Messages in this thread |  | | Date | Fri, 13 Dec 2013 11:10:27 +0000 | From | One Thousand Gnomes <> | Subject | Re: [PATCH 1/9] Known exploit detection |
| |
> Totally true, but there's a million way to DoS a local machine. At > least this way shows who's doing it. It's the DoSes that don't include > attribution that I worry about. :)
So long as they compile out to nothingness for all the systems where this stuff is useless (ie most of them because they are embedded or phones etc) I don't see a big problem.
Note however if you trip one of those in any code with the console lock held and your log goes to the consoles due to printk level or similar you'll probably hang the box.
There are some other lock sequences that are going to do that too, so it can't be placed arbitarily but will need the locking assumptions for each non-obvious one documented clearly.
Alan
|  |