My only concern with concurrencykit is the use of hazard pointers, it's my understanding that hazard pointers are patented by IBM: http://www.google.com/patents/US20040107227?printsec=description#v=onepage&q&f=false
Brian On Fri, Mar 8, 2013 at 9:43 AM, Igor Galić <i.ga...@brainsware.org> wrote: > > > ----- Original Message ----- > > On 3/8/13 9:50 AM, Brian Geffon wrote: > > > I'm all for this also, i think at minimum we should take advantage > > > of > > > what we get for free from c++11, especially since we have c++11 > > > checks > > > in our autoconf script already anyway. I had never actually heard > > > of > > > concurrencykit until now. I know TBB isn't really an option, but > > > has > > > anyone compared the performance of the data structures of TBB to > > > concurrencykit? > > > > Something to look into. I think can be a solid "goal" for v3.5/3.6, > > to clean > > up these areas, and perhaps finish all the NUMA stuff as well. I'm > > hesitant > > to a huge change like this so late in the v3.3/3.4 release cycle > > though, but > > it's up to the community to decide (of course). > > > > > > But don't let any of that stop you guys from reviewing this patch > > > ;) > > > > Absolutely. I think this is a good incremental improvement to land > > for > > v3.3.2, for inclusion with v3.4. > > +1 on ck in 3.5 > > I'd suggest we wrap up what have into 3.4 and see to stabelizing that > If you want to release 3.4.0 in June, we could branch it off in May, > to do so, for one, and to allow those who are eager to make fixes > to continue working on trunk. > > > Cheers, > > > > -- leif > > > -- i > Igor Galić > > Tel: +43 (0) 664 886 22 883 > Mail: i.ga...@brainsware.org > URL: http://brainsware.org/ > GPG: 6880 4155 74BD FD7C B515 2EA5 4B1D 9E08 A097 C9AE > >
