aserio 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 has quit [Ping timeout: 248 seconds]
zbyerly_ has joined #ste||ar
Rodario has joined #ste||ar
Rodario1 has quit [Ping timeout: 240 seconds]
Rodario1 has joined #ste||ar
Rodario has quit [Ping timeout: 248 seconds]
mcopik has quit [Ping timeout: 264 seconds]
<K-ballo>
hkaiser: here
<hkaiser>
K-ballo: hey
<hkaiser>
just wanted to confirm the best way to extend hpx::util::begin et.al.
<hkaiser>
for a type namespace foo { struct A {}; } I would have to add a begin(A&) in namespace foo
<K-ballo>
extend? how? likely using the same patterns the existing facilities use
<hkaiser>
is that the right way?
<K-ballo>
that's the way that works on all the targeted compilers
<hkaiser>
ok, thanks
Rodario has joined #ste||ar
Rodario1 has quit [Ping timeout: 248 seconds]
bikineev has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
Rodario1 has joined #ste||ar
Rodario has quit [Ping timeout: 240 seconds]
EverYoung has quit [Ping timeout: 276 seconds]
hkaiser has quit [Quit: bye]
vamatya has quit [Ping timeout: 248 seconds]
Rodario has joined #ste||ar
Rodario1 has quit [Ping timeout: 248 seconds]
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
Rodario1 has joined #ste||ar
vamatya has joined #ste||ar
Rodario has quit [Ping timeout: 248 seconds]
EverYoung has joined #ste||ar
EverYoung has quit [Ping timeout: 246 seconds]
Rodario has joined #ste||ar
Rodario1 has quit [Ping timeout: 246 seconds]
Rodario1 has joined #ste||ar
patg has joined #ste||ar
patg is now known as Guest9383
Rodario has quit [Ping timeout: 240 seconds]
zbyerly__ has joined #ste||ar
zbyerly_ has quit [Ping timeout: 276 seconds]
K-ballo has quit [Quit: K-ballo]
Guest9383 is now known as patg
Rodario has joined #ste||ar
Rodario1 has quit [Ping timeout: 248 seconds]
EverYoung has joined #ste||ar
patg has quit [Read error: Connection reset by peer]
EverYoung has quit [Ping timeout: 246 seconds]
Rodario has quit [Ping timeout: 240 seconds]
quaz0r has joined #ste||ar
Matombo has joined #ste||ar
Matombo has quit [Remote host closed the connection]
vamatya has quit [Ping timeout: 246 seconds]
Matombo has joined #ste||ar
david_pfander has joined #ste||ar
<jbjnr>
heller: yt?
zbyerly__ has quit [Ping timeout: 248 seconds]
<heller>
jbjnr: hey
<heller>
jbjnr: we need to fix master ;
<heller>
)
bikineev has joined #ste||ar
<github>
[hpx] hendrx force-pushed fixing_threadmanager_hpp_headers from d228f75 to cfd1f27: https://git.io/v5sXA
<github>
hpx/fixing_threadmanager_hpp_headers cfd1f27 Arne Hendricks: Fixing unmatched header defines and recursive inclusion of threadmanager.hpp in threadmanager.hpp
mcopik has joined #ste||ar
Matombo has quit [Ping timeout: 240 seconds]
hkaiser has joined #ste||ar
bikineev has quit [Remote host closed the connection]
bikineev has joined #ste||ar
bikineev has quit [Ping timeout: 246 seconds]
<jbjnr>
heller: what's wrong with master?
<jbjnr>
have you informed the others about hackathon?
<heller>
not yet
<jbjnr>
CSCS admin iformaned now. see email
<jbjnr>
^informed.
<jbjnr>
Had someone from google here talking about "Go". Not as awesome as I had thought.
<zao>
Cute in some niches, but generally dead-end and the Google control over it is strong.
<zao>
Got a legitimate bug that they don't feel is worth fixing? You're hosed.
<zao>
(I'm bitter as heck over mine)
<jbjnr>
what bug?
<hkaiser>
jbjnr: Go has channels and a lightweight threading model - should sound familiar
<jbjnr>
yes. I know we stole the idea of channels off them, but ours are better etce tc, and go seems just a bit shit overall.
<zao>
jbjnr: They rejected TLS handshakes over a particular size as a hardening measure against DoS.
<zao>
jbjnr: When shown that whenever you deal with GRID or CERN, you have handshake messages way over their self-imposed limit and below any RFC recommendations.
<zao>
They just felt that our use cases were utterly irrelevant and closed the bug.
<hkaiser>
jbjnr: and master has some tests broken after the rpo stuff was merged over
<hkaiser>
but that was expected as not everything has been implemented yet
<zao>
As it's in the standard library, you have to hack your own build of the toolchain to fix the bug, which is _great_ if you want to make something source-distributable.
<jbjnr>
ok. I shall start working on hpx again this week, so I'll have a look
<zao>
It's the Google way. Universal across all their products from Chrome to Go and everything inbetween.
<hkaiser>
jbjnr: there is a concrete ticket
<hkaiser>
hold on...
<zao>
Involve them in any way in your stacks and you're likely to be sad.
<jbjnr>
:(
<hkaiser>
jbjnr: #2862, would you have time to look into that?
Rodario has joined #ste||ar
<hkaiser>
jbjnr: I really think we shouldn't use 'parcelport'-logging in the scheduler, btw
<jbjnr>
no. That's just me getting stuff to work locally and it ended up being merged cos I forgot about it,.
<hkaiser>
nod
<hkaiser>
np
<jbjnr>
at some point I must figure how to use the hpx logging. is it worth the effort?
<hkaiser>
jbjnr: the logging we currently have is not the ultimate solution, but it usually works
Rodario has quit [Quit: Leaving.]
Rodario has joined #ste||ar
bikineev has joined #ste||ar
<jbjnr>
hkaiser: the thing I'd like to do, is disable all the logging -except- the stuff I am working on. if that can be done, then I'll invest some time in using it.
<hkaiser>
jbjnr: ok, let's think about how we can achieve this
<jbjnr>
ok, but not today :)
<hkaiser>
sure, just fix #2862, pls
<jbjnr>
ok
<heller>
#2860 is fixed.
<hkaiser>
disable the stuff for osx in the simplest case (like it's disable d for windows)
<hkaiser>
heller: what was it?
<heller>
hkaiser: dirty build
<hkaiser>
uhh
<heller>
as usual :/
<hkaiser>
heller: did he tell you that it works?
<heller>
yes
<heller>
he is sitting right next to me at the moment
<hkaiser>
ahh
<hkaiser>
didn't know he works with you
<heller>
yeah
<heller>
we will get into trouble soon once we need to have HPX resilient to failures ;)
<hkaiser>
resilient to whatfailures?
<zao>
Malfunctioning nodes? Disappearing nodes?
<jbjnr>
what's the correct #ifdef for linux?
<hkaiser>
user's stupidity?
<jbjnr>
(detecting it)
<hkaiser>
grep for --linux__
<hkaiser>
__linux__
<jbjnr>
ta
<heller>
hkaiser: failing processes
<heller>
disappearing nodes, more or less
<hkaiser>
failing localities?
<heller>
yes
<hkaiser>
that requires to make AGAS redundant
<heller>
yup
<hkaiser>
good luck
<hkaiser>
requires ensuing consistency
<hkaiser>
ensuring, even
<heller>
yup
<heller>
as I said, this will cause trouble ;)
<hkaiser>
heller: please discuss things before implementing
Rodario1 has joined #ste||ar
<hkaiser>
heller: has ther ebeen some conceptual discussion f what needs to be implemented already?
Rodario has quit [Ping timeout: 255 seconds]
<heller>
hkaiser: not on the HPX level
<hkaiser>
heller: ok, pls keep me in the loop
<heller>
sure
<heller>
hkaiser: still about half a year or so to get there
<hkaiser>
anyways
<heller>
just keeping you in the loop :P
<hkaiser>
thanks
<github>
[hpx] biddisco created fixing_2862 (+1 new commit): https://git.io/v5s59
<github>
hpx/fixing_2862 7788fdd John Biddiscombe: Only use sched_getcpu on linux machines
<jbjnr>
oops
<github>
[hpx] biddisco force-pushed fixing_2862 from 7788fdd to 6a34d9b: https://git.io/v5s5N
<github>
hpx/fixing_2862 6a34d9b John Biddiscombe: Only use sched_getcpu on linux machines
<github>
[hpx] biddisco opened pull request #2865: Only use sched_getcpu on linux machines (master...fixing_2862) https://git.io/v5sdv
<zao>
Five failed tests on master, two of which are timeouts. Nice.
<hkaiser>
zao: unhandled_exception_582 is a new one, the two thread_pool_os_executors tests are expected
<hkaiser>
path_elements is unexpected as well - need to look - I probably screwed up while adding the pool related counters
mcopik has quit [Ping timeout: 240 seconds]
<hkaiser>
overall not too bad after such a significant change, though
<zao>
debian 9, gcc 6.3.0
<hkaiser>
thanks
<zao>
relwithdebinfo build, btw.
eschnett has quit [Quit: eschnett]
<hkaiser>
ok
Rodario has joined #ste||ar
Matombo has quit [Ping timeout: 276 seconds]
Rodario1 has quit [Ping timeout: 246 seconds]
K-ballo has joined #ste||ar
bikineev has quit [Ping timeout: 264 seconds]
Rodario1 has joined #ste||ar
Rodario has quit [Ping timeout: 248 seconds]
bikineev has joined #ste||ar
denis_blank has joined #ste||ar
hkaiser has quit [Quit: bye]
eschnett has joined #ste||ar
Matombo has joined #ste||ar
diehlpk_work has joined #ste||ar
<diehlpk_work>
taeguk[m], Your final evaluation is still misisng and you forgot to send the abstract before the deadline!
aserio has joined #ste||ar
Rodario has joined #ste||ar
Rodario1 has quit [Ping timeout: 240 seconds]
<david_pfander>
heller: Do you want to join us in skype?
Rodario1 has joined #ste||ar
Rodario has quit [Ping timeout: 255 seconds]
<jbjnr>
heller: we are very disappointed in you! First you gib out of the hackathon, and now you're going to not even do skype. I've made David team captain.
<jbjnr>
(I've asked for an extra laptop for your face to be on all week.)
<jbjnr>
(and if you're not there, then I'll ask for a dartboard and we'll put it up in the min hackathon room with your face on it for everyone to take shots at.)
<zao>
jbjnr: Glue it onto a rubber ducky, get some debugging help.
<jbjnr>
zao: I don't understand - won't a rubber ducky puncture and deflate?