News for hackers. Stuff that matters.
Re:About the security content of iOS 7.0.6 by d0 on Saturday February 22, 2014 @04:43AM (#7281574)
You are 100% right. LLVM should fire a dead code warning here. However later code is used as a jump target for goto so it's quite hard for the compiler to infer this (or is it because its a different scope?).

Either way, definitely avoidable.

Looks typical of a merge cock up to me as the indentation is preserved suggesting it's a duplicate line or there was an if statement on the previous line that was removed.

Re:About the security content of iOS 7.0.6 by jey on Saturday February 22, 2014 @06:03AM (#7281731)
The basic blocks from just after the unconditional goto up to the "fail:" label are all unreachable, so a dead code warning could be issued.

> However later code is used as a jump target for goto so it's quite hard for the compiler to infer this

Nope, because a jump target always starts a new basic block.

Re:About the security content of iOS 7.0.6 by fit2rule on Saturday February 22, 2014 @02:43PM (#7283284)
Looks to me like Apple aren't running Code-Coverage tests on their OS. That's kind of very scary.