From: Vladimir Oltean <vladimir.oltean@xxxxxxx>
commit 10b6bb62ae1a49ee818fc479cf57b8900176773e upstream.
Ido Schimmel points out that since commit 52cff74eef5d ("dcbnl : Disable
software interrupts before taking dcb_lock"), the DCB API can be called
by drivers from softirq context.
One such in-tree example is the chelsio cxgb4 driver:
dcb_rpl
-> cxgb4_dcb_handle_fw_update
-> dcb_ieee_setapp
If the firmware for this driver happened to send an event which resulted
in a call to dcb_ieee_setapp() at the exact same time as another
DCB-enabled interface was unregistering on the same CPU, the softirq
would deadlock, because the interrupted process was already holding the
dcb_lock in dcbnl_flush_dev().
Fix this unlikely event by using spin_lock_bh() in dcbnl_flush_dev() as
in the rest of the dcbnl code.
Fixes: 91b0383fef06 ("net: dcb: flush lingering app table entries for unregistered devices")
Reported-by: Ido Schimmel <idosch@xxxxxxxxxx>
Signed-off-by: Vladimir Oltean <vladimir.oltean@xxxxxxx>
Link: https://lore.kernel.org/r/20220302193939.1368823-1-vladimir.oltean@xxxxxxx
Signed-off-by: Jakub Kicinski <kuba@xxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
net/dcb/dcbnl.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
--- a/net/dcb/dcbnl.c
+++ b/net/dcb/dcbnl.c
@@ -2058,7 +2058,7 @@ static void dcbnl_flush_dev(struct net_d
{
struct dcb_app_type *itr, *tmp;
- spin_lock(&dcb_lock);
+ spin_lock_bh(&dcb_lock);
list_for_each_entry_safe(itr, tmp, &dcb_app_list, list) {
if (itr->ifindex == dev->ifindex) {
@@ -2067,7 +2067,7 @@ static void dcbnl_flush_dev(struct net_d
}
}
- spin_unlock(&dcb_lock);
+ spin_unlock_bh(&dcb_lock);
}
static int dcbnl_netdevice_event(struct notifier_block *nb,
