Hacker News Reader

>Nobody will try to pull that shit again for quite some time.

Watch me! (not really, but if I wanted to troll Linus, rants like that would serve as encouragement)

Also using the compiler builtins is the most secure, most readable, least error-prone and fastest way of implementing overflow checks.

Linus's rant is literally entirely based only on why that is wrong.

He makes a lot of claims without backing any of them up.

Are you saying that the claim that the new function "is nonstandard" is false?

No, he's right about that. But it's a red herring: The Linux kernel already uses a lot of GCC-specific, i.e. nonstandard, extensions.