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
Yorlik has quit [Ping timeout: 240 seconds]
nanmiao11 has quit [Remote host closed the connection]
diehlpk has quit [Ping timeout: 240 seconds]
hkaiser has quit [Quit: bye]
kale[m] has quit [Ping timeout: 240 seconds]
kale[m] has joined #ste||ar
kale[m] has quit [Ping timeout: 240 seconds]
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: 244 seconds]
kale[m] has joined #ste||ar
kale[m] has quit [Ping timeout: 244 seconds]
kale[m] has joined #ste||ar
kale[m] has quit [Ping timeout: 244 seconds]
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
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
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
Yorlik has joined #ste||ar
kale[m] has quit [Ping timeout: 256 seconds]
kale[m] has joined #ste||ar
hkaiser has joined #ste||ar
kale[m] has quit [Ping timeout: 260 seconds]
kale[m] has joined #ste||ar
akheir has joined #ste||ar
diehlpk_work has joined #ste||ar
diehlpk_work has quit [Changing host]
diehlpk_work has joined #ste||ar
<diehlpk_work> h/join #fedora-devel
<diehlpk_work> hkaiser, Can you please do the GSoC evaluation today?
<diehlpk_work> ms[m], jbjnr Can you do the GSoC evaluation today as well?
<diehlpk_work> parsa, Same for you
<hkaiser> diehlpk_work: will do
<diehlpk_work> ms[m], HPX 1.5.0-rc1 compiled on all arch on Fedora
<diehlpk_work> We just see warning, we already now about and hkaiser will fix them for the rc2
<ms[m]> hooray, we haven't regressed on that front at least
<ms[m]> ok, thanks (both of you)
weilewei has quit [Remote host closed the connection]
weilewei has joined #ste||ar
nanmiao11 has joined #ste||ar
kale[m] has quit [Read error: Connection reset by peer]
kale[m] has joined #ste||ar
<hkaiser> diehlpk_work: done
Yorlik has quit [Ping timeout: 264 seconds]
Yorlik has joined #ste||ar
ct-clmsn has joined #ste||ar
weilewei has quit [Remote host closed the connection]
weilewei has joined #ste||ar
Yorlik has quit [Ping timeout: 256 seconds]
ct-clmsn has quit [Read error: Connection reset by peer]
ct-clmsn has joined #ste||ar
ct-clmsn is now known as Guest3209
<weilewei> apex with cuda_support works on Summit for DCA++ hkaiser
<weilewei> also Wombat machine at ORNL...
<weilewei> I will try Cori soon, it seems I still get hdf5 issue, even though now it compiles on Cori
Guest3209 has quit [Read error: Connection reset by peer]
Guest3209 has joined #ste||ar
Guest3209 is now known as ct_clmsn_
ct_clmsn_ has quit [Quit: Leaving]
Yorlik has joined #ste||ar
kale[m] has quit [Ping timeout: 260 seconds]
kale[m] has joined #ste||ar
nanmiao11 has quit [Remote host closed the connection]
nanmiao11 has joined #ste||ar
noise[m] has quit [Ping timeout: 244 seconds]
kimbo[m] has quit [Ping timeout: 244 seconds]
freifrau_von_ble has quit [Ping timeout: 244 seconds]
fred[m] has quit [Ping timeout: 244 seconds]
diehlpk_mobile[m has quit [Ping timeout: 244 seconds]
noise[m] has joined #ste||ar
freifrau_von_ble has joined #ste||ar
diehlpk_mobile[m has joined #ste||ar
kimbo[m] has joined #ste||ar
<kale[m]> hkaiser: Since the size of Phylanx wheel is greater than 100 MB, and I dont have an enterprise edition of github. I am having problem with testing the wheel. The git LFS data limit is very small for Student edition.
fred[m] has joined #ste||ar
<kale[m]> I have a working wheel but testing the wheel is difficult as Build and Test environments are different.
<kale[m]> Would it be possible for you to create a repo on stellar and grant me access so that I can push everything there.
<hkaiser> kale[m]: the wheel doesn't belong into a git repository
<hkaiser> kale[m]: but I'm sure akheir can help with an ftp access for you to upload and test the files, if needed
<K-ballo> what makes it so big?
<gnikunj[m]> K-ballo: it contains all dependencies
<K-ballo> like half of boost?
<gnikunj[m]> HPX, Boost, hwloc etc. etc.
<K-ballo> yeah, that'll do it
<gnikunj[m]> it is 360mb as of now
<gnikunj[m]> we thought of git lfs as a solution
<hkaiser> gnikunj[m]: do we need boost binaries?
<gnikunj[m]> hkaiser: we have only the necessary one
<hkaiser> which one?
<gnikunj[m]> not all
<hkaiser> I think we don't link with boost anymore
<gnikunj[m]> kale[m]: list the boost libraries
<gnikunj[m]> hkaiser: there were boost dependencies in rpath for sure
<K-ballo> there's always the posibility of linking boost statically
<gnikunj[m]> that I remember quite well
<hkaiser> gnikunj[m]: with c++17 we don't have any boost linktime dependencies
<K-ballo> program options is still opt-out, no?
<gnikunj[m]> hkaiser: we use phylanx pre-requisite docker image to build phylanx. It does have dependencies.
<hkaiser> should kick in as well
<gnikunj[m]> There hpx is built with c++14 iirc
<hkaiser> gnikunj[m]: shrug
<hkaiser> stop complaining, then ;-)
<gnikunj[m]> I'm not complaining :p
<hkaiser> sure you do
<gnikunj[m]> if we have C++17 that gets us rid of size, then we can tone down the size by a good margin
<hkaiser> nod, try that
<gnikunj[m]> <hkaiser "sure you do"> old habits don't die easily I guess :P
<kale[m]> The size is due to Phylanx plugins boost libraries combined take less than 1 MB
<hkaiser> urgs
<kale[m]> hkaiser: libbosst_filesystem, libboost_program_options, libboost_system are the only boost libraries in the wheel
<gnikunj[m]> we have hpx program options that replaced boost program options
<K-ballo> is that an old boost build? system went header only several releases ago
<hkaiser> kale[m]: neither are still needed (if compiled with c++17, i.e. <filesystem> is available)
<kale[m]> Oh! I am creating the wheel from phylanx_base so the Phylanx was having a dependency on Boost
<weilewei> ok dca runs on Cori now. next step is building hpx+apex on Cori and run dca with them
<hkaiser> weilewei: great job!
nanmiao11 has quit [Remote host closed the connection]
<gnikunj[m]> hkaiser: yt?
kale[m] has quit [Ping timeout: 260 seconds]
kale[m] has joined #ste||ar
<hkaiser> here
diehlpk has joined #ste||ar
<gnikunj[m]> hkaiser: how do I get templated arguments to work with plain actions?
<hkaiser> gnikunj[m]: what's the problem?
<gnikunj[m]> it is not compiling
<hkaiser> heh
<hkaiser> show me the code
<gnikunj[m]> wait that's not it
<gnikunj[m]> updated
<hkaiser> gnikunj[m]: well, the compiler doesn't know which of the infinite number of functions you might generate from the template it should bind to the action
<gnikunj[m]> so I can't use a variadic template with an action?
Yorlik has quit [Ping timeout: 246 seconds]
<hkaiser> you can, give me a sec
<hkaiser> mind, it's untested
<hkaiser> well, put a typename in there as the compiler requests
<gnikunj[m]> ohh it's a typename error lol
<hkaiser> as said, this is from the top of my head, use your brain
<gnikunj[m]> aah well, these are issues I can solve
<hkaiser> nod
<gnikunj[m]> let me correct these and return
<hkaiser> use int (*)(int) instead of int(int)
<gnikunj[m]> yes
<hkaiser> I can play with it later, if needed
<gnikunj[m]> I'll let you know. If this works that's good enough
<gnikunj[m]> this is an experiment for the replicate api's
<hkaiser> essentially what you want in teh end is make_action<decltype(&f), &f>::type, where f is the function to be bound
<gnikunj[m]> I want to check if there's a performance benefit of creating another listener for interruption
<gnikunj[m]> hkaiser: thanks!
diehlpk has quit [Ping timeout: 244 seconds]
<gnikunj[m]> hkaiser: actions don't support raw pointers. int (*)(int) won't work
<gnikunj[m]> std::function and bind may work. Let me try
<hkaiser> gnikunj[m]: I think you need to bind the function pointer to the universal answer to the action as well
<hkaiser> or create a separate plain action for the universal answer and bind that
<hkaiser> let me try to sketch something
<gnikunj[m]> hkaiser: thanks.