| |||||||||||||
|
| |||||||||||||
Fun with NULL pointers, part 2Fun with NULL pointers, part 2Posted Jul 21, 2009 23:32 UTC (Tue) by johill (subscriber, #25196)In reply to: Fun with NULL pointers, part 2 by ebiederm Parent article: Fun with NULL pointers, part 2 Besides, depending on the platform the compiler could elide the BUG_ON(ptr == NULL) anyway, no? It's often just
if (unlikely(ptr == NULL))
out_of_line_bug();
or something like that?
(Log in to post comments)
Fun with NULL pointers, part 2 Posted Jul 22, 2009 4:12 UTC (Wed) by gmaxwell (subscriber, #30048) [Link]
Yes, if the compiler can prove that its impossible for the condition to be true it can and will omit the check. Sadly because the compiler's visibility is so narrow it usually can't do this except in the most obvious of situations.
Fun with NULL pointers, part 2 Posted Jul 22, 2009 9:53 UTC (Wed) by mb (subscriber, #50428) [Link]
Well, for most architectures the BUG() functions are __attribute__((noreturn)) or something equivalent. So the compiler cannot assume that the execution flow reaches to after the if (!ptr) check at all, if ptr is NULL. So it should not optimize it out in that case.
Fun with NULL pointers, part 2 Posted Jul 22, 2009 11:23 UTC (Wed) by johill (subscriber, #25196) [Link]
Really? I thought the out_of_line_bug() in my example would have that attribute, but not the BUG() itself. But powerpc for example is different though, it just inserts a trap (twi or such).
|
|
||||||||||||