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/
K-ballo has quit [Quit: K-ballo]
hkaiser has quit [Quit: bye]
diehlpk_work has quit [Remote host closed the connection]
bita has quit [Ping timeout: 240 seconds]
jehelset has joined #ste||ar
jejune has joined #ste||ar
jejune has quit [Client Quit]
bita has joined #ste||ar
bita has quit [Ping timeout: 240 seconds]
<gnikunj[m]> and all this time, I used to think hpx::reduce returns a future<T> instead of T
<gnikunj[m]> ms: is there a reason to why we don't return a future<T>?
<gnikunj[m]> I thought we return future<void> for for_each and similar algorithms and future<T> for reduce and similar others
<srinivasyadav227> current c++ standard uses these(https://en.cppreference.com/w/cpp/algorithm/execution_policy_tag) as execution policies right?
<srinivasyadav227> and in #5157 it is mentioned that hpx parallel algorithms use datapar execution policy, so after googling I found this (http://www.open-std.org/jtc1/sc22/wg21//docs/papers/2016/p0350r0.pdf), is this datapar same as the one hpx uses?
<gnikunj[m]> srinivasyadav227: yes. I don't think we have the unseq policies in HPX though as they require vectorization. Iirc there was a gsoc project to implement par_unseq policy in HPX as well.
<srinivasyadav227> yes #2271 says about adding vec support (par_unseq)
peltonp1 has quit [Quit: Lost terminal]
<gnikunj[m]> right, you need to vectorize the loop body.
<srinivasyadav227> ok, one more doubt please, in #5157 it says currently we use datapar right? which is older, so now we should adapt the datapar to new standard (https://wg21.link/n4755)?
<gnikunj[m]> yes, that's what Hartmut was talking about the other day
<gnikunj[m]> it's essentially refactoring things
<srinivasyadav227> oh, just started to understand it better now, :)
<gnikunj[m]> nice!
<gnikunj[m]> gonidelis[m]: yt?
<srinivasyadav227> gnikunj: thanks! 🙂
<jedi18[m]1> @hkaiser Did you forget to convert the tag_invoke in container_algorithms/for_each.hpp to tag_fallback_invoke? Since you asked us to do that for adjacent_find and copy? Should I do that in my seperating for_each_n PR?
<jedi18[m]1> I've created a PR for the seperation of the for_each_n, please let me know changes I need to make to it
jehelset has quit [Remote host closed the connection]
jehelset has joined #ste||ar
<jedi18[m]1> Shall I remove this since the segmented implementation has been moved to segmented_algorithms/for_each.hpp ? https://github.com/STEllAR-GROUP/hpx/blob/830f05b14d8cfa5d73c7e4406cf54466b1e4483d/libs/parallelism/algorithms/include/hpx/parallel/algorithms/for_each.hpp#L514
<jedi18[m]1> * Shall I remove this since the segmented implementation has been moved to segmented_algorithms/for_each.hpp ? https://github.com/STEllAR-GROUP/hpx/blob/830f05b14d8cfa5d73c7e4406cf54466b1e4483d/libs/parallelism/algorithms/include/hpx/parallel/algorithms/for_each.hpp#L514-L552
<rori> I think you are safe to leave it here as I think this same deprecation warning is used for the [non segmented impl](https://github.com/STEllAR-GROUP/hpx/blob/830f05b14d8cfa5d73c7e4406cf54466b1e4483d/libs/parallelism/algorithms/include/hpx/parallel/algorithms/for_each.hpp#L713) but ask hkaiser when he comes back to be sure (he is in the US timezone btw)
iiiiiiiqr[m] has joined #ste||ar
<rori> clang-format step has been failing for a few days since the code coverage docker image has been merged, I've merged a fix to the image, it should be fixed later today
iiiiiiiqr[m] has left #ste||ar ["User left"]
<jedi18[m]1> I had remove it since it uses the for_each_n_ which has been removed. What I wanted to know was is it safe to remove it or do I have to shift it to segmented_algorithms somehow?
<jedi18[m]1> Oh ok thanks
<rori> jpinto[m]: For what I see from your CMakeLists.txt, you still have the [INTERNAL_FLAGS](https://github.com/JAPPinto/HPX-Performance-Counters/blob/master/full_counter/CMakeLists.txt#L23) argument to the `add_hpx_executable` and `add_hpx_component` can you try removing those? :)
K-ballo has joined #ste||ar
hkaiser has joined #ste||ar
hkaiser has quit [Quit: bye]
diehlpk_work has joined #ste||ar
jehelset has quit [Remote host closed the connection]
jehelset has joined #ste||ar
nanmiao has joined #ste||ar
<diehlpk_work> ms[m], Hi, would you be interested to serve in the program committee for the Euro-Par workshop?
bita has joined #ste||ar
parsa has quit [Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net]
parsa has joined #ste||ar
<rori> diehlpk_work: ms is on vacation this week
<diehlpk_work> rori, thanks. Would you be interested too?
nanmiao has quit [Quit: Connection closed]
nanmiao has joined #ste||ar
hkaiser has joined #ste||ar
<rori> I'm not sure what it does imply to be in the committee..
<diehlpk_work> rori, You would need to review one paper about AMTs
<rori> Not sure I'm good enough for this ^^'
<rori> ms would fit better haha
<diehlpk_work> rori, Sure
<gonidelis[m]> gnikunj[m]: sorry just saw that
jejune has joined #ste||ar
jehelset has quit [Remote host closed the connection]
jejune has quit [Quit: "What are you trying to say? That I can dodge bullets?" "No Neo, what I'm trying to say, is that when you are ready.....you won't have to"]
<gonidelis[m]> K-ballo someone is playing with me
<K-ballo> ?
<gonidelis[m]> `and` as a rref operator
<K-ballo> that's not an operator
<K-ballo> `and` is just an alternative spelling of `&&`
<K-ballo> it can act as an operator, qualifier, specifier.. whatever `&&` does
jejune has joined #ste||ar
jaafar_ is now known as jaafar