Skip to contentSkip to conversation

e100_shutdown: netif_poll_disable hang

PATCH NAK Daniel Walker 3 participants 6 messages Oct 20, 2006 – Oct 21, 2006

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 e100_shutdown: netif_poll_disable hang

My machine annoyingly hangs while rebooting. I tracked it down to e100-fix-reboot-f-with-netconsole-enabled.patch in 2.6.18-rc2-mm2

I review the changes and it seemed to be calling netif_poll_disable one too many time. Once in e100_down(), and again in e100_shutdown().

The second one in e100_shutdown() caused the hang. So this patch removes it.

---
 drivers/net/e100.c |    1 -
 1 files changed, 1 deletion(-)
drivers/net/e100.c +0 -1
Index: linux-2.6.18/drivers/net/e100.c
===================================================================
@@ -2709,7 +2709,6 @@ static void e100_shutdown(struct pci_dev
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct nic *nic = netdev_priv(netdev);
 
-	netif_poll_disable(nic->netdev);
 	del_timer_sync(&nic->watchdog);
 	netif_carrier_off(nic->netdev);
 

--

#1 lore raw
Quoted 20 lines · Daniel Walker

My machine annoyingly hangs while rebooting. I tracked it down to e100-fix-reboot-f-with-netconsole-enabled.patch in 2.6.18-rc2-mm2

I review the changes and it seemed to be calling netif_poll_disable one too many time. Once in e100_down(), and again in e100_shutdown().

The second one in e100_shutdown() caused the hang. So this patch removes it.

---
 drivers/net/e100.c |    1 -
 1 files changed, 1 deletion(-)

Index: linux-2.6.18/drivers/net/e100.c
===================================================================

drivers/net/e100.c static void e100_shutdown(struct pci_dev

Quoted 8 lines
drivers/net/e100.c +0 -2
@@ -2709,7 +2709,6 @@ static void e100_shutdown(struct pci_dev
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct nic *nic = netdev_priv(netdev);
 
-	netif_poll_disable(nic->netdev);
 	del_timer_sync(&nic->watchdog);
 	netif_carrier_off(nic->netdev);
 
--

this won't help netconsole shutdown/reboot -f, probably locking it up again!

NAK

Also missing is the NAPI conditional, which I left out. Also missing is the same code for suspend.

Here's a better approach, allowing both normal shutdown code path and reboot -f to disable polling.

Cheers,

Auke
---

e100: disable polling only when up during suspend and shutdown

Signed-off-by Auke Kok <auke-jan.h.kok>
Cc Daniel Walker <[email protected]>
Cc Andrew Morton <[email protected]>
---
 drivers/net/e100.c |   10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

---

drivers/net/e100.c +8 -2
index d4a2572..815eb29 100644
@@ -2719,7 +2719,10 @@ static int e100_suspend(struct pci_dev *
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct nic *nic = netdev_priv(netdev);
 
-	netif_poll_disable(nic->netdev);
+#ifdef CONFIG_E100_NAPI
+	if (netif_running(netdev))
+		netif_poll_disable(nic->netdev);
+#endif
 	del_timer_sync(&nic->watchdog);
 	netif_carrier_off(nic->netdev);
 
@@ -2763,7 +2766,10 @@ static void e100_shutdown(struct pci_dev
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct nic *nic = netdev_priv(netdev);
 
-	netif_poll_disable(nic->netdev);
+#ifdef CONFIG_E100_NAPI
+	if (netif_running(netdev))
+		netif_poll_disable(nic->netdev);
+#endif
 	del_timer_sync(&nic->watchdog);
 	netif_carrier_off(nic->netdev);
#2 lore raw
Quoted 9 lines · Auke Kok
Quoted 8 lines · Daniel Walker

My machine annoyingly hangs while rebooting. I tracked it down to e100-fix-reboot-f-with-netconsole-enabled.patch in 2.6.18-rc2-mm2

I review the changes and it seemed to be calling netif_poll_disable one too many time. Once in e100_down(), and again in e100_shutdown().

The second one in e100_shutdown() caused the hang. So this patch removes it.

it doesn't even do harm to netif_poll_disable() twice as far as I can see, as it merely calls test_and_set_bit(), which will instantly succeed on the first attempt if the bit was already set.

did this change actually fix it for you? I'm wondering if the netif_carrier_off might not be the culprit here...

Auke

#3 lore raw
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines

My machine annoyingly hangs while rebooting. I tracked it down to e100-fix-reboot-f-with-netconsole-enabled.patch in 2.6.18-rc2-mm2 I review the changes and it seemed to be calling netif_poll_disable one too many time. Once in e100_down(), and again in e100_shutdown().
The second one in e100_shutdown() caused the hang. So this patch removes it.

Quoted 6 lines · Auke Kok

it doesn't even do harm to netif_poll_disable() twice as far as I can see, as it merely calls test_and_set_bit(), which will instantly succeed on the first attempt if the bit was already set.

did this change actually fix it for you? I'm wondering if the netif_carrier_off might not be the culprit here...

I can confirm the proposed original change of D. Walker fixed the problem for me. I did not test the change you proposed as a followup.

Signature · 1 line
Damien Wyart
#4 lore raw
Quoted 13 lines · Damien Wyart
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines

My machine annoyingly hangs while rebooting. I tracked it down to e100-fix-reboot-f-with-netconsole-enabled.patch in 2.6.18-rc2-mm2 I review the changes and it seemed to be calling netif_poll_disable one too many time. Once in e100_down(), and again in e100_shutdown().
The second one in e100_shutdown() caused the hang. So this patch removes it.

Quoted 3 lines · Auke Kok

it doesn't even do harm to netif_poll_disable() twice as far as I can see, as it merely calls test_and_set_bit(), which will instantly succeed on the first attempt if the bit was already set.

Quoted 5 lines
Quoted 2 lines

did this change actually fix it for you? I'm wondering if the netif_carrier_off might not be the culprit here...

I can confirm the proposed original change of D. Walker fixed the problem for me. I did not test the change you proposed as a followup.

his change breaks something else (a reboot with netconsole, possibly suspend). Please give the latest version I sent a try. Daniel confirmed me that it works, but it's always nice to hear it from more people.

Thanks

Auke

#5 lore raw

* Auke Kok <[email protected]> [2006-10-21 10:54]: his change

Quoted 3 lines

breaks something else (a reboot with netconsole, possibly suspend). Please give the latest version I sent a try. Daniel confirmed me that it works, but it's always nice to hear it from more people.

Yes, your version works here too.

Signature · 1 line
Damien Wyart