hkaiser changed the topic of #ste||ar to: STE||AR: Systems Technology, Emergent Parallelism, and Algorithm Research | stellar-group.org | HPX: A cure for performance impaired parallel applications | github.com/STEllAR-GROUP/hpx | This channel is logged: irclog.cct.lsu.edu
diehlpk_work has quit [Remote host closed the connection]
K-ballo has quit [Quit: K-ballo]
LorenDB[m] has left #ste||ar [#ste||ar]
hkaiser has quit [Quit: Bye!]
K-ballo has joined #ste||ar
hkaiser has joined #ste||ar
K-ballo has quit [Quit: K-ballo]
K-ballo has joined #ste||ar
<gnikunj[m]> hkaiser @rori @srinivas : pls go ahead with the meeting without me. I have my final exam at 9:30 today
<zao> Enjoy the exam! ;)
diehlpk_work has joined #ste||ar
<hkaiser> gnikunj[m]: we did cancel meetings up until Jan 4
<gnikunj[m]> hkaiser: yeah, so the email right after I wrote that
<gnikunj[m]> Anyway, gtg now
<hkaiser> ms[m]: yt?
<hkaiser> ms[m]: could you access the full logs for this run, please: https://cdash.cscs.ch/buildSummary.php?buildid=206297
<hkaiser> I can't reproduce this locally and am at a loss which HPX_INVOKE is setting off gcc 7
<ms[m]> hkaiser: sure, just a sec
<hkaiser> thanks
<hkaiser> that will help
<ms[m]> note that it could just be a gcc 7 bug and ifdefing the use of HPX_FORWARD may be necessary :/
<ms[m]> let me know if you want me to have a look
<hkaiser> ms[m]: there are more spots where later gcc versions fail to see through the HPX_INVOKE, I have just reverted those to util::invoke()
<hkaiser> I plan to do the same here
<ms[m]> yep, sounds good
diehlpk_work has quit [Read error: Connection reset by peer]
diehlpk_work has joined #ste||ar
<ms[m]> dkaratza and everyone else involved in gsod, thank you! https://developers.google.com/season-of-docs/docs/participants
<diehlpk_work> Nice
<diehlpk_work> GSoC 2022 is annoucned
<diehlpk_work> Orgs need to register in Feb 22
<hkaiser> ms[m]: \o/ thanks for everything you did to make this happen!
<gonidelis[m]> ms: yay!
<diehlpk_work> ms[m], Do you intent to write some blog post?
<hkaiser> diehlpk_work: dkaratza[m] can do that
<hkaiser> well, can and should
<diehlpk_work> Ok, that would be great
<hkaiser> diehlpk_work: we already asked her to do that
K-ballo has quit [Quit: K-ballo]
K-ballo has joined #ste||ar
<dkaratza[m]> <ms[m]> "dkaratza and everyone else..." <- very happy to hear that, its always nice working with you!
<dkaratza[m]> <hkaiser> "well, can and should" <- sure, I will start looking at this once Christmas vacation starts and I wont have any school
<hkaiser> dkaratza[m]: +1
<ms[m]> hkaiser: sorry for dragging out that atomic thread out on the fork-join pr... we can have a call about it if needed (not anymore tonight though)
<hkaiser> ms[m]: np, I think it is ok now
<ms[m]> hkaiser: no, that's what I'm apologizing about: https://github.com/STEllAR-GROUP/hpx/pull/5697#discussion_r768979199 :P
<hkaiser> ms[m]: I definitely appreciate your comments
<hkaiser> apparently I misunderstood you :/
<ms[m]> I most likely wasn't very clear...
<hkaiser> the thread state load in the wait function?
<ms[m]> yep, that one exactly
<hkaiser> but it would be sufficient to make the first of the loads acquire, right?
<ms[m]> hmm, no I don't think so
<ms[m]> because that might not return the wanted state yet
<ms[m]> the last one would be enough, but that doesn't make much sense
<hkaiser> well, let's discuss this Thursday, I don't you to stay awake too long...
<hkaiser> want*
<hkaiser> I'll think about it some more
<ms[m]> yeah, let's do that
<ms[m]> it's not that late :P but it's definitely time to stop working...