aserio changed the topic of #ste||ar to: 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/
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 276 seconds]
EverYoun_ has quit [Ping timeout: 276 seconds]
EverYoung has joined #ste||ar
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 240 seconds]
EverYoung has joined #ste||ar
parsa has joined #ste||ar
EverYoun_ has quit [Ping timeout: 276 seconds]
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
hkaiser_ has joined #ste||ar
hkaiser has quit [Read error: Connection reset by peer]
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 240 seconds]
ct_clmsn has quit [Quit: Leaving]
EverYoun_ has quit [Ping timeout: 250 seconds]
eschnett has joined #ste||ar
aserio has joined #ste||ar
diehlpk has joined #ste||ar
jaafar_ has quit [Ping timeout: 248 seconds]
parsa has quit [Quit: Zzzzzzzzzzzz]
parsa has joined #ste||ar
hkaiser_ has quit [Quit: bye]
EverYoung has joined #ste||ar
K-ballo has quit [Quit: K-ballo]
EverYoung has quit [Ping timeout: 240 seconds]
aserio has quit [Quit: aserio]
diehlpk has quit [Ping timeout: 276 seconds]
mcopik has quit [Ping timeout: 268 seconds]
parsa has quit [Quit: Zzzzzzzzzzzz]
nanashi55 has quit [Ping timeout: 276 seconds]
nanashi55 has joined #ste||ar
eschnett has quit [Ping timeout: 260 seconds]
jaafar_ has joined #ste||ar
jaafar_ has quit [Ping timeout: 260 seconds]
EverYoung has joined #ste||ar
EverYoung has quit [Ping timeout: 255 seconds]
david_pfander has joined #ste||ar
<simbergm> heller: I guess it's not possible to run even a subset of our tests on circleci? it seems to have only two cores per instance...
EverYoung has joined #ste||ar
EverYoung has quit [Ping timeout: 240 seconds]
<heller> simbergm: yeah ...
<heller> simbergm: circle-ci is just too slow
<heller> simbergm: I really would like to utilize our resources more efficiently. Everyone one of us has a large variety of different compute resources available ... it's a shame that we don't use those...
EverYoung has joined #ste||ar
<heller> simbergm: and yes ... I currently spend most of time looking into failures coming from PRs where nobody actually cared to run the unit test suite...
EverYoung has quit [Ping timeout: 252 seconds]
K-ballo has joined #ste||ar
<simbergm> heller: yeah, it's not good (and I'm guilty of wasting your time as well...)
<simbergm> I was asking because the cscs ci stuff is progressing very slowly
<heller> yup...
<simbergm> how much du you know about buildbot on rostam? would it be possible to even build and test a single configuration there for each branch?
<simbergm> or reduce the number of supported configurations
<heller> simbergm: buildbot on rostam uses an old buildbot version
<heller> the problem with buildbot really is to integrate other machines not located on the LSU network
<simbergm> right, so we wait for cscs :/
<heller> or implement something on our own ;)
<heller> I have machines and public IPs to run the basic service on
<heller> I just need the technology to dispatch to the different machines
<simbergm> hmm, so what we should get from cscs is jenkins which seems to have the option to accept results from remote machines, then we could use cscs and rostam for testing and I imagine that should be enough
* zao perks up
<simbergm> jenkins is open source though, so you can set that up if you want ;) but probably better to have cscs manage it
<heller> yeah, then you only have to have jenkins slaves running on the machines you include
<heller> does the CSCS jenkins support machines that are *not* in the CSCS network?
<simbergm> well, I think at the moment not and this is one of our requirements to them
<simbergm> we have a meeting next week which is supposed to get this going somehow
<heller> ok
david_pfander has quit [Remote host closed the connection]
hkaiser has joined #ste||ar
eschnett has joined #ste||ar
<jbjnr> hkaiser: yt?
<hkaiser> jbjnr: here
<hkaiser> g'morning
<jbjnr> hi
<jbjnr> if I have a futures_factory(blah) and I call, fact.apply(stuff) and then return fact.get_future() - that creates a task for me - does it go onto the pending (ready) queue or one of the staged/suspended queues -
<jbjnr> if I know it has no futures, it is ready to run - can I add a flag to say so?
<jbjnr> (not sure if I need it, didn't look closely yet)
<hkaiser> hold on, checking
<hkaiser> jbjnr: I'm almost done with the changes we discussed, btw - things should get better for you there
<jbjnr> thanks
david_pfander has joined #ste||ar
<jbjnr> david_pfander: How is gregor getting on in Louisisna? Did you find somewhere for him to live?
<hkaiser> jbjnr: futures_factory currently does schedule a staged task: https://github.com/STEllAR-GROUP/hpx/blob/master/hpx/lcos/local/futures_factory.hpp#L132-L137
<jbjnr> staged = ready to run yes?
<hkaiser> jbjnr: Nina was very helpful in finding a nice place for him
<jbjnr> good
<hkaiser> staged means ready to run, but not scheduled yet
<jbjnr> we'll keep her contact info!
<jbjnr> is it possible for me to schedule it directly - I mean add it straight to the scheduler?
<hkaiser> absolutely - we did lunch together last Monday - very nice lady
<jbjnr> great
<david_pfander> jbjnr: Hi John, only heard good things from him. He is making good progress at converting/porting the remaining OctoTiger kernels
<jbjnr> hkaiser: great - I'll add an apply that takes an extra flag if necessary - if my experiments work and show it helps.
hkaiser has quit [Read error: Connection reset by peer]
<jbjnr> david_pfander: good to hear
hkaiser has joined #ste||ar
<david_pfander> jbjnr: BTW, will there be access to Piz Daint for some OctoTiger performance runs early next year, or do our accounts expire in the near future?
hkaiser has quit [Read error: Connection reset by peer]
hkaiser has joined #ste||ar
<hkaiser> jbjnr: we could also extend future_factory to support executors, that puts it back into your hands
<jbjnr> david_pfander: the hpx test account already exired and when I asked to renew it - I got no reply - so I'd better ping them again
<jbjnr> hkaiser: interesting idea. I will look at that.
<david_pfander> jbjnr: I haven't done any runs lately, but I at least can still login into daint and tave. Would be extremely useful if those machines remain usable for Gregor and myself :)
<jbjnr> I'll look into it
<david_pfander> jbjnr: thanks a lot!
eschnett has quit [Quit: eschnett]
eschnett has joined #ste||ar
hkaiser has quit [Quit: bye]
jaafar_ has joined #ste||ar
aserio has joined #ste||ar
aserio has quit [Client Quit]
aserio has joined #ste||ar
Smasher has quit [Ping timeout: 240 seconds]
daissgr has quit [Ping timeout: 240 seconds]
daissgr has joined #ste||ar
mcopik has joined #ste||ar
hkaiser has joined #ste||ar
diehlpk_work has quit [Quit: Leaving]
EverYoung has joined #ste||ar
<hkaiser> jbjnr: I think I can add executor support to futures_factory easily while I'm at it
EverYoung has quit [Ping timeout: 276 seconds]
mcopik has quit [Ping timeout: 248 seconds]
aserio has quit [Remote host closed the connection]
aserio has joined #ste||ar
<wash[m]> aserio: I'll call in today
<aserio> Cool, do you have the new conferencing software link?
<aserio> wash[m]: ^^
<wash[m]> aserio: no I do not
<aserio> wash[m]: see pm
mcopik has joined #ste||ar
EverYoung has joined #ste||ar
david_pfander has quit [Ping timeout: 240 seconds]
diehlpk_work has joined #ste||ar
aserio has quit [Ping timeout: 276 seconds]
simbergm2 has joined #ste||ar
jaafar_ has quit [Remote host closed the connection]
jaafar has joined #ste||ar
simbergm2 has quit [Quit: WeeChat 1.9.1]
jaafar has quit [Remote host closed the connection]
simbergm2 has joined #ste||ar
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 250 seconds]
EverYoun_ has quit [Remote host closed the connection]
aserio has joined #ste||ar
EverYoung has joined #ste||ar
jaafar has joined #ste||ar
<jbjnr> hkaiser: please do add executor support if it doesn't take much work - that will allow a wide range of extra customizations that might prove useful in the future (pun)
<diehlpk_work> --hpx:print-counter=/threads{locality#0}/idle-rate What is the replacement for this counter?
<hkaiser> diehlpk_work: you need to enable this counter at configure time
<hkaiser> HPX_WITH_THREAD_IDLE_RATES=On
<hkaiser> jbjnr: on it
<diehlpk_work> hkaiser, Thanks, for got about that
jaafar has quit [Remote host closed the connection]
jaafar has joined #ste||ar
<jbjnr> hkaiser: please let me know when your branch is ready to be tested again - last time I tried it still wasn't compiling for me, but I'm ready to have another go.
<hkaiser> jbjnr: will do - still some problems - but I'm closing in
<hkaiser> jbjnr: also, Ihave the futures_fatory changes implemented on that branch as well
simbergm2 has quit [Quit: WeeChat 1.9.1]
simbergm2 has joined #ste||ar
<simbergm2> aserio: friday at 9 am your time ok?
<simbergm2> also jbjnr, 4 pm here, are you available then if needed?
<simbergm2> or if you're interested...
<aserio> simbergm2: I have another meeting at 9am
<aserio> simbergm2: but I would be free between 10am and 1:30pm
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 276 seconds]
<simbergm2> aserio: ok, then sometime around 10-10:30 am? whenever you're free and ready after the previous meeting
<aserio> simbergm2: Ok, Skype? or did you have something else in mind?
<simbergm2> aserio: yeah, I think that's easiest(?)
<aserio> Sure, see pm
simbergm2 has quit [Quit: WeeChat 1.9.1]
hkaiser has quit [Quit: bye]
<github> [hpx] hkaiser pushed 1 new commit to fixing_3027: https://git.io/vbkSz
<github> hpx/fixing_3027 43b1f20 Hartmut Kaiser: Fixing futures_factory
<jbjnr> simbergm: friday 4pm is fine - but for what?
<jbjnr> meeting with aserio ?
<jbjnr> ok
EverYoun_ has quit [Remote host closed the connection]
<aserio> jbjnr: Do you have my SkypeID?
<jbjnr> probly
EverYoung has joined #ste||ar
<aserio> jbjnr: I don't think I have yours
<jbjnr> aserio: mine is johnbiddiscombe
<jbjnr> I think
<aserio> Request sent
<jbjnr> got it
<aserio> jbjnr, simbergm: 10amCST is 5pmCET
<jbjnr> ok
mcopik has quit [Ping timeout: 258 seconds]
Smasher has joined #ste||ar
Smasher has quit [Changing host]
Smasher has joined #ste||ar
hkaiser has joined #ste||ar
mcopik has joined #ste||ar
hkaiser has quit [Read error: Connection reset by peer]
pagrubel has joined #ste||ar
pagrubel is now known as patg[[w]]
Smasher has quit [Remote host closed the connection]
hkaiser has joined #ste||ar
eschnett has quit [Quit: eschnett]
<aserio> hkaiser: yt?
<aserio> hkaiser: nvm
<aserio> :D
<jbjnr> hkaiser: question?
<jbjnr> can I make hpx_main run as a high priority task - or should I just spawn a new function as HP from main?
aserio has quit [Quit: aserio]
eschnett has joined #ste||ar
<hkaiser> jbjnr: you can't directly do that, just spawn a new thread
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 276 seconds]
nanashi55 has quit [Ping timeout: 248 seconds]
nanashi55 has joined #ste||ar
patg[[w]] has quit [Quit: Leaving]
mcopik has quit [Ping timeout: 248 seconds]