Home » Mailing list archives » linux-netdev » 2009 » June » 30

[3/7] net1080: Use netdev stats structure

Previous thread: Re: [Bugme-new] [Bug 13568] New: Intel e1000 4-port NIC - unable to communicate, slowly blinking by Andrew Morton on Monday, June 29, 2009 - 4:00 pm. (2 messages)

Next thread: (no subject) by Mrs Dianne Thompson on Monday, June 29, 2009 - 6:28 pm. (1 message)
[view original message]
From: Andrew Morton
Subject: Re: [Bugme-new] [Bug 13589] New: usbnet and /proc/net/dev stats
Date: Monday, June 29, 2009 - 5:23 pm

(switched to email.  Please respond via emailed reply-to-all, not via the
bugzilla web interface).

On Sat, 20 Jun 2009 16:24:13 GMT

This has turned into a bit of a mess, as the fix which I typed in needs
the revert of another patch, and I don't know what actually went wrong
with usbnet stats anyway.

Could someone (Stephen?) please take a look?

As it appears to be a 2.6.29->2.6.30 regression, please remember to cc
stable@kernel.org on any fixes, thanks.

--


[ message continues ]
[view original message]
From: Ben Kibbey
Subject: Re: [Bugme-new] [Bug 13589] New: usbnet and /proc/net/dev stats
Date: Monday, June 29, 2009 - 5:42 pm

On Mon, Jun 29, 2009 at 05:23:41PM -0700, Andrew Morton wrote:

All usbnet stats weren't getting updated (ifconfig, was always showing 0).
When applied to 2.6.30 this fixes it. Commit 805aaa29fa was the culprit and
reverting it re-adds usbnet_get_stats() but needed to be modified to add whats
in usbnet_netdev_ops:

--- /tmp/usbnet.c	2009-06-29 20:35:59.000000000 -0400
+++ usbnet.c	2009-06-20 11:40:41.000000000 -0400
@@ -1106,9 +1106,16 @@
 }
 EXPORT_SYMBOL_GPL(usbnet_disconnect);
 
+static struct net_device_stats *usbnet_get_stats (struct net_device *net)
+{
+       struct usbnet   *dev = netdev_priv(net);
+       return &dev->stats;
+}
+
 static const struct net_device_ops usbnet_netdev_ops = {
 	.ndo_open		= usbnet_open,
 	.ndo_stop		= usbnet_stop,
+	.ndo_get_stats		= usbnet_get_stats,
 	.ndo_start_xmit		= usbnet_start_xmit,
 	.ndo_tx_timeout		= usbnet_tx_timeout,
 	.ndo_change_mtu		= usbnet_change_mtu,

-- 
Ben Kibbey (bjk) @ FreeNode/OFTC/Jabber
--


[ message continues ]
[view original message]
From: Herbert Xu
Subject: [1/7] cdc_eem: Use netdev stats structure
Date: Monday, June 29, 2009 - 7:49 pm

This series should fix the problem:

cdc_eem: Use netdev stats structure

Now that netdev has its own stats structure we should use that
instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

diff --git a/drivers/net/usb/cdc_eem.c b/drivers/net/usb/cdc_eem.c
index 80e0177..cd35d50 100644
--- a/drivers/net/usb/cdc_eem.c
+++ b/drivers/net/usb/cdc_eem.c
@@ -319,7 +319,7 @@ static int eem_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 				return crc == crc2;
 
 			if (unlikely(crc != crc2)) {
-				dev->stats.rx_errors++;
+				dev->net->stats.rx_errors++;
 				dev_kfree_skb_any(skb2);
 			} else
 				usbnet_skb_return(dev, skb2);

Cheers,
-- 
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
--


[ message continues ]
[view original message]
From: Herbert Xu
Subject: [2/7] dm9601: Use netdev stats structure
Date: Monday, June 29, 2009 - 7:50 pm

dm9601: Use netdev stats structure

Now that netdev has its own stats structure we should use that
instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

diff --git a/drivers/net/usb/dm9601.c b/drivers/net/usb/dm9601.c
index 6fc4f82..9acd574 100644
--- a/drivers/net/usb/dm9601.c
+++ b/drivers/net/usb/dm9601.c
@@ -513,11 +513,11 @@ static int dm9601_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 	len = (skb->data[1] | (skb->data[2] << 8)) - 4;
 
 	if (unlikely(status & 0xbf)) {
-		if (status & 0x01) dev->stats.rx_fifo_errors++;
-		if (status & 0x02) dev->stats.rx_crc_errors++;
-		if (status & 0x04) dev->stats.rx_frame_errors++;
-		if (status & 0x20) dev->stats.rx_missed_errors++;
-		if (status & 0x90) dev->stats.rx_length_errors++;
+		if (status & 0x01) dev->net->stats.rx_fifo_errors++;
+		if (status & 0x02) dev->net->stats.rx_crc_errors++;
+		if (status & 0x04) dev->net->stats.rx_frame_errors++;
+		if (status & 0x20) dev->net->stats.rx_missed_errors++;
+		if (status & 0x90) dev->net->stats.rx_length_errors++;
 		return 0;
 	}

Cheers,
-- 
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
--


[ message continues ]
[view original message]
From: Herbert Xu
Subject: [3/7] net1080: Use netdev stats structure
Date: Monday, June 29, 2009 - 7:51 pm

net1080: Use netdev stats structure

Now that netdev has its own stats structure we should use that
instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

diff --git a/drivers/net/usb/net1080.c b/drivers/net/usb/net1080.c
index 034e8a7..aeb1ab0 100644
--- a/drivers/net/usb/net1080.c
+++ b/drivers/net/usb/net1080.c
@@ -433,7 +433,7 @@ static int net1080_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 		dbg("rx framesize %d range %d..%d mtu %d", skb->len,
 			net->hard_header_len, dev->hard_mtu, net->mtu);
 #endif
-		dev->stats.rx_frame_errors++;
+		dev->net->stats.rx_frame_errors++;
 		nc_ensure_sync(dev);
 		return 0;
 	}
@@ -442,12 +442,12 @@ static int net1080_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 	hdr_len = le16_to_cpup(&header->hdr_len);
 	packet_len = le16_to_cpup(&header->packet_len);
 	if (FRAMED_SIZE(packet_len) > NC_MAX_PACKET) {
-		dev->stats.rx_frame_errors++;
+		dev->net->stats.rx_frame_errors++;
 		dbg("packet too big, %d", packet_len);
 		nc_ensure_sync(dev);
 		return 0;
 	} else if (hdr_len < MIN_HEADER) {
-		dev->stats.rx_frame_errors++;
+		dev->net->stats.rx_frame_errors++;
 		dbg("header too short, %d", hdr_len);
 		nc_ensure_sync(dev);
 		return 0;
@@ -465,21 +465,21 @@ static int net1080_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 
 	if ((packet_len & 0x01) == 0) {
 		if (skb->data [packet_len] != PAD_BYTE) {
-			dev->stats.rx_frame_errors++;
+			dev->net->stats.rx_frame_errors++;
 			dbg("bad pad");
 			return 0;
 		}
 		skb_trim(skb, skb->len - 1);
 	}
 	if (skb->len != packet_len) {
-		dev->stats.rx_frame_errors++;
+		dev->net->stats.rx_frame_errors++;
 		dbg("bad packet len %d (expected %d)",
 			skb->len, packet_len);
 		nc_ensure_sync(dev);
 		return 0;
 	}
 	if (header->packet_id != get_unaligned(&trailer->packet_id)) {
-		dev->stats.rx_fifo_errors++;
+		dev->net->stats.rx_fifo_errors++;
 		dbg("(2+ dropped) rx packet_id mismatch 0x%x 0x%x",
 			le16_to_cpu(header->packet_id),
 ...
[ message continues ]
[view original message]
From: Herbert Xu
Subject: [4/7] rndis_host: Use netdev stats structure
Date: Monday, June 29, 2009 - 7:52 pm

rndis_host: Use netdev stats structure

Now that netdev has its own stats structure we should use that
instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

diff --git a/drivers/net/usb/rndis_host.c b/drivers/net/usb/rndis_host.c
index 1bf243e..2232232 100644
--- a/drivers/net/usb/rndis_host.c
+++ b/drivers/net/usb/rndis_host.c
@@ -487,7 +487,7 @@ int rndis_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 		if (unlikely(hdr->msg_type != RNDIS_MSG_PACKET
 				|| skb->len < msg_len
 				|| (data_offset + data_len + 8) > msg_len)) {
-			dev->stats.rx_frame_errors++;
+			dev->net->stats.rx_frame_errors++;
 			devdbg(dev, "bad rndis message %d/%d/%d/%d, len %d",
 				le32_to_cpu(hdr->msg_type),
 				msg_len, data_offset, data_len, skb->len);

Cheers,
-- 
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
--


[ message continues ]
[view original message]
From: Herbert Xu
Subject: [5/7] smsc95xx: Use netdev stats structure
Date: Monday, June 29, 2009 - 7:53 pm

smsc95xx: Use netdev stats structure

Now that netdev has its own stats structure we should use that
instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

diff --git a/drivers/net/usb/smsc95xx.c b/drivers/net/usb/smsc95xx.c
index 5a72833..fc7f640 100644
--- a/drivers/net/usb/smsc95xx.c
+++ b/drivers/net/usb/smsc95xx.c
@@ -1108,18 +1108,18 @@ static int smsc95xx_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 		if (unlikely(header & RX_STS_ES_)) {
 			if (netif_msg_rx_err(dev))
 				devdbg(dev, "Error header=0x%08x", header);
-			dev->stats.rx_errors++;
-			dev->stats.rx_dropped++;
+			dev->net->stats.rx_errors++;
+			dev->net->stats.rx_dropped++;
 
 			if (header & RX_STS_CRC_) {
-				dev->stats.rx_crc_errors++;
+				dev->net->stats.rx_crc_errors++;
 			} else {
 				if (header & (RX_STS_TL_ | RX_STS_RF_))
-					dev->stats.rx_frame_errors++;
+					dev->net->stats.rx_frame_errors++;
 
 				if ((header & RX_STS_LE_) &&
 					(!(header & RX_STS_FT_)))
-					dev->stats.rx_length_errors++;
+					dev->net->stats.rx_length_errors++;
 			}
 		} else {
 			/* ETH_FRAME_LEN + 4(CRC) + 2(COE) + 4(Vlan) */

Cheers,
-- 
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
--


[ message continues ]
[view original message]
From: Herbert Xu
Subject: [6/7] usbnet: Use netdev stats structure
Date: Monday, June 29, 2009 - 7:53 pm

usbnet: Use netdev stats structure

Now that netdev has its own stats structure we should use that
instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
index f3a2fce..f27ad07 100644
--- a/drivers/net/usb/usbnet.c
+++ b/drivers/net/usb/usbnet.c
@@ -203,8 +203,8 @@ void usbnet_skb_return (struct usbnet *dev, struct sk_buff *skb)
 	int	status;
 
 	skb->protocol = eth_type_trans (skb, dev->net);
-	dev->stats.rx_packets++;
-	dev->stats.rx_bytes += skb->len;
+	dev->net->stats.rx_packets++;
+	dev->net->stats.rx_bytes += skb->len;
 
 	if (netif_msg_rx_status (dev))
 		devdbg (dev, "< rx, len %zu, type 0x%x",
@@ -366,7 +366,7 @@ static inline void rx_process (struct usbnet *dev, struct sk_buff *skb)
 		if (netif_msg_rx_err (dev))
 			devdbg (dev, "drop");
 error:
-		dev->stats.rx_errors++;
+		dev->net->stats.rx_errors++;
 		skb_queue_tail (&dev->done, skb);
 	}
 }
@@ -389,8 +389,8 @@ static void rx_complete (struct urb *urb)
 	case 0:
 		if (skb->len < dev->net->hard_header_len) {
 			entry->state = rx_cleanup;
-			dev->stats.rx_errors++;
-			dev->stats.rx_length_errors++;
+			dev->net->stats.rx_errors++;
+			dev->net->stats.rx_length_errors++;
 			if (netif_msg_rx_err (dev))
 				devdbg (dev, "rx length %d", skb->len);
 		}
@@ -402,7 +402,7 @@ static void rx_complete (struct urb *urb)
 	 * storm, recovering as needed.
 	 */
 	case -EPIPE:
-		dev->stats.rx_errors++;
+		dev->net->stats.rx_errors++;
 		usbnet_defer_kevent (dev, EVENT_RX_HALT);
 		// FALLTHROUGH
 
@@ -420,7 +420,7 @@ static void rx_complete (struct urb *urb)
 	case -EPROTO:
 	case -ETIME:
 	case -EILSEQ:
-		dev->stats.rx_errors++;
+		dev->net->stats.rx_errors++;
 		if (!timer_pending (&dev->delay)) {
 			mod_timer (&dev->delay, jiffies + THROTTLE_JIFFIES);
 			if (netif_msg_link (dev))
@@ -434,12 +434,12 @@ block:
 
 	/* data overrun ... flush fifo? */
 	case ...
[ message continues ]
[view original message]
From: Herbert Xu
Subject: [7/7] usbnet: Remove private stats structure
Date: Monday, June 29, 2009 - 7:54 pm

usbnet: Remove private stats structure

Now that nothing uses the private stats structure we can remove it.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

diff --git a/include/linux/usb/usbnet.h b/include/linux/usb/usbnet.h
index 36fabb9..3b10097 100644
--- a/include/linux/usb/usbnet.h
+++ b/include/linux/usb/usbnet.h
@@ -42,7 +42,6 @@ struct usbnet {
 
 	/* protocol/interface state */
 	struct net_device	*net;
-	struct net_device_stats	stats;
 	int			msg_enable;
 	unsigned long		data [5];
 	u32			xid;

Cheers,
-- 
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
--


[ message continues ]
Previous thread: Re: [Bugme-new] [Bug 13568] New: Intel e1000 4-port NIC - unable to communicate, slowly blinking by Andrew Morton on Monday, June 29, 2009 - 4:00 pm. (2 messages)

Next thread: (no subject) by Mrs Dianne Thompson on Monday, June 29, 2009 - 6:28 pm. (1 message)