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/
nanashi55 has joined #ste||ar
nanashi55 has quit [Ping timeout: 256 seconds]
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
nanashi55 has joined #ste||ar
EverYoung has quit [Ping timeout: 276 seconds]
EverYoung has joined #ste||ar
EverYoung has quit [Client Quit]
ct-clmsn has joined #ste||ar
EverYoung has joined #ste||ar
kisaacs has quit [Ping timeout: 246 seconds]
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
ct-clmsn has quit [Read error: Connection reset by peer]
ct-clmsn has joined #ste||ar
ct-clmsn is now known as Guest16906
EverYoung has quit [Ping timeout: 240 seconds]
Guest16906 has quit [Quit: Leaving]
kisaacs has joined #ste||ar
hkaiser has quit [Quit: bye]
parsa has joined #ste||ar
diehlpk has joined #ste||ar
eschnett has joined #ste||ar
kisaacs has quit [Ping timeout: 265 seconds]
kisaacs has joined #ste||ar
parsa has quit [Quit: Zzzzzzzzzzzz]
katywilliams has joined #ste||ar
diehlpk has quit [Ping timeout: 260 seconds]
eschnett has quit [Quit: eschnett]
K-ballo has quit [Quit: K-ballo]
kisaacs has quit [Ping timeout: 256 seconds]
kisaacs has joined #ste||ar
katywilliams has quit [Ping timeout: 256 seconds]
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 256 seconds]
katywilliams has joined #ste||ar
nanashi55 has quit [Ping timeout: 246 seconds]
nanashi55 has joined #ste||ar
nikunj_ has joined #ste||ar
katywilliams has quit [Ping timeout: 248 seconds]
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 256 seconds]
daissgr has quit [Ping timeout: 240 seconds]
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 248 seconds]
kisaacs has quit [Ping timeout: 256 seconds]
nikunj_ has quit [Ping timeout: 260 seconds]
Anushi1998 has joined #ste||ar
anushi has quit [Ping timeout: 240 seconds]
jaafar has quit [Ping timeout: 256 seconds]
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 260 seconds]
katywilliams has joined #ste||ar
Anushi1998 has quit [Ping timeout: 245 seconds]
katywilliams has quit [Ping timeout: 260 seconds]
kisaacs has joined #ste||ar
anushi has joined #ste||ar
kisaacs has quit [Ping timeout: 246 seconds]
<jbjnr>
heller: yt?
<github>
[hpx] biddisco force-pushed remove_wait_or_add_new from 098b544 to 8444608: https://git.io/vxWVe
<github>
hpx/remove_wait_or_add_new 5788edb Thomas Heller: Changing the coroutine implementations to do a lazy init...
<github>
hpx/remove_wait_or_add_new 8444608 John Biddiscombe: Remove wait_or_add_new from scheduling loop, thread_queue and schedulers
<github>
hpx/remove_wait_or_add_new 33e340d John Biddiscombe: Remove staged queue from thread map and run_now param from create_thread api...
<github>
[hpx] biddisco force-pushed remove_wait_or_add_new from 8444608 to 5e13a67: https://git.io/vxWVe
<github>
hpx/remove_wait_or_add_new 5e13a67 John Biddiscombe: Remove wait_or_add_new from scheduling loop, thread_queue and schedulers
<github>
[hpx] biddisco force-pushed lazy_thread_init from fe5b5f1 to 5788edb: https://git.io/vAeeA
<jbjnr>
simbergm: warning - I just rebased those two branches - remove_wait_or_add_new and lazy_thread_init onto latest master (that includes the scheduler removal)
<jbjnr>
I'm glad those schedulers have gone. thanks for that
Anushi1998 has joined #ste||ar
<jbjnr>
tests.unit.parallel.executors.executor_parameters still not working for me - I thought that was fixed ...
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 263 seconds]
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 263 seconds]
Anushi1998 has quit [*.net *.split]
anushi has quit [*.net *.split]
jbjnr has quit [*.net *.split]
Anushi1998 has joined #ste||ar
anushi_ has joined #ste||ar
jakub_golinowski has joined #ste||ar
jbjnr has joined #ste||ar
<jbjnr>
anyone in?
Anushi1998 is now known as anushi
anushi_ is now known as Anushi1998
<simbergm>
hey jbjnr, thanks for the heads up on your branch
<simbergm>
I didn't fix executor_parameters, it just didn't fail for me
<simbergm>
but will try again now with the remove-schedulers branch merged
kisaacs has joined #ste||ar
Anushi1998 has quit [Quit: Leaving]
<jbjnr>
simbergm: I ran tests on the remove_wait_or_Add_new rebased branch and the executor params was the only one that failed for me (I actually didn't run all tests, but pycicle will soon)
<simbergm>
yep, there were some build failures, but it looks like they might just be some run_now arguments that sneaked back in
<github>
[hpx] msimberg force-pushed remove_wait_or_add_new from c132c82 to fa00ffe: https://git.io/vxWVe
<github>
hpx/remove_wait_or_add_new fa00ffe Mikael Simberg: Remove a few of run_now arguments to register_thread*
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 248 seconds]
<github>
[hpx] msimberg pushed 1 new commit to remove_wait_or_add_new: https://git.io/vxWpj
<github>
hpx/remove_wait_or_add_new c6011bf Mikael Simberg: Small change in scheduling_loop when not running
Smasher has joined #ste||ar
Anushi1998 has joined #ste||ar
hkaiser has quit [Quit: bye]
katywilliams has joined #ste||ar
Anushi1998 has quit [Quit: Leaving]
katywilliams has quit [Ping timeout: 260 seconds]
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 264 seconds]
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 264 seconds]
katywilliams has joined #ste||ar
kisaacs has joined #ste||ar
katywilliams has quit [Ping timeout: 256 seconds]
kisaacs has quit [Ping timeout: 265 seconds]
aserio has joined #ste||ar
diehlpk has joined #ste||ar
<jbjnr>
apologies simbergm when I pushed the wait_or_add_new, I missed a couple of commits from another branch that you have now fixed again.
<jbjnr>
I have too many branches. and merges of them. awful
<simbergm>
jbjnr: no worries! I'm removing everything related to staged tasks now (so far I haven't found a place where they would still be needed)
<simbergm>
I hope you're not doing the same right now...
<simbergm>
and I still can't get executor_parameters to fail, neither can pycicle it seems...
<simbergm>
only reinit counters
Anushi1998 has joined #ste||ar
eschnett has joined #ste||ar
kisaacs has joined #ste||ar
hkaiser has joined #ste||ar
diehlpk has quit [Ping timeout: 276 seconds]
<jbjnr>
simbergm: great. no. I'm trying to merge my stuff with the rebased branches. Really annyed by some bastard adding a fallback thread param to the scheduler
<jbjnr>
screws up my schedule hint stuff. millions of conflicts now
<jbjnr>
no ide why it's there either
<simbergm>
jbjnr: sorry... it's there to deal with thread suspension, otherwise it's not used
<jbjnr>
please explain what it is for
<simbergm>
if you have suspended threads (elasticity enabled) I take locks to make sure you don't schedule work on suspended threads (or ones about to be suspended)
<simbergm>
the fallback says: I couldn't take the locks so just use the fallback instead of looping for ages
<simbergm>
and I set it to thread_num when calling schedule_thread from inside the scheduling_loop (the scheduling_loop itself can schedule work on itself, it's not going to suspend while doing that)
<simbergm>
I'll happily change it to something cleaner
<diehlpk_work>
Compilation on my local machine took real0m0.461s and on carson real 9m16.163s
<jbjnr>
I don't like the fact that you're adding a new parameter to a function that I'm trying to remove stuff from and that 99% of the time it isn't used
<zao>
diehlpk_work: Ah, didn't see you were building some third party code, I was talking about HPX itself for me.
<jbjnr>
anyway, it's merged now, so I have to work with it
<jbjnr>
K-ballo: surely the old N version are older than the new Pversions?
<K-ballo>
the old N version is the actual TS, the new P-versions are proposals to make changes to the next TS
aserio1 has joined #ste||ar
<jbjnr>
ok
<K-ballo>
there should be some new N version soon with maybe some of those changes integrated
aserio1 has quit [Client Quit]
<diehlpk_work>
zao, Yes, I am working to improve the performance of blaze with HPX
<jbjnr>
except that N4723 is the newer version of 4680 ....
<jbjnr>
(apparently)
eschnett_ has joined #ste||ar
<K-ballo>
that's the new draft for the next TS
aserio has quit [Ping timeout: 276 seconds]
eschnett has quit [Ping timeout: 264 seconds]
<diehlpk_work>
hkaiser, return 4 * hpx::get_os_thread_count(); Is there any reason why you multiply the thrad count by 4 in blaze?
<diehlpk_work>
zao, The slow fs depends on the node. On carson it is slow, but on reno it is normal
<jbjnr>
diehlpk_work: because you want to create a few more tasks than there ar cores, but not too many - 4* is the default multiplier used in the parallel algorithms when generating tasks
<diehlpk_work>
Ok, thanks
katywilliams has joined #ste||ar
eschnett_ has quit [Quit: eschnett_]
aserio has joined #ste||ar
katywilliams has quit [Ping timeout: 248 seconds]
katywilliams has joined #ste||ar
<jbjnr>
K-ballo: the doc just contains the edits to the standard. I was hoping for something more readable that outlines usage and examples etc
katywilliams has quit [Ping timeout: 240 seconds]
<K-ballo>
jbjnr: if you want to restrict yourself to papers you could try the earliest papers of the most recent design
<K-ballo>
however a blog article or one of gor's talks would likely be much more informative
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 264 seconds]
<jbjnr>
K-ballo: good point. I'll have a look
aserio has quit [Remote host closed the connection]
aserio has joined #ste||ar
Anushi1998 has joined #ste||ar
anushi_ has joined #ste||ar
Anushi1998 has quit [Ping timeout: 276 seconds]
victor_ludorum has quit [Ping timeout: 260 seconds]
Anushi1998 has joined #ste||ar
<simbergm>
hkaiser: would you be around tomorrow/saturday to upload the release packages to the website?
<simbergm>
or can I do it somehow? it looks like I can't do it via wordpress, right?
EverYoung has joined #ste||ar
anushi_ has quit [Ping timeout: 276 seconds]
<jbjnr>
simbergm: what do you normally set this fallthrough value to? can you point me to somewhere it is passed in with a real value that's not the same as the num thread. (I have a problem with this because I have removed the thread number passing in)
<simbergm>
jbjnr: it's -1 if you don't specify it, and I only set it to thread_num explicitly in the scheduling_loop
<simbergm>
(I never set it to something else, which I admit is not very nice...)
jbjnr_ has joined #ste||ar
<jbjnr_>
why do I keep getting disconnected ...
<simbergm>
you can try removing it and see if the tests pass, I couldn't back when I added it
<hkaiser>
simbergm: I am around, yes
jbjnr has quit [Ping timeout: 252 seconds]
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
jaafar has joined #ste||ar
parsa[[w]] has joined #ste||ar
parsa[w] has quit [Ping timeout: 276 seconds]
EverYoung has quit [Remote host closed the connection]
kisaacs has quit [Ping timeout: 246 seconds]
daissgr has joined #ste||ar
kisaacs has joined #ste||ar
Anushi1998 has quit [Remote host closed the connection]
david_pfander has quit [Ping timeout: 256 seconds]
<simbergm>
jbjnr_: you removed passing the thread number because don't do round robin scheduling?
Anushi1998 has joined #ste||ar
parsa| has joined #ste||ar
aserio has quit [Ping timeout: 265 seconds]
parsa has quit [Ping timeout: 276 seconds]
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
<jbjnr_>
simbergm: no. I moved it into the thread_data beause the guided executor does things slightly differntly
Anushi1998 has quit [Remote host closed the connection]
Anushi1998 has joined #ste||ar
kisaacs has quit [Ping timeout: 240 seconds]
aserio has joined #ste||ar
CaptainRubik has joined #ste||ar
katywilliams has joined #ste||ar
kisaacs has joined #ste||ar
<github>
[hpx] biddisco force-pushed guided_pool_executor from 07f6ab1 to 349ca7b: https://git.io/vxkTv
<github>
hpx/guided_pool_executor 1f32ab1 John Biddiscombe: Move type info into hpx::debug namespace and add print helper functions
<github>
hpx/guided_pool_executor ce5fd38 John Biddiscombe: Add new guided_pool_executor that passes thread_schedule_hint to the scheduler...
<github>
hpx/guided_pool_executor dbba495 John Biddiscombe: Add overload for pool_numa_hint that can be used with lambdas
<Anushi1998>
Is there any documentation available to demonstrate examples on credits system for AGAS or the only way is to go through codebase?
hkaiser has quit [Quit: bye]
parsa| has quit [Quit: Zzzzzzzzzzzz]
parsa has joined #ste||ar
aserio1 has joined #ste||ar
aserio has quit [Ping timeout: 276 seconds]
aserio1 is now known as aserio
CaptainRubik has quit [Ping timeout: 260 seconds]
parsa has quit [Quit: Zzzzzzzzzzzz]
katywilliams has quit [Ping timeout: 256 seconds]
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 276 seconds]
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 260 seconds]
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 264 seconds]
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 268 seconds]
EverYoun_ has quit [Remote host closed the connection]
aserio has quit [Ping timeout: 276 seconds]
EverYoung has joined #ste||ar
kisaacs has quit [Quit: Lost terminal]
aserio has joined #ste||ar
parsa[[w]] is now known as parsa[w]
<aserio>
Anushi1998: I think you will have to brave the code
<aserio>
You could try googling it first though
<Anushi1998>
aserio: Okay,thanks
<Anushi1998>
yes I tried
<Anushi1998>
but I didn't get something useful
<aserio>
Template magic?
<Anushi1998>
What you meant?
<aserio>
Was the code hard to understand due to template magic?
<zao>
jbjnr_: I like how you cleverly sneak away from all GSoC responsibility by directing all traffic to the list :P
<Anushi1998>
No actually the logic behind specifically the case when we have to copy it to another locality and hence add it to AGAS (since it is now global)
<Anushi1998>
*(since it now need to be handled globally)
<aserio>
You mean when you instantiate a component?
<aserio>
or are you talking about migration?
<Anushi1998>
yes then we have to add that object to global object list
<Anushi1998>
no instantiate
<aserio>
ah, incrementing the AGAS counter
<Anushi1998>
yeah
<aserio>
You can probably find a test which check's its functionality
<aserio>
from there you could step through the code
<Anushi1998>
Hmm, that's a good idea :)
mcopik has joined #ste||ar
<jbjnr_>
zao: it's a genuine plea - I've had dozens of emails over the last few weeks asking about opencv, concurrent stuff, task plots, all-to-all and I can't guarantee I'll manage whilst I'm away on travel
<zao>
Hehe.
<jbjnr_>
^last week I mean
aserio has quit [Quit: aserio]
katywilliams has joined #ste||ar
katywilliams has quit [Ping timeout: 276 seconds]
<github>
[hpx] khuck created apex_refactoring_with_guids (+2 new commits): https://git.io/vx8qf
<github>
hpx/apex_refactoring_with_guids 3b22ae9 Kevin Huck: Adding parent dependencies for hpx threads.
<github>
hpx/apex_refactoring_with_guids ff4f604 Kevin Huck: Merge branch 'master' of https://github.com/STEllAR-GROUP/hpx into apex_refactoring_with_guids
diehlpk has joined #ste||ar
<github>
[hpx] khuck opened pull request #3250: Apex refactoring with guids (master...apex_refactoring_with_guids) https://git.io/vx8qG
katywilliams has joined #ste||ar
diehlpk has quit [Ping timeout: 264 seconds]
katywilliams has quit [Ping timeout: 240 seconds]
katywilliams has joined #ste||ar
<zao>
I like `tests.unit.lcos.local_dataflow_executor`. It completes in around 0.09 sec, so I can run thousands of it.
katywilliams has quit [Ping timeout: 264 seconds]
hkaiser has joined #ste||ar
Smasher has quit [Remote host closed the connection]
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 240 seconds]
EverYoun_ has quit [Ping timeout: 276 seconds]
<parsa[w]>
hkaiser: make_vector is the only test that fails. i don't know where it's coming from... expression_tree.eval(self.__src__, self.cs) is called and i don't see anything suspicious in the execution_tree.cpp
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
parsa has joined #ste||ar
<hkaiser>
parsa[w]: ok
<hkaiser>
I can have a look
<parsa>
thanks
katywilliams has joined #ste||ar
<zao>
It's a lovely test, fails every few hundred or thousand runs.
<zao>
I wonder how many files I can create in a directory...