hkaiser 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/ | GSoC: https://github.com/STEllAR-GROUP/hpx/wiki/Google-Summer-of-Code-%28GSoC%29-2020
nanmiao11 has quit [Remote host closed the connection]
hkaiser_ has joined #ste||ar
hkaiser has quit [Ping timeout: 244 seconds]
hkaiser_ has quit [Quit: bye]
Yorlik has quit [Ping timeout: 265 seconds]
shahrzad has quit [Quit: Leaving]
akheir has quit [Quit: Leaving]
norbert[m] has left #ste||ar ["Kicked by @appservice-irc:matrix.org : Idle for 30+ days"]
Yorlik has joined #ste||ar
parsa has quit [Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net]
parsa has joined #ste||ar
hkaiser has joined #ste||ar
akheir has joined #ste||ar
bita has joined #ste||ar
parsa has quit [Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net]
parsa has joined #ste||ar
weilewei has joined #ste||ar
diehlpk_work_ has joined #ste||ar
diehlpk_work has quit [Remote host closed the connection]
<weilewei> how to turn on hpx mpi async feature nowadays?
<weilewei> but without distributed version
<weilewei> hkaiser ms[m] ^^
diehlpk_work__ has joined #ste||ar
nanmiao11 has joined #ste||ar
<weilewei> HPX_WITH_ASYNC_MPI ?
diehlpk_work_ has quit [Ping timeout: 244 seconds]
diehlpk_work__ has quit [Ping timeout: 240 seconds]
diehlpk_work has joined #ste||ar
nanmiao11 has quit [Remote host closed the connection]
nanmiao11 has joined #ste||ar
<hkaiser> weilewei: I think so, hold on
<weilewei> hkaiser ok
<hkaiser> weilewei: yes, it's HPX_WITH_ASYNC_MPI
<weilewei> hkaiser nice, I will start using it to overlap things
<hkaiser> cool
weilewei has quit [Remote host closed the connection]
weilewei has joined #ste||ar
nanmiao11 has quit [Remote host closed the connection]
diehlpk_work_ has joined #ste||ar
nanmiao11 has joined #ste||ar
diehlpk_work has quit [Ping timeout: 244 seconds]
<diehlpk_work_> hkaiser, Can you please remove GSoC from the welcome message?
K-ballo 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/
<diehlpk_work_> K-ballo, Thanks
parsa has quit [Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net]
parsa has joined #ste||ar
weilewei has quit [Remote host closed the connection]
weilewei has joined #ste||ar
parsa has quit [Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net]
parsa has joined #ste||ar
parsa has quit [Client Quit]
parsa has joined #ste||ar
parsa has quit [Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net]
parsa has joined #ste||ar
weilewei has quit [Remote host closed the connection]
parsa has quit [Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net]
parsa has joined #ste||ar
nanmiao11 has quit [Remote host closed the connection]
khuck has joined #ste||ar
khuck has left #ste||ar [#ste||ar]
khuck has joined #ste||ar
<khuck> hkaiser this might be faster
<hkaiser> khuck: hey
<hkaiser> sorry for all the trouble
<khuck> not at all
<hkaiser> not sure what's going on, though
<khuck> I don't know how to C++, that's what's going on
<khuck> as I said in the email, everything seems to work until I try to get the result from the future returned by the gather_here/gather_there calls
<khuck> then both localities hang
<hkaiser> strange
<hkaiser> khuck: if gather hangs on two localities in a simple hello world setting, can you give me the code?
<hkaiser> I'd be happy to have a look
<khuck> how do I build the example in libs/full/collectives/tests/unit/gather.cpp?
<hkaiser> make tests.unit.modules.collectives.gather
<khuck> fun! That test hangs too
<khuck> whew
<hkaiser> grrr
<hkaiser> khuck: ok, you step back for a moment, I'll have a look
<khuck> "everybody clear the room"
<khuck> I'll poke it with a stick while you see if you can reproduce it
nanmiao11 has joined #ste||ar
<khuck> hkaiser wait a minute, it hung because it was in APEX OTF2 shutdown
<khuck> it doesn't hang on its own
<hkaiser> ok, I was already wondering
<khuck> I am modifying the example to see if it works with strings
<khuck> but another question - the thread that calls finalize() - does HPX consider it a worker thread? Can I make requests from it?
<khuck> I remember we had this problem before with the APEX asynchronous background processing
<khuck> wait, things are progressing again... I rebuilt everything and now I am getting that same weird error from the broadcast_to() call - that there are 0 ranks in the communicator
<khuck> I think I know how to fix that
Yorlik has quit [Ping timeout: 260 seconds]
bita_ has joined #ste||ar
parsa has quit [Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net]
parsa has joined #ste||ar
bita has quit [Ping timeout: 260 seconds]
<khuck> hkaiser good news, bad news - it works, but intermittently crashes
<khuck> and sometimes hangs
<hkaiser> race condition>
<hkaiser> ?
parsa has quit [Client Quit]
<khuck> don't know. the question is, does thread 0 have permission to make these calls? does HPX consider it a worker?
parsa has joined #ste||ar
<hkaiser> the thread that was executing main()?
<hkaiser> I don't think that's a good idea
<khuck> yeah, confirmed it's the master thread that is making these calls
<khuck> should I schedule it as a "pre shutdown" task?
<khuck> is that possible?
<khuck> yes, it's possible
<khuck> I'm going to try that
<hkaiser> khuck: yes, please
<khuck> that seems to work. I
<khuck> I'll do some more testing
<khuck> getting random deadlocks and crashes when I run more than 2 localities
<khuck> but I think that's an openmpi problem
parsa has quit [Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net]
parsa has joined #ste||ar