On Thu, Apr 20, 2017 at 11:33:19AM -0700, Eric Biggers wrote: > Like I suggested months ago, how about doing an efficient implementation of > refcount_t which doesn't use the bloated cmpxchg loop? Then there would be no > need for endless performance arguments. In fact, in PaX there are already > example implementations for several architectures. It's unfortunate that > they're still being ignored for some reason. Yes. I will also NAK other conversions until this is done, which is a bit sad as the recount_t API itself is very useful.
