[PATCH] ieee802154: hwsim: serialize pib updates to fix double-free

[PATCH] ieee802154: hwsim: serialize pib updates to fix double-free

[David Carlier]

hwsim_update_pib() does an unserialized read-swap-free of phy->pib:

	pib_old = rtnl_dereference(phy->pib);
	...
	rcu_assign_pointer(phy->pib, pib);
	kfree_rcu(pib_old, rcu);

It assumes the RTNL is held, but ->set_channel is not always called
under it: the mac802154 scan worker changes channels via
drv_set_channel() without the RTNL. Such an update can race an
RTNL-held one on the same phy; both read the same pib_old and both
kfree_rcu() it, double-freeing the object. With SLUB percpu sheaves
batching kfree_rcu(), this surfaces as a KASAN invalid-free in
rcu_free_sheaf().

struct hwsim_phy has no lock for pib. Add one and make the swap atomic
with rcu_replace_pointer() under it, dropping the misleading
rtnl_dereference().

Reported-by: syzbot+60332fd095f8bb2946ad@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=60332fd095f8bb2946ad
Fixes: f25da51fdc38 ("ieee802154: hwsim: add replacement for fakelb")
Signed-off-by: David Carlier <devnexen@gmail.com>
Cc: <stable@vger.kernel.org>
---
 drivers/net/ieee802154/mac802154_hwsim.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ieee802154/mac802154_hwsim.c b/drivers/net/ieee802154/mac802154_hwsim.c
index 6daa0f198b9f..a9bd1555d2dc 100644
--- a/drivers/net/ieee802154/mac802154_hwsim.c
+++ b/drivers/net/ieee802154/mac802154_hwsim.c
@@ -72,6 +72,8 @@ struct hwsim_phy {
 	struct ieee802154_hw *hw;
 	u32 idx;
 
+	/* Serializes phy->pib_updates. */
+	spinlock_t pib_lock;
 	struct hwsim_pib __rcu *pib;
 
 	bool suspended;
@@ -102,8 +104,6 @@ static int hwsim_update_pib(struct ieee802154_hw *hw, u8 page, u8 channel,
 	if (!pib)
 		return -ENOMEM;
 
-	pib_old = rtnl_dereference(phy->pib);
-
 	pib->page = page;
 	pib->channel = channel;
 	pib->filt.short_addr = filt->short_addr;
@@ -112,7 +112,10 @@ static int hwsim_update_pib(struct ieee802154_hw *hw, u8 page, u8 channel,
 	pib->filt.pan_coord = filt->pan_coord;
 	pib->filt_level = filt_level;
 
-	rcu_assign_pointer(phy->pib, pib);
+	spin_lock_bh(&phy->pib_lock);
+	pib_old = rcu_replace_pointer(phy->pib, pib,
+				      lockdep_is_held(&phy->pib_lock));
+	spin_unlock_bh(&phy->pib_lock);
 	kfree_rcu(pib_old, rcu);
 	return 0;
 }
@@ -952,6 +955,7 @@ static int hwsim_add_one(struct genl_info *info, struct device *dev,
 		goto err_pib;
 	}
 
+	spin_lock_init(&phy->pib_lock);
 	pib->channel = 13;
 	pib->filt.short_addr = cpu_to_le16(IEEE802154_ADDR_BROADCAST);
 	pib->filt.pan_id = cpu_to_le16(IEEE802154_PANID_BROADCAST);
-- 
2.53.0

Re: [PATCH] ieee802154: hwsim: serialize pib updates to fix double-free

[Miquel Raynal]

Hello David,

On 09/07/2026 at 23:18:58 +01, David Carlier <devnexen@gmail.com> wrote:

> hwsim_update_pib() does an unserialized read-swap-free of phy->pib:
>
> 	pib_old = rtnl_dereference(phy->pib);
> 	...
> 	rcu_assign_pointer(phy->pib, pib);
> 	kfree_rcu(pib_old, rcu);
>
> It assumes the RTNL is held, but ->set_channel is not always called
> under it: the mac802154 scan worker changes channels via
> drv_set_channel() without the RTNL. Such an update can race an
> RTNL-held one on the same phy; both read the same pib_old and both
> kfree_rcu() it, double-freeing the object. With SLUB percpu sheaves
> batching kfree_rcu(), this surfaces as a KASAN invalid-free in
> rcu_free_sheaf().
>
> struct hwsim_phy has no lock for pib. Add one and make the swap atomic
> with rcu_replace_pointer() under it, dropping the misleading
> rtnl_dereference().
>
> Reported-by: syzbot+60332fd095f8bb2946ad@syzkaller.appspotmail.com
> Closes: https://syzkaller.appspot.com/bug?extid=60332fd095f8bb2946ad
> Fixes: f25da51fdc38 ("ieee802154: hwsim: add replacement for fakelb")
> Signed-off-by: David Carlier <devnexen@gmail.com>
> Cc: <stable@vger.kernel.org>

Thank you for the patch, but I think Yousef already provided a similar
patch:

https://lore.kernel.org/all/20260627235805.17310-1-alhouseenyousef@gmail.com/

Yousef, can you confirm you will send v2 soon?

Thanks,
Miquèl