Skip to contentSkip to conversation

cfg80211: fix race in netlink owner interface destruction

PATCH In review Johannes Berg 3 participants 4 messages Feb 1, 2022 802.11 (including cfg80211/NL80211)

lore
Share & tools
lore

Johannes Berg

author maintainer
#0 lore raw

PATCH 1/1 cfg80211: fix race in netlink owner interface destruction

From: Johannes Berg <[email protected]>

My previous fix here to fix the deadlock left a race where the exact same deadlock (see the original commit referenced below) can still happen if cfg80211_destroy_ifaces() already runs while nl80211_netlink_notify() is still marking some interfaces as nl_owner_dead.

The race happens because we have two loops here - first we dev_close() all the netdevs, and then we destroy them. If we also have two netdevs (first one need only be a wdev though) then we can find one during the first iteration, close it, and go to the second iteration -- but then find two, and try to destroy also the one we didn't close yet.

Fix this by only iterating once.

Change-Id Ie56cd0ef3f0d2108bb8a25c8bb5efced15e6a909
Fixes ea6b2098dd02 ("cfg80211: fix locking in netlink owner interface destruction")
---
 net/wireless/core.c | 17 ++++-------------
 1 file changed, 4 insertions(+), 13 deletions(-)
net/wireless/core.c +4 -13
index ff74549b1022..d151a433388c 100644
@@ -5,7 +5,7 @@
  * Copyright 2006-2010		Johannes Berg <[email protected]>
  * Copyright 2013-2014  Intel Mobile Communications GmbH
  * Copyright 2015-2017	Intel Deutschland GmbH
- * Copyright (C) 2018-2021 Intel Corporation
+ * Copyright (C) 2018-2022 Intel Corporation
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
@@ -332,29 +332,20 @@ static void cfg80211_event_work(struct work_struct *work)
 void cfg80211_destroy_ifaces(struct cfg80211_registered_device *rdev)
 {
 	struct wireless_dev *wdev, *tmp;
-	bool found = false;
 
 	ASSERT_RTNL();
 
-	list_for_each_entry(wdev, &rdev->wiphy.wdev_list, list) {
+	list_for_each_entry_safe(wdev, tmp, &rdev->wiphy.wdev_list, list) {
 		if (wdev->nl_owner_dead) {
 			if (wdev->netdev)
 				dev_close(wdev->netdev);
-			found = true;
-		}
-	}
-
-	if (!found)
-		return;
 
-	wiphy_lock(&rdev->wiphy);
-	list_for_each_entry_safe(wdev, tmp, &rdev->wiphy.wdev_list, list) {
-		if (wdev->nl_owner_dead) {
+			wiphy_lock(&rdev->wiphy);
 			cfg80211_leave(rdev, wdev);
 			rdev_del_virtual_intf(rdev, wdev);
+			wiphy_unlock(&rdev->wiphy);
 		}
 	}
-	wiphy_unlock(&rdev->wiphy);
 }
 
 static void cfg80211_destroy_iface_wk(struct work_struct *work)
Signature · 1 line
2.34.1
Tested-by on v1 1/1
#1 lore raw
Quoted 15 lines · Johannes Berg

From: Johannes Berg <[email protected]>

My previous fix here to fix the deadlock left a race where the exact same deadlock (see the original commit referenced below) can still happen if cfg80211_destroy_ifaces() already runs while nl80211_netlink_notify() is still marking some interfaces as nl_owner_dead.

The race happens because we have two loops here - first we dev_close() all the netdevs, and then we destroy them. If we also have two netdevs (first one need only be a wdev though) then we can find one during the first iteration, close it, and go to the second iteration -- but then find two, and try to destroy also the one we didn't close yet.

Quoted 6 lines

Fix this by only iterating once.

Change-Id Ie56cd0ef3f0d2108bb8a25c8bb5efced15e6a909
Fixes ea6b2098dd02 ("cfg80211: fix locking in netlink owner interface destruction")

Looks like this fixed the crash I was seeing - thanks!

#2 lore raw
In reply to Johannes Berg
Quoted 15 lines · Johannes Berg

From: Johannes Berg <[email protected]>

My previous fix here to fix the deadlock left a race where the exact same deadlock (see the original commit referenced below) can still happen if cfg80211_destroy_ifaces() already runs while nl80211_netlink_notify() is still marking some interfaces as nl_owner_dead.

The race happens because we have two loops here - first we dev_close() all the netdevs, and then we destroy them. If we also have two netdevs (first one need only be a wdev though) then we can find one during the first iteration, close it, and go to the second iteration -- but then find two, and try to destroy also the one we didn't close yet.

Quoted 3 lines

Fix this by only iterating once.

Change-Id: Ie56cd0ef3f0d2108bb8a25c8bb5efced15e6a909

You forgot the Change-Id.

Signature · 3 lines
https://patchwork.kernel.org/project/linux-wireless/list/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
author maintainer
#3 lore raw
Quoted 17 lines · Kalle Valo
Quoted 15 lines · Johannes Berg

From: Johannes Berg <[email protected]>

My previous fix here to fix the deadlock left a race where the exact same deadlock (see the original commit referenced below) can still happen if cfg80211_destroy_ifaces() already runs while nl80211_netlink_notify() is still marking some interfaces as nl_owner_dead.

The race happens because we have two loops here - first we dev_close() all the netdevs, and then we destroy them. If we also have two netdevs (first one need only be a wdev though) then we can find one during the first iteration, close it, and go to the second iteration -- but then find two, and try to destroy also the one we didn't close yet.

Quoted 6 lines
Quoted 3 lines

Fix this by only iterating once.

Change-Id: Ie56cd0ef3f0d2108bb8a25c8bb5efced15e6a909

You forgot the Change-Id.

Wtf, sorry, guess I sent it from a place with the wrong script. I'll strip it when I apply it, I guess

johannes