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 has joined #ste||ar
diehlpk has quit [Client Quit]
Yorlik_ has joined #ste||ar
Yorlik has quit [Ping timeout: 268 seconds]
hkaiser has quit [Quit: Bye!]
<zao> Hah, I managed to trick myself the other day. The constexpr variable name is simply not defined in the #if and evaluates to integer zero.
tufei has quit [Remote host closed the connection]
tufei has joined #ste||ar
hkaiser has joined #ste||ar
hkaiser has quit [Quit: Bye!]
hkaiser has joined #ste||ar
K-ballo has joined #ste||ar
K-ballo has quit [Read error: Connection reset by peer]
K-ballo has joined #ste||ar
hkaiser has quit [Quit: Bye!]
hkaiser has joined #ste||ar
K-ballo has quit [Ping timeout: 264 seconds]
K-ballo has joined #ste||ar
<gonidelis[m]> hkaiser: i have a question regarding combining two .cpp files into a single target
<gonidelis[m]> i have the tc_test.cpp which includes the catch2 tests and a custom_main.cpp which has the main() function
<gonidelis[m]> my god i answered my own question while typing it... I wanted to ask, if custom_main.cpp (which includes the main) should have its own target
<gonidelis[m]> hkaiser: https://prnt.sc/kBOSzqWclgir i cannot believe it took me two days!
<gonidelis[m]> need to read the cmake book
<hkaiser> gonidelis[m]: nice
<hkaiser> gonidelis[m]: you don't need a separate target for the other source
<hkaiser> a target is either an executable or some library module
<gonidelis[m]> tests pass which means nothing was screwed during parallelization
<gonidelis[m]> shift of parallelization runtime^^
<hkaiser> gonidelis[m]: you could have added that main function to the tc_test.cpp file as well, but separating it will simplify things for the other tests and reduces the hpx specifics in the sources
<gonidelis[m]> somehow catch wouldn't allow me. i tried to
<hkaiser> yah, if the tests pass the next step is to do perf tests
<gonidelis[m]> they specify that main should not be at the same file where TEST_CASE(){} is
<hkaiser> no no, separating it is fine
<gonidelis[m]> yup... so how do we measure performance now?
<hkaiser> let's call the file hpx_main.cpp or somesuch to make it clear it's HPX specific
<hkaiser> use a stop watch and run the tests ;-)
<gonidelis[m]> i could use VS perf preliminary. ah yeah sure. i ll tidy up the place
<hkaiser> I think the perf tests should use larger data sets
<gonidelis[m]> huh true
<hkaiser> there are perf tests ready made somewhere in the repo
<hkaiser> this one has a main ;0)
<gonidelis[m]> god....
<gonidelis[m]> at least we verify correctness its ok
<hkaiser> we should ask Andrew where the benchmark scripts are
<hkaiser> there were some in the bgl17 repo, I believe
<hkaiser> hah funny, they didn't even change the name here: https://github.com/pnnl/NWGraph/blob/master/bench/tc.cpp#L18
<gonidelis[m]> well they have this cmake option -DNW_GRAPH_BUILD_BENCH=ON
<gonidelis[m]> so everything should be here