wireless: Fix ethtool stats and other ops.

Commit Message

Bjørn Mork Dec. 7, 2012, 11:30 a.m. UTC

Stanislaw Gruszka <[e-mail]> writes:
> On Thu, Dec 06, 2012 at 12:15:28PM -0800, Ben Greear wrote:
>> I don't have time to work on this today...maybe not tomorrow either.
>> 
>> So, if someone else feels like fixing it, please feel free.
>
> Ok, I'll try to fix that. However this could require modification
> of all cfg80211 drivers, so could not be fast.

I wonder if that strategy isn't just adding unnecessary complexity?  How
about using something like the (completely untested) patch below instead?


Bjørn

Comments

Stanislaw Gruszka Dec. 7, 2012, 11:40 a.m. UTC | #1

On Fri, Dec 07, 2012 at 12:30:12PM +0100, Bjørn Mork wrote:
> Stanislaw Gruszka <[e-mail]> writes:
> > On Thu, Dec 06, 2012 at 12:15:28PM -0800, Ben Greear wrote:
> >> I don't have time to work on this today...maybe not tomorrow either.
> >> 
> >> So, if someone else feels like fixing it, please feel free.
> >
> > Ok, I'll try to fix that. However this could require modification
> > of all cfg80211 drivers, so could not be fast.
> 
> I wonder if that strategy isn't just adding unnecessary complexity?  How
> about using something like the (completely untested) patch below instead?

Yeah, I wrote very similar patch and I'm testing it locally. Plan to
post RFC to net folks ...

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

Patch

From 6b6ce488f64d5530e1886072f64b034d5a65e5f7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Bj=C3=B8rn=20Mork?= <[e-mail]>
Date: Fri, 7 Dec 2012 12:14:02 +0100
Subject: [RFC] net: fix ethtool_ops for wireless
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

commit 2c60db03 ("net: provide a default dev->ethtool_ops") made
tests for unset ethtool_ops always fail.  This caused a regression
for wireless drivers relying on the wireless specific ethtool_ops
default.

Adding a new test for code which need to know whether ethtool_ops
was changed after netdev allocation, and update the wireless core
to use it.

Signed-off-by: Bjørn Mork <[e-mail]>
---
 include/linux/netdevice.h |   14 ++++++++++++++
 net/core/dev.c            |    1 +
 net/wireless/core.c       |    2 +-
 3 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 18c5dc9..95741f1 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2707,6 +2707,20 @@  static inline bool netif_supports_nofcs(struct net_device *dev)
 
 extern struct pernet_operations __net_initdata loopback_net_ops;
 
+extern struct ethtool_ops default_ethtool_ops;
+
+/**
+ *	netdev_has_default_ethtool_ops - test for default ethtool_ops
+ *	@dev: device
+ *
+ *	This function returns true if dev is using the generic default
+ *	ethtool_ops
+ */
+static inline bool netdev_has_default_ethtool_ops(struct net_device *dev)
+{
+	return dev->ethtool_ops && (dev->ethtool_ops == &default_ethtool_ops);
+}
+
 /* Logging, debugging and troubleshooting/diagnostic helpers. */
 
 /* netdev_printk helpers, similar to dev_printk */
diff --git a/net/core/dev.c b/net/core/dev.c
index 0aea3fe..0b20be5 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -6107,6 +6107,7 @@  struct netdev_queue *dev_ingress_queue_create(struct net_device *dev)
 }
 
 static const struct ethtool_ops default_ethtool_ops;
+EXPORT_SYMBOL_GPL(default_ethtool_ops);
 
 /**
  *	alloc_netdev_mqs - allocate network device
diff --git a/net/wireless/core.c b/net/wireless/core.c
index 14d9904..95467d0 100644
--- a/net/wireless/core.c
+++ b/net/wireless/core.c
@@ -866,7 +866,7 @@  static int cfg80211_netdev_notifier_call(struct notifier_block *nb,
 		/* allow mac80211 to determine the timeout */
 		wdev->ps_timeout = -1;
 
-		if (!dev->ethtool_ops)
+		if (netdev_has_default_ethtool_ops(dev)
 			dev->ethtool_ops = &cfg80211_ethtool_ops;
 
 		if ((wdev->iftype == NL80211_IFTYPE_STATION ||
-- 
1.7.10.4