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/
eschnett has joined #ste||ar
jaafar_ has joined #ste||ar
jaafar has joined #ste||ar
jaafar has quit [Read error: Connection reset by peer]
jaafar has joined #ste||ar
jaafar_ has quit [Ping timeout: 272 seconds]
Anushi1998 has joined #ste||ar
anushi_ has quit [Ping timeout: 240 seconds]
anushi has joined #ste||ar
Anushi1998 has quit [Ping timeout: 245 seconds]
anushi_ has joined #ste||ar
anushi has quit [Ping timeout: 252 seconds]
jaafar has quit [Ping timeout: 252 seconds]
anushi_ has quit [Quit: Bye]
jaafar has joined #ste||ar
eschnett has quit [Quit: eschnett]
eschnett has joined #ste||ar
hkaiser has quit [Quit: bye]
jaafar has quit [Ping timeout: 260 seconds]
nanashi55 has quit [Ping timeout: 272 seconds]
nanashi55 has joined #ste||ar
jaafar has joined #ste||ar
jaafar has quit [Ping timeout: 240 seconds]
jaafar has joined #ste||ar
jaafar has quit [Ping timeout: 240 seconds]
david_pfander has joined #ste||ar
nikunj has joined #ste||ar
nikunj97 has joined #ste||ar
nikunj has quit [Ping timeout: 240 seconds]
heller has quit [Read error: Connection reset by peer]
hkaiser has quit [Read error: Connection reset by peer]
hkaiser has joined #ste||ar
<zao>
hkaiser: Had lunch with my department head today, made sure to mention HPX while we were discussing their H2020 project and their use of StarPU :)
<K-ballo>
failing *after*? ok, I'll have a look
<hkaiser>
zao: you didn't face much understanding, I'd venture
<hkaiser>
K-ballo: yes
<hkaiser>
StarPU is the elephant in the room in Europe in this domain
<zao>
They seem to be quite determined that you need some sort of model that's better than stop-the-world chatter over MPI, and they seem reasonably sold on the value of task graphs and more granular dependencies.
<zao>
I didn't cover the futurization and other fancypants stuff, mostly described how it had the cluster transparency, implementation of the SC++L-like algorithms, components and stuff.
<zao>
In any way, the term "HPX" is now seeded in their mind :)
<zao>
That reminds me, I need to compile a newer version of it.
<hkaiser>
zao: ok, thanks
<zao>
The guy's a leading authority in numeric linear algebra with focus on some fun niche.
<zao>
Something something eigenblaha problems, in this case.
nikunj97 has joined #ste||ar
eschnett has quit [Ping timeout: 252 seconds]
hkaiser has quit [Quit: bye]
aserio has joined #ste||ar
Anushi1998 has joined #ste||ar
_diers_1 has joined #ste||ar
_diers_ has quit [Ping timeout: 252 seconds]
_diers_1 is now known as _diers_
hkaiser has joined #ste||ar
<K-ballo>
hkaiser: I can reproduce the failure *before* the merging
<K-ballo>
looks like a global barrier is involved, and it throws an interruption exception from its destructor while unwinding another interruption exception
* K-ballo
doubts the whole thread interruption machinery now
aserio has quit [Ping timeout: 260 seconds]
<K-ballo>
the barrier is called `b`
<K-ballo>
I can't see where it comes from
<zao>
What a good name :)
eschnett has joined #ste||ar
<hkaiser>
K-ballo: ok, I have not seen it before the merge, so sorry for blaming it on you
<K-ballo>
WOW, and now vs auto-rebooted while the debug session was active :|
<K-ballo>
I've never seen that happen before
<zao>
Been there, done that, considered throwing computer out window.
<K-ballo>
really? VS restarting?
<zao>
It self-restarts when it "crashes".
<K-ballo>
I've had windows restart and take everything with it, but never just VS
<zao>
Or maybe turns unresponsive.
<K-ballo>
stupid VS :@
<zao>
Not sure why, but it sure does it.
<K-ballo>
I lost my trace now :/
<zao>
Dumps are your friend :)
<K-ballo>
I wasn't even using it... it was paused and in the background
<zao>
But yes, it's completely stupid, whatever it's doing.
<zao>
I consider Windows and VS extremely unreliable nowadays.
<zao>
(lost a HPX multi-process trace that took days to reproduce due to WU a while ago :( )
<akheir>
galabc: what compiler you want to use and which version?
<galabc>
I want to use clang6.0.0
<akheir>
did you used modules? run $ module load clang/6.0.0
<akheir>
it should set all the environment variables for you
<galabc>
yes
<galabc>
when I load the module it is the variable $CPLUS_INCLUDE_PATH is empty
<akheir>
in the error message you posted yesterday you ran g++ while loaded the clang module
<zao>
I've never seen that variable before, normally a compiler has its paths compiled in.
<zao>
And if not, the module sets whatever state is needed.
<akheir>
you see when you load the clang module it set the -stdlib=libc++
<akheir>
but g++ does not work with libc++. it has its own libstdc++
<galabc>
okok
<galabc>
but Im not sure why adding a path to $CPLUS_INCLUDE_PATH would bring an error since it has its path compiled in as you say
<akheir>
I don't know about that one
<galabc>
Its ok its just wierd
<galabc>
So I wont manually modify $CPLUS_INCLUDE_PATH when using module clang in the futur :D
<akheir>
yes. and you cannot use gcc when clang module is loaded.
<galabc>
Okok I will know in the futur
<akheir>
good ;-)
<galabc>
All this compiler stuff is new to me but I will learn in time
<galabc>
In my bachelor degree we mostly used python and matlab where compilation was a no brainer
<zao>
galabc: Those are the worst things for me as a cluster sysadmin ;)
<zao>
Sometimes modules break the outside world, and it's hard to mitigate.
<zao>
(which is why we at my site recommend that users only load modules they need when they need them, and don't do it in their login .bashrc profile and whatnot)
<zao>
galabc: Considering that I run part of our support on my cluster, your questions are not remotely as basic as some I see, if it makes you feel better :)
anushi has joined #ste||ar
aserio has quit [Ping timeout: 252 seconds]
Anushi1998 has quit [Ping timeout: 272 seconds]
<galabc>
zao: its good to know haha
<K-ballo>
hkaiser: took a lot longer, but I could reproduce before atomic futures as well
Anushi1998 has joined #ste||ar
Anushi1998 has quit [Remote host closed the connection]
<K-ballo>
updated the gist with a trace
Anushi1998 has joined #ste||ar
jaafar has joined #ste||ar
Anushi1998 has quit [Ping timeout: 260 seconds]
Anushi1998 has joined #ste||ar
mcopik has quit [Ping timeout: 250 seconds]
Anushi1998 has quit [Ping timeout: 240 seconds]
Anushi1998 has joined #ste||ar
Anushi1998 has quit [Ping timeout: 252 seconds]
<hkaiser>
K-ballo: thanks
jaafar has quit [Ping timeout: 250 seconds]
Anushi1998 has joined #ste||ar
Anushi1998 has quit [Remote host closed the connection]
Anushi1998 has joined #ste||ar
aserio has joined #ste||ar
jaafar has joined #ste||ar
david_pfander has quit [Ping timeout: 252 seconds]
jaafar has quit [Ping timeout: 240 seconds]
Anushi1998 has quit [Ping timeout: 245 seconds]
Anushi1998 has joined #ste||ar
Anushi1998 has quit [Remote host closed the connection]
Anushi1998 has joined #ste||ar
Anushi1998 has quit [Ping timeout: 252 seconds]
Anushi1998 has joined #ste||ar
jaafar has joined #ste||ar
Anushi1998 has quit [Remote host closed the connection]
Anushi1998 has joined #ste||ar
Anushi1998 has quit [Remote host closed the connection]
Anushi1998 has joined #ste||ar
heller__ has joined #ste||ar
<heller__>
I'm really getting annoyed by freenode...
aserio has quit [Ping timeout: 260 seconds]
<zao>
What shenanigans are they up to now?
<zao>
Same old registration/spam stuff?
Anushi1998 has quit [Ping timeout: 252 seconds]
Anushi1998 has joined #ste||ar
Anushi1998 has quit [Ping timeout: 245 seconds]
Anushi1998 has joined #ste||ar
Anushi1998 has quit [Ping timeout: 260 seconds]
anushi has quit [Quit: Bye]
Anushi1998 has joined #ste||ar
aserio has joined #ste||ar
hkaiser has quit [Read error: Connection reset by peer]
hkaiser has joined #ste||ar
nikunj97 has quit [Ping timeout: 272 seconds]
<heller__>
zao: yeah... I'm mostly not reconnecting properly