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: 240 seconds]
katywilliams has joined #ste||ar
EverYoung has joined #ste||ar
EverYoung has quit [Ping timeout: 240 seconds]
jakub_golinowski has quit [Quit: Ex-Chat]
diehlpk has joined #ste||ar
prashantjha has quit [Read error: Connection reset by peer]
ct_clmsn has joined #ste||ar
mcopik_ has quit [Ping timeout: 268 seconds]
Smasher has quit [Remote host closed the connection]
jaafar has quit [Ping timeout: 240 seconds]
daissgr has quit [Ping timeout: 240 seconds]
katywilliams has quit [Ping timeout: 263 seconds]
vamatya has joined #ste||ar
diehlpk has quit [Ping timeout: 240 seconds]
K-ballo has quit [Quit: K-ballo]
jaafar has joined #ste||ar
jaafar has quit [Ping timeout: 256 seconds]
jaafar has joined #ste||ar
jaafar has quit [Ping timeout: 263 seconds]
jaafar has joined #ste||ar
jaafar has quit [Ping timeout: 256 seconds]
ct_clmsn has quit [Ping timeout: 260 seconds]
parsa has quit [Quit: Zzzzzzzzzzzz]
hkaiser has quit [Quit: bye]
jaafar has joined #ste||ar
nanashi55 has quit [Ping timeout: 264 seconds]
nanashi55 has joined #ste||ar
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 264 seconds]
Anushi1998 has joined #ste||ar
vamatya has quit [Ping timeout: 256 seconds]
jaafar has quit [Ping timeout: 240 seconds]
Anushi1998 has quit [Quit: Leaving]
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 240 seconds]
verganz has joined #ste||ar
verganz has quit [Ping timeout: 260 seconds]
Guest44028 has joined #ste||ar
Guest44028 is now known as V||r
V||r has left #ste||ar ["Konversation terminated!"]
david_pfander has joined #ste||ar
<github>
[hpx] msimberg pushed 1 new commit to msimberg-patch-1: https://git.io/vxsfw
<github>
hpx/msimberg-patch-1 d6a2555 Mikael Simberg: Remove trailing whitespace
<github>
hpx/master f1a3454 Mikael Simberg: Merge pull request #3242 from STEllAR-GROUP/msimberg-patch-1...
<github>
[hpx] msimberg deleted msimberg-patch-1 at d6a2555: https://git.io/vxs8V
<simbergm>
hkaiser, heller, jbjnr: I'm planning on making an RC today or tomorrow if all goes well on rostam (I think I will still merge 3186 as well)
<heller>
simbergm: sounds good to me
<simbergm>
any problems with that?
<hkaiser>
simbergm: nice
<hkaiser>
simbergm: let me know if I can do anything to help
<K-ballo>
\o/
<simbergm>
hkaiser: thanks, I'll probably need your help with the later steps (where to upload files, blog post etc) but I'm hoping I'll manage to tag the rc without problems first ;)
<simbergm>
btw, K-ballo: why make? because ninja seems to be messed up...? see #3036
<K-ballo>
runs out of memory?
<simbergm>
yeah, looks like it somehow manages to do that
victor_ludorum has quit [Ping timeout: 260 seconds]
nikunj has quit [Client Quit]
nikunj has joined #ste||ar
<nikunj>
@hkaiser: does hpx::finalize() de-initializes global objects as well?
eschnett has joined #ste||ar
<hkaiser>
hpx::finalize tells the runtime to exit which should make the application exit and the global objects being destructed
<hkaiser>
the other example works that way, doesn't it?
<nikunj>
yes it did
<nikunj>
but then I thought, I might have to add some initialization sequences myself
<nikunj>
so that might hinder with the working of hpx::finalize()
<nikunj>
That's why I implemented it that way
<hkaiser>
finalize is already called from the global object, isn't it?
<nikunj>
oh yes! Sorry I forgot.
<nikunj>
I'll make changes to the code
victor_ludorum has joined #ste||ar
<nikunj>
@hkaiser: one important thing I found while investigating for hpx runtime system. All commandline options outputs on the screen after hpx::init() is called (Correct me if I'm wrong)
<nikunj>
So calling commandline information regarding hpx:info does not work if you don't pass in the arguments into init.
<hkaiser>
nikunj: yes, that's correct
<nikunj>
Is this implemented this way, or is it an issue?
<hkaiser>
we could probably work around that in a similar way as __argc/__arv are initialized for the init_globally example
<nikunj>
@hkaiser: yes, but in current systems it doesn't show outputs related to hpx:info, hpx:version and type
<hkaiser>
nikunj: not sure what you mean by 'All commandline options outputs on the screen after hpx::init() is called'
<nikunj>
Oh, let me share a gist of what I mean. That might clear up what I mean
<nikunj>
@hkaiser: try running --hpx:info or --hpx:version or similar info type commandline options
<nikunj>
You will not get any output for them
<nikunj>
@hkaiser: by no output, I mean you will just get Hello Worl
<nikunj>
*Hello World
diehlpk_work has joined #ste||ar
<zao>
I wonder if I made life worse for nikunj with my FreeBSD init stuff :)
<hkaiser>
nikunj: yes, as I said, this is a known issue - we could work around similarily to what's done with __argc/__argv in init_globally
<hkaiser>
nikunj: feel free to modify the existing code for hpx::init/hpx::start
<nikunj>
@hkaiser: ok, I'll look into the code for it. Also it shouldn't be a problem for init_globally since it will call init with __argc, __argv
<hkaiser>
yes
<hkaiser>
we also could simplify init_globally if we changed hpx::init to use __argc/__argv whenever needed
<zao>
If the FBSD stuff gets in the way, I'd recommend ripping it out.
<zao>
I don't think we have any actual users for it.
<victor_ludorum>
@hkaiser Hello Sir!! Sorry to interruption , I just want to know that is there any unit test file for new performance counters .
<hkaiser>
zao: I don't thin kthat will be a problem
<hkaiser>
victor_ludorum: what 'new' counters?
<victor_ludorum>
Actually, Sir as I am planning to implement new arithemetic and statistics performance counter. So for its testing
<victor_ludorum>
I need to know
<victor_ludorum>
How can I test them ?
<hkaiser>
victor_ludorum: create a test?
<victor_ludorum>
hkaiser: Yeah sir!! I have to test. Is there any previous file where such testing is done for performance counters
<victor_ludorum>
so that I can take help
<nikunj>
@hkaiser: ok, so I'll check the code and try to work my way out. Also, I have found a few initialization sequence issues in the code. Should I share it here or through email?
<hkaiser>
victor_ludorum: there are some tests, yes - see tests/unit/performance_counters
<hkaiser>
nikunj: sharing things here is fine
<victor_ludorum>
Oh.. Okay Sir thanks!!
<nikunj>
@hkaiser: ok so I'll share it here.
<nikunj>
@hkaiser: so first I went ahead and checked for sequencing issues in commandline options
<nikunj>
I found that hpx options (such as --hpx:info, --hpx:version) gave initialization sequence issues
<nikunj>
also the performance counters gave such issues as well
<hkaiser>
nikunj: what init-sequencing issues?
<nikunj>
@hkaiser: I don't remember exactly and currently I'm not using my laptop so I can't exactly tell
<nikunj>
They were all segmentation faults though
<nikunj>
as far as I can remember
<nikunj>
I didn't find any parsing issues ( or maybe there was one ). I could tell you exact issues with them once I get hold of my laptop ( should take another 1-1.5h)
<hkaiser>
how do you plan to work over the summer if you aren't even able to find a type definition?
<hkaiser>
type definition*
<victor_ludorum>
@hkaiser: Sorry Sir :(
nikunj has quit [Ping timeout: 260 seconds]
irlE60L58 has joined #ste||ar
irlE60L58 has quit [Client Quit]
daissgr has joined #ste||ar
EverYoung has joined #ste||ar
gablab has quit [Quit: Leaving]
EverYoung has quit [Ping timeout: 245 seconds]
victor_ludorum has quit [Quit: Page closed]
nikunj has joined #ste||ar
<verganz>
I'll duplicate question that I yesterday: In LibDeoDecomp coordinate vectors used to be a Coord<DIM> objects, is there some objects, that implements DIM*DIM matrix, I didn't found in doc, but maybe it have some specific name
<verganz>
Surely in some projects they were necessary, or for such needs we should create custom classes?
katywilliams has joined #ste||ar
Anushi1998 has joined #ste||ar
<nikunj>
@hkaiser: sorry for the delay. Here is the error links for all the sequence error I found.