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/
parsa has quit [Quit: Zzzzzzzzzzzz]
jaafar_ has quit [Ping timeout: 260 seconds]
<K-ballo>
ugh, pre17 std::regex doesn't match beginning/end of line
<K-ballo>
I'll have to modify the include related inspects to operate per line
kisaacs has quit [Ping timeout: 260 seconds]
diehlpk has joined #ste||ar
<diehlpk>
I do like that the parts of the hpx algorithms I looked at were not in standardese style which is hard to read(for good reasons)
<diehlpk>
This we got as feedback from one guy looking at hpx
<diehlpk>
hkaiser,
<K-ballo>
someone on slack reports hpx::sort crashing on a vector of integers
<K-ballo>
diehlpk: what did he mean by standardese style? the specification documentation? the uglified reserved names? other?
<diehlpk>
K-ballo, One guy I meet a the tlak of Bjarne sent me this
<diehlpk>
He started ti implement parallel algortihms by jis one and i recommend him to use hpx
<diehlpk>
Have to ask him what he means precisely
EverYoung has quit [Ping timeout: 255 seconds]
<hkaiser>
K-ballo: what channel?
<K-ballo>
#general
<K-ballo>
turns out is diehlpk's guy
<hkaiser>
diehlpk: pls ask him to provide a test case if he run sinto problems
<K-ballo>
turns out he called it outside of an HPX thread
<hkaiser>
nod
<hkaiser>
I need to improve diagnostics for this case
hkaiser has quit [Quit: bye]
K-ballo has quit [Quit: K-ballo]
eschnett has joined #ste||ar
denis_blank has quit [Quit: denis_blank]
diehlpk has quit [Remote host closed the connection]
vamatya_ has joined #ste||ar
parsa has joined #ste||ar
pree has joined #ste||ar
vamatya_ has quit [Ping timeout: 240 seconds]
jaafar_ has joined #ste||ar
quaz0r has quit [Ping timeout: 240 seconds]
quaz0r has joined #ste||ar
jaafar_ has quit [Ping timeout: 258 seconds]
<heller>
jbjnr: already there?
jaafar_ has joined #ste||ar
jaafar_ has quit [Ping timeout: 246 seconds]
<github>
[hpx] biddisco force-pushed fixing_2924 from 65f3710 to a9f1ea0: https://git.io/vdBON
<github>
hpx/fixing_2924 a9f1ea0 John Biddiscombe: Add #include files needed to set _POSIX_VERSION for debug check...
<github>
hpx/master 954818a Adrian Serio: Merge pull request #2917 from STEllAR-GROUP/add_checkpoint...
<heller>
hkaiser: I got a hang on the failing tests now
<heller>
oh, he's not here
<heller>
hkaiser_: ^^
<hkaiser_>
heller: hey
<heller>
hkaiser_: I am just not entirely sure what to do about the attached thread pool executors, they are more or less obsolete now, aren't they?
<hkaiser_>
heller: it's a nullptr
<heller>
yes, and a bit more
<hkaiser_>
well, yes - the executors need to be rethought and changed/removed
<heller>
it's a nullptr and the termination check in the scheduling loop i introduced
<hkaiser_>
heller: yes
<heller>
I got the nullptr under control now
<hkaiser_>
the this_thread_executor is still useful
<heller>
it is
<heller>
that's the one I am fixing right now
<hkaiser_>
the attached_executors are obsolete/superseeded by Johns stuff
<heller>
right
<heller>
the non-attached ones not quite yet
<hkaiser_>
so replacing them by a typedef for compatibility, removing them later
<hkaiser_>
etc.
<heller>
a simple typedef won't cut it
<hkaiser_>
the rp changes have not been completely propagated into the executors yet
<heller>
sure
aserio has quit [Ping timeout: 258 seconds]
aserio has joined #ste||ar
<heller>
hkaiser_: I want to focus on getting back a green dashboard again first
<hkaiser_>
heller: sure
<heller>
hkaiser_: nice one on the blaze backend btw
eschnett has quit [Ping timeout: 240 seconds]
eschnett has joined #ste||ar
rod_t has left #ste||ar [#ste||ar]
<hkaiser_>
heller: Klaus says hello, btw
<heller>
did you meet him seattle?
<heller>
I guess I have to catch up with him and go to user group meeting with him
<hkaiser_>
yes
<hkaiser_>
he said something like that
<hkaiser_>
;)
<heller>
:P
<heller>
and his company does HPC ...
parsa has quit [Quit: Zzzzzzzzzzzz]
<hkaiser_>
right
<heller>
see pm
hkaiser_ has quit [Read error: Connection reset by peer]
parsa has joined #ste||ar
hkaiser has joined #ste||ar
pree has quit [Ping timeout: 248 seconds]
pree has joined #ste||ar
hkaiser_ has joined #ste||ar
hkaiser has quit [Ping timeout: 248 seconds]
<jbjnr>
hello world definitely has issues, apex craps out with it, I get extra text output on random runs and the help issue I submitted might be related. very odd.
<jbjnr>
look like it might be related to iostreams and messages coming after shutdown has begun
<jbjnr>
s/messages/threads/g
pree has quit [Ping timeout: 240 seconds]
EverYoun_ has joined #ste||ar
<heller>
jbjnr: not good
<heller>
jbjnr: so it might be a different bug then
<heller>
do the other examples show similar behavior?
<jbjnr>
not so far
<heller>
ok
<jbjnr>
we still have apex crashes, but differnt
<heller>
it might have been me that messed it up
<heller>
the thing with hello_world
<jbjnr>
what
<jbjnr>
?
<heller>
I changed the termination detection in the scheduling loop
<heller>
fixing it right now as I type
<jbjnr>
ok
<jbjnr>
that may solve many odd things we're seeing
EverYoung has quit [Ping timeout: 258 seconds]
<K-ballo>
removing executors sounds nice
<heller>
jbjnr: indeed
<jbjnr>
K-ballo: have you lost your mind?
<K-ballo>
heh, not all of them, the obsolete ones
<jbjnr>
ah, you just mean the dodgy ones they were talking about earlier
parsa has quit [Quit: Zzzzzzzzzzzz]
parsa has joined #ste||ar
pree has joined #ste||ar
mbremer has quit [Quit: Page closed]
hkaiser_ has quit [Read error: Connection reset by peer]
rod_t has joined #ste||ar
aserio has quit [Ping timeout: 255 seconds]
pree has quit [Quit: AaBbCc]
EverYoun_ has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
<github>
[hpx] sithhell created fix_this_thread_executor (+3 new commits): https://git.io/vdRMB
<github>
hpx/fix_this_thread_executor ac72555 Thomas Heller: Exposing a function to return the pool of the current HPX thread
<github>
hpx/fix_this_thread_executor 0fe50ba Thomas Heller: Fixing thread executors to use current pool as parent pool
<github>
hpx/fix_this_thread_executor bd35a2f Thomas Heller: Fixing termination detection of scheduling loop...
<heller>
jbjnr: just commited a potential fix
<heller>
jbjnr: see above
<heller>
all unit tests pass again
<jbjnr>
ok, will test thanks
aserio has joined #ste||ar
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 258 seconds]
aserio has quit [Quit: aserio]
EverYoun_ has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
<jbjnr>
heller: I think your patch has fixed the apex hello worl problem. not 100% certain but looks better
hkaiser has joined #ste||ar
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 258 seconds]
EverYoun_ has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
parsa has quit [Quit: Zzzzzzzzzzzz]
parsa has joined #ste||ar
parsa has quit [Ping timeout: 246 seconds]
eschnett has quit [Quit: eschnett]
EverYoun_ has joined #ste||ar
rod_t has left #ste||ar [#ste||ar]
EverYoung has quit [Ping timeout: 246 seconds]
EverYoun_ has quit [Ping timeout: 258 seconds]
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]