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/ | GSoD: https://developers.google.com/season-of-docs/
K-ballo has quit [Quit: K-ballo]
hkaiser has quit [Quit: bye]
JClave has joined #ste||ar
<JClave> Hi I'm looking to contribute to the github repo. Especially interested in parallel algorithms. Anyone familiar with who I need to talk to before picking up items from the Github tickets? Looking to pick an item from https://github.com/STEllAR-GROUP/hpx/projects/1
mdiers_1 has joined #ste||ar
mdiers_ has quit [Remote host closed the connection]
mdiers_1 is now known as mdiers_
zao has quit [Ping timeout: 272 seconds]
zao has joined #ste||ar
JClave has quit [Ping timeout: 256 seconds]
daissgr5 has quit [Remote host closed the connection]
quaz0r has quit [Ping timeout: 248 seconds]
auriane has joined #ste||ar
quaz0r has joined #ste||ar
auriane has quit [Quit: WeeChat 1.9.1]
auriane[m] has joined #ste||ar
heller has quit [Quit: http://quassel-irc.org - Chat comfortably. Anywhere.]
heller has joined #ste||ar
hkaiser has joined #ste||ar
<hkaiser> simbergm: hey
<simbergm> hkaiser: hiho
<hkaiser> hey, welcome back!
<hkaiser> you want to talk?
<hkaiser> when would be a good time?
<simbergm> thanks! yeah, why not
<simbergm> if you have time today that would be great, I'm away again thu-tue
<hkaiser> sure, I could talk at 9:30amCDT/16:30CEDT
<simbergm> hkaiser: sure, sounds good
<hkaiser> ok, let's play the venue by ear, later
<simbergm> yep
<simbergm> btw, hkaiser, have you received any emails about trying to link travis with stellar-group on github?
<hkaiser> yes, I did approve everything
<simbergm> ok, thanks, have to see why nothing shows up yet on the travis pr
<hkaiser> simbergm: I might have missed something, but I pushed all the buttons...
<simbergm> :P
K-ballo has joined #ste||ar
<simbergm> no worries, it'll sort itself out somehow soon, I think...
<simbergm> might just need another trigger
hkaiser has quit [Ping timeout: 250 seconds]
hkaiser has joined #ste||ar
hkaiser has quit [Quit: bye]
hkaiser has joined #ste||ar
<hkaiser> simbergm: I'm ready whenever you are
<simbergm> hkaiser: 10 min?
<simbergm> appear.in?
<hkaiser> sure
diehlpk_mobile has joined #ste||ar
<diehlpk_mobile> hkaiser parsa So we need have a room for the meeting tomorrow?
<diehlpk_mobile> or do you attend from home
<parsa> i told the AV guy after the Monday meeting that we'd be having a metting at 8am tomorrow
<diehlpk_mobile> yes. they know
<diehlpk_mobile> they want to know if one of us will be physically there
<diehlpk_mobile> or if all of us just join the meeting remotely
<diehlpk_mobile> I think we all can join remotely and Darren do not need to set up the webcam and conference Mic
diehlpk_mobile has quit [Ping timeout: 252 seconds]
<heller> simbergm: did you get the participants list?
<simbergm> heller yep, thanks
<simbergm> Looks promising, more than 2 people
<heller> Indeed ;)
<heller> Also the accounts are there... You might want to tell him your IP to connect
<heller> hkaiser: I know It's a little earlier than we agreed, but I am available since now
hkaiser has quit [Quit: bye]
hkaiser has joined #ste||ar
<hkaiser> heller: sorry, can
<hkaiser> t
<hkaiser> do it today :/
<heller> oh, ok :/
<hkaiser> tomorrow?
<heller> tomorrow is bad for me...
<heller> friday?
<hkaiser> k
<hkaiser> sounds like a plan
<hkaiser> thanks!
<heller> great, no problem
diehlpk_mobile has joined #ste||ar
diehlpk_mobile has quit [Ping timeout: 252 seconds]
hkaiser has quit [Quit: bye]
jaafar has quit [Ping timeout: 252 seconds]
jaafar has joined #ste||ar
jaafar has quit [Ping timeout: 250 seconds]
Coldblackice|2 has quit [Ping timeout: 258 seconds]
jaafar has joined #ste||ar
Coldblackice has joined #ste||ar
hkaiser has joined #ste||ar
<nikunj> hkaiser: yt?
<hkaiser> here
<nikunj> hkaiser: just realized that the implementation requires us to not forward anything by reference
<nikunj> otherwise we will not get correct results
<hkaiser> why?
<nikunj> check this https://pastebin.com/Tmeqnt8d
<nikunj> if we pass anything by reference and want to play with the variable as well, we will not get the expected results
<nikunj> answer comes out to be 9 when it had to be 6
<hkaiser> nothing we can do about this, can we?
<hkaiser> we can't protect the user from itself
<nikunj> we should be able to mitigate if we send in the copied arguments
<hkaiser> do we want to do that?
<nikunj> no, a lot more of allocations
<hkaiser> users have to know what they are doing
<hkaiser> the same is true for normal async, btw
<nikunj> but if it's not the users fault, they will observe incorrect results
<hkaiser> it's always the user's fault, even more as your example has a race
<nikunj> it's replay
<nikunj> so another thread won't be invoked until one fails
<nikunj> so I don't think any race is happening
<hkaiser> ok
<nikunj> but yes, we need to do something about it
<hkaiser> I'd say, let's not worry about this, we can mention it in the docs that the user should be careful with ref-args
<nikunj> btw I'm reading what GaTech did when Sandia asked them to do the same. They don't have such a thing as well
<nikunj> so they experience the same with replay
<nikunj> I just wanted to differentiate our work from them, coz literally everything seems same as of now xD
<nikunj> hkaiser: let me try to tweak it works either ways and let me check if there's a significant overhead to it
<nikunj> *tweak it to work
jaafar has quit [Ping timeout: 252 seconds]
jaafar has joined #ste||ar
diehlpk_mobile has joined #ste||ar
<diehlpk_mobile> hkaiser Talk went well
<diehlpk_mobile> people asked if hpx is free
<diehlpk_mobile> and does it run on the national lab resources
<diehlpk_mobile> and if hpx is easy to install ?
<hkaiser> :D
<hkaiser> diehlpk_mobile: great!
<diehlpk_mobile> they were impressed by the libfabric results
<diehlpk_mobile> and asked if this is available as well
<diehlpk_mobile> but most mechanical or computational engineering folks
<diehlpk_mobile> gtg to the coffee break and battery dies soon
nikunj97 has joined #ste||ar
diehlpk_mobile has quit [Ping timeout: 252 seconds]
jaafar has quit [Ping timeout: 252 seconds]
jaafar has joined #ste||ar