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/
patrick22 has quit [Ping timeout: 272 seconds]
weilewei has quit [Ping timeout: 245 seconds]
Coldblackice_ has joined #ste||ar
Coldblackice|2 has joined #ste||ar
Coldblackice|2 has quit [Max SendQ exceeded]
Coldblackice|2 has joined #ste||ar
Coldblackice_ has quit [Ping timeout: 240 seconds]
Coldblackice|2 is now known as Coldblackice
hkaiser has quit [Quit: bye]
shahrzad has quit [Quit: Leaving]
akheir has quit [Quit: Leaving]
bita_ has quit [Ping timeout: 260 seconds]
patrick22 has joined #ste||ar
freifrau_von_ble has quit [Quit: Idle for 30+ days]
patrick22 has quit [Ping timeout: 272 seconds]
<gonidelis[m]> Does anyone kow why this book is so freaking expensive while I cannot find any reference at all on the web about it?
<gonidelis[m]> I found like just two reviews
patrick22 has joined #ste||ar
patrick22 has quit [Ping timeout: 272 seconds]
<gonidelis[m]> is `result::get(std::make_pair(std::move(last), std::move(first)))` equivalent to `result::get(util::in_out_result{std::move(last), std::move(first)})` ?
<gonidelis[m]> I am of the impression that `result::get()` is redundant in the second expression
<tarzeau> what's your favourite terminal monospace font for coding?
hkaiser has joined #ste||ar
patrick22 has joined #ste||ar
<gonidelis[m]> this is my unary segmented transform adaptation and i am trying to test it on a single segmented test
<gonidelis[m]> here
<gonidelis[m]> and i get this weird error
<gonidelis[m]> actually not weird. I can't find out why `get_second_element` is not recognized
<gonidelis[m]> on the parallel algorithms it is recognized just fine
<gonidelis[m]> ALSO
<gonidelis[m]> there is no difference between `partitioned_vector_transform1.cpp` and `partitioned_vector_transform2.cpp` tests
<ms[m]> gonidelis[m]: all good in the end, or still getting an error?
<ms[m]> the difference between partitioned_vector_transform1.cpp and 2 is that one tests for int and the other for double
jehelset has quit [Remote host closed the connection]
<ms[m]> they take a lot of memory to compile and used to be in one file before, but got split up to reduce the memory requirement a bit
shahrzad has joined #ste||ar
<gonidelis[m]> ms[m]: hmmm that's weird. `git diff partitioned_vector_transform1.cpp partitioned_vector_transform2.cpp` produces no output
<gonidelis[m]> ms[m]: sorry my bad
<gonidelis[m]> you are right
<gonidelis[m]> ms[m]: on the `get_second_element` thing I still get the error
<ms[m]> there's at least one line different... but it would be nice if they didn't duplicate so much...
<gonidelis[m]> yy you are right
<gonidelis[m]> anyways
<gonidelis[m]> I did `git diff` instead of `diff`. That was my mistake
<ms[m]> np :)
<ms[m]> looking through your errors now...
<gonidelis[m]> thanks
<gonidelis[m]> please notice that I have commented out most of the partioned vector transform test in order to see if these single calls compile
<ms[m]> the error messages and and the diffs you posted don't line up, right? can you push either a matching commit somewhere or update the gist please?
<ms[m]> sorry, they might...
<gonidelis[m]> ahh... yeah. I think I have uploeaded the correct version of my code
<gonidelis[m]> let me double check
<ms[m]> my bad, was looking at the wrong transform.hpp
<ms[m]> you'll need to update that one as well
<gonidelis[m]> Aaahhhh
<gonidelis[m]> Can't believe I missed that
<gonidelis[m]> Thank you so much
<gonidelis[m]> But how did you find it?? Does the compiler indicate anything about it ??
<ms[m]> the compiler tells me that you were trying to pass a tagged_pair to get_second_element, the only way that can happen is if you have either a definition or a declaration of transform_ that returns tagged_pair
<ms[m]> I did happen to know that there are forward declarations for the segmented overloads, so that helped me find where it actually was, but the error message told me that there is one somewhere
teonnik has quit [Quit: Idle for 30+ days]
<gonidelis[m]> ms[m]: IT WORKED
<gonidelis[m]> god I missed like two days in this stupid thing
<gonidelis[m]> and I was aware of the forward declaration I just didn't thing of them
<gonidelis[m]> thank you very very much
<hkaiser> gonidelis[m]: do you want to meet today?
<gonidelis[m]> if you have a 10-15' gap it would be great
<gonidelis[m]> just for a mini brief
<hkaiser> ok, let's do it then
<hkaiser> ms[m]: sorry for my confusing ramblings on the PR wrt hpx::get
<hkaiser> I got myself utterly confused, mostly - that reflected on my comments
<ms[m]> hkaiser: :D
<ms[m]> no problem, I think that was more or less what I understood from your original comment, but my reply may have confused you as well :P
<ms[m]> I will not be adding the get<T> overloads in that PR though
<ms[m]> I suppose adding the support for std things is not much work, it's not clear to me how easy it is to do that with e.g. our tuple (quick and easy for you guys...)
<hkaiser> ms[m]: sure, that's a separate issue
<ms[m]> it's on my todo list now ;)
<hkaiser> glad to be of service by filling that list ;-)
<ms[m]> then we are (and were) on the same page
<ms[m]> much appreciated :P
<hkaiser> thanks for everything!
patrick22 has quit [Ping timeout: 272 seconds]
bita_ has joined #ste||ar
<gonidelis[m]> Do I need to chage `hpx::tuple` in here
<gonidelis[m]> to `util::in_in_out_result` ?
<gonidelis[m]> I think the return value depends on them but I am not sure whether i can just leave and for some "typecastin" to `in_in_out_result` to take place afterwards
<gonidelis[m]> leave it ^^
<gonidelis[m]> plus should I use some custom (ours) `dispatch` / `async_dispatch` ?