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
akheir has quit [Quit: Leaving]
hkaiser has quit [Quit: bye]
kale[m] has quit [Ping timeout: 258 seconds]
kale[m] has joined #ste||ar
weilewei has quit [Remote host closed the connection]
weilewei has joined #ste||ar
kale[m] has quit [Ping timeout: 256 seconds]
weilewei has quit [Remote host closed the connection]
kale[m] has joined #ste||ar
kale[m] has quit [Ping timeout: 260 seconds]
kale[m] has joined #ste||ar
kale[m] has quit [Ping timeout: 260 seconds]
kale[m] has joined #ste||ar
nikunj97 has joined #ste||ar
kale[m] has quit [Ping timeout: 260 seconds]
kale[m] has joined #ste||ar
Nikunj__ has joined #ste||ar
nikunj97 has quit [Ping timeout: 256 seconds]
Nikunj__ has quit [Ping timeout: 260 seconds]
Nikunj__ has joined #ste||ar
nikunj97 has joined #ste||ar
Nikunj__ has quit [Ping timeout: 260 seconds]
Nikunj__ has joined #ste||ar
nikunj97 has quit [Ping timeout: 256 seconds]
wash[m] has quit [Ping timeout: 272 seconds]
wash[m] has joined #ste||ar
karame_ has quit [Remote host closed the connection]
hkaiser has joined #ste||ar
<hkaiser> ms[m]: yt?
weilewei has joined #ste||ar
<ms[m]> hkaiser: here
<hkaiser> hey
<hkaiser> #4821 is ready, I think - there are some builders failing but I think those refer to older commits
<hkaiser> I'm pretty pleased with that, btw
<hkaiser> that should enable easy extension of the algorithms
<ms[m]> hey, are you sure the results are for an older commit? :/ the logs tell me they built 039696c which looks like the latest commit
<ms[m]> in any case, that pr is very nice! thank you for doing that?
<ms[m]> could it be that there's still a pair somewhere where there should be in_out_result now?
<hkaiser> ms[m]: I looked at my local files, those are ok, let me see if I forgot to commit
<hkaiser> right!
<hkaiser> I didn't look at CUDA - that's why :/
<hkaiser> will do right away
<hkaiser> ms[m]: will you join the gsoc meeting now?
K-ballo has quit [*.net *.split]
K-ballo has joined #ste||ar
<ms[m]> yes
<kale[m]> zao: I was testing the wheel and I'm getting the same error as you. (https://gist.github.com/zao/82965bdd4189c9652b460b89badbc3d6). Can you tell me what was the issue?
<hkaiser> kale[m]: that means that Phylanx was not able to find/load the plugins implementing the primitives
<hkaiser> (those in the libs/phylanx subdirectory)
diehlpk__ has joined #ste||ar
rtohid has joined #ste||ar
<kale[m]> hkaiser: I'm running python3 -v <test.py>. And the output shows that it is able to find those plugins (https://imgur.com/a/kP1dHVI)
<hkaiser> where do you see that those have been found?
nanmiao11 has joined #ste||ar
<kale[m]> hkaiser: It is not showing any errors after import statement in output. How can I check that all the plugins are loaded ??
<hkaiser> ms[m]: finally, I'm ready whenever you are: https://lsu.zoom.us/j/3340410194
diehlpk__ has quit [Ping timeout: 256 seconds]
<hkaiser> kale[m]: sure, it's designed to ignore those ;-)
<hkaiser> kale[m]: run the application with --hpx:hpx_debug_log=<file>, the logs should tell you what got loaded and what not
<hkaiser> it's --hpx:hpx-debug-log, I think
<kale[m]> Thanks!
<hkaiser> kale[m]: if it's a python script then you will need to add the option either to the Phylanx.Session configuration or an environment variable
<hkaiser> kale[m]: btw, it's --hpx:debug-hpx-log=<file>, sorry for the confusion
<rtohid> kale[m] you may also want to make sure you have correct paths set in your environment.
<kale[m]> rtohid: Are you talking about the environmental variables or paths in library ?
<hkaiser> kale[m]: you should also be able to set the base path of the phylanx installation by defining PHYLANX_PLUGINS_PATH as an environment variable
<rtohid> kale[m] environment variables
karame_ has joined #ste||ar
diehlpk__ has joined #ste||ar
diehlpk__ has quit [Ping timeout: 260 seconds]
<hkaiser> kale[m]: you can set the HPX_COMMANDLINE_OPTIONS environment variable, its value will be prepended verbatim to the commandline HPX sees before command line handling
nikunj97 has joined #ste||ar
Nikunj__ has quit [Ping timeout: 240 seconds]
<hkaiser> gonidelis[m]: yt?
nikunj97 has quit [Ping timeout: 240 seconds]
<parsa> hkaiser: https://www.olcf.ornl.gov/wp-content/uploads/2018/02/SummitDev_Spectrum-and-CUDA-awere-MPI.pdf#page=11 this is one of the references i was talking about that refers to "inter-node" GPU Direct. will ask weilewei
<weilewei> parsa this is a great reference! Thanks for sharing. Do you have any questions btw?
<parsa> weilewei: yeah. can you actually do move things in GPU memories between nodes on Summit?
<weilewei> yes, we can
<parsa> without using MPI yourself?
<weilewei> I haven't tried without MPI
<parsa> is it something anyone you know that uses Summit does? i thought this was supposed to be a feature Summit has that no one else yet had
<weilewei> I don't know if that is possible to do gpudirect without mpi?
<parsa> okay. good to know. thank you
<weilewei> in what scenarios do you want to have gpudirect without mpi? Just curious
<parsa> i don't care if MPI is used to implement it or not underneath. just using GPU Direct between GPUs to move data between GPUs
<weilewei> parsa I see, then I really don't know
<kale[m]> hkaiser: I added the option to output hpx debug log in Phylanx. And as you suspected the plugins are not loading. Can you tell me how plugins load in Phylanx ?
nikunj97 has joined #ste||ar
K-ballo has quit [Quit: K-ballo]
kale[m] has quit [Ping timeout: 258 seconds]
K-ballo has joined #ste||ar
kale[m] has joined #ste||ar
<hkaiser> kale[m]: what does it report?
<hkaiser> file not found? does it look in the right directory?
<ms[m]> hkaiser: finally... :P thanks for all the explanations again, I wouldn't have known where to even start otherwise
<ms[m]> the archive data gets set from one library and accessed from another, and the static function has different addresses with hidden visibility
<ms[m]> now I can sleep again :)
<kale[m]> hkaiser: https://gist.github.com/git-kale/33c99864f2323b1521b6e46b1407b197 This is the log generated. 17:58:01:370 showing No plugin found/loaded
kale[m] has quit [Ping timeout: 264 seconds]
kale[m] has joined #ste||ar
weilewei has quit [Remote host closed the connection]
weilewei has joined #ste||ar
<hkaiser> kale[m]: yah, that's not too useful
<hkaiser> :/
<hkaiser> kale[m]: try adding --hpx:dump-cfg to the command line - that might give us some more information
<kale[m]> hkaiser: Setting PHYLANX_PLUGINS_LIBRARY isnt working. But when I move the plugins to /usr/local/lib/phylanx the test pass
<kale[m]> *PHYLANX_PLUGINS_PATH
<rtohid> kale[m] try LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/path/to/plugins
<rtohid> export *...
<kale[m]> rtohid: Seems like I misspelled the PHYLANX_LIBRARY_PATH before. :-!
diehlpk__ has joined #ste||ar
diehlpk__ has quit [Ping timeout: 256 seconds]
hkaiser has quit [Quit: bye]
diehlpk__ has joined #ste||ar
K-ballo has quit [Ping timeout: 240 seconds]
K-ballo has joined #ste||ar
hkaiser has joined #ste||ar
diehlpk__ has quit [Ping timeout: 256 seconds]
kale[m] has quit [Ping timeout: 240 seconds]
kale[m] has joined #ste||ar
rtohid has left #ste||ar [#ste||ar]
Nikunj__ has joined #ste||ar
Nikunj__ has quit [Remote host closed the connection]
nikunj97 has quit [Ping timeout: 240 seconds]
<weilewei> hkaiser may I ask where should I move functions to, in your comments: https://github.com/STEllAR-GROUP/hpx/pull/4823#discussion_r453975423
<weilewei> like move to threading_base?
<weilewei> and libs/coroutines/include/hpx/coroutines/detail/context_base.hpp ?
<weilewei> :') since I blindly follow original hpx thread data, and add codes following that...
<hkaiser> weilewei: ahh, my bad - these are independent types - all is well
<hkaiser> pls ignore that comment
<weilewei> oh ok
<weilewei> another thing is I am not aware of what error handling is needed
<weilewei> I did cmake following apex cmake stuff