net: core: synchronize link-watch when carrier is queried

There are multiple ways to query for the carrier state: through
rtnetlink, sysfs, and (possibly) ethtool. Synchronize linkwatch
work before these operations so that we don't have a situation
where userspace queries the carrier state between the driver's
carrier off->on transition and linkwatch running and expects it
to work, when really (at least) TX cannot work until linkwatch
has run.

I previously posted a longer explanation of how this applies to
wireless [1] but with this wireless can simply query the state
before sending data, to ensure the kernel is ready for it.

[1] https://lore.kernel.org/all/346b21d87c69f817ea3c37caceb34f1f56255884.camel@sipsolutions.net/

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://lore.kernel.org/r/20231204214706.303c62768415.I1caedccae72ee5a45c9085c5eb49c145ce1c0dd5@changeid
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

include/linux/netdevice.h

@@ -4229,6 +4229,15 @@ static inline void netdev_ref_replace(struct net_device *odev,
  */
 void linkwatch_fire_event(struct net_device *dev);
 
+/**
+ * linkwatch_sync_dev - sync linkwatch for the given device
+ * @dev: network device to sync linkwatch for
+ *
+ * Sync linkwatch for the given device, removing it from the
+ * pending work list (if queued).
+ */
+void linkwatch_sync_dev(struct net_device *dev);
+
 /**
  *	netif_carrier_ok - test if carrier present
  *	@dev: network device

net/core/dev.c

@@ -10548,7 +10548,7 @@ void netdev_run_todo(void)
 		write_lock(&dev_base_lock);
 		dev->reg_state = NETREG_UNREGISTERED;
 		write_unlock(&dev_base_lock);
-		linkwatch_forget_dev(dev);
+		linkwatch_sync_dev(dev);
 	}
 
 	while (!list_empty(&list)) {

net/core/dev.h

@@ -30,7 +30,6 @@ int __init dev_proc_init(void);
 #endif
 
 void linkwatch_init_dev(struct net_device *dev);
-void linkwatch_forget_dev(struct net_device *dev);
 void linkwatch_run_queue(void);
 
 void dev_addr_flush(struct net_device *dev);

net/core/link_watch.c

@@ -245,7 +245,7 @@ static void __linkwatch_run_queue(int urgent_only)
 	spin_unlock_irq(&lweventlist_lock);
 }
 
-void linkwatch_forget_dev(struct net_device *dev)
+void linkwatch_sync_dev(struct net_device *dev)
 {
 	unsigned long flags;
 	int clean = 0;

net/core/net-sysfs.c

@@ -194,8 +194,14 @@ static ssize_t carrier_show(struct device *dev,
 {
 	struct net_device *netdev = to_net_dev(dev);
 
-	if (netif_running(netdev))
+	if (netif_running(netdev)) {
+		/* Synchronize carrier state with link watch,
+		 * see also rtnl_getlink().
+		 */
+		linkwatch_sync_dev(netdev);
+
 		return sysfs_emit(buf, fmt_dec, !!netif_carrier_ok(netdev));
+	}
 
 	return -EINVAL;
 }

net/core/rtnetlink.c

@@ -3853,6 +3853,14 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	if (nskb == NULL)
 		goto out;
 
+	/* Synchronize the carrier state so we don't report a state
+	 * that we're not actually going to honour immediately; if
+	 * the driver just did a carrier off->on transition, we can
+	 * only TX if link watch work has run, but without this we'd
+	 * already report carrier on, even if it doesn't work yet.
+	 */
+	linkwatch_sync_dev(dev);
+
 	err = rtnl_fill_ifinfo(nskb, dev, net,
 			       RTM_NEWLINK, NETLINK_CB(skb).portid,
 			       nlh->nlmsg_seq, 0, 0, ext_filter_mask,

net/ethtool/ioctl.c

@@ -58,6 +58,9 @@ static struct devlink *netdev_to_devlink_get(struct net_device *dev)
 
 u32 ethtool_op_get_link(struct net_device *dev)
 {
+	/* Synchronize carrier state with link watch, see also rtnl_getlink() */
+	linkwatch_sync_dev(dev);
+
 	return netif_carrier_ok(dev) ? 1 : 0;
 }
 EXPORT_SYMBOL(ethtool_op_get_link);