submit_rx, at76_work_submit_rx);
     INIT_DELAYED_WORK(&priv->dwork_hw_scan, at76_dwork_hw_scan);

-    priv->rx_tasklet.func = at76_rx_tasklet;
-    priv->rx_tasklet.data = 0;
+    tasklet_init(&priv->rx_tasklet, at76_rx_tasklet, 0);

     priv->pm_mode = AT76_PM_OFF;
     priv->pm_period = 0;
@@ -2243,6 +2252,7 @@ static int at76_init_new_device(struct at76_priv *priv,
     priv->scan_min_time = DEF_SCAN_MIN_TIME;
     priv->scan_max_time = DEF_SCAN_MAX_TIME;
     priv->scan_mode = SCAN_TYPE_ACTIVE;
+    priv->device_unplugged = 0;

     /* mac80211 initialisation */
     priv->hw->wiphy->max_scan_ssids = 1;
@@ -2284,13 +2294,12 @@ static void at76_delete_device(struct at76_priv *priv)
     /* The device is gone, don't bother turning it off */
     priv->device_unplugged = 1;

-    if (priv->mac80211_registered)
-        ieee80211_unregister_hw(priv->hw);
+    tasklet_kill(&priv->rx_tasklet);

-    /* assuming we used keventd, it must quiesce too */
-    flush_scheduled_work();
-
-    kfree(priv->bulk_out_buffer);
+    if (priv->mac80211_registered) {
+        flush_workqueue(priv->hw->workqueue);
+        ieee80211_unregister_hw(priv->hw);
+    }

     if (priv->tx_urb) {
         usb_kill_urb(priv->tx_urb);
@@ -2303,6 +2312,8 @@ static void at76_delete_device(struct at76_priv *priv)

     at76_dbg(DBG_PROC_ENTRY, "%s: unlinked urbs", __func__);

+    kfree(priv->bulk_out_buffer);
+
     if (priv->rx_skb)
         kfree_skb(priv->rx_skb);

^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH] at76c50x-usb: fix Oops on disconnect
  2009-02-17  1:39 [PATCH] at76c50x-usb: fix Oops on disconnect Jason Andryuk
@ 2009-02-18 20:45 ` Kalle Valo
  2009-02-19  3:09   ` Jason Andryuk
  0 siblings, 1 reply; 4+ messages in thread
From: Kalle Valo @ 2009-02-18 20:45 UTC (permalink / raw)
  To: Jason Andryuk; +Cc: linux-wireless@vger.kernel.org

Jason Andryuk <jandryuk@gmail.com> writes:

> flush_workqueue needs to be called instead of the generic one and the
> associated functions need to be modified to prevent re-adding
> themselves to the workqueue.
>
> The rx_tasklet is also killed in the small (?) chance it is scheduled.

Unfortunately the patch is corrupted:

patching file drivers/net/wireless/at76c50x-usb.c
Hunk #1 FAILED at 1860.
Hunk #2 FAILED at 1900.
patch: **** malformed patch at line 80: ieee80211_hw *hw,

Can you resend it?

-- 
Kalle Valo

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] at76c50x-usb: fix Oops on disconnect
  2009-02-18 20:45 ` Kalle Valo
@ 2009-02-19  3:09   ` Jason Andryuk
  2009-02-19  6:09     ` Kalle Valo
  0 siblings, 1 reply; 4+ messages in thread
From: Jason Andryuk @ 2009-02-19  3:09 UTC (permalink / raw)
  To: Kalle Valo; +Cc: linux-wireless@vger.kernel.org

On Wed, 2009-02-18 at 22:45 +0200, Kalle Valo wrote:
> Jason Andryuk <jandryuk@gmail.com> writes:
> 
> > flush_workqueue needs to be called instead of the generic one and the
> > associated functions need to be modified to prevent re-adding
> > themselves to the workqueue.
> >
> > The rx_tasklet is also killed in the small (?) chance it is scheduled.
> 
> Unfortunately the patch is corrupted:
> 
> patching file drivers/net/wireless/at76c50x-usb.c
> Hunk #1 FAILED at 1860.
> Hunk #2 FAILED at 1900.
> patch: **** malformed patch at line 80: ieee80211_hw *hw,
> 
> Can you resend it?

I the problem was line wrapping from gmail's web interface.

---

flush_workqueue needs to be called instead of the generic one and the
associated functions need to be modified to prevent re-adding
themselves to the workqueue.

The rx_tasklet is also killed in the small (?) chance it is scheduled.

Signed-off-by: Jason Andryuk <jandryuk@gmail.com>
--

diff --git a/drivers/net/wireless/at76c50x-usb.c b/drivers/net/wireless/at76c50x-usb.c
index 8d508ca..7a03251 100644
--- a/drivers/net/wireless/at76c50x-usb.c
+++ b/drivers/net/wireless/at76c50x-usb.c
@@ -1860,6 +1860,9 @@ static void at76_dwork_hw_scan(struct work_struct *work)
 					      dwork_hw_scan.work);
 	int ret;
 
+	if (priv->device_unplugged)
+		return;
+
 	mutex_lock(&priv->mtx);
 
 	ret = at76_get_cmd_status(priv->udev, CMD_SCAN);
@@ -1897,6 +1900,9 @@ static int at76_hw_scan(struct ieee80211_hw *hw,
 
 	at76_dbg(DBG_MAC80211, "%s():", __func__);
 
+	if (priv->device_unplugged)
+		return 0;
+
 	mutex_lock(&priv->mtx);
 
 	ieee80211_stop_queues(hw);
@@ -2003,6 +2009,10 @@ static void at76_configure_filter(struct ieee80211_hw *hw,
 	flags = changed_flags & AT76_SUPPORTED_FILTERS;
 	*total_flags = AT76_SUPPORTED_FILTERS;
 
+	/* Bail out after updating flags to prevent a WARN_ON in mac80211. */
+      	if (priv->device_unplugged)
+		return;
+
 	/* FIXME: access to priv->promisc should be protected with
 	 * priv->mtx, but it's impossible because this function needs to be
 	 * atomic */
@@ -2103,8 +2113,7 @@ static struct at76_priv *at76_alloc_new_device(struct usb_device *udev)
 	INIT_WORK(&priv->work_submit_rx, at76_work_submit_rx);
 	INIT_DELAYED_WORK(&priv->dwork_hw_scan, at76_dwork_hw_scan);
 
-	priv->rx_tasklet.func = at76_rx_tasklet;
-	priv->rx_tasklet.data = 0;
+	tasklet_init(&priv->rx_tasklet, at76_rx_tasklet, 0);
 
 	priv->pm_mode = AT76_PM_OFF;
 	priv->pm_period = 0;
@@ -2243,6 +2252,7 @@ static int at76_init_new_device(struct at76_priv *priv,
 	priv->scan_min_time = DEF_SCAN_MIN_TIME;
 	priv->scan_max_time = DEF_SCAN_MAX_TIME;
 	priv->scan_mode = SCAN_TYPE_ACTIVE;
+	priv->device_unplugged = 0;
 
 	/* mac80211 initialisation */
 	priv->hw->wiphy->max_scan_ssids = 1;
@@ -2284,13 +2294,12 @@ static void at76_delete_device(struct at76_priv *priv)
 	/* The device is gone, don't bother turning it off */
 	priv->device_unplugged = 1;
 
-	if (priv->mac80211_registered)
-		ieee80211_unregister_hw(priv->hw);
+	tasklet_kill(&priv->rx_tasklet);
 
-	/* assuming we used keventd, it must quiesce too */
-	flush_scheduled_work();
-
-	kfree(priv->bulk_out_buffer);
+	if (priv->mac80211_registered) {
+		flush_workqueue(priv->hw->workqueue);
+		ieee80211_unregister_hw(priv->hw);
+	}
 
 	if (priv->tx_urb) {
 		usb_kill_urb(priv->tx_urb);
@@ -2303,6 +2312,8 @@ static void at76_delete_device(struct at76_priv *priv)
 
 	at76_dbg(DBG_PROC_ENTRY, "%s: unlinked urbs", __func__);
 
+	kfree(priv->bulk_out_buffer);
+
 	if (priv->rx_skb)
 		kfree_skb(priv->rx_skb);
 


^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH] at76c50x-usb: fix Oops on disconnect
  2009-02-19  3:09   ` Jason Andryuk
@ 2009-02-19  6:09     ` Kalle Valo
  0 siblings, 0 replies; 4+ messages in thread
From: Kalle Valo @ 2009-02-19  6:09 UTC (permalink / raw)
  To: Jason Andryuk; +Cc: linux-wireless@vger.kernel.org

Jason Andryuk <jandryuk@gmail.com> writes:

>> Unfortunately the patch is corrupted:
>> 
>> patching file drivers/net/wireless/at76c50x-usb.c
>> Hunk #1 FAILED at 1860.
>> Hunk #2 FAILED at 1900.
>> patch: **** malformed patch at line 80: ieee80211_hw *hw,
>> 
>> Can you resend it?
>
> I the problem was line wrapping from gmail's web interface.
>
> ---
>
> flush_workqueue needs to be called instead of the generic one and the
> associated functions need to be modified to prevent re-adding
> themselves to the workqueue.
>
> The rx_tasklet is also killed in the small (?) chance it is scheduled.

Thanks, the patch was now ok. I'll test the patch and if I found no
issues, I'll send this to John. This might take few days, though.

-- 
Kalle Valo

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2009-02-19  6:09 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-02-17  1:39 [PATCH] at76c50x-usb: fix Oops on disconnect Jason Andryuk
2009-02-18 20:45 ` Kalle Valo
2009-02-19  3:09   ` Jason Andryuk
2009-02-19  6:09     ` Kalle Valo

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).