K-ballo 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/
nanmiao has joined #ste||ar
<srinivasyadav227> hkaiser: can you please help me with this (https://app.codacy.com/gh/STEllAR-GROUP/hpx/pullRequest?prid=7425496), I could not understand why Codacy is giving an error for this template specialization but it was ok with others?
<hkaiser> thanks zao
<hkaiser> srinivasyadav227: I marked those as to be ignored
hkaiser has quit [Quit: bye]
<srinivasyadav227> thank you
diehlpk_work has quit [Ping timeout: 245 seconds]
parsa| has joined #ste||ar
parsa has quit [Ping timeout: 258 seconds]
parsa| is now known as parsa
nanmiao has quit [Ping timeout: 240 seconds]
<rachitt_shah[m]> ms: gnikunj I wouldn't be able to make it to the call today, my city hasn't had electricity since yesterday, can we meet tomorrow?
<rachitt_shah[m]> I have figured out a way to arrange the API docs in a more refined way, it was to do with our sphinx theme.
<zao> Automatic code review ought to just go "shrug, this is too fancy C++ for my AI mind" given any HPX ;)
<rachitt_shah[m]> I am assuming these areas aren't updated as frequently as the APIs? We could always redirect them to Sphinx
<ms[m]> rachitt_shah: what would be the benefit of using docosaurus?
<rachitt_shah[m]> I would like to showcase some demos at our call, or perhaps build a page for HPX as proof of concept.
hkaiser has joined #ste||ar
<ms[m]> rachitt_shah: all right, I'm open to hear your suggestions, but I think moving to another platform for the non-API docs is most likely overkill and may be focusing on the wrong parts
<ms[m]> let's discuss in the meeting though!
<rachitt_shah[m]> <ms[m] "rachitt_shah: all right, I'm ope"> Makes sense, perhaps I would need to direct my focus on APIs first. Let's take this up on the meeting.
<gonidelis[m]> hkaiser: ms[m] gnikunj[m] Apologies, I won't be able to attend gsod meeting today. I am cought up in some schoolwork....
<ms[m]> gonidelis[m]: no worries
<hkaiser> sure, np
<hkaiser> ms[m]: since you asked: for V1.7 I would like to get #5357 in
<hkaiser> also, I would like to fix the c++17 github builder problems - btw, was that the reason for deciding not to move to c++17 yet?
<hkaiser> (#4994)
<ms[m]> hkaiser: yep, I only meant for rc1 at this point, we can definitely get that in 1.7.0
<hkaiser> #5328 looks to be ready - almost as well
<ms[m]> I don't know if I can give a useful review on that but I can definitely have a high level look if you'd like
<hkaiser> ahh ok - cool
<hkaiser> much appreciated
<ms[m]> uhm, I think c++17 itself was fine but we tried adding some other flags as well which made everything terrible
<ms[m]> don't remember the details, but I think there might be some links to logs in the discussion on the pr
<hkaiser> ms[m]: for #5303, I'd rather move serializable_function into the serialization module, we've done this for serializable_any already
<ms[m]> that makes sense
<hkaiser> I plan to do that
<ms[m]> we could do that and move the invoke stuff to a separate module as well
<hkaiser> sure, but would that be necessary?
<ms[m]> ok 👍️
<ms[m]> not necessary, mostly just possible...
<hkaiser> ms[m]: gnikunj[m] promised to finish #5353
<hkaiser> #5362 would be nice to have as well - I could use that for one of our projects
<hkaiser> btw
<hkaiser> the future_sender you created (forgot the name), I think it needs a keep_alive as well
<ms[m]> sure, anything marked 1.7.0 is still on track for 1.7.0 unless they get badly delayed
<ms[m]> hmm, could you link to where
<hkaiser> the captured `this` should be a intrusive_ptr
<ms[m]> hrm, the future should be moved into the lambda?
<ms[m]> good catch!
<hkaiser> sorry, not the `this`, but the state
<hkaiser> I assume that `this` is kept alive long enough somehow
<ms[m]> ah, yes, the operation_state is guaranteed to be kept alive until set_value is called
<hkaiser> on a second look, everything might be fine as the state sits on the stack
<hkaiser> and if the operation_state is guaranteed to be alive, the the future will be too
<ms[m]> yes, not necessarily on the stack, but the lifetime is guaranteed
<hkaiser> I would probably have moved the future into the capture, though
<hkaiser> anyways, food for thought
<ms[m]> it's for sure clearer
hkaiser has quit [Quit: bye]
hkaiser has joined #ste||ar
nanmiao has joined #ste||ar
nanmiao has quit [Client Quit]
nanmiao has joined #ste||ar
hkaiser has quit [Ping timeout: 245 seconds]
chuanqiu has joined #ste||ar
hkaiser has joined #ste||ar
diehlpk_work has joined #ste||ar
chuanqiu has quit [Ping timeout: 240 seconds]
Vir has joined #ste||ar
Vir_libera_chat has quit [*.net *.split]
hkaiser has quit [Ping timeout: 245 seconds]
hkaiser has joined #ste||ar
<diehlpk_work> gnikunj[m]: How is GSoc going for your student?
<diehlpk_work> gonidelis[m]: How is GSoC going for your student?
diehlpk_work has quit [Quit: Leaving.]
<gonidelis[m]> akhil is doing great
<gonidelis[m]> he is very proactive
diehlpk_work has joined #ste||ar
diehlpk_work has quit [Quit: Leaving.]
nanmiao has quit [Quit: Connection closed]