Re: 2.6.23-rc6-mm1: Build failure on ppc64 drivers/net/ehea/ehea_main.c

Re: 2.6.23-rc6-mm1: Build failure on ppc64 drivers/net/ehea/ehea_main.c

[Mel Gorman]

My apologies for the repost, this should have gone to netdev and Dave Miller
as well.

On (18/09/07 17:20), Mel Gorman didst pronounce:
> Hi Andrew,
> 
> PPC64 failed to build with the driver drivers/net/ehea with the
> following error
> 
>   CC [M]  drivers/net/ehea/ehea_main.o
> drivers/net/ehea/ehea_main.c: In function `ehea_open':
> drivers/net/ehea/ehea_main.c:2322: error: implicit declaration of function `port_napi_enable'
> drivers/net/ehea/ehea_main.c: At top level:
> drivers/net/ehea/ehea_main.c:2340: error: conflicting types for 'port_napi_enable'
> drivers/net/ehea/ehea_main.c:2322: error: previous implicit declaration of 'port_napi_enable' was here
> make[1]: *** [drivers/net/ehea/ehea_main.o] Error 1
> 
> It's buried in git-net.patch and I've cc'd a potential owner based simply
> on the mention of EHEA in a leader. I've included a candidate fix for the
> error. It seems to be a simple case of port_napi_enable being defined in
> the wrong place.
> 
> Signed-off-by: Mel Gorman <mel@csn.ul.ie>
> 
> ---
>  ehea_main.c |   32 ++++++++++++++++----------------
>  1 file changed, 16 insertions(+), 16 deletions(-)
> 
> diff -rup -X /usr/src/patchset-0.6/bin//dontdiff linux-2.6.23-rc6-mm1-005_fix-rpadlpar_sysfs/drivers/net/ehea/ehea_main.c linux-2.6.23-rc6-mm1-010_fix_ehea_main/drivers/net/ehea/ehea_main.c
> --- linux-2.6.23-rc6-mm1-005_fix-rpadlpar_sysfs/drivers/net/ehea/ehea_main.c	2007-09-18 11:29:28.000000000 +0100
> +++ linux-2.6.23-rc6-mm1-010_fix_ehea_main/drivers/net/ehea/ehea_main.c	2007-09-18 17:15:27.000000000 +0100
> @@ -2307,6 +2307,22 @@ out:
>  	return ret;
>  }
>  
> +static void port_napi_disable(struct ehea_port *port)
> +{
> +	int i;
> +
> +	for (i = 0; i < port->num_def_qps; i++)
> +		napi_disable(&port->port_res[i].napi);
> +}
> +
> +static void port_napi_enable(struct ehea_port *port)
> +{
> +	int i;
> +
> +	for (i = 0; i < port->num_def_qps; i++)
> +		napi_enable(&port->port_res[i].napi);
> +}
> +
>  static int ehea_open(struct net_device *dev)
>  {
>  	int ret;
> @@ -2328,22 +2344,6 @@ static int ehea_open(struct net_device *
>  	return ret;
>  }
>  
> -static void port_napi_disable(struct ehea_port *port)
> -{
> -	int i;
> -
> -	for (i = 0; i < port->num_def_qps; i++)
> -		napi_disable(&port->port_res[i].napi);
> -}
> -
> -static void port_napi_enable(struct ehea_port *port)
> -{
> -	int i;
> -
> -	for (i = 0; i < port->num_def_qps; i++)
> -		napi_enable(&port->port_res[i].napi);
> -}
> -
>  static int ehea_down(struct net_device *dev)
>  {
>  	int ret;
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

-- 
-- 
Mel Gorman
Part-time Phd Student                          Linux Technology Center
University of Limerick                         IBM Dublin Software Lab

Re: 2.6.23-rc6-mm1

[Andrew Morton]

On Thu, 20 Sep 2007 21:42:44 +0530
"Kamalesh Babulal" <kamalesh.babulal@gmail.com> wrote:

> ...
>
> > i have tested the change with cross compiler for power405 with the same
> > .config
> > with which the build problem is solved, but the build fails with another
> > error
> >
> >   CC [M]  drivers/net/mace.o
> > drivers/net/mace.c: In function 'mace_handle_misc_intrs':
> > drivers/net/mace.c:642: error: 'dev' undeclared (first use in this
> > function)
> > drivers/net/mace.c:642: error: (Each undeclared identifier is reported
> > only once
> > drivers/net/mace.c:642: error: for each function it appears in.)
> > make[2]: *** [drivers/net/mace.o] Error 1
> > make[1]: *** [drivers/net] Error 2
> > make: *** [drivers] Error 2
> >
> > This patch fixes the build failure
> >
> > Signed-off-by: Kamalesh Babulal <kamalesh@linux.vnet.ibm.com>
> > ---
> > --- linux-2.6.23-rc6 /drivers/net/mace.c 2007-09-20 17:16:50.000000000+0530
> > +++ linux-2.6.23-rc6/drivers/net/~mace.c        2007-09-20 17:12:
> > 47.000000000 +0530
> > @@ -633,7 +633,7 @@ static void mace_set_multicast(struct ne
> >      spin_unlock_irqrestore(&mp->lock, flags);
> >  }
> >
> > -static void mace_handle_misc_intrs(struct mace_data *mp, int intr)
> > +static void mace_handle_misc_intrs(struct mace_data *mp, int intr, struct
> > net_device *dev)
> >  {
> >      volatile struct mace __iomem *mb = mp->mace;
> >      static int mace_babbles, mace_jabbers;
> > @@ -669,7 +669,7 @@ static irqreturn_t mace_interrupt(int ir
> >      spin_lock_irqsave(&mp->lock, flags);
> >      intr = in_8(&mb->ir);              /* read interrupt register */
> >      in_8(&mb->xmtrc);                  /* get retries */
> > -    mace_handle_misc_intrs(mp, intr);
> > +    mace_handle_misc_intrs(mp, intr, dev);
> >
> >      i = mp->tx_empty;
> >      while (in_8(&mb->pr) & XMTSV) {
> > @@ -682,7 +682,7 @@ static irqreturn_t mace_interrupt(int ir
> >          */
> >         intr = in_8(&mb->ir);
> >         if (intr != 0)
> > -           mace_handle_misc_intrs(mp, intr);
> > +           mace_handle_misc_intrs(mp, intr, dev);
> >         if (mp->tx_bad_runt) {
> >             fs = in_8(&mb->xmtfs);
> >             mp->tx_bad_runt = 0;
> > @@ -817,7 +817,7 @@ static void mace_tx_timeout(unsigned lon
> >         goto out;
> >
> >      /* update various counters */
> > -    mace_handle_misc_intrs(mp, in_8(&mb->ir));
> > +    mace_handle_misc_intrs(mp, in_8(&mb->ir), dev);
> >
> >      cp = mp->tx_cmds + NCMDS_TX * mp->tx_empty;

Thanks, I will fix the wordwrapping in your patch and shall send it in to
David. 

> 
> Hi,
> 
> The build fails when compiling with the same .config over cross compiler for
> powerpc405
> 
> drivers/net/mv643xx_eth.c: In function 'mv643xx_eth_int_handler':
> drivers/net/mv643xx_eth.c:564: error: 'bp' undeclared (first use in this
> function)
> drivers/net/mv643xx_eth.c:564: error: (Each undeclared identifier is
> reported only once
> drivers/net/mv643xx_eth.c:564: error: for each function it appears in.)
> drivers/net/mv643xx_eth.c: At top level:
> drivers/net/mv643xx_eth.c:1010: error: conflicting types for 'mv643xx_poll'
> drivers/net/mv643xx_eth.c:68: error: previous declaration of 'mv643xx_poll'
> was here
> make[2]: *** [drivers/net/mv643xx_eth.o] Error 1
> make[1]: *** [drivers/net] Error 2
> make: *** [drivers] Error 2

Yes, rather a lot of net drivers got broken in easy-to-fix ways.

Re: 2.6.23-rc6-mm1: Build failures on ppc64_defconfig

[Satyam Sharma]

On Thu, 20 Sep 2007, Satyam Sharma wrote:
> 
> BTW ppc64_defconfig didn't quite like 2.6.23-rc6-mm1 either ...
> IIRC I got build failures in:

> drivers/net/spider_net.c


[PATCH -mm] spider_net: Misc build fixes after recent netdev stats changes

Unbreak the following:

drivers/net/spider_net.c: In function 'spider_net_release_tx_chain':
drivers/net/spider_net.c:818: error: 'dev' undeclared (first use in this function)
drivers/net/spider_net.c:818: error: (Each undeclared identifier is reported only once
drivers/net/spider_net.c:818: error: for each function it appears in.)
drivers/net/spider_net.c: In function 'spider_net_xmit':
drivers/net/spider_net.c:922: error: 'dev' undeclared (first use in this function)
drivers/net/spider_net.c: In function 'spider_net_pass_skb_up':
drivers/net/spider_net.c:1018: error: 'dev' undeclared (first use in this function)
drivers/net/spider_net.c: In function 'spider_net_decode_one_descr':
drivers/net/spider_net.c:1215: error: 'dev' undeclared (first use in this function)
make[2]: *** [drivers/net/spider_net.o] Error 1

Signed-off-by: Satyam Sharma <satyam@infradead.org>

---

 drivers/net/spider_net.c |   24 +++++++++++-------------
 1 file changed, 11 insertions(+), 13 deletions(-)

diff -ruNp a/drivers/net/spider_net.c b/drivers/net/spider_net.c
--- a/drivers/net/spider_net.c	2007-09-22 06:26:39.000000000 +0530
+++ b/drivers/net/spider_net.c	2007-09-22 12:12:23.000000000 +0530
@@ -795,6 +795,7 @@ spider_net_set_low_watermark(struct spid
 static int
 spider_net_release_tx_chain(struct spider_net_card *card, int brutal)
 {
+	struct net_device *dev = card->netdev;
 	struct spider_net_descr_chain *chain = &card->tx_chain;
 	struct spider_net_descr *descr;
 	struct spider_net_hw_descr *hwdescr;
@@ -919,7 +920,7 @@ spider_net_xmit(struct sk_buff *skb, str
 	spider_net_release_tx_chain(card, 0);
 
 	if (spider_net_prepare_tx_descr(card, skb) != 0) {
-		dev->stats.tx_dropped++;
+		netdev->stats.tx_dropped++;
 		netif_stop_queue(netdev);
 		return NETDEV_TX_BUSY;
 	}
@@ -979,16 +980,12 @@ static void
 spider_net_pass_skb_up(struct spider_net_descr *descr,
 		       struct spider_net_card *card)
 {
-	struct spider_net_hw_descr *hwdescr= descr->hwdescr;
-	struct sk_buff *skb;
-	struct net_device *netdev;
-	u32 data_status, data_error;
-
-	data_status = hwdescr->data_status;
-	data_error = hwdescr->data_error;
-	netdev = card->netdev;
+	struct spider_net_hw_descr *hwdescr = descr->hwdescr;
+	struct sk_buff *skb = descr->skb;
+	struct net_device *netdev = card->netdev;
+	u32 data_status = hwdescr->data_status;
+	u32 data_error = hwdescr->data_error;
 
-	skb = descr->skb;
 	skb_put(skb, hwdescr->valid_size);
 
 	/* the card seems to add 2 bytes of junk in front
@@ -1015,8 +1012,8 @@ spider_net_pass_skb_up(struct spider_net
 	}
 
 	/* update netdevice statistics */
-	dev->stats.rx_packets++;
-	dev->stats.rx_bytes += skb->len;
+	netdev->stats.rx_packets++;
+	netdev->stats.rx_bytes += skb->len;
 
 	/* pass skb up to stack */
 	netif_receive_skb(skb);
@@ -1184,6 +1181,7 @@ static int spider_net_resync_tail_ptr(st
 static int
 spider_net_decode_one_descr(struct spider_net_card *card)
 {
+	struct net_device *dev = card->netdev;
 	struct spider_net_descr_chain *chain = &card->rx_chain;
 	struct spider_net_descr *descr = chain->tail;
 	struct spider_net_hw_descr *hwdescr = descr->hwdescr;
@@ -1210,7 +1208,7 @@ spider_net_decode_one_descr(struct spide
 	     (status == SPIDER_NET_DESCR_PROTECTION_ERROR) ||
 	     (status == SPIDER_NET_DESCR_FORCE_END) ) {
 		if (netif_msg_rx_err(card))
-			dev_err(&card->netdev->dev,
+			dev_err(&dev->dev,
 			       "dropping RX descriptor with state %d\n", status);
 		dev->stats.rx_dropped++;
 		goto bad_desc;

Re: 2.6.23-rc6-mm1: Build failures on ppc64_defconfig

[Satyam Sharma]

On Thu, 20 Sep 2007, Satyam Sharma wrote:
> 
> BTW ppc64_defconfig didn't quite like 2.6.23-rc6-mm1 either ...
> IIRC I got build failures in:

> drivers/net/spider_net.c

Fixing the above showed up another problem in another file of the
same driver (drivers/net/spider_net_ethtool.c)


[PATCH -mm] spider_net_ethtool: Keep up with recent netdev stats changes

Unbreak the following:

drivers/net/spider_net_ethtool.c: In function 'spider_net_get_ethtool_stats':
drivers/net/spider_net_ethtool.c:160: error: structure has no member named 'netdev_stats'
drivers/net/spider_net_ethtool.c:161: error: structure has no member named 'netdev_stats'
drivers/net/spider_net_ethtool.c:162: error: structure has no member named 'netdev_stats'
drivers/net/spider_net_ethtool.c:163: error: structure has no member named 'netdev_stats'
drivers/net/spider_net_ethtool.c:164: error: structure has no member named 'netdev_stats'
drivers/net/spider_net_ethtool.c:165: error: structure has no member named 'netdev_stats'
drivers/net/spider_net_ethtool.c:166: error: structure has no member named 'netdev_stats'
make[2]: *** [drivers/net/spider_net_ethtool.o] Error 1

Also do another ARRAY_SIZE() cleanup while at it.

Signed-off-by: Satyam Sharma <satyam@infradead.org>

---

 drivers/net/spider_net_ethtool.c |   18 ++++++++----------
 1 file changed, 8 insertions(+), 10 deletions(-)

diff -ruNp a/drivers/net/spider_net_ethtool.c b/drivers/net/spider_net_ethtool.c
--- a/drivers/net/spider_net_ethtool.c	2007-09-22 06:26:39.000000000 +0530
+++ b/drivers/net/spider_net_ethtool.c	2007-09-22 12:43:51.000000000 +0530
@@ -28,8 +28,6 @@
 #include "spider_net.h"
 
 
-#define SPIDER_NET_NUM_STATS 13
-
 static struct {
 	const char str[ETH_GSTRING_LEN];
 } ethtool_stats_keys[] = {
@@ -149,7 +147,7 @@ spider_net_ethtool_get_ringparam(struct 
 
 static int spider_net_get_stats_count(struct net_device *netdev)
 {
-	return SPIDER_NET_NUM_STATS;
+	return ARRAY_SIZE(ethtool_stats_keys);
 }
 
 static void spider_net_get_ethtool_stats(struct net_device *netdev,
@@ -157,13 +155,13 @@ static void spider_net_get_ethtool_stats
 {
 	struct spider_net_card *card = netdev->priv;
 
-	data[0] = card->netdev_stats.tx_packets;
-	data[1] = card->netdev_stats.tx_bytes;
-	data[2] = card->netdev_stats.rx_packets;
-	data[3] = card->netdev_stats.rx_bytes;
-	data[4] = card->netdev_stats.tx_errors;
-	data[5] = card->netdev_stats.tx_dropped;
-	data[6] = card->netdev_stats.rx_dropped;
+	data[0] = netdev->stats.tx_packets;
+	data[1] = netdev->stats.tx_bytes;
+	data[2] = netdev->stats.rx_packets;
+	data[3] = netdev->stats.rx_bytes;
+	data[4] = netdev->stats.tx_errors;
+	data[5] = netdev->stats.tx_dropped;
+	data[6] = netdev->stats.rx_dropped;
 	data[7] = card->spider_stats.rx_desc_error;
 	data[8] = card->spider_stats.tx_timeouts;
 	data[9] = card->spider_stats.alloc_rx_skb_error;

[PATCH -mm] pasemi_mac: Build fix after recent netdev stats changes

[Satyam Sharma]

On Thu, 20 Sep 2007, Satyam Sharma wrote:
> 
> BTW ppc64_defconfig didn't quite like 2.6.23-rc6-mm1 either ...
> IIRC I got build failures in:

> drivers/net/pasemi_mac.c


[PATCH -mm] pasemi_mac: Build fix after recent netdev stats changes

Unbreak the following:

drivers/net/pasemi_mac.c: In function 'pasemi_mac_clean_rx':
drivers/net/pasemi_mac.c:533: error: 'dev' undeclared (first use in this function)
drivers/net/pasemi_mac.c:533: error: (Each undeclared identifier is reported only once
drivers/net/pasemi_mac.c:533: error: for each function it appears in.)

And remove an unused static function:

drivers/net/pasemi_mac.c: At top level:
drivers/net/pasemi_mac.c:89: warning: 'read_iob_reg' defined but not used

Signed-off-by: Satyam Sharma <satyam@infradead.org>

---

 drivers/net/pasemi_mac.c |    9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff -ruNp a/drivers/net/pasemi_mac.c b/drivers/net/pasemi_mac.c
--- a/drivers/net/pasemi_mac.c	2007-09-22 06:26:39.000000000 +0530
+++ b/drivers/net/pasemi_mac.c	2007-09-22 13:03:04.000000000 +0530
@@ -85,11 +85,6 @@ MODULE_PARM_DESC(debug, "PA Semi MAC bit
 
 static struct pasdma_status *dma_status;
 
-static unsigned int read_iob_reg(struct pasemi_mac *mac, unsigned int reg)
-{
-	return in_le32(mac->iob_regs+reg);
-}
-
 static void write_iob_reg(struct pasemi_mac *mac, unsigned int reg,
 			  unsigned int val)
 {
@@ -530,8 +525,8 @@ static int pasemi_mac_clean_rx(struct pa
 		} else
 			skb->ip_summed = CHECKSUM_NONE;
 
-		dev->stats.rx_bytes += len;
-		dev->stats.rx_packets++;
+		mac->netdev->stats.rx_bytes += len;
+		mac->netdev->stats.rx_packets++;
 
 		skb->protocol = eth_type_trans(skb, mac->netdev);
 		netif_receive_skb(skb);

[PATCH -mm] mv643xx_eth: Remove redundant multiple initialization

[Satyam Sharma]

Of ethtool_ops->get_stats_count and ethtool_ops->get_ethtool_stats.

Signed-off-by: Satyam Sharma <satyam@infradead.org>

---

 drivers/net/mv643xx_eth.c |    2 --
 1 file changed, 2 deletions(-)

diff -ruNp a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c
--- a/drivers/net/mv643xx_eth.c	2007-09-22 06:26:39.000000000 +0530
+++ b/drivers/net/mv643xx_eth.c	2007-09-22 13:15:41.000000000 +0530
@@ -2740,8 +2740,6 @@ static const struct ethtool_ops mv643xx_
 	.get_stats_count        = mv643xx_get_stats_count,
 	.get_ethtool_stats      = mv643xx_get_ethtool_stats,
 	.get_strings            = mv643xx_get_strings,
-	.get_stats_count        = mv643xx_get_stats_count,
-	.get_ethtool_stats      = mv643xx_get_ethtool_stats,
 	.nway_reset		= mv643xx_eth_nway_restart,
 };
 

[PATCH -mm] iseries_veth: Kill unused variable

[Satyam Sharma]

drivers/net/iseries_veth.c: In function 'veth_transmit_to_many':
drivers/net/iseries_veth.c:1174: warning: unused variable 'port'

Signed-off-by: Satyam Sharma <satyam@infradead.org>

---

 drivers/net/iseries_veth.c |    1 -
 1 file changed, 1 deletion(-)

diff -ruNp a/drivers/net/iseries_veth.c b/drivers/net/iseries_veth.c
--- a/drivers/net/iseries_veth.c	2007-09-22 06:26:39.000000000 +0530
+++ b/drivers/net/iseries_veth.c	2007-09-22 11:17:29.000000000 +0530
@@ -1171,7 +1171,6 @@ static void veth_transmit_to_many(struct
 					  HvLpIndexMap lpmask,
 					  struct net_device *dev)
 {
-	struct veth_port *port = (struct veth_port *) dev->priv;
 	int i, success, error;
 
 	success = error = 0;

Re: 2.6.23-rc6-mm1: Build failures on ppc64_defconfig

[Mel Gorman]

On (22/09/07 12:24), Satyam Sharma didst pronounce:
> 
> 
> On Thu, 20 Sep 2007, Satyam Sharma wrote:
> > 
> > BTW ppc64_defconfig didn't quite like 2.6.23-rc6-mm1 either ...
> > IIRC I got build failures in:
> 
> > drivers/net/spider_net.c
> 
> 
> [PATCH -mm] spider_net: Misc build fixes after recent netdev stats changes
> 
> Unbreak the following:
> 
> drivers/net/spider_net.c: In function 'spider_net_release_tx_chain':
> drivers/net/spider_net.c:818: error: 'dev' undeclared (first use in this function)
> drivers/net/spider_net.c:818: error: (Each undeclared identifier is reported only once
> drivers/net/spider_net.c:818: error: for each function it appears in.)
> drivers/net/spider_net.c: In function 'spider_net_xmit':
> drivers/net/spider_net.c:922: error: 'dev' undeclared (first use in this function)
> drivers/net/spider_net.c: In function 'spider_net_pass_skb_up':
> drivers/net/spider_net.c:1018: error: 'dev' undeclared (first use in this function)
> drivers/net/spider_net.c: In function 'spider_net_decode_one_descr':
> drivers/net/spider_net.c:1215: error: 'dev' undeclared (first use in this function)
> make[2]: *** [drivers/net/spider_net.o] Error 1
> 
> Signed-off-by: Satyam Sharma <satyam@infradead.org>

I've confirmed that this patch fixes the build error in question.

Acked-by: Mel Gorman <mel@csn.ul.ie>

> 
> ---
> 
>  drivers/net/spider_net.c |   24 +++++++++++-------------
>  1 file changed, 11 insertions(+), 13 deletions(-)
> 
> diff -ruNp a/drivers/net/spider_net.c b/drivers/net/spider_net.c
> --- a/drivers/net/spider_net.c	2007-09-22 06:26:39.000000000 +0530
> +++ b/drivers/net/spider_net.c	2007-09-22 12:12:23.000000000 +0530
> @@ -795,6 +795,7 @@ spider_net_set_low_watermark(struct spid
>  static int
>  spider_net_release_tx_chain(struct spider_net_card *card, int brutal)
>  {
> +	struct net_device *dev = card->netdev;
>  	struct spider_net_descr_chain *chain = &card->tx_chain;
>  	struct spider_net_descr *descr;
>  	struct spider_net_hw_descr *hwdescr;
> @@ -919,7 +920,7 @@ spider_net_xmit(struct sk_buff *skb, str
>  	spider_net_release_tx_chain(card, 0);
>  
>  	if (spider_net_prepare_tx_descr(card, skb) != 0) {
> -		dev->stats.tx_dropped++;
> +		netdev->stats.tx_dropped++;
>  		netif_stop_queue(netdev);
>  		return NETDEV_TX_BUSY;
>  	}
> @@ -979,16 +980,12 @@ static void
>  spider_net_pass_skb_up(struct spider_net_descr *descr,
>  		       struct spider_net_card *card)
>  {
> -	struct spider_net_hw_descr *hwdescr= descr->hwdescr;
> -	struct sk_buff *skb;
> -	struct net_device *netdev;
> -	u32 data_status, data_error;
> -
> -	data_status = hwdescr->data_status;
> -	data_error = hwdescr->data_error;
> -	netdev = card->netdev;
> +	struct spider_net_hw_descr *hwdescr = descr->hwdescr;
> +	struct sk_buff *skb = descr->skb;
> +	struct net_device *netdev = card->netdev;
> +	u32 data_status = hwdescr->data_status;
> +	u32 data_error = hwdescr->data_error;
>  
> -	skb = descr->skb;
>  	skb_put(skb, hwdescr->valid_size);
>  
>  	/* the card seems to add 2 bytes of junk in front
> @@ -1015,8 +1012,8 @@ spider_net_pass_skb_up(struct spider_net
>  	}
>  
>  	/* update netdevice statistics */
> -	dev->stats.rx_packets++;
> -	dev->stats.rx_bytes += skb->len;
> +	netdev->stats.rx_packets++;
> +	netdev->stats.rx_bytes += skb->len;
>  
>  	/* pass skb up to stack */
>  	netif_receive_skb(skb);
> @@ -1184,6 +1181,7 @@ static int spider_net_resync_tail_ptr(st
>  static int
>  spider_net_decode_one_descr(struct spider_net_card *card)
>  {
> +	struct net_device *dev = card->netdev;
>  	struct spider_net_descr_chain *chain = &card->rx_chain;
>  	struct spider_net_descr *descr = chain->tail;
>  	struct spider_net_hw_descr *hwdescr = descr->hwdescr;
> @@ -1210,7 +1208,7 @@ spider_net_decode_one_descr(struct spide
>  	     (status == SPIDER_NET_DESCR_PROTECTION_ERROR) ||
>  	     (status == SPIDER_NET_DESCR_FORCE_END) ) {
>  		if (netif_msg_rx_err(card))
> -			dev_err(&card->netdev->dev,
> +			dev_err(&dev->dev,
>  			       "dropping RX descriptor with state %d\n", status);
>  		dev->stats.rx_dropped++;
>  		goto bad_desc;
> 

-- 
-- 
Mel Gorman
Part-time Phd Student                          Linux Technology Center
University of Limerick                         IBM Dublin Software Lab

Re: [PATCH] Remove broken netfilter binary sysctls from bridging code

[Patrick McHardy]

Stephen Hemminger wrote:
> On Mon, 24 Sep 2007 18:55:38 +0200
> Patrick McHardy <kaber@trash.net> wrote:
> 
>>Eric W. Biederman wrote:
>>
>>>A really good fix would be to remove the binary side and then to
>>>modify brnf_sysctl_call_tables to allocate a temporary ctl_table and
>>>integer on the stack and only set ctl->data after we have normalized
>>>the written value.  But since in practice nothing cares about
>>>the race a better fix probably isn't worth it.
>>
>>
>>I seem to be missing something, the entire brnf_sysctl_call_tables
>>thing looks purely cosmetic to me, wouldn't it be better to simply
>>remove it?
> 
> 
> I agree, removing seems like a better option.  But probably need to go
> through a 3-6mo warning period, since sysctl's are technically an API.


I meant removing brnf_sysctl_call_tables function, not the sysctls
themselves, all it does is change values != 0 to 1. Or did you
actually mean that something in userspace might depend on reading
back the value 1 after writing a value != 0?

Re: [PATCH] Remove broken netfilter binary sysctls from bridging code

[Stephen Hemminger]

On Tue, 25 Sep 2007 06:07:24 +0200
Patrick McHardy <kaber@trash.net> wrote:

> Stephen Hemminger wrote:
> > On Mon, 24 Sep 2007 18:55:38 +0200
> > Patrick McHardy <kaber@trash.net> wrote:
> > 
> >>Eric W. Biederman wrote:
> >>
> >>>A really good fix would be to remove the binary side and then to
> >>>modify brnf_sysctl_call_tables to allocate a temporary ctl_table
> >>>and integer on the stack and only set ctl->data after we have
> >>>normalized the written value.  But since in practice nothing cares
> >>>about the race a better fix probably isn't worth it.
> >>
> >>
> >>I seem to be missing something, the entire brnf_sysctl_call_tables
> >>thing looks purely cosmetic to me, wouldn't it be better to simply
> >>remove it?
> > 
> > 
> > I agree, removing seems like a better option.  But probably need to
> > go through a 3-6mo warning period, since sysctl's are technically
> > an API.
> 
> 
> I meant removing brnf_sysctl_call_tables function, not the sysctls
> themselves, all it does is change values != 0 to 1. Or did you
> actually mean that something in userspace might depend on reading
> back the value 1 after writing a value != 0?

I was going farther, because don't really see the value of having
a sysctl for this. It seems better to just not load filters if
they aren't going to be used. Having another enable/disable hook
just adds needless complexity.

Re: [PATCH] Remove broken netfilter binary sysctls from bridging code

[Patrick McHardy]

Stephen Hemminger wrote:
> On Tue, 25 Sep 2007 06:07:24 +0200
> Patrick McHardy <kaber@trash.net> wrote:
>
>   
>> I meant removing brnf_sysctl_call_tables function, not the sysctls
>> themselves, all it does is change values != 0 to 1. Or did you
>> actually mean that something in userspace might depend on reading
>> back the value 1 after writing a value != 0?
>>     
>
> I was going farther, because don't really see the value of having
> a sysctl for this. It seems better to just not load filters if
> they aren't going to be used. Having another enable/disable hook
> just adds needless complexity.
>   

These sysctls control whether bridged packets will be handled
by iptables and friends. The bridge netfilter code always
handles bridged packets, and iptables might be loaded for
different reasons. So I don't see how that would work.

I think it should be specified in the ebtables ruleset, but
the current netfilter infrastructure doesn't allow to do that
cleanly.