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/
khuck has quit [Remote host closed the connection]
khuck has joined #ste||ar
khuck has quit [Ping timeout: 252 seconds]
khuck has joined #ste||ar
nanashi55 has quit [Ping timeout: 244 seconds]
nanashi55 has joined #ste||ar
khuck has quit [Ping timeout: 252 seconds]
rtohid has joined #ste||ar
rtohid has left #ste||ar ["Leaving"]
jbjnr_ has quit [Read error: Connection reset by peer]
khuck has joined #ste||ar
jbjnr has joined #ste||ar
khuck has quit [Ping timeout: 244 seconds]
khuck has joined #ste||ar
khuck has quit [Ping timeout: 252 seconds]
eschnett has quit [Read error: Connection reset by peer]
eschnett has joined #ste||ar
khuck has joined #ste||ar
david_pfander has joined #ste||ar
khuck_ has joined #ste||ar
khuck has quit [Read error: Connection reset by peer]
khuck_ has quit [Ping timeout: 252 seconds]
nikunj has quit [Ping timeout: 260 seconds]
nikunj has joined #ste||ar
K-ballo has quit [Ping timeout: 246 seconds]
khuck has joined #ste||ar
khuck has quit [Read error: Connection reset by peer]
khuck has joined #ste||ar
khuck has quit [Ping timeout: 246 seconds]
khuck has joined #ste||ar
khuck has quit [Ping timeout: 252 seconds]
khuck has joined #ste||ar
khuck has quit [Ping timeout: 250 seconds]
<jbjnr>
my builds on daint all fail because of link errors to lib atomic - surely we have that in our cmake somewhere?
khuck has joined #ste||ar
<jbjnr>
khuck: I saw that last night you asked about compute resources - we actually have a test project for HPX on cscs machines that gives us some time
khuck has quit [Ping timeout: 250 seconds]
<heller__>
jbjnr: I fixed the dataflow problem locally
<ste||ar-github>
hpx/fix_dataflow 72f2fe4 Thomas Heller: Fixing dataflow...
ste||ar-github has left #ste||ar [#ste||ar]
<heller__>
I really like the circle failure reporting
khuck has joined #ste||ar
khuck has quit [Read error: Connection reset by peer]
khuck_ has joined #ste||ar
khuck_ has quit [Read error: Connection reset by peer]
khuck has joined #ste||ar
khuck has quit [Read error: Connection reset by peer]
khuck has joined #ste||ar
khuck has quit [Ping timeout: 245 seconds]
khuck has joined #ste||ar
khuck has quit [Ping timeout: 272 seconds]
khuck has joined #ste||ar
khuck has quit [Ping timeout: 246 seconds]
mcopik has joined #ste||ar
khuck has joined #ste||ar
khuck has quit [Read error: Connection reset by peer]
khuck_ has joined #ste||ar
<zao>
I talked with a CS department colleague the other day, they've gotten a nice 288 core NUMA machine.
<zao>
It's apparently six machines daisy-chained with duct tape. I wonder how well HPX will balance across that architecture, if it at all understands hwloc's world view :)
khuck_ has quit [Ping timeout: 252 seconds]
khuck has joined #ste||ar
<jbjnr>
someone tell me what boost thread needs -latomic and we don't do it?
<jbjnr>
s/what/why/
<jbjnr>
is this another casualty of the cmake changes?
khuck has quit [Ping timeout: 250 seconds]
<heller__>
jbjnr: no, its a fault in the cmake detection to add -latomic
khuck has joined #ste||ar
khuck has quit [Ping timeout: 272 seconds]
hkaiser has joined #ste||ar
<hkaiser>
heller__: I might not be able to talk today, driving all day
<heller__>
hkaiser: alright. have a save trip!
<hkaiser>
sorry...
<heller__>
np
<hkaiser>
heller__: btw, Klaus changed blaze, fixing our issue
<heller__>
hkaiser: saw it
<heller__>
hkaiser: pretty fast
<heller__>
i'd still like to upgrade to clang7 ;)
<hkaiser>
sure, np
<heller__>
hkaiser: btw, do we have access to any larger scale cluster that's not a cray these days?
<hkaiser>
define 'large'
<heller__>
>256 nodes
<hkaiser>
there are still the (older) LSU clusters
<hkaiser>
we could have access there if needed
<hkaiser>
we could get access to stampede and lonestar, not sure if those are cray
<heller__>
no, dell
<heller__>
which timeframes would we get access to those?
<hkaiser>
shrug, is it urgent?
<heller__>
more or less ;)
<hkaiser>
LSU should be easy enough
<heller__>
ok, let me try to ponder in my basement for a little more
<hkaiser>
TACC resources: talk to Max (do you know him)?
<heller__>
no
<hkaiser>
Maximilian Bremer <bremerm31@gmail.com>, graduate student of Clint Dawson, works with HPX
<heller__>
lonestar 5 is a cray
<heller__>
thanks
<zao>
Bah, forgot that my build system maxes out -j by default... hello swap my old friend.
<zao>
== Results of the build can be found in the log file(s) /home/zao/.local/easybuild/software/HPX/1.1.0-496-g72f2fe42fbd-foss-2018b-cxx14/easybuild/easybuild-HPX-1.1.0-496-g72f2fe42fbd-20181016.141434.log
<zao>
heller__: fix_dataflow branch passed the test suite :D
<heller__>
zao: woohoo! :D
<zao>
heller__: Managed to craft EasyBuild recipes for C++14 and C++17 builds of HPX with GCC 7.3.0 and OpenMPI 3.