Null pointers, one month later [LWN.net]

Null pointers, one month later

Posted Aug 18, 2009 15:57 UTC (Tue) by christian.convey (guest, #39159)
Parent article: Null pointers, one month later

"One month ago we got an undeniable warning that this kind of vulnerability was coming."

Actually, didn't we get it *many* months ago, in the Coverity scan reports that people ignored?

I'm not complaining - the kernel developers don't owe me anything. I'm just saying that the warnings were sounded far earlier if only people were willing to work through the Coverity output.

to post comments

Null pointers, one month later

Posted Aug 18, 2009 16:11 UTC (Tue) by JoeBuck (subscriber, #2330) [Link]

No, not really. When Coverity reports that a deference of a pointer precedes, rather than follows, a test to see if that pointer is null, this doesn't tell you whether or not it is possible for a real null pointer to reach that point. It's possible that the compare against null is redundant.

On the other hand, it does give a code auditor, or a black hat, a starting point to look for an exploitable problem.

Null pointers, one month later

Posted Aug 19, 2009 11:51 UTC (Wed) by spender (guest, #23067) [Link]

Technically you got it in 2007 when I released the first null ptr deref exploit for Linux ;)

-Brad