Messages in this thread | | | From | Matthias Urlichs <> | Subject | Re: 2.0.21 crashes with ISDN | Date | Sat, 21 Sep 1996 13:13:03 +0100 |
| |
In linux.dev.kernel, article <199609202025.WAA00771@wildsau.idv.uni-linz.ac.at>, Herbert Rosmanith <herp@wildsau.idv.uni-linz.ac.at> writes: > > without this patch, a frame which is neither 1TR6 nor EDSS1 can > a) crash the machine and if it does not b) will be misinterprated > by the driver, which is IMO even worse than crashing (the error > is more subtle then). > Problem (a) cannot be fixed by checking for non-EDSS1 frames.
_No_ incoming frame, however strange or broken it is, should ever be able to crash the driver or the kernel. This is a requirement for compliance with the EDSS1 standards (just like not sending broken frames is, but that's the sender's problem).
We all know what happens when the receiver doesn't check for _everything_. For a quick demonstration of this, consider Appletalk -- I can crash the Mac AFP client quite easily by sending even perfectly legal replies which just happen not to match the sender's expectations. :-(
-- Lysistrata had a good idea. -- Matthias Urlichs \ noris network GmbH / Xlink-POP Nürnberg Schleiermacherstraße 12 \ Linux+Internet / EMail: urlichs@noris.de 90491 Nürnberg (Germany) \ Consulting+Programming+Networking+etc'ing PGP: 1024/4F578875 1B 89 E2 1C 43 EA 80 44 15 D2 29 CF C6 C7 E0 DE Click <A HREF="http://info.noris.de/~smurf/finger">here</A>. 42
| |