I'd be interested too. A good lock-free wait-free data structure is a powerful abstraction in terms of the programmer's quality of life.
Likewise, afaik, hardware lock Ellision is only a real win in the no / low contention case. You still need to provide a code path to fall back on if there's any contention. On Thursday, December 26, 2013, comex wrote: > On Wed, Dec 25, 2013 at 4:23 AM, Daniel Micay > <danielmi...@gmail.com<javascript:;>> > wrote: > > It makes sense to allocate it up-front for some lock-free queues but > > not a classical one built on a mutex and two condition variables. I > > doubt that a lock-free algorithm with have any significant wins > > against a Linux mutex or Windows critical section, and on Haswell lock > > elision erases the theoretical benefits too. > > I'm skeptical of the claim that a lock-free algorithm would not have > any significant benefits. Do you have a citation? > _______________________________________________ > Rust-dev mailing list > Rust-dev@mozilla.org <javascript:;> > https://mail.mozilla.org/listinfo/rust-dev >
_______________________________________________ Rust-dev mailing list Rust-dev@mozilla.org https://mail.mozilla.org/listinfo/rust-dev