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 joined #ste||ar
<khuck>
hkaiser: there?
<hkaiser>
khuck: here
<khuck>
that eval() failure with Clang 5 on POWER...
<hkaiser>
yah
<khuck>
it is stack related.
<hkaiser>
nod
<khuck>
I played with the definitions in eval.py, and if I change to fib(9), it's ok, but fib(10) = boom
<khuck>
but I don't know enough about the phylanx / hpx internals to figure out what's wrong beyond this
<khuck>
wait - should I hard code it or pass it in as an argument?
jaafar has joined #ste||ar
<hkaiser>
if you run python code you'd have to hard code it for now, no way to pass it dynamically
<khuck>
there might be a reasonable limit for clang/power8. something other than that 1/2 second threshold. Although for fibonacci... no limit is too low.
<khuck>
don't think that had an affect - stack is 820 deep in GDB
<hkaiser>
yah, then it didn't have effect
<khuck>
HOWEVER
<khuck>
I changed the default from 500000 to 0...and it worked
<khuck>
\o/
<hkaiser>
K-ballo: if you add a "--hpx:dump-config" there is should print the whole configuration database, letting you see whether your settings have been recognized
<hkaiser>
khuck: ^^
<hkaiser>
khuck: ahh, good
<hkaiser>
so it switched to async execution only
<khuck>
so...do we need a "depth" parameter as well?
<hkaiser>
khuck: no way to keep track of 'depth', currently
<khuck>
ok
<hkaiser>
khuck: but I think I know what to do
<khuck>
so...should I check for clang/power8 and disable direct actions in this way? And submit an issue, of course
jaafar has quit [Ping timeout: 240 seconds]
<hkaiser>
khuck: yah, might be the best for now
<khuck>
will do
<khuck>
I'd like the power8 tests to pass :)
<khuck>
(even if I have to rig them)
<hkaiser>
khuck: but those settings will affect the python tests only
<khuck>
that's the only test that is failing
<hkaiser>
do c++ tests fail as well?
<hkaiser>
ahh, cool
<khuck>
no, just this one python test
<khuck>
awesome. it's been over 2 months since a power8 succeeded
<hkaiser>
good job!
<khuck>
can't believe it took me this long to find it
<khuck>
I'll just update the current issue to explain the real problem. OK?
<hkaiser>
khuck: ok, feel free to add that setting for power only for now
<khuck>
done
<khuck>
need to test it with other platforms, too - then I'll do a pull request
<khuck>
(need to make sure I get the macros right)
<hkaiser>
thanks
<khuck>
no, thank you
<khuck>
by the way, was there a hwloc bug introduced in the last 24 hours, then fixed?
<khuck>
hkaiser: I saw some odd behavior with hwloc 1.11.5 yesterday, but then worked around it.
<hkaiser>
khuck: simbergm had some PR merged end of last week
<khuck>
no, this was more recent. no biggie.
jaafar has joined #ste||ar
anushi_ has quit [Quit: Bye]
jaafar has quit [Ping timeout: 252 seconds]
<khuck>
hkaiser: pr created, have a good night!
<khuck>
thanks again
<hkaiser>
g'night
<hkaiser>
thanks!
jaafar has joined #ste||ar
hkaiser has quit [Quit: bye]
khuck has quit [Remote host closed the connection]
jaafar has quit [Ping timeout: 252 seconds]
jaafar has joined #ste||ar
jaafar has quit [Ping timeout: 246 seconds]
jaafar has joined #ste||ar
jaafar has quit [Quit: Konversation terminated!]
jaafar has joined #ste||ar
jaafar has quit [Remote host closed the connection]
jaafar has joined #ste||ar
jaafar_ has joined #ste||ar
jaafar has quit [Remote host closed the connection]
Anushi1998 has joined #ste||ar
Anushi1998 has quit [Quit: Bye]
<zao>
You people have competition in the build domain... "I noticed that you need more than 100GB of memory when building Trilinos 12.12.1 on 20 cores"
anushi has quit [Remote host closed the connection]
anushi has joined #ste||ar
hkaiser has joined #ste||ar
anushi has quit [Remote host closed the connection]
anushi has joined #ste||ar
anushi has quit [Remote host closed the connection]
<simbergm>
hkaiser: yt?
<hkaiser>
simbergm: here
<simbergm>
would you have time for a short call later today?
<hkaiser>
simbergm: I can try, but it might be possible only later in the day :/
<hkaiser>
tomorrow is better
<simbergm>
we're planning internship projects and we were thinking about some map-reduce type of things with hpx, and figured you'd be a good person to talk about any possible existing work on that
<simbergm>
python and phylanx also came up
<simbergm>
ok, no worries
<simbergm>
what would later mean if still today?
<hkaiser>
evening for you 8/9pm
<hkaiser>
I'd like to talk about other things as well, so this is fitting
<simbergm>
all right, tomorrow is not so good for me so I'll check if this other guy would be free around that time tonight
<simbergm>
I'll get back to you soon
<hkaiser>
Friday?
<simbergm>
well, I go on holidays so in principle I have lots of time :P
<simbergm>
but can't promise anything
<hkaiser>
ok
<hkaiser>
let me kno
<hkaiser>
w
<simbergm>
sure
<simbergm>
if you do have time tonight I think that could work
<ste||ar-github>
hpx/fix_3439_2 190b5af John Biddiscombe: Merge branch 'master' into fix_3439_2
ste||ar-github has left #ste||ar [#ste||ar]
anushi has joined #ste||ar
anushi_ has quit [Ping timeout: 245 seconds]
anushi_ has joined #ste||ar
anushi has quit [Ping timeout: 252 seconds]
Anushi1998 has joined #ste||ar
Anushi1998 has quit [Remote host closed the connection]
hkaiser has joined #ste||ar
Anushi1998 has joined #ste||ar
jaafar_ has joined #ste||ar
<simbergm>
hkaiser: yt? so no need for a call tonight, we talked a bit more and concluded it's not so urgent anymore but I still want to pick your brain about phylanx at some point...
<simbergm>
do you want to have a call still before I come over there? I'm back from holidays on the 7th and fly to the us on the 10th
parsa[w] has quit [Read error: Connection reset by peer]
parsa[w] has joined #ste||ar
Anushi1998 has quit [Ping timeout: 272 seconds]
eschnett_ has quit [Quit: eschnett_]
_diers_ has quit [Ping timeout: 246 seconds]
eschnett_ has joined #ste||ar
eschnett_ has quit [Client Quit]
<hkaiser>
simbergm: should be fine
<hkaiser>
thanks
daissgr has quit [Ping timeout: 250 seconds]
nikunj has joined #ste||ar
eschnett_ has joined #ste||ar
Anushi1998 has joined #ste||ar
<simbergm>
hkaiser: all right, let's arrange it in the next few weeks if there's time then
<simbergm>
thanks and sorry for the false alarm
aserio has joined #ste||ar
Anushi1998 has quit [Remote host closed the connection]
<ste||ar-github>
hpx/fix_startup_hang_exception f52f6de Hartmut Kaiser: Avoid hang if there is an exception thrown during startup
ste||ar-github has left #ste||ar [#ste||ar]
ste||ar-github has joined #ste||ar
<ste||ar-github>
[hpx] hkaiser opened pull request #3468: Avoid hang if there is an exception thrown during startup (master...fix_startup_hang_exception) https://github.com/STEllAR-GROUP/hpx/pull/3468