hkaiser changed the topic of #ste||ar to: STE||AR: Systems Technology, Emergent Parallelism, and Algorithm Research | stellar-group.org | HPX: A cure for performance impaired parallel applications | github.com/STEllAR-GROUP/hpx | This channel is logged: irclog.cct.lsu.edu
hkaiser has quit [Quit: Bye!]
diehlpk_work_ has joined #ste||ar
diehlpk_work has quit [Ping timeout: 240 seconds]
diehlpk_work__ has joined #ste||ar
diehlpk_work_ has quit [Ping timeout: 240 seconds]
diehlpk_work_ has joined #ste||ar
diehlpk_work__ has quit [Ping timeout: 240 seconds]
diehlpk_work has joined #ste||ar
diehlpk_work_ has quit [Ping timeout: 240 seconds]
diehlpk_work_ has joined #ste||ar
diehlpk_work has quit [Ping timeout: 240 seconds]
diehlpk_work_ has quit [Ping timeout: 240 seconds]
diehlpk_work has joined #ste||ar
diehlpk_work_ has joined #ste||ar
diehlpk_work has quit [Ping timeout: 240 seconds]
jehelset has joined #ste||ar
hkaiser has joined #ste||ar
diehlpk_work_ has quit [Ping timeout: 240 seconds]
<gnikunj[m]> hkaiser: just saw the hpx+lci meeting notification :/
<gnikunj[m]> Will join next meeting now
<hkaiser> gnikunj[m]: sure
diehlpk_work_ has joined #ste||ar
<gnikunj[m]> hkaiser: btw I wrote the abstract yesterday. Mind going over it in your free time today? If it looks good I'll submit it.
<hkaiser> gnikunj[m]: is it in overleaf?
<gonidelis[m]> hkaiser: pm
johnny_tester has joined #ste||ar
<johnny_tester> test
<gonidelis[m]> test
<gonidelis[m]> hkaiser: do we keep this?
johnny_tester has quit [Quit: Client closed]
<gonidelis[m]> ms: hpx does have mac support underlying infastructure right? it's just that it is broken or sth?
<gnikunj[m]> hkaiser: yes, it's in overleaf
<gonidelis[m]> I changed it
<gonidelis[m]> At some point
<gonidelis[m]> Will change the other case too
<gnikunj[m]> Thanks. It seems like the page hasn't been updated. We already have a TaskBench port now :P
<gnikunj[m]> I'll edit the page next weekend. Hope that's not too late.
<gonidelis[m]> gnikunj: I am on it
<gonidelis[m]> I talked with hartmut and already remove reordering stuff
<gnikunj[m]> nice
<ms[m]> gonidelis: it's working about as well as on linux: https://github.com/STEllAR-GROUP/hpx/runs/5069100847?check_suite_focus=true, i.e. occasional flaky tests, but otherwise ok
<ms[m]> the number of tested configurations is a lot smaller though
<gonidelis[m]> as well as on linux? dammit
<hkaiser> gonidelis[m]: that's MacOS I think that's different from iOS
<gonidelis[m]> but when we talk about M1 we talk about mac, don't we?
<gnikunj[m]> gonidelis: I think M1 is separate coz it's Arm so there may be issues building it there.
<gnikunj[m]> MacOS is not the problem there. We can build on x86 MacOS.
<gonidelis[m]> but there is no ios - m1 correspondance
<gonidelis[m]> also apple droped intel...
<gnikunj[m]> iOS is separate. That's building HPX on iPad or iPhone.
<gonidelis[m]> that's what i am saying
<gnikunj[m]> M1 is not iOS
<gnikunj[m]> what the project is about is to port it to M1 AND iOS
<gonidelis[m]> gonidelis[m]: .
<gnikunj[m]> It builds on x86 Macs. We don't know about M1 Macs yet.
<gnikunj[m]> Macs come in different architecture flavors.
<gonidelis[m]> i kno
<gonidelis[m]> that's why it's a project
<gnikunj[m]> right. So what's the question?
<gonidelis[m]> whether mac (of course x86 mac that is) is broken or not
<gnikunj[m]> nope it isn't (at least in the configurations we check on)
<gnikunj[m]> having said that, I think static HPX builds may be something that's broken there (it was back in 2019)
jehelset has quit [Ping timeout: 250 seconds]
<gnikunj[m]> hkaiser: europar deadline extended by a week. Now we should be able to get acceptance from Sandia before submitting!
<hkaiser> gnikunj[m]: cool
<hkaiser> gonidelis[m]: pm, please
<diehlpk_work_> ms[m], Are you aware of any updates on Piz Daint?
<diehlpk_work_> I am running the same hpx and octo as I did in September and now it it 4 times slower
jehelset has joined #ste||ar