kernel BUG at net/core/dev.c:1328!

BUG In review Alessandro Guido 5 participants 16 messages Jul 21, 2008 – Jul 24, 2008

Subsystems (3)

Networking Network drivers Networking drivers (wireless)

lore

Share & tools

lore

Alessandro Guido

author

2008-07-21 16:14 UTC #0 lore raw

this is what I get when try to det up a wireless connection with current git (v2.6.26-5253-g14b395e):

kernel BUG at net/core/dev.c:1328!
invalid opcode: 0000 [#1] PREEMPT

Modules linked in: ipw2200

Pid: 5, comm: events/0 Not tainted (2.6.26-05253-g14b395e #1)
EIP: 0060:[<c02e58ca>] EFLAGS: 00010246 CPU: 0
EIP is at __netif_schedule+0x3a/0x40
EAX: c043d380 EBX: eed84b40 ECX: c043d380 EDX: ffffffff
ESI: eed84b44 EDI: ef808140 EBP: ef82bf64 ESP: ef82bf60
DS: 007b ES: 007b FS: 0000 GS: 0000 SS: 0068
Process events/0 (pid: 5, ti=ef82b000 task=ef82cd40 task.ti=ef82b000)

Stack: eed84b40 ef82bf80 f0939850 ef808140 ef82bf78 c02ef6c2 eed8584c ef808140
       ef82bfa8 c012b02f ef82ce9c ee24fc00 ef82ce9c ffff007b f09396f0 ef808148
       ef808140 ef82bfb0 ef82bfd0 c012b53f 00000000 ef82cd40 c012eaf0 ef82bfbc
Call Trace:
 [<f0939850>] ? ipw_bg_link_up+0x160/0x1f0 [ipw2200]
 [<c02ef6c2>] ? rtnl_unlock+0x12/0x20
 [<c012b02f>] ? run_workqueue+0x6f/0x160
 [<f09396f0>] ? ipw_bg_link_up+0x0/0x1f0 [ipw2200]
 [<c012b53f>] ? worker_thread+0x7f/0xe0
 [<c012eaf0>] ? autoremove_wake_function+0x0/0x50
 [<c012b4c0>] ? worker_thread+0x0/0xe0
 [<c012e67a>] ? kthread+0x3a/0x70
 [<c012e640>] ? kthread+0x0/0x70
 [<c0103c8f>] ? kernel_thread_helper+0x7/0x18
 =======================
Code: 24 0f ba 28 01 19 d2 85 d2 75 1d 9c 5b fa a1 c0 11 4a c0 89 41

3c b8 02 00 00 00 89 0d c0 11 4a c0 e8 4b ab e3 ff 53 9d 5b 5d c3 <0f> 0b eb fe 66 90 55 89 e5 53 89 c3 8d 40 2c 0f ba 28 01 19 d2
EIP: [<c02e58ca>] __netif_schedule+0x3a/0x40 SS:ESP 0068:ef82bf60

David Miller

maintainer

2008-07-21 16:20 UTC #1 lore raw

From: "Alessandro Guido" <[email protected]>
Date: Mon, 21 Jul 2008 18:14:28 +0200

linux-wireless and netdev added to CC:

Quoted 39 lines

this is what I get when try to det up a wireless connection with current git (v2.6.26-5253-g14b395e):
kernel BUG at net/core/dev.c:1328!
invalid opcode: 0000 [#1] PREEMPT
Modules linked in: ipw2200
Pid: 5, comm: events/0 Not tainted (2.6.26-05253-g14b395e #1)
EIP: 0060:[<c02e58ca>] EFLAGS: 00010246 CPU: 0
EIP is at __netif_schedule+0x3a/0x40
EAX: c043d380 EBX: eed84b40 ECX: c043d380 EDX: ffffffff
ESI: eed84b44 EDI: ef808140 EBP: ef82bf64 ESP: ef82bf60
DS: 007b ES: 007b FS: 0000 GS: 0000 SS: 0068
Process events/0 (pid: 5, ti=ef82b000 task=ef82cd40 task.ti=ef82b000)
Stack: eed84b40 ef82bf80 f0939850 ef808140 ef82bf78 c02ef6c2 eed8584c ef808140
       ef82bfa8 c012b02f ef82ce9c ee24fc00 ef82ce9c ffff007b f09396f0 ef808148
       ef808140 ef82bfb0 ef82bfd0 c012b53f 00000000 ef82cd40 c012eaf0 ef82bfbc
Call Trace:
 [<f0939850>] ? ipw_bg_link_up+0x160/0x1f0 [ipw2200]
 [<c02ef6c2>] ? rtnl_unlock+0x12/0x20
 [<c012b02f>] ? run_workqueue+0x6f/0x160
 [<f09396f0>] ? ipw_bg_link_up+0x0/0x1f0 [ipw2200]
 [<c012b53f>] ? worker_thread+0x7f/0xe0
 [<c012eaf0>] ? autoremove_wake_function+0x0/0x50
 [<c012b4c0>] ? worker_thread+0x0/0xe0
 [<c012e67a>] ? kthread+0x3a/0x70
 [<c012e640>] ? kthread+0x0/0x70
 [<c0103c8f>] ? kernel_thread_helper+0x7/0x18
 =======================
Code: 24 0f ba 28 01 19 d2 85 d2 75 1d 9c 5b fa a1 c0 11 4a c0 89 41
3c b8 02 00 00 00 89 0d c0 11 4a c0 e8 4b ab e3 ff 53 9d 5b 5d c3 <0f> 0b eb fe 66 90 55 89 e5 53 89 c3 8d 40 2c 0f ba 28 01 19 d2
EIP: [<c02e58ca>] __netif_schedule+0x3a/0x40 SS:ESP 0068:ef82bf60 --
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Alessandro Guido

author

2008-07-22 09:54 UTC #2 lore raw

Got a WARNING this morning (2.6.26-05752-g93ded9b) and I think it's related.

WARNING: at net/core/dev.c:1330 __netif_schedule+0x53/0x60()
Modules linked in:
Pid: 5, comm: events/0 Not tainted 2.6.26-05752-g93ded9b #1
 [<c011bc64>] warn_on_slowpath+0x54/0xa0
 [<c0135483>] ? getnstimeofday+0x53/0x100
 [<c0132d1c>] ? ktime_get_ts+0x4c/0x50
 [<c012ba67>] ? insert_work+0x57/0x70
 [<c012c129>] ? queue_work+0x39/0x60
 [<c012c1e5>] ? queue_delayed_work+0x25/0x30
 [<c012c201>] ? schedule_delayed_work+0x11/0x20
 [<c030398a>] ? linkwatch_schedule_work+0x3a/0xb0
 [<c02f78d3>] __netif_schedule+0x53/0x60
 [<c0262a80>] ipw_bg_link_up+0x160/0x1f0
 [<c035ae4e>] ? schedule+0x1de/0x3f0
 [<c012bdef>] run_workqueue+0x6f/0x160
 [<c0262920>] ? ipw_bg_link_up+0x0/0x1f0
 [<c012c08f>] worker_thread+0x7f/0xe0
 [<c012f5d0>] ? autoremove_wake_function+0x0/0x50
 [<c012c010>] ? worker_thread+0x0/0xe0
 [<c012f15a>] kthread+0x3a/0x70
 [<c012f120>] ? kthread+0x0/0x70
 [<c0103d77>] kernel_thread_helper+0x7/0x10
 =======================

Alessandro Suardi

2008-07-22 20:18 UTC #3 lore raw

Quoted 18 lines · Alessandro Guido

Got a WARNING this morning (2.6.26-05752-g93ded9b) and I think it's related.

WARNING: at net/core/dev.c:1330 __netif_schedule+0x53/0x60()
Modules linked in:
Pid: 5, comm: events/0 Not tainted 2.6.26-05752-g93ded9b #1
 [<c011bc64>] warn_on_slowpath+0x54/0xa0
 [<c0135483>] ? getnstimeofday+0x53/0x100
 [<c0132d1c>] ? ktime_get_ts+0x4c/0x50
 [<c012ba67>] ? insert_work+0x57/0x70
 [<c012c129>] ? queue_work+0x39/0x60
 [<c012c1e5>] ? queue_delayed_work+0x25/0x30
 [<c012c201>] ? schedule_delayed_work+0x11/0x20
 [<c030398a>] ? linkwatch_schedule_work+0x3a/0xb0
 [<c02f78d3>] __netif_schedule+0x53/0x60
 [<c0262a80>] ipw_bg_link_up+0x160/0x1f0
 [<c035ae4e>] ? schedule+0x1de/0x3f0
 [<c012bdef>] run_workqueue+0x6f/0x160

Quoted 8 lines

 [<c0262920>] ? ipw_bg_link_up+0x0/0x1f0
 [<c012c08f>] worker_thread+0x7f/0xe0
 [<c012f5d0>] ? autoremove_wake_function+0x0/0x50
 [<c012c010>] ? worker_thread+0x0/0xe0
 [<c012f15a>] kthread+0x3a/0x70
 [<c012f120>] ? kthread+0x0/0x70
 [<c0103d77>] kernel_thread_helper+0x7/0x10
 =======================

I also hit a very similar one on my Dell D610 with onboard
ipw2200 wireless (already kerneloops'd, BTW):

WARNING: at net/core/dev.c:1330 __netif_schedule+0x2a/0x63()
Modules linked in: fuse sunrpc iptable_filter ip_tables

ip6table_filter ip6_tables x_tables ipv6 cpufreq_ondemand ppdev pcmcia dcdbas parport_pc parport snd_intel8x0 snd_intel8x0m snd_ac97_codec snd_seq_dummy ac97_bus snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device snd_pcm_oss pcspkr snd_mixer_oss snd_pcm snd_timer snd ipw2200 soundcore snd_page_alloc i2c_i801 ieee80211 ieee80211_crypt yenta_socket rsrc_nonstatic pcmcia_core video output pata_acpi ata_piix uhci_hcd ehci_hcd [last unloaded: microcode]
Pid: 5, comm: events/0 Not tainted 2.6.26-git9 #5

 [<c0118cda>] warn_on_slowpath+0x46/0x80
 [<c0113a91>] ? __enqueue_entity+0xe3/0xeb
 [<c0113754>] ? __wake_up+0x31/0x3b
 [<c0125ebb>] ? __queue_work+0x2d/0x32
 [<c0125f05>] ? queue_work+0x2a/0x34
 [<c01261ff>] ? queue_delayed_work+0x11/0x23
 [<c0126227>] ? schedule_delayed_work+0x16/0x1b
 [<c02ce1d5>] ? linkwatch_schedule_work+0x58/0x99
 [<c02ce278>] ? linkwatch_fire_event+0x62/0x66
 [<c02c567f>] __netif_schedule+0x2a/0x63
 [<f8990009>] ipw_bg_link_up+0xae/0x173 [ipw2200]
 [<c0125a0f>] run_workqueue+0x7d/0xf6
 [<f898ff5b>] ? ipw_bg_link_up+0x0/0x173 [ipw2200]
 [<c0125cec>] worker_thread+0xb7/0xc3
 [<c0128862>] ? autoremove_wake_function+0x0/0x38
 [<c0125c35>] ? worker_thread+0x0/0xc3
 [<c01285d3>] kthread+0x3e/0x63
 [<c0128595>] ? kthread+0x0/0x63
 [<c0103003>] kernel_thread_helper+0x7/0x10

But it works - to the point that I'm connected to my wireless
router without apparent issues.

--alessandro

"Give me love / Or give me hate
Give me anything that's not just ok"

(Sophia, 'Weightless')

David Miller

maintainer

2008-07-22 22:56 UTC #4 lore raw

In reply to Alessandro Guido

From: "Alessandro Guido" <[email protected]>
Date: Tue, 22 Jul 2008 11:54:26 +0200

Quoted 1 line

Got a WARNING this morning (2.6.26-05752-g93ded9b) and I think it's related.

Please try this patch:

ipw2200: Call netif_*_queue() interfaces properly.

netif_carrier_{on,off}() handles starting and stopping packet flow into the driver. So there is no reason to invoke netif_stop_queue() and netif_wake_queue() in response to link status events.

Signed-off-by David S. Miller <[email protected]>

drivers/net/wireless/ipw2200.c +2 ~~-31~~

index 6e70460..1acfbcd 100644
@@ -4972,8 +4972,7 @@ static int ipw_queue_tx_reclaim(struct ipw_priv *priv,
 	}
       done:
 	if ((ipw_tx_queue_space(q) > q->low_mark) &&
-	    (qindex >= 0) &&
-	    (priv->status & STATUS_ASSOCIATED) && netif_running(priv->net_dev))
+	    (qindex >= 0))
 		netif_wake_queue(priv->net_dev);
 	used = q->first_empty - q->last_used;
 	if (used < 0)

@@ -10154,14 +10153,8 @@ static  void init_sys_config(struct ipw_sys_config *sys_config)
 
 static int ipw_net_open(struct net_device *dev)
 {
-	struct ipw_priv *priv = ieee80211_priv(dev);
 	IPW_DEBUG_INFO("dev->open\n");
-	/* we should be verifying the device is ready to be opened */
-	mutex_lock(&priv->mutex);
-	if (!(priv->status & STATUS_RF_KILL_MASK) &&
-	    (priv->status & STATUS_ASSOCIATED))
-		netif_start_queue(dev);
-	mutex_unlock(&priv->mutex);
+	netif_start_queue(dev);
 	return 0;
 }

@@ -10481,13 +10474,6 @@ static int ipw_net_hard_start_xmit(struct ieee80211_txb *txb,
 	IPW_DEBUG_TX("dev->xmit(%d bytes)\n", txb->payload_size);
 	spin_lock_irqsave(&priv->lock, flags);
 
-	if (!(priv->status & STATUS_ASSOCIATED)) {
-		IPW_DEBUG_INFO("Tx attempt while not associated.\n");
-		priv->ieee->stats.tx_carrier_errors++;
-		netif_stop_queue(dev);
-		goto fail_unlock;
-	}
-
 #ifdef CONFIG_IPW2200_PROMISCUOUS
 	if (rtap_iface && netif_running(priv->prom_net_dev))
 		ipw_handle_promiscuous_tx(priv, txb);

@@ -10499,10 +10485,6 @@ static int ipw_net_hard_start_xmit(struct ieee80211_txb *txb,
 	spin_unlock_irqrestore(&priv->lock, flags);
 
 	return ret;
-
-      fail_unlock:
-	spin_unlock_irqrestore(&priv->lock, flags);
-	return 1;
 }
 
 static struct net_device_stats *ipw_net_get_stats(struct net_device *dev)

@@ -10703,13 +10685,6 @@ static void ipw_link_up(struct ipw_priv *priv)
 	priv->last_packet_time = 0;
 
 	netif_carrier_on(priv->net_dev);
-	if (netif_queue_stopped(priv->net_dev)) {
-		IPW_DEBUG_NOTIF("waking queue\n");
-		netif_wake_queue(priv->net_dev);
-	} else {
-		IPW_DEBUG_NOTIF("starting queue\n");
-		netif_start_queue(priv->net_dev);
-	}
 
 	cancel_delayed_work(&priv->request_scan);
 	cancel_delayed_work(&priv->request_direct_scan);

@@ -10739,7 +10714,6 @@ static void ipw_link_down(struct ipw_priv *priv)
 {
 	ipw_led_link_down(priv);
 	netif_carrier_off(priv->net_dev);
-	netif_stop_queue(priv->net_dev);
 	notify_wx_assoc_event(priv);
 
 	/* Cancel any queued work ... */

@@ -11419,7 +11393,6 @@ static void ipw_down(struct ipw_priv *priv)
 	/* Clear all bits but the RF Kill */
 	priv->status &= STATUS_RF_KILL_MASK | STATUS_EXIT_PENDING;
 	netif_carrier_off(priv->net_dev);
-	netif_stop_queue(priv->net_dev);
 
 	ipw_stop_nic(priv);

@@ -11522,7 +11495,6 @@ static int ipw_prom_open(struct net_device *dev)
 
 	IPW_DEBUG_INFO("prom dev->open\n");
 	netif_carrier_off(dev);
-	netif_stop_queue(dev);
 
 	if (priv->ieee->iw_mode != IW_MODE_MONITOR) {
 		priv->sys_config.accept_all_data_frames = 1;

@@ -11558,7 +11530,6 @@ static int ipw_prom_stop(struct net_device *dev)
 static int ipw_prom_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	IPW_DEBUG_INFO("prom dev->xmit\n");
-	netif_stop_queue(dev);
 	return -EOPNOTSUPP;
 }

Alessandro Guido

author

2008-07-23 07:57 UTC #5 lore raw

I'm testing the latest Linus' git which includes your patch (2.6.26-06077-gc010b2f) and the problem is fixed.

Thank you!
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

Alessandro Suardi

2008-07-23 13:16 UTC #6 lore raw

Quoted 4 lines · Alessandro Guido

I'm testing the latest Linus' git which includes your patch (2.6.26-06077-gc010b2f) and the problem is fixed.
Thank you!

-git10 also has this fixed for me. Thanks,

--alessandro

"Give me love / Or give me hate
Give me anything that's not just ok"

(Sophia, 'Weightless')
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

Markus Trippelsdorf

2008-07-23 05:40 UTC #7 lore raw

Re: WARNING: at net/core/dev.c:1328!

In reply to Alessandro Guido

Quoted 1 line · Alessandro Guido

Got a WARNING this morning (2.6.26-05752-g93ded9b) and I think it's related.

Same thing here (latest git):

skge eth1: enabling interface
skge eth1: disabling interface

------------[ cut here ]------------
WARNING: at net/core/dev.c:1344 __netif_schedule+0x24/0x6d()

Pid: 1904, comm: ip Not tainted 2.6.26-06077-gc010b2f #33

Call Trace:
 [<ffffffff8022e5f4>] warn_on_slowpath+0x4c/0x87
 [<ffffffff802fc5cb>] delay_tsc+0x0/0x55
 [<ffffffff802fc5cb>] delay_tsc+0x0/0x55
 [<ffffffff802fc5e7>] delay_tsc+0x1c/0x55
 [<ffffffff8037e2ee>] gm_phy_write+0x50/0x90
 [<ffffffff8037e2f8>] gm_phy_write+0x5a/0x90
 [<ffffffff8037f03f>] skge_led+0x1fb/0x20b
 [<ffffffff804202e8>] __netif_schedule+0x24/0x6d
 [<ffffffff80381298>] skge_down+0x440/0x4cf
 [<ffffffff803814be>] skge_change_mtu+0x3d/0x61
 [<ffffffff80420775>] dev_set_mtu+0x46/0x79
 [<ffffffff80427a51>] do_setlink+0x1c5/0x31f
 [<ffffffff804287d3>] rtnl_newlink+0x2b9/0x3e3
 [<ffffffff804285a8>] rtnl_newlink+0x8e/0x3e3
 [<ffffffff80428d69>] rtnetlink_rcv_msg+0x5a/0x1ea
 [<ffffffff80428d0f>] rtnetlink_rcv_msg+0x0/0x1ea
 [<ffffffff8042efde>] netlink_rcv_skb+0x34/0x7e
 [<ffffffff80428d09>] rtnetlink_rcv+0x1f/0x25
 [<ffffffff8042eb97>] netlink_unicast+0x119/0x17f
 [<ffffffff8041b35d>] __alloc_skb+0x61/0x123
 [<ffffffff8042ee37>] netlink_sendmsg+0x23a/0x24d
 [<ffffffff804152e4>] sock_sendmsg+0xcb/0xe3
 [<ffffffff8023f670>] autoremove_wake_function+0x0/0x2e
 [<ffffffff80225643>] need_resched+0x1e/0x28
 [<ffffffff804147d8>] move_addr_to_kernel+0x25/0x36
 [<ffffffff8041c090>] verify_iovec+0x46/0x82
 [<ffffffff80415513>] sys_sendmsg+0x217/0x28a
 [<ffffffff8042e01e>] netlink_insert+0xfe/0x121
 [<ffffffff80414b40>] sockfd_lookup_light+0x1a/0x52
 [<ffffffff80261fe5>] __vma_link+0x58/0x61
 [<ffffffff80262062>] vma_link+0x74/0x99
 [<ffffffff802630ce>] do_brk+0x2c1/0x319
 [<ffffffff8020b3eb>] system_call_after_swapgs+0x7b/0x80

---[ end trace 92936ef183e09876 ]---

skge eth1: enabling interface
skge eth1: Link is up at 100 Mbps, full duplex, flow control both

Signature · 5 lines

Markus
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Wang Chen

2008-07-23 08:51 UTC #8 lore raw

PATCH 1/1 skge: resolve tx multiqueue bug

Markus Trippelsdorf said the following on 2008-7-23 13:40:

Quoted 47 lines · Alessandro Guido

Quoted 1 line

Got a WARNING this morning (2.6.26-05752-g93ded9b) and I think it's related.

Same thing here (latest git):

skge eth1: enabling interface
skge eth1: disabling interface

------------[ cut here ]------------
WARNING: at net/core/dev.c:1344 __netif_schedule+0x24/0x6d()

Pid: 1904, comm: ip Not tainted 2.6.26-06077-gc010b2f #33

Call Trace:
 [<ffffffff8022e5f4>] warn_on_slowpath+0x4c/0x87
 [<ffffffff802fc5cb>] delay_tsc+0x0/0x55
 [<ffffffff802fc5cb>] delay_tsc+0x0/0x55
 [<ffffffff802fc5e7>] delay_tsc+0x1c/0x55
 [<ffffffff8037e2ee>] gm_phy_write+0x50/0x90
 [<ffffffff8037e2f8>] gm_phy_write+0x5a/0x90
 [<ffffffff8037f03f>] skge_led+0x1fb/0x20b
 [<ffffffff804202e8>] __netif_schedule+0x24/0x6d
 [<ffffffff80381298>] skge_down+0x440/0x4cf
 [<ffffffff803814be>] skge_change_mtu+0x3d/0x61
 [<ffffffff80420775>] dev_set_mtu+0x46/0x79
 [<ffffffff80427a51>] do_setlink+0x1c5/0x31f
 [<ffffffff804287d3>] rtnl_newlink+0x2b9/0x3e3
 [<ffffffff804285a8>] rtnl_newlink+0x8e/0x3e3
 [<ffffffff80428d69>] rtnetlink_rcv_msg+0x5a/0x1ea
 [<ffffffff80428d0f>] rtnetlink_rcv_msg+0x0/0x1ea
 [<ffffffff8042efde>] netlink_rcv_skb+0x34/0x7e
 [<ffffffff80428d09>] rtnetlink_rcv+0x1f/0x25
 [<ffffffff8042eb97>] netlink_unicast+0x119/0x17f
 [<ffffffff8041b35d>] __alloc_skb+0x61/0x123
 [<ffffffff8042ee37>] netlink_sendmsg+0x23a/0x24d
 [<ffffffff804152e4>] sock_sendmsg+0xcb/0xe3
 [<ffffffff8023f670>] autoremove_wake_function+0x0/0x2e
 [<ffffffff80225643>] need_resched+0x1e/0x28
 [<ffffffff804147d8>] move_addr_to_kernel+0x25/0x36
 [<ffffffff8041c090>] verify_iovec+0x46/0x82
 [<ffffffff80415513>] sys_sendmsg+0x217/0x28a
 [<ffffffff8042e01e>] netlink_insert+0xfe/0x121
 [<ffffffff80414b40>] sockfd_lookup_light+0x1a/0x52
 [<ffffffff80261fe5>] __vma_link+0x58/0x61
 [<ffffffff80262062>] vma_link+0x74/0x99
 [<ffffffff802630ce>] do_brk+0x2c1/0x319
 [<ffffffff8020b3eb>] system_call_after_swapgs+0x7b/0x80

---[ end trace 92936ef183e09876 ]---

Quoted 3 lines

skge eth1: enabling interface
skge eth1: Link is up at 100 Mbps, full duplex, flow control both

Markus, please try this.

- Add netif_start_queue() in ->open()
- netif_carrier_*() is enough, remove netif_*_queue()

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

---

drivers/net/skge.c +1 -4

index 2e26dce..7507585 100644
@@ -1069,7 +1069,6 @@ static void skge_link_up(struct skge_port *skge)
 		    LED_BLK_OFF|LED_SYNC_OFF|LED_ON);
 
 	netif_carrier_on(skge->netdev);
-	netif_wake_queue(skge->netdev);
 
 	if (netif_msg_link(skge)) {
 		printk(KERN_INFO PFX

@@ -1084,7 +1083,6 @@ static void skge_link_down(struct skge_port *skge)
 {
 	skge_write8(skge->hw, SK_REG(skge->port, LNK_LED_REG), LED_OFF);
 	netif_carrier_off(skge->netdev);
-	netif_stop_queue(skge->netdev);
 
 	if (netif_msg_link(skge))
 		printk(KERN_INFO PFX "%s: Link is down.\n", skge->netdev->name);

@@ -2450,7 +2448,6 @@ static void skge_phy_reset(struct skge_port *skge)
 	int port = skge->port;
 	struct net_device *dev = hw->dev[port];
 
-	netif_stop_queue(skge->netdev);
 	netif_carrier_off(skge->netdev);
 
 	spin_lock_bh(&hw->phy_lock);

@@ -2640,6 +2637,7 @@ static int skge_up(struct net_device *dev)
 	spin_unlock_irq(&hw->hw_lock);
 
 	napi_enable(&skge->napi);
+	netif_start_queue(dev);
 	return 0;
 
  free_rx_ring:

@@ -3863,7 +3861,6 @@ static struct net_device *skge_devinit(struct skge_hw *hw, int port,
 
 	/* device is off until link detection */
 	netif_carrier_off(dev);
-	netif_stop_queue(dev);
 
 	return dev;
 }

Markus Trippelsdorf

2008-07-23 10:18 UTC #9 lore raw

Re: [PATCH] skge: resolve tx multiqueue bug

Quoted 12 lines · Wang Chen

Markus Trippelsdorf said the following on 2008-7-23 13:40:
Quoted 11 lines · Alessandro Guido
Quoted 1 line
Got a WARNING this morning (2.6.26-05752-g93ded9b) and I think it's related.
Same thing here (latest git):
skge eth1: enabling interface
skge eth1: disabling interface
------------[ cut here ]------------
WARNING: at net/core/dev.c:1344 __netif_schedule+0x24/0x6d()
Pid: 1904, comm: ip Not tainted 2.6.26-06077-gc010b2f #33
[<ffffffff8020b3eb>] system_call_after_swapgs+0x7b/0x80

...

Quoted 9 lines

Quoted 4 lines
---[ end trace 92936ef183e09876 ]---
skge eth1: enabling interface
skge eth1: Link is up at 100 Mbps, full duplex, flow control both
Markus, please try this.
- Add netif_start_queue() in ->open()
- netif_carrier_*() is enough, remove netif_*_queue()

Unfortunately, your patch does not fix this. I still get the same warning.

Signature · 1 line

Markus

Markus Trippelsdorf

2008-07-23 14:03 UTC #10 lore raw

Re: [PATCH] skge: resolve tx multiqueue bug

Quoted 19 lines · Markus Trippelsdorf

Quoted 12 lines · Wang Chen
Markus Trippelsdorf said the following on 2008-7-23 13:40:
Quoted 11 lines · Alessandro Guido
Quoted 1 line
Got a WARNING this morning (2.6.26-05752-g93ded9b) and I think it's related.
Same thing here (latest git):
skge eth1: enabling interface
skge eth1: disabling interface
------------[ cut here ]------------
WARNING: at net/core/dev.c:1344 __netif_schedule+0x24/0x6d()
Pid: 1904, comm: ip Not tainted 2.6.26-06077-gc010b2f #33
[<ffffffff8020b3eb>] system_call_after_swapgs+0x7b/0x80
...
Quoted 5 lines
Quoted 4 lines
---[ end trace 92936ef183e09876 ]---
skge eth1: enabling interface
skge eth1: Link is up at 100 Mbps, full duplex, flow control both

Quoted 7 lines

Quoted 4 lines
Markus, please try this.
- Add netif_start_queue() in ->open()
- netif_carrier_*() is enough, remove netif_*_queue()
Unfortunately, your patch does not fix this. I still get the same warning.

This patch works for me:

drivers/net/skge.c +1 -6

index 2e26dce..d761296 100644
@@ -1069,7 +1069,6 @@ static void skge_link_up(struct skge_port *skge)
                    LED_BLK_OFF|LED_SYNC_OFF|LED_ON);
 
        netif_carrier_on(skge->netdev);
-       netif_wake_queue(skge->netdev);
 
        if (netif_msg_link(skge)) {
                printk(KERN_INFO PFX

@@ -1084,7 +1083,6 @@ static void skge_link_down(struct skge_port *skge)
 {
        skge_write8(skge->hw, SK_REG(skge->port, LNK_LED_REG), LED_OFF);
        netif_carrier_off(skge->netdev);
-       netif_stop_queue(skge->netdev);
 
        if (netif_msg_link(skge))
                printk(KERN_INFO PFX "%s: Link is down.\n", skge->netdev->name);

@@ -2450,7 +2448,6 @@ static void skge_phy_reset(struct skge_port *skge)
        int port = skge->port;
        struct net_device *dev = hw->dev[port];
 
-       netif_stop_queue(skge->netdev);
        netif_carrier_off(skge->netdev);
 
        spin_lock_bh(&hw->phy_lock);

@@ -2640,6 +2637,7 @@ static int skge_up(struct net_device *dev)
        spin_unlock_irq(&hw->hw_lock);
 
        napi_enable(&skge->napi);
+       netif_start_queue(dev);
        return 0;
 
  free_rx_ring:

@@ -2673,8 +2671,6 @@ static int skge_down(struct net_device *dev)
        if (netif_msg_ifdown(skge))
                printk(KERN_INFO PFX "%s: disabling interface\n", dev->name);
 
-       netif_stop_queue(dev);
-
        if (hw->chip_id == CHIP_ID_GENESIS && hw->phy_type == SK_PHY_XMAC)
                del_timer_sync(&skge->link_timer);

@@ -3863,7 +3859,6 @@ static struct net_device *skge_devinit(struct skge_hw *hw, int port,
 
        /* device is off until link detection */
        netif_carrier_off(dev);
-       netif_stop_queue(dev);
 
        return dev;
 }

Signature · 1 line

Markus

Wang Chen

Tested-by

2008-07-23 15:22 UTC #11 lore raw

Re: [PATCH] skge: resolve tx multiqueue bug

Markus Trippelsdorf said the following on 2008-7-23 22:03:

Quoted 18 lines · Markus Trippelsdorf

Quoted 17 lines · Wang Chen
Quoted 11 lines
Markus Trippelsdorf said the following on 2008-7-23 13:40:
Quoted 10 lines · Alessandro Guido
Quoted 1 line
Got a WARNING this morning (2.6.26-05752-g93ded9b) and I think it's related.
Same thing here (latest git):
skge eth1: enabling interface
skge eth1: disabling interface
------------[ cut here ]------------
WARNING: at net/core/dev.c:1344 __netif_schedule+0x24/0x6d()
Pid: 1904, comm: ip Not tainted 2.6.26-06077-gc010b2f #33
[<ffffffff8020b3eb>] system_call_after_swapgs+0x7b/0x80
...
Quoted 4 lines
Quoted 4 lines
---[ end trace 92936ef183e09876 ]---
skge eth1: enabling interface
skge eth1: Link is up at 100 Mbps, full duplex, flow control both

Quoted 8 lines

Quoted 6 lines
Quoted 4 lines
Markus, please try this.
- Add netif_start_queue() in ->open()
- netif_carrier_*() is enough, remove netif_*_queue()
Unfortunately, your patch does not fix this. I still get the same warning.
This patch works for me:

Your patch works for me too. So I think it's better than mine. :)

Tested-by Wang Chen <[email protected]>

Dave, since Markus and me tested this patch, would you please apply it?

David Miller

maintainer

2008-07-23 22:30 UTC #12 lore raw

Re: [PATCH] skge: resolve tx multiqueue bug

From: Wang Chen <[email protected]>
Date: Wed, 23 Jul 2008 23:21:14 +0800

Quoted 26 lines

Markus Trippelsdorf said the following on 2008-7-23 22:03:
Quoted 25 lines · Markus Trippelsdorf
Quoted 23 lines · Wang Chen
Quoted 11 lines
Markus Trippelsdorf said the following on 2008-7-23 13:40:
Quoted 10 lines · Alessandro Guido
Quoted 1 line
Got a WARNING this morning (2.6.26-05752-g93ded9b) and I think it's related.
Same thing here (latest git):
skge eth1: enabling interface
skge eth1: disabling interface
------------[ cut here ]------------
WARNING: at net/core/dev.c:1344 __netif_schedule+0x24/0x6d()
Pid: 1904, comm: ip Not tainted 2.6.26-06077-gc010b2f #33
[<ffffffff8020b3eb>] system_call_after_swapgs+0x7b/0x80
...
Quoted 8 lines
Quoted 4 lines
---[ end trace 92936ef183e09876 ]---
skge eth1: enabling interface
skge eth1: Link is up at 100 Mbps, full duplex, flow control both
Markus, please try this.
- Add netif_start_queue() in ->open()
- netif_carrier_*() is enough, remove netif_*_queue()
Unfortunately, your patch does not fix this. I still get the same warning.

Quoted 7 lines

Quoted 1 line
This patch works for me:
Your patch works for me too. So I think it's better than mine. :)
Tested-by Wang Chen <[email protected]>
Dave, since Markus and me tested this patch, would you please apply it?

I can't, it's whitespace damaged. All the tabs are turned into spaces, also there is no signoff from Markus.

Wang Chen

2008-07-24 01:01 UTC #13 lore raw

Re: [PATCH] skge: resolve tx multiqueue bug

David Miller said the following on 2008-7-24 6:30:

Quoted 29 lines

From: Wang Chen <[email protected]>
Date: Wed, 23 Jul 2008 23:21:14 +0800
Quoted 26 lines
Markus Trippelsdorf said the following on 2008-7-23 22:03:
Quoted 25 lines · Markus Trippelsdorf
Quoted 17 lines · Wang Chen
Quoted 11 lines
Markus Trippelsdorf said the following on 2008-7-23 13:40:
Quoted 10 lines · Alessandro Guido
Quoted 1 line
Got a WARNING this morning (2.6.26-05752-g93ded9b) and I think it's related.
Same thing here (latest git):
skge eth1: enabling interface
skge eth1: disabling interface
------------[ cut here ]------------
WARNING: at net/core/dev.c:1344 __netif_schedule+0x24/0x6d()
Pid: 1904, comm: ip Not tainted 2.6.26-06077-gc010b2f #33
[<ffffffff8020b3eb>] system_call_after_swapgs+0x7b/0x80
...
Quoted 4 lines
Quoted 4 lines
---[ end trace 92936ef183e09876 ]---
skge eth1: enabling interface
skge eth1: Link is up at 100 Mbps, full duplex, flow control both
Quoted 6 lines
Quoted 4 lines
Markus, please try this.
- Add netif_start_queue() in ->open()
- netif_carrier_*() is enough, remove netif_*_queue()
Unfortunately, your patch does not fix this. I still get the same warning.
This patch works for me:

Quoted 9 lines

Quoted 5 lines
Your patch works for me too. So I think it's better than mine. :)
Tested-by Wang Chen <[email protected]>
Dave, since Markus and me tested this patch, would you please apply it?
I can't, it's whitespace damaged. All the tabs are turned into spaces, also there is no signoff from Markus.

OK. So let it be, since you removed the warning from __netif_schedule(). I will wait for your netif_tx_{free,unfreeze}() :)

--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

Markus Trippelsdorf

Tested-by Wang Chen

2008-07-24 06:19 UTC #14 lore raw

Re: [PATCH] skge: resolve tx multiqueue bug

In reply to David Miller

Quoted 31 lines · David Miller

From: Wang Chen <[email protected]>
Date: Wed, 23 Jul 2008 23:21:14 +0800
Quoted 28 lines
Markus Trippelsdorf said the following on 2008-7-23 22:03:
Quoted 26 lines · Markus Trippelsdorf
Quoted 17 lines · Wang Chen
Quoted 11 lines
Markus Trippelsdorf said the following on 2008-7-23 13:40:
Quoted 10 lines · Alessandro Guido
Quoted 1 line
Got a WARNING this morning (2.6.26-05752-g93ded9b) and I think it's related.
Same thing here (latest git):
skge eth1: enabling interface
skge eth1: disabling interface
------------[ cut here ]------------
WARNING: at net/core/dev.c:1344 __netif_schedule+0x24/0x6d()
Pid: 1904, comm: ip Not tainted 2.6.26-06077-gc010b2f #33
[<ffffffff8020b3eb>] system_call_after_swapgs+0x7b/0x80
...
Quoted 4 lines
Quoted 4 lines
---[ end trace 92936ef183e09876 ]---
skge eth1: enabling interface
skge eth1: Link is up at 100 Mbps, full duplex, flow control both
Quoted 6 lines
Quoted 4 lines
Markus, please try this.
- Add netif_start_queue() in ->open()
- netif_carrier_*() is enough, remove netif_*_queue()
Unfortunately, your patch does not fix this. I still get the same warning.
This patch works for me:

Quoted 8 lines

Quoted 5 lines
Your patch works for me too. So I think it's better than mine. :)
Tested-by Wang Chen <[email protected]>
Dave, since Markus and me tested this patch, would you please apply it?
I can't, it's whitespace damaged. All the tabs are turned into spaces, also there is no signoff from Markus.

Sorry, here is a correct version:

- Add netif_start_queue() in ->open()
- netif_carrier_*() is enough, remove netif_*_queue()

Signed-off-by Markus Trippelsdorf <[email protected]>

Tested-by Wang Chen <[email protected]>

Signature · 1 line

Markus

drivers/net/skge.c +1 -6

index 2e26dce..d761296 100644
@@ -1069,7 +1069,6 @@ static void skge_link_up(struct skge_port *skge)
 		    LED_BLK_OFF|LED_SYNC_OFF|LED_ON);
 
 	netif_carrier_on(skge->netdev);
-	netif_wake_queue(skge->netdev);
 
 	if (netif_msg_link(skge)) {
 		printk(KERN_INFO PFX

@@ -1084,7 +1083,6 @@ static void skge_link_down(struct skge_port *skge)
 {
 	skge_write8(skge->hw, SK_REG(skge->port, LNK_LED_REG), LED_OFF);
 	netif_carrier_off(skge->netdev);
-	netif_stop_queue(skge->netdev);
 
 	if (netif_msg_link(skge))
 		printk(KERN_INFO PFX "%s: Link is down.\n", skge->netdev->name);

@@ -2450,7 +2448,6 @@ static void skge_phy_reset(struct skge_port *skge)
 	int port = skge->port;
 	struct net_device *dev = hw->dev[port];
 
-	netif_stop_queue(skge->netdev);
 	netif_carrier_off(skge->netdev);
 
 	spin_lock_bh(&hw->phy_lock);

@@ -2640,6 +2637,7 @@ static int skge_up(struct net_device *dev)
 	spin_unlock_irq(&hw->hw_lock);
 
 	napi_enable(&skge->napi);
+	netif_start_queue(dev);
 	return 0;
 
  free_rx_ring:

@@ -2673,8 +2671,6 @@ static int skge_down(struct net_device *dev)
 	if (netif_msg_ifdown(skge))
 		printk(KERN_INFO PFX "%s: disabling interface\n", dev->name);
 
-	netif_stop_queue(dev);
-
 	if (hw->chip_id == CHIP_ID_GENESIS && hw->phy_type == SK_PHY_XMAC)
 		del_timer_sync(&skge->link_timer);

@@ -3863,7 +3859,6 @@ static struct net_device *skge_devinit(struct skge_hw *hw, int port,
 
 	/* device is off until link detection */
 	netif_carrier_off(dev);
-	netif_stop_queue(dev);
 
 	return dev;
 }

Alessandro Guido

author

2008-07-21 17:08 UTC #15 lore raw

In reply to Alessandro Guido

attacching config