On 11.04.24 17:35, Alan Stern wrote:
On Thu, Apr 11, 2024 at 02:42:59PM +0200, Oliver Neukum wrote:
We have to ignore the higher bits in bEndpointAddress
Mention that these bits are reserved. That's why we ought to ignore
them.
OK
Also, this is not really an example of hardening; it's more like
future-proofing the code. The existing code will work fine with a
malicious device; your real concern is about possible changes to the
spec in the future.
It seemed to me like a vector for a DoS, but in hindsight, yes I'll
take it out of the series.
Regards
Oliver
