Subj : Re: [RFC] Boost.Threads on its way to C++0x To : comp.programming.threads From : Joe Seigh Date : Fri Apr 29 2005 05:33 pm On Fri, 29 Apr 2005 16:37:52 +0300, Peter Dimov wrote: > Joe Seigh wrote: >> On Thu, 28 Apr 2005 21:35:59 -0400, Joe Seigh >> wrote: >>>> . >>>> >>> >>> Well, that explains why it's been relatively quiet here lately. >> >> I don't think I'm going to attempt to participate or offer any input, >> however. It takes time to put together input on this kind of stuff, for me >> at least. It took me over a year to come up with an initial memory model >> (mostly in background, you don't work full time on that kind of >> thing). Again, hopefully they don't screw up too much. > > I think that you definitely should keep an eye on things to make sure they > don't screw up too much. ;-) > The operative phrase being "too much". The formal definition of the semantics should be distinct from the implementation. It's kind of difficult to do that with a meta-implementation as your semantics. And it becomes difficult to add in a new set of better and improved atomic primatives if the semantic definition is stated in terms of the older primatives. -- Joe Seigh When you get lemons, you make lemonade. When you get hardware, you make software. .