hkaiser changed the topic of #ste||ar to: The topic is 'STE||AR: Systems Technology, Emergent Parallelism, and Algorithm Research | stellar.cct.lsu.edu | HPX: A cure for performance impaired parallel applications | github.com/STEllAR-GROUP/hpx | Buildbot: http://rostam.cct.lsu.edu/ | Log: http://irclog.cct.lsu.edu/
hkaiser has joined #ste||ar
hkaiser has quit [Quit: bye]
K-ballo has quit [Remote host closed the connection]
diehlpk has joined #ste||ar
diehlpk has quit [Ping timeout: 260 seconds]
<jbjnr> in our source code files, does "\cond NOINTERNAL" turn off doc generation?
kordejong has joined #ste||ar
rori has joined #ste||ar
simbergm has joined #ste||ar
<simbergm> jbjnr: looks like the limiting executor does the same thing as the async customization executor did, i.e. schedule continuations before the future is ready
<simbergm> also, yes \cond NOINTERNAL does exactly that
<jbjnr> simbergm: yes. my cut'n'paste of executor stuff needs investigating
<jbjnr> but the limiting executor is being redesigned anyway at the moment
<jbjnr> so I'll have a new PR soon
<simbergm> sure, no worries
K-ballo has joined #ste||ar
hkaiser has joined #ste||ar
<simbergm> btw, now that we've delayed the release if you get the mpi stuff running soon enough we can add that as well
<simbergm> but I'm not delaying it further because of that
<jbjnr> I keep experimenting with things like the sync launch of continuations and I must have disabled it when testing and then forgotten to renable it
<jbjnr> I think that originally, there was a sync task that was short that could run on the existing task thread, but I redesigned things and I need to rethink it
<jbjnr> the main idea was that if tasks complete and then we want to schedule the next based on the results, we could do it sync, but doing it sync might invalidate what the hint returns.
<jbjnr> the hint might say use numa 1, but this task might be on numa 0
<jbjnr> so using sync would be wrong.
<jbjnr> I never finished experimenting
<jbjnr> it should go in hpx::threads::executors::experimental :)
<jbjnr> the mpi stuff is going to be awesome and I don't mind waiting until after the release as I might have quite a lot of new stuff to go in. However, it will all be in a new module, so if over christmas I make a lot of progress, maybe you can merge it in.
<jbjnr> (without breaking anything)
<K-ballo> why was the release delayed?
<jbjnr> Becasue nobody wil notice the difference if it goes out just before xmas, or at the start of new year I think
<jbjnr> gtg
<hkaiser> simbergm: I'd suggest rather waiting with merging the mpi stuff, not sure it's mature enough at this point
<simbergm> sure, only if the pr stays the same size it is now and it really works, otherwise it'll definitely wait until after the release
<hkaiser> simbergm: I think we should wait for the executor being in place
<simbergm> yep, fair enough
<simbergm> btw, hkaiser, I forgot to ask yesterday: the minisymposium was accepted for pasc
<hkaiser> ohh nice
<simbergm> correction, it was accepted through the first round
<simbergm> ...
<simbergm> but now I need names and presentation titles
<simbergm> can I put you on the list?
<hkaiser> when is it?
<simbergm> june 29th to july 1st
<hkaiser> and where?
<simbergm> geneva
<hkaiser> nod
<hkaiser> yah, count me in - thanks!
<simbergm> john or I can always be a backup if it turns out you can't make it
<simbergm> nice, thanks!
<simbergm> do you want to come up with a talk title? :)
<simbergm> we'll hopefully have someone from kokkos there as well
<simbergm> something on the execution context-y stuff and interoperability would probably be a good topic
<hkaiser> simbergm: I don't want to but I will ;-)
<simbergm> :P
<simbergm> the deadline is jan 17th for the title, abstract comes after that if it's accepted through the final round
<hkaiser> ok, pls ping me again if I forget
<simbergm> oh, I will...
<simbergm> thanks!
<hkaiser> heh
hkaiser has quit [Quit: bye]
mdiers_ has quit [Remote host closed the connection]
kordejong has quit [Quit: WeeChat 2.7]
aserio has joined #ste||ar
aserio has quit [Client Quit]
aserio has joined #ste||ar
khuck_ has joined #ste||ar
khuck has quit [Ping timeout: 250 seconds]
aserio has quit [Read error: Connection reset by peer]
aserio has joined #ste||ar
rori has quit [Quit: WeeChat 1.9.1]
hkaiser has joined #ste||ar
aserio has quit [Ping timeout: 260 seconds]
<K-ballo> someone on cpplang #general is pondering whether to use HPX over TBB
aserio has joined #ste||ar
hkaiser has quit [Quit: bye]
aserio has quit [Quit: aserio]
hkaiser has joined #ste||ar