Re: net: integer overflow in ip_idents_reserve

On Wed, 2014-12-17 at 00:09 +0100, Hannes Frederic Sowa wrote:

Also we compile the whole kernel with -fno-strict-overflow, so every
report of signed overflow leading to undefined behavior is probably a
false positive. I don't know if it is worth to try to get rid of them, I
doubt it.

Presumably we could have uatomic_t , or atomic_u32_t, whatever...

This particular xadd() is heavily hit in some cases, we really do not
want a cmpxchg()