By: Patrick Chase (patrickjchase.delete@this.gmail.com), August 25, 2014 11:32 pm
Room: Moderated Discussions
nksingh (none.delete@this.none.non) on August 25, 2014 9:26 pm wrote:
> > 2. If used naively LL/SC is indeed prone to livelock as others have
> > pointed out, but that's a pretty simple problem to solve in SW.
>
> I'm curious, what are some of the software solutions? I'm
> aware of exponential backoff. Is there anything else?
Randomized exponential backoff is the easy one that just about everybody uses. Note that exponential backoff alone (without a separate randomized starting delay for each thread) doesn't provably resolve livelock, though.
> > 2. If used naively LL/SC is indeed prone to livelock as others have
> > pointed out, but that's a pretty simple problem to solve in SW.
>
> I'm curious, what are some of the software solutions? I'm
> aware of exponential backoff. Is there anything else?
Randomized exponential backoff is the easy one that just about everybody uses. Note that exponential backoff alone (without a separate randomized starting delay for each thread) doesn't provably resolve livelock, though.
