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/
Vir has quit [*.net *.split]
Vir has joined #ste||ar
hkaiser has quit [Quit: bye]
diehlpk_work has quit [Remote host closed the connection]
syrex has quit [Quit: syrex]
hkaiser has joined #ste||ar
<rachitt_shah[m]>
<ms[m] "might be worth either asking abo"> I'll refloat and edit the form? Sorry about the abscence recently.
<ms[m]>
rachitt_shah[m]: refloat? in any case, it might be easiest to just make the change and ask people when you interview them what they think, let's not overcomplicate it
<ms[m]>
have you set up any meetings yet?
<rachitt_shah[m]>
<ms[m] "rachitt_shah[m]: refloat? in any"> Oh, I meant share again. Sorry about that.
hkaiser has quit [Quit: bye]
<rachitt_shah[m]>
ms[m] also, I was looking at alternatives, we can look at docusaurous as well. It has a refreshing UI as well.
diehlpk_work has joined #ste||ar
<ms[m]>
rachitt_shah[m]: ok, the meetings are up to you, but set them up as early as possible so that you can start making a concrete plan earlyer
<ms[m]>
is docosaurus just for api docs or all the docs? if the latter, I think we should avoid changing sphinx unless there's a really strong case for it; for the api docs we're more open since the current situation is quite bad
<ms[m]>
could you please collect a list of pros and cons of these other systems so that we can evaluate whether it would actually be worth it?
<ms[m]>
gnikunj, others, kokkos call starting soonish now if interested, we'll keep it short
<rachitt_shah[m]>
<ms[m] "rachitt_shah[m]: ok, the meeting"> I would start from Monday. I should have began this week itself, sorry about the delay.
<rachitt_shah[m]>
<ms[m] "rachitt_shah[m]: ok, the meeting"> For all the docs, but for API docs I'm thinking of sticking with Sphinx. This would require a detailed analysis before we take action.
<rachitt_shah[m]>
<ms[m] "could you please collect a list "> I'm evaluating them at the moment. Because this needs to be sorted on priority so we're good to go for the rest of the tasks.
hkaiser has joined #ste||ar
nanmiao has quit [Quit: Connection closed]
<srinivasyadav227>
hkaiser: can you please look at this https://github.com/STEllAR-GROUP/hpx/pull/5328, i could not understand why `HPX_WITH_CXX20_EXPERIMENTAL_SIMD` passes for clang compilers in Jenkins checks ,but the test fails correctly (as expected) locally and on circle ci with clang. And passed with GCC 11.1 both locally and on Jenkins
<jedi18[m]>
<gonidelis[m] ""random greek there""> What does it stand for :D
<jedi18[m]>
Oh right "hey"
<gonidelis[m]>
it's just the mapping of hey
<gonidelis[m]>
yes
<jedi18[m]>
Right so I don't understand what's the use of the second sentinel value
<jedi18[m]>
Also the base implementations only take in two iterators and a sentinel value
<jedi18[m]>
So I'll have to deal with it in the tag_fallback_invoke right?
<gonidelis[m]>
where do you see the two sentinel values?
<jedi18[m]>
uninitialized_copy_result<I, O>
<jedi18[m]>
uninitialized_copy(I ifirst, S1 ilast, O ofirst, S2 olast);
<gonidelis[m]>
ok now i see it
<jedi18[m]>
Why would the output iterator need a sentinel? Does that mean we also check to see if the place it's writing to has that value, so whichever sentinel we reach first decides the end?
<gonidelis[m]>
nice catch
<gonidelis[m]>
I won't claim that I am a standard expert but your explanation seems reasonable
<gonidelis[m]>
@k-ballo is usually around answering those things. but he is offline for a couple of days now
<gonidelis[m]>
let me see
<gonidelis[m]>
jedi18: well
<jedi18[m]>
Oh ok no problem, I'll add the overloads for now
<gonidelis[m]>
according to this `N = min(ranges::distance(ifirst, ilast), ranges::distance(ofirst, olast))` you are right
<jedi18[m]>
Ohh ok
<jedi18[m]>
Other*
<gonidelis[m]>
we are just copying all the elements if the output range is equal or bigger
<gonidelis[m]>
or we copy a part of the input range if the output is smaller
<jedi18[m]>
Ohh ok thanks!
<jedi18[m]>
<gonidelis[m] "according to this `N = min(range"> Where did you get this from? I dont see it on that page