Skip to contentSkip to conversation

staging-p80211: Kill directly reference of netdev->priv

PATCH Applied Wang Chen 3 participants 7 messages Dec 5, 2008 – Dec 7, 2008

Subsystems (3)
Networking Network drivers Staging
lore
Share & tools
lore

Wang Chen

author
#0 lore raw

PATCH 1/1 staging-p80211: Kill directly reference of netdev->priv

In this driver, netdev's private data is wlandevice_t.
And the wlandev(type of wlandevice_t) is exist before netdev be allocated. So use netdev->ml_priv to point to the private data.

I am not sure whether I should consider the kernel version older than 2.3.38. Because in those kernels, netdevice_t is "structure dev" instead of "structure net_device" and of course "dev->ml_priv" will cause compile error. But before my patch, in function wlan_setup(), there is a ether_setup(net_device) which already broke kernels which older than 2.3.38.

Compile test only.

Signed-off-by Wang Chen <[email protected]>
---
 drivers/staging/wlan-ng/p80211netdev.c |   20 +++++-----
 drivers/staging/wlan-ng/p80211wext.c   |   58 ++++++++++++++++----------------
 2 files changed, 39 insertions(+), 39 deletions(-)
drivers/staging/wlan-ng/p80211netdev.c +10 -10
index 11f84a8..2b705ea 100644
@@ -244,7 +244,7 @@ static int p80211knetdev_init( netdevice_t *netdev)
 static struct net_device_stats*
 p80211knetdev_get_stats(netdevice_t *netdev)
 {
-	wlandevice_t	*wlandev = (wlandevice_t*)netdev->priv;
+	wlandevice_t	*wlandev = netdev->ml_priv;
 	DBFENTER;
 
 	/* TODO: review the MIB stats for items that correspond to

@@ -272,7 +272,7 @@ p80211knetdev_get_stats(netdevice_t *netdev)
 static int p80211knetdev_open( netdevice_t *netdev )
 {
 	int 		result = 0; /* success */
-	wlandevice_t	*wlandev = (wlandevice_t*)(netdev->priv);
+	wlandevice_t	*wlandev = netdev->ml_priv;
 
 	DBFENTER;
 

@@ -315,7 +315,7 @@ static int p80211knetdev_open( netdevice_t *netdev )
 static int p80211knetdev_stop( netdevice_t *netdev )
 {
 	int		result = 0;
-	wlandevice_t	*wlandev = (wlandevice_t*)(netdev->priv);
+	wlandevice_t	*wlandev = netdev->ml_priv;
 
 	DBFENTER;
 

@@ -460,7 +460,7 @@ static int p80211knetdev_hard_start_xmit( struct sk_buff *skb, netdevice_t *netd
 {
 	int		result = 0;
 	int		txresult = -1;
-	wlandevice_t	*wlandev = (wlandevice_t*)netdev->priv;
+	wlandevice_t	*wlandev = netdev->ml_priv;
 	p80211_hdr_t    p80211_hdr;
 	p80211_metawep_t p80211_wep;
 

@@ -603,7 +603,7 @@ static int p80211knetdev_hard_start_xmit( struct sk_buff *skb, netdevice_t *netd
 ----------------------------------------------------------------*/
 static void p80211knetdev_set_multicast_list(netdevice_t *dev)
 {
-	wlandevice_t	*wlandev = (wlandevice_t*)dev->priv;
+	wlandevice_t	*wlandev = dev->ml_priv;
 
 	DBFENTER;
 

@@ -696,7 +696,7 @@ static int p80211knetdev_do_ioctl(netdevice_t *dev, struct ifreq *ifr, int cmd)
 {
 	int			result = 0;
 	p80211ioctl_req_t	*req = (p80211ioctl_req_t*)ifr;
-	wlandevice_t		*wlandev = (wlandevice_t*)dev->priv;
+	wlandevice_t		*wlandev = dev->ml_priv;
 	UINT8			*msgbuf;
 	DBFENTER;
 

@@ -812,7 +812,7 @@ static int p80211knetdev_set_mac_address(netdevice_t *dev, void *addr)
 	dot11req.msgcode = DIDmsg_dot11req_mibset;
 	dot11req.msglen = sizeof(p80211msg_dot11req_mibset_t);
 	memcpy(dot11req.devname,
-		((wlandevice_t*)(dev->priv))->name,
+		((wlandevice_t *)dev->ml_priv)->name,
 		WLAN_DEVNAMELEN_MAX - 1);
 
 	/* Set up the mibattribute argument */

@@ -833,7 +833,7 @@ static int p80211knetdev_set_mac_address(netdevice_t *dev, void *addr)
 	resultcode->data = 0;
 
 	/* now fire the request */
-	result = p80211req_dorequest(dev->priv, (UINT8*)&dot11req);
+	result = p80211req_dorequest(dev->ml_priv, (UINT8 *)&dot11req);
 
 	/* If the request wasn't successful, report an error and don't
 	 * change the netdev address

@@ -917,7 +917,7 @@ int wlan_setup(wlandevice_t *wlandev)
 		memset( dev, 0, sizeof(netdevice_t));
 		ether_setup(dev);
 		wlandev->netdev = dev;
-		dev->priv = wlandev;
+		dev->ml_priv = wlandev;
 		dev->hard_start_xmit =	p80211knetdev_hard_start_xmit;
 		dev->get_stats =	p80211knetdev_get_stats;
 #ifdef HAVE_PRIVATE_IOCTL

@@ -1487,7 +1487,7 @@ void    p80211_resume(wlandevice_t *wlandev)
 
 static void p80211knetdev_tx_timeout( netdevice_t *netdev)
 {
-	wlandevice_t	*wlandev = (wlandevice_t*)netdev->priv;
+	wlandevice_t	*wlandev = netdev->ml_priv;
 	DBFENTER;
 
 	if (wlandev->tx_timeout) {
drivers/staging/wlan-ng/p80211wext.c +29 -29

232 lines not on this page · raw message

Signature · 1 line
1.5.3.4
#1 lore raw
Quoted 9 lines · Wang Chen

In this driver, netdev's private data is wlandevice_t.
And the wlandev(type of wlandevice_t) is exist before netdev be allocated. So use netdev->ml_priv to point to the private data.

I am not sure whether I should consider the kernel version older than 2.3.38. Because in those kernels, netdevice_t is "structure dev" instead of "structure net_device" and of course "dev->ml_priv" will cause compile error. But before my patch, in function wlan_setup(), there is a ether_setup(net_device) which already broke kernels which older than 2.3.38.

Heh, no we only need to support the current mainline kernel with this driver, no old 2.3.x versions :)

thanks for the patch, I'll queue it up,

greg k-h

author
#2 lore raw

Re: [PATCH next] staging-p80211: Kill directly reference of c

Greg KH said the following on 2008-12-5 23:44:

Quoted 5 lines · Wang Chen
Quoted 4 lines

In this driver, netdev's private data is wlandevice_t.
And the wlandev(type of wlandevice_t) is exist before netdev be allocated. So use netdev->ml_priv to point to the private data.

Quoted 9 lines
Quoted 5 lines

I am not sure whether I should consider the kernel version older than 2.3.38. Because in those kernels, netdevice_t is "structure dev" instead of "structure net_device" and of course "dev->ml_priv" will cause compile error. But before my patch, in function wlan_setup(), there is a ether_setup(net_device) which already broke kernels which older than 2.3.38.

Heh, no we only need to support the current mainline kernel with this driver, no old 2.3.x versions :)

Thanks for your explanation.

Quoted 2 lines

thanks for the patch, I'll queue it up,

I'm doing this "kill netdev->priv" work on Dave's net-next tree now. So, will it be better to let Dave apply this to his net-next instead you apply it?
Just suggestion :)

maintainer
#3 lore raw

Re: [PATCH next] staging-p80211: Kill directly reference of c

From: Wang Chen <[email protected]>
Date: Sat, 06 Dec 2008 10:28:45 +0800

Quoted 7 lines

Greg KH said the following on 2008-12-5 23:44:

Quoted 1 line

thanks for the patch, I'll queue it up,

I'm doing this "kill netdev->priv" work on Dave's net-next tree now. So, will it be better to let Dave apply this to his net-next instead you apply it?
Just suggestion :)

Yes, I would prefer it be done this way so I can do the final kill of the netdev->priv member without having to wait for a dependency on the staging tree :)

Acked-by
#4 lore raw

Re: [PATCH next] staging-p80211: Kill directly reference of c

Quoted 6 lines · David Miller

From: Wang Chen <[email protected]>
Date: Sat, 06 Dec 2008 10:28:45 +0800

Quoted 3 lines

Greg KH said the following on 2008-12-5 23:44:

Quoted 1 line

thanks for the patch, I'll queue it up,

Quoted 8 lines
Quoted 4 lines

I'm doing this "kill netdev->priv" work on Dave's net-next tree now. So, will it be better to let Dave apply this to his net-next instead you apply it?
Just suggestion :)

Yes, I would prefer it be done this way so I can do the final kill of the netdev->priv member without having to wait for a dependency on the staging tree :)

Feel free to take it yourself, I have no objection to anyone else taking drivers/staging/ patches if they want to. You're the first one to be so brave :)

You can add a:

Acked-by Greg Kroah-Hartman <[email protected]>

to the patch.

thanks,

greg k-h

maintainer
#5 lore raw

Re: [PATCH next] staging-p80211: Kill directly reference of c

From: Greg KH <[email protected]>
Date: Fri, 5 Dec 2008 18:41:23 -0800

Quoted 16 lines · David Miller
Quoted 6 lines

From: Wang Chen <[email protected]>
Date: Sat, 06 Dec 2008 10:28:45 +0800

Quoted 3 lines

Greg KH said the following on 2008-12-5 23:44:

Quoted 1 line

thanks for the patch, I'll queue it up,

Quoted 8 lines
Quoted 4 lines

I'm doing this "kill netdev->priv" work on Dave's net-next tree now. So, will it be better to let Dave apply this to his net-next instead you apply it?
Just suggestion :)

Yes, I would prefer it be done this way so I can do the final kill of the netdev->priv member without having to wait for a dependency on the staging tree :)

Quoted 7 lines

Feel free to take it yourself, I have no objection to anyone else taking drivers/staging/ patches if they want to. You're the first one to be so brave :)

You can add a:

Acked-by Greg Kroah-Hartman <[email protected]>

to the patch.

Great, I'll do that, thanks Greg!

maintainer Applied
#6 lore raw
In reply to Wang Chen

From: Wang Chen <[email protected]>
Date: Fri, 05 Dec 2008 23:14:53 +0800

Quoted 10 lines

In this driver, netdev's private data is wlandevice_t.
And the wlandev(type of wlandevice_t) is exist before netdev be allocated. So use netdev->ml_priv to point to the private data.

I am not sure whether I should consider the kernel version older than 2.3.38. Because in those kernels, netdevice_t is "structure dev" instead of "structure net_device" and of course "dev->ml_priv" will cause compile error. But before my patch, in function wlan_setup(), there is a ether_setup(net_device) which already broke kernels which older than 2.3.38.

Quoted 3 lines

Compile test only.

Signed-off-by Wang Chen <[email protected]>

Applied.