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/ | GSoC2018: https://wp.me/p4pxJf-k1
EverYoung has joined #ste||ar
diehlpk has joined #ste||ar
EverYoun_ has quit [Ping timeout: 276 seconds]
hkaiser has joined #ste||ar
EverYoung has quit [Ping timeout: 260 seconds]
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
eschnett has joined #ste||ar
EverYoung has quit [Ping timeout: 240 seconds]
nikunj has joined #ste||ar
eschnett has quit [Quit: eschnett]
EverYoung has joined #ste||ar
parsa has joined #ste||ar
EverYoung has quit [Ping timeout: 240 seconds]
_gedaj_ has quit [Read error: Connection reset by peer]
__gedaj__ has joined #ste||ar
parsa has quit [Quit: Zzzzzzzzzzzz]
diehlpk has quit [Remote host closed the connection]
diehlpk has joined #ste||ar
K-ballo has quit [Quit: K-ballo]
nikunj has quit [Ping timeout: 260 seconds]
diehlpk has quit [Ping timeout: 240 seconds]
EverYoung has joined #ste||ar
EverYoung has quit [Ping timeout: 276 seconds]
nanashi55 has quit [Ping timeout: 264 seconds]
nanashi55 has joined #ste||ar
V|r has quit [Ping timeout: 265 seconds]
parsa has joined #ste||ar
V|r has joined #ste||ar
V|r has quit [Changing host]
V|r has joined #ste||ar
titzi has quit [Ping timeout: 265 seconds]
titzi has joined #ste||ar
parsa[[w]] has joined #ste||ar
bibek_ has joined #ste||ar
rod_ has joined #ste||ar
bibek has quit [Ping timeout: 276 seconds]
RostamLog_ has quit [Ping timeout: 265 seconds]
parsa[w] has quit [Ping timeout: 276 seconds]
RostamLog has joined #ste||ar
rtohid has quit [Ping timeout: 276 seconds]
jbjnr_ has joined #ste||ar
parsa has quit [Quit: Zzzzzzzzzzzz]
jbjnr__ has quit [Ping timeout: 255 seconds]
nikunj has joined #ste||ar
jbjnr_ has quit [Quit: Leaving]
jbjnr has joined #ste||ar
__gedaj__ has quit [Read error: Connection reset by peer]
__gedaj__ has joined #ste||ar
jaafar has quit [Ping timeout: 248 seconds]
nikunj has quit [Ping timeout: 260 seconds]
mcopik has joined #ste||ar
CaptainRubik has joined #ste||ar
CaptainRubik has quit [Client Quit]
RostamLog has quit [Ping timeout: 276 seconds]
RostamLog has joined #ste||ar
mcopik has quit [Ping timeout: 240 seconds]
mcopik has joined #ste||ar
heller has quit [Quit: http://quassel-irc.org - Chat comfortably. Anywhere.]
heller has joined #ste||ar
<jbjnr> yay \o/
<github> [hpx] StellarBot pushed 1 new commit to gh-pages: https://git.io/vpu11
<github> hpx/gh-pages 1b30af9 StellarBot: Updating docs
<heller> yay?
<jbjnr> heller: because today I received a letter from the university addressed to "Dr John" and officially conferring the title on me. All the paperwork has now been completed and it is official. Took months - but it's almost over - just the cerremony left now ....
<heller> yay!
<heller> good for you
<jbjnr> You're next!
<jbjnr> have you submitted yet?
<heller> don't want to talk about it :/
<jbjnr> <sigh>
<jbjnr> just submit and be done. you must have enough matreial in there already. You were 'almost' ready back in Sept!
<heller> I still am
<heller> just can't find the bloody time to finalize it
<jbjnr> make time - the joy of being free is worth it.
<jbjnr> I'm floating.
<jbjnr> There's no weight on my shoulders .....
<jbjnr> <flap flap flap>
<jbjnr> :)
<heller> /quit
<jbjnr> lol
<jbjnr> are you home or still swanning about the USA somewhere?
<heller> home
<jbjnr> job?
<heller> university
<heller> but yes
<zao> $ make defense install
<heller> rather: this = nullptr;
mcopik has quit [Ping timeout: 265 seconds]
mcopik has joined #ste||ar
eschnett has joined #ste||ar
K-ballo has joined #ste||ar
<hkaiser> jbjnr: congrats!
<jbjnr> thanks.
EverYoung has joined #ste||ar
<hkaiser> jbjnr: the deadline for P1017 is next Monday
<jbjnr> really? ok. I'd better get writing
<jbjnr> want a chat at all?
<hkaiser> jbjnr: sure, might be good
<jbjnr> now? later? tomorrow?
<hkaiser> later or tomorrow - not now
<jbjnr> tomorrow then. Pick a time. I'm, in all day
<hkaiser> what about 10 my time? would that be too late for you?
<hkaiser> 9:30 should work as well
<hkaiser> that's 4:30pm
<jbjnr> not sure what time that is for me. hold on....
<hkaiser> 4:30
EverYoung has quit [Ping timeout: 265 seconds]
<jbjnr> that's fine. Tomorrow 16:30 my time. ok
<hkaiser> jbjnr: ok
mcopik has quit [Ping timeout: 256 seconds]
mcopik has joined #ste||ar
hkaiser has quit [Ping timeout: 248 seconds]
hkaiser has joined #ste||ar
apsknight has joined #ste||ar
eschnett has quit [Quit: eschnett]
apsknight has left #ste||ar [#ste||ar]
aserio has joined #ste||ar
K-ballo has quit [Quit: K-ballo]
EverYoung has joined #ste||ar
EverYoung has quit [Client Quit]
K-ballo has joined #ste||ar
aserio1 has joined #ste||ar
aserio has quit [Ping timeout: 265 seconds]
aserio1 is now known as aserio
parsa has joined #ste||ar
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
eschnett has joined #ste||ar
parsa has quit [Quit: Zzzzzzzzzzzz]
eschnett has quit [Client Quit]
parsa has joined #ste||ar
aserio has quit [Ping timeout: 240 seconds]
<jbjnr> hkaiser: heller K-ballo do either of you know what changes wash is planning on making to futures in his proposal for iso c++ this june?
<K-ballo> I have not been following closely
<K-ballo> my impression so far is that they are still all over the place
<jbjnr> any idea what a/the fundamental change will be?
<jbjnr> break hpx much?
<K-ballo> they are still discussing interface related changes
<K-ballo> .then and such
<jbjnr> ok, but nothing fundamental to the synchronization activity then
<K-ballo> like what? what could that be?
aserio has joined #ste||ar
<jbjnr> no idea, but CJ told me today that big changes were afoot from Bryce that would change futures completely. That's not to say he knows what we do in hpx or anything, but ... it got me wondering and hence my question
<K-ballo> it's quite likely that they change the entire `future` interface in a breaking way
<K-ballo> drop get, drop wait, redesign then, etc
<jbjnr> chaging the api is ok, as long as they don't change what they do
<jbjnr> drop get?
<jbjnr> hmmm.
<jbjnr> that might make it tricky to 'get' stuff from the future!
<K-ballo> it can be reimplemented on top of .then
<jbjnr> that's ok, as long as the future still passes the dat through
<jbjnr> nobody ever calls .get except in the continuation anyway. mostly.
parsa has quit [Quit: Zzzzzzzzzzzz]
eschnett has joined #ste||ar
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
diehlpk has joined #ste||ar
david_pfander has quit [Ping timeout: 276 seconds]
Anushi1998 has joined #ste||ar
__gedaj__ has quit [Remote host closed the connection]
_gedaj_ has joined #ste||ar
diehlpk has quit [Ping timeout: 264 seconds]
<K-ballo> hkaiser: there's been advancements in packing variadics in lambdas
<K-ballo> this isn't valid: [f = forward<F>(f), args = forward<Args>(args)...]() mutable
<K-ballo> but now this is: [f = forward<F>(f), ...args = forward<Args>(args)...]() mutable
<K-ballo> uhm, one sec
<K-ballo> iiuc it would be: [f = forward<F>(f), ...args = forward<Args>(args)]() mutable
aserio has quit [Ping timeout: 264 seconds]
<hkaiser> K-ballo: thanks for this info, wasn't aware of this
<zao> o.O
jaafar has joined #ste||ar
hkaiser has quit [Ping timeout: 255 seconds]
EverYoun_ has joined #ste||ar
EverYoun_ has quit [Ping timeout: 255 seconds]
EverYoung has quit [Ping timeout: 276 seconds]
parsa has joined #ste||ar
mcopik has quit [Ping timeout: 248 seconds]
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
Anushi1998 has quit [Remote host closed the connection]
parsa has quit [Quit: Zzzzzzzzzzzz]
parsa has joined #ste||ar
parsa has quit [Quit: Zzzzzzzzzzzz]
parsa has joined #ste||ar
parsa has quit [Client Quit]
nikunj has joined #ste||ar
parsa has joined #ste||ar
hkaiser has joined #ste||ar
eschnett has quit [Quit: eschnett]
aserio has joined #ste||ar
__gedaj__ has joined #ste||ar
_gedaj_ has quit [Remote host closed the connection]
__gedaj__ has quit [Remote host closed the connection]
__gedaj__ has joined #ste||ar
_gedaj_ has joined #ste||ar
__gedaj__ has quit [Ping timeout: 268 seconds]
<heller> jbjnr: it will mostly be a different API, IIUC. Making composition easier
<heller> at least that's the intention. I don't agree with dropping .get(). But people are offended by the shear presence of it...
<hkaiser> heller: there will be a global get(f) instead, which is equivalent
<heller> how should that work?
<hkaiser> in the simplest case using .then()
<heller> IMHO, the usecase of calling .get() (even if it's just once in an application) is common enough to deserve to keep it in
<heller> sure, but that's still a member function on a future, no?
<heller> is there a document to read up on this?
<hkaiser> what? .get()?
<hkaiser> it's Wip, though
<heller> yeah, the whole rationale
<hkaiser> I mostly don't like it myself, but there too much testosteron involved atm
<heller> auto r = std::this_thread::get(f2);
_gedaj_ has quit [Read error: Connection reset by peer]
__gedaj__ has joined #ste||ar
<heller> I understand that the intention is to let a blocking get stand out more to get a "don't do this" mark
<hkaiser> yes
<hkaiser> that's mostly it
<heller> what should be done instead is to let the asynchronous providers/consumer interoperate better, especially between execution contexts
<hkaiser> heller: that is the most important rationale behind the current design, at least that's what they claim
<heller> this should be the focus, not this stupid bike shed color
<heller> ok, at least something ;)
<hkaiser> heller: I agree - but as said - too much testosteron
<hkaiser> imagine Bryce, Michal, and David in the same room, I gave up after 30 minutes
<heller> lol
<heller> David Stone?
<hkaiser> David Hollman
<hkaiser> the blue haired guy
<heller> ah, never met him
<heller> currently pulling my hair over a strange deadlock of mine ... is there anything in future_data that could be potentially racy at the moment?
<hkaiser> heller: on master?
<heller> yes
<hkaiser> don't think so
<heller> yeah, my application works perfectly when running on one node, but hangs, deterministically on two
<heller> the sanitizer don't give me anything
EverYoun_ has joined #ste||ar
<heller> the hanging threads both point to the same parent (which doesn't seem to be alive anymore)
EverYoung has quit [Ping timeout: 265 seconds]
__gedaj__ has quit [Remote host closed the connection]
__gedaj__ has joined #ste||ar
aserio has quit [Quit: aserio]
<parsa> hkaiser: i don't think we can use circleci's docker executor for phylanx, or workflows
<parsa> there's no way i can cache the prerequisites docker image without deploying it first
<parsa> so i can't reuse it between steps
EverYoun_ has quit [Remote host closed the connection]
nanashi55 has quit [Ping timeout: 268 seconds]
nanashi55 has joined #ste||ar
EverYoung has joined #ste||ar
<heller> parsa: prebuild it
<heller> in the same way we do it for HPX
<parsa> heller: depends on HPX… that's the issue
<heller> so?
<heller> we can implement a hook such that it gets rebuild whenever a new HPX image gets pushed
<parsa> prebuild it when? is has to be built every time HPX's master is built
<heller> yes
nikunj has quit [Quit: Page closed]
<heller> do you have admin rights for dockerhub?
<parsa> i test with my own account
<parsa> who would trigger it? it's a successful hpx build that should trigger that call back
<heller> so, multiple options
<heller> the prereqs for phylanx, without HPX, won't change that often. That can be easily prebuilt
<parsa> yes
<heller> and then you just use multiple base images, which is perfectly fine
EverYoun_ has joined #ste||ar
<parsa> multiple base images are used to run services needed a stage. they won't merge into a single container. only the primary container is used for running commands
<parsa> :s/stage/job
<heller> docker-compose might be an option here
<parsa> triggering a build of phylanx's base image inside hpx's own CI workflow is the only option
EverYoung has quit [Ping timeout: 255 seconds]
<heller> which is a nogo, IMHO
<parsa> that's what i though
<parsa> t
<heller> so let's take a step back
<heller> what are the requirements we need for a phylanx docker image?
<heller> 1) Compile phylanx C++ applications?
<heller> 2) Run phylanx python code?
<heller> 3) Use the profiling tools?
<parsa> the requirement for a container is that it can be used to build and test phyla
<parsa> nx
<heller> for circleci
<heller> yes
<heller> which can use multiple base containers
<parsa> that's the one we're talking about here, i don't care for other images
<parsa> heller: "all steps are executed in the container created by the first image listed"
<heller> ok, that's for multiple instanced
<heller> s
<parsa> yes
<heller> so what we need is an automated step to merge latest HPX and the phylanx prereqs
<parsa> i don't get it the merging part merging images creates a new image anyway. wherever that happens we can just run a bunch of extra commands
<heller> I would be surprised if that isn't possible within dockerhub
<heller> sure
<heller> but let's suppose that phylanx changes more frequently than HPX
<heller> rebuilding the phylanx base image on every phylanx change is a waste of resources then
<parsa> that's a given not an assumption
<parsa> takes 1-2 minutes… doesn't matter
<heller> still
<heller> it happens on every PR, every merge
<parsa> Repository Links
EverYoun_ has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
<heller> parsa: yeah, something like that
EverYoung has quit [Remote host closed the connection]
<parsa> i've set things up…. to test it i need someone to push to hpx to see if it works
EverYoung has joined #ste||ar
EverYoung has quit [Ping timeout: 276 seconds]
jaafar has quit [Ping timeout: 240 seconds]
EverYoung has joined #ste||ar
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 265 seconds]
EverYoun_ has quit [Remote host closed the connection]
EverYoung has joined #ste||ar