aserio 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/
bikineev has joined #ste||ar
K-ballo has quit [Quit: K-ballo]
bikineev has quit [Remote host closed the connection]
vamatya has joined #ste||ar
hkaiser has quit [Quit: bye]
parsa has joined #ste||ar
vamatya has quit [Ping timeout: 248 seconds]
parsa has quit [Quit: Zzzzzzzzzzzz]
vamatya has joined #ste||ar
pree has joined #ste||ar
vamatya has quit [Client Quit]
pree has quit [Read error: Connection reset by peer]
pree has joined #ste||ar
pree has quit [Read error: Connection reset by peer]
Matombo has joined #ste||ar
bikineev has joined #ste||ar
bikineev has quit [Remote host closed the connection]
bikineev has joined #ste||ar
jaafar has joined #ste||ar
Matombo has quit [Remote host closed the connection]
<jbjnr>
builds fin ewith clang, but I cannot find any combination of cxx flags that works with my gcc 7
StefanLSU has joined #ste||ar
<zao>
jbjnr: Can you for our sanity run a verbose make/ninja to see what exact flags we pass to the compiler?
<zao>
Haven't booted my easybuild box in a while, so what branch and what gcc is this?
<jbjnr>
ouch - there's a c++17 flag right at the end of the flags. not sure where that's coming from
<jbjnr>
thanks zao I should not have assumed that my flags were the only ones in there
<jbjnr>
rats. it must be vc library
<jbjnr>
I forgot I used that for hackathon
<zao>
Ooh, nice.
<zao>
hkaiser: What's our endgame regarding this workaround? Getting some sort of fix into future Boosts and waiting for the future to arrive? Clamp HPX's use of C++ to 14 based on which version of Boost we detect?
<zao>
Kind of sucks if this causes everyone ever consuming HPX to be stuck in the dark ages of C++14 going forward.
<zao>
(as I assume that this affects end-user code)
<zao>
Or can you reliably build end-user software with C++17/20 if HPX is built with C++14, and do we emit any std flags into our CMake/pkgconfig?
jaafar has joined #ste||ar
jaafar has quit [Ping timeout: 246 seconds]
<jbjnr>
zao: it builds ok using c++17 with clang
<zao>
Yeah, but that doesn't help the GCC users.
<jbjnr>
well, you said that " everyone ever consuming HPX to be stuck in the dark ages of C++14 going forward" so I was just reassuring you that clang is ok
Aalice has quit [Quit: Leaving.]
<zao>
Indeed a good point and a possible local workaround.
<zao>
But not quite sustainable if we intend to claim GCC as a supported compiler.
<zao>
Particularly as there may be quite some other deps built with GCC.
<zao>
In a perfect world, HPX would be a trendsetter and kill off GCC :)
<hkaiser>
zao: no idea - I'tired of trying to fix boost
<hkaiser>
I'm tired, even
<K-ballo>
icl in particular is abandoned
<zao>
*nod*
<hkaiser>
wemight be able to rip it out, it's not used too heavily
<jbjnr>
I have a new bug that didn't exist before I did a fetch/pull just now
<jbjnr>
the -latomic thing is very annoying indeed. nothing links without manual intervention. Will try to track it down tomorrow.
<K-ballo>
we were adding -latomic unconditionally but now we check whether std::atomic actually needs it.. this leaked boost::atomic might need it, although it's curious that cmake isn't bringing the indirect dependency in
Aalice has quit [Quit: Leaving.]
parsa has joined #ste||ar
parsa has quit [Read error: Connection reset by peer]