On Wed, Apr 07, 2021 at 10:19:03AM +0800, Du Cheng wrote: > A WARN_ON(wdev->conn) would trigger in cfg80211_sme_connect(), if > multiple send_msg() system calls are made from the userland, which > should be anticipated and handled by the wireless driver. Convert this > WARN() to pr_warn to prevent a kernel panic if kernel is configured to > "panic on warn". > > Bug reported by syzbot. > > Reported-by: syzbot+5f9392825de654244975@xxxxxxxxxxxxxxxxxxxxxxxxx > Signed-off-by: Du Cheng <ducheng2@xxxxxxxxx> > --- > link to syzkaller: > https://syzkaller.appspot.com/bug?extid=5f9392825de654244975 > this patch has passed syzbot test. > net/wireless/sme.c | 4 +++- > 1 file changed, 3 insertions(+), 1 deletion(-) > > diff --git a/net/wireless/sme.c b/net/wireless/sme.c > index 07756ca5e3b5..87a65a4c40ae 100644 > --- a/net/wireless/sme.c > +++ b/net/wireless/sme.c > @@ -529,8 +529,10 @@ static int cfg80211_sme_connect(struct wireless_dev *wdev, > cfg80211_sme_free(wdev); > } > > - if (WARN_ON(wdev->conn)) > + if (wdev->conn) { > + pr_warn("%s: wdev->conn != NULL, sme connect in progress", __func__); You have a real device, please always use 'dev_warn() for stuff like this, or the netdev equivalent. Also no need for __func__ for stuff like this, that's just clutter. Also, what can a user do if they get this information? What does it mean to them? Try making the text more informative. thanks, greg k-h
