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/
<github>
hpx/master 0b269d1 Hartmut Kaiser: Jemalloc is now supported for MSVC if built using VcPkg
quaz0r has joined #ste||ar
<github>
[hpx] hkaiser created noreturn_attribute (+1 new commit): https://git.io/v7yB7
<github>
hpx/noreturn_attribute 2bf98d6 Hartmut Kaiser: Properly support [[noreturn]] attribute if available...
<github>
[hpx] hkaiser opened pull request #2813: Properly support [[noreturn]] attribute if available (master...noreturn_attribute) https://git.io/v7yBb
K-ballo has quit [Quit: K-ballo]
hkaiser has quit [Quit: bye]
twwright_ has joined #ste||ar
twwright has quit [Ping timeout: 260 seconds]
twwright_ is now known as twwright
EverYoung has quit [Ping timeout: 240 seconds]
Matombo has joined #ste||ar
parsa[w] has quit [Read error: Connection reset by peer]
parsa[w] has joined #ste||ar
bikineev has quit [Remote host closed the connection]
mcopik has quit [Ping timeout: 248 seconds]
bikineev has joined #ste||ar
Matombo has quit [Remote host closed the connection]
david_pfander has joined #ste||ar
david_pfander has quit [Remote host closed the connection]
david_pfander has joined #ste||ar
jbjnr has joined #ste||ar
bikineev has quit [Remote host closed the connection]
bikineev has joined #ste||ar
<github>
[hpx] AntonBikineev force-pushed nondefault_ctor_ser from 2e90fd5 to bd92900: https://git.io/v7ySh
<github>
hpx/nondefault_ctor_ser bd92900 AntonBikineev: Reserving space before collection deserialization
<github>
hpx/master 1416e1f Andreas Schäfer: reorder forward declarations to get rid of C++14-only auto return types
<github>
hpx/master 0c56fbc Andreas Schäfer: add test for C++14 auto compiler feature
<github>
hpx/master ee0f107 Andreas Schäfer: revert customization points, conditionally to the availability of C++14
bikineev has quit [Remote host closed the connection]
bikineev has joined #ste||ar
<jbjnr>
taeguk[m]: -sorry - I forgot the gsoc calla gain. I need to fix my calendar.
<jbjnr>
I'm a terrible mentor :(
<wash>
Paper's submitted.
<wash>
hkaiser: why are you awake o.O
<hkaiser>
you're awake as well ;)
<jbjnr>
wash well done. and good job. Thanks again. you are a hereo :)
<wash>
Paper :p
<jbjnr>
^hero
<hkaiser>
wash: you are a hero indeed
<wash>
heller, hkaiser, jbjnr: I'd like to add Brandon Cook to the author list, if everyone is okay with that. 1.) we already have a million authors 2.) he helped out during the runs and 3.) he is one of the few people who actually sent me feedback when I sent out a copy of the GB paper
<wash>
thanks guys :). It's a good paper, I think it will get accepted...
<hkaiser>
whatever it takes - one more doesn't hurt
<jbjnr>
we have so many already, it feel like we work for CERN
<hkaiser>
lol
<hkaiser>
yah
<hkaiser>
success has many fathers, failure always only one
<hkaiser>
doesn't really matter - this will be a seminal paper, so everybody wants to be part of it
<wash>
hkaiser: also, if you think of a one or two sentence definition of "Asynchronous Many Task" models in the next 18 hours or so, lemme know.
<hkaiser>
wash: ok
<hkaiser>
I wouldn't hold my breath, though
<hkaiser>
wash: I'd say (quoting John): just submit the bloody thing
<hkaiser>
let go
<wash>
I already did :p
<hkaiser>
good
<wash>
hkaiser: btw some bad news
<hkaiser>
ok?
<wash>
hkaiser: the notification deadline for this paper is after the IPDPS deadline, by about two weeks
<hkaiser>
not really bad news
<wash>
hkaiser: so even with whatever extension IPDPS will give, if it gets rejected from this, we cant do ipdps
<hkaiser>
there will be always more conferences
<wash>
hkaiser: btw, we should try to submit a paper to IPDPS
<hkaiser>
if we really want to submit to ipdps later, we'll find a way ;)
<hendrx>
anynone using windows with vcpkg to use hpx ?
<jbjnr>
hkaiser: Should we support multiple localities on one node - doesn't seem very useful to me (apart from testing maybe?)
hendrx has quit [Quit: Page closed]
<hkaiser>
jbjnr: we have supported it so far
<hkaiser>
I think we should
<jbjnr>
So what doe the RP break that used to work? AFACT all it does is bind stuff to cores and having two of them (with different command line parms?) shouldn't chan ge that
<hkaiser>
jbjnr: it binds the threads used by all localities on the same node to the same cores
<jbjnr>
so what? if that's what the user asks for ...
<hkaiser>
we used to 'move' the second locality up to the first free core not used by the first locality
<jbjnr>
hmmm
<hkaiser>
jbjnr: I think we can still do that
<jbjnr>
loc 0 : --hpx:bind=compact --hpx:threads=4 | loc 1 : --hpx:bind=compact --hpx:threads=4 --hpx::thread-offset=4 ?
<jbjnr>
just add an offset so we can move the threads on the command line?
<jbjnr>
or does it have to be command line free
<hkaiser>
commandline will not work with slurm etc.
<jbjnr>
(it will, it's just not as simple)
<hkaiser>
nod
<hkaiser>
why not leave it as automatic as it is right now?
<jbjnr>
I'm not sure I understand. the problem is that the app doesn't know it has 2 localities until after the runtime gets going
<jbjnr>
but the binding is done before the runtime init if RP is used
<jbjnr>
ask me about it next week. I must work on thesis tody :(
<hkaiser>
jbjnr: ok, I'll fix it
denis_blank has joined #ste||ar
bikineev has joined #ste||ar
Matombo has joined #ste||ar
mcopik has joined #ste||ar
Matombo has quit [Ping timeout: 255 seconds]
eschnett has quit [Quit: eschnett]
denis_blank has quit [Ping timeout: 258 seconds]
K-ballo has joined #ste||ar
auviga has quit [*.net *.split]
ABresting has quit [*.net *.split]
ABresting has joined #ste||ar
auviga has joined #ste||ar
Matombo has joined #ste||ar
eschnett has joined #ste||ar
bikineev has quit [Read error: Connection reset by peer]
bikineev_ has joined #ste||ar
mcopik has quit [Ping timeout: 260 seconds]
hkaiser has quit [Quit: bye]
Matombo has quit [Ping timeout: 240 seconds]
Matombo has joined #ste||ar
mbremer has quit [Quit: Page closed]
hkaiser has joined #ste||ar
aserio has joined #ste||ar
Matombo has quit [Ping timeout: 258 seconds]
Matombo has joined #ste||ar
Matombo has quit [Ping timeout: 260 seconds]
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
Matombo has joined #ste||ar
david_pfander has quit [Ping timeout: 246 seconds]
mcopik has joined #ste||ar
aserio has quit [Ping timeout: 246 seconds]
bikineev_ has quit [Ping timeout: 248 seconds]
Matombo has quit [Quit: Leaving]
mcopik has quit [Ping timeout: 246 seconds]
<zao>
hkaiser: Didn't do a full rebuild of partitioned_vector, but I seem to only get errors on partitioned_vector<int> now, in one test.
<zao>
Oh nevermind, the build stopped due to the error.
<zao>
Everything's still fubar.
<hkaiser>
zao: yah, seeing it myself as well - down to experimentation now
<zao>
Figured I'd give it a go as I saw the commit in my mailbox.