On Thu, 2024-01-25 at 12:07 +0300, Dmitry Antipov wrote: > Well, the literal approach to "not even use the bitrate mask for scanning" > may look like the following. It fixes an original WARN_ONCE() as reported > by syzkaller, but I'm definitely interesting on how to check whether it > doesn't break something else. I will always run the hwsim tests in hostap, but I'm not really concerned it would break anything - this was never really _intended_ to cover scanning in the first place. johannes
