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/
hkaiser has quit [Quit: bye]
<gonidelis[m]> Thanks a lot
K-ballo has quit [Quit: K-ballo]
parsa has quit [Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net]
hkaiser has joined #ste||ar
parsa has joined #ste||ar
parsa has quit [Client Quit]
parsa has joined #ste||ar
parsa has quit [Client Quit]
parsa has joined #ste||ar
parsa has quit [Client Quit]
parsa has joined #ste||ar
hkaiser has quit [Quit: bye]
bita has joined #ste||ar
bita has quit [Ping timeout: 268 seconds]
bita has joined #ste||ar
bita has quit [Ping timeout: 260 seconds]
K-ballo has joined #ste||ar
hkaiser has joined #ste||ar
<gonidelis[m]> hkaiser: yt?
<hkaiser> here
<gonidelis[m]> the `set(transform_range_binary_FLAGS COMPILE_DEFINITIONS HPX_HAVE_DEPRECATION_WARNINGS_V1_6=0) ` you told me to add
<gonidelis[m]> gives me this cmake error
<gonidelis[m]> "cannot find source file"
<gonidelis[m]> "/home/giannis/adapt_transform/libs/parallelism/algorithms/tests/unit/container_algorithms/COMPILE_DEFINITIONS "
<hkaiser> hmmm
<hkaiser> gonidelis[m]: try set(transform_range_binary_FLAGS COMPILE_FLAGS -DHPX_HAVE_DEPRECATION_WARNINGS_V1_6=0)
<gonidelis[m]> hkaiser: ok that's great
<gonidelis[m]> So there is only this minor problem remaining
<gonidelis[m]> It's the thing I was talking about with ms[m]. If we leave the test that runs the deprecated transform overload, then doesn't that mean that I should change the result type of the deprecated algo, also?
surbhi has joined #ste||ar
<surbhi> Hey there! I have a doubt. I have a windows laptop. So is it possible to build hpx on that or dual-booting with ubuntu will work? Thanks.
<K-ballo> why wouldn't it work? either?
<hkaiser> gonidelis[m]: I think we should leave the deprecated APIs unchanged
<zao> surbhi: unless someone has broken something or an update to VS breaks the build it’s a top tier platform.
<surbhi> Okay thank you.
<zao> Realistically, there might be some surprises in less tested/used parts of the build, like if all the devs on Windows never build the examples.
<zao> (has happened :D )
<gnikunj[m]> Lol when did that happen?
surbhi has quit [Ping timeout: 256 seconds]
<zao> It happens occasionally when I get around building HPX for some reason :)
<K-ballo> I never build the examples
<hkaiser> me neither ;-)
<gnikunj[m]> Really curious, do you build tests?
<hkaiser> only the ones I'm working on
<K-ballo> only if I happen to be working on one
<hkaiser> gnikunj[m]: I'm planning to clean up the resiliency PR over the next couple of days
<hkaiser> could you please make sure your latest version is on that PR?
<gnikunj[m]> hkaiser sure. Let me add everything to it today
<hkaiser> thanks
<jaafar> hkaiser: I'm trying out the PR :)
<jaafar> Just to verify: tasks added to a specific queue via the hint can still be stolen into other queues?
<hkaiser> yes
<jaafar> (My first experiment was to put everything in the same queue)
<jaafar> OK
<gnikunj[m]> hkaiser: yt?
<hkaiser> here
<gnikunj[m]> How do I ssh into loni? I don't have it in my config for some reason
surbhi has joined #ste||ar
<gnikunj[m]> hkaiser: ok, found: http://hpc.loni.org/docs/guides.php?system=Qb2
<gnikunj[m]> don't remember what my password was :/
<gnikunj[m]> hkaiser: just to make sure, password and username is the same as the one used at: https://allocations.loni.org/uselink.php?id=998cdf5fc878f2d9d48ee2d57c1e9a9e3b10e6e5
<gnikunj[m]> right?
<hkaiser> shrug
<hkaiser> probably
<gnikunj[m]> ughh
<gnikunj[m]> does anyone else has access to it? I'll talk to them
<jaafar> Is there an API to get the current scheduler queue? I don't see one...
<hkaiser> yes
<hkaiser> hpx::threads::get_thread_um() or somesuch, give me a sec
<jaafar> I ask because the relevant data member doesn't seem to get exported in any manner from the scheduler
<hkaiser> jaafar: hpx::get_worker_thread_num()
<hkaiser> #include <hpx/modules/threading_base.hpp>
<gnikunj[m]> hkaiser: finally accessed! Let me work on the PR today. I'm really sorry, I couldn't work on it earlier :/
<hkaiser> gnikunj[m]: no need, really
<hkaiser> I can do the rest, just wanted to make sure to have your latest version
<gnikunj[m]> are you sure?
<hkaiser> gnikunj[m]: I need to report on something in Jan to Keita ;-)
<hkaiser> yah, sure
<hkaiser> no worries
<gnikunj[m]> I do have everything updated there
<hkaiser> ok, thanks
<gnikunj[m]> just need to add cmake configurations to build them successfully, that's all
<hkaiser> I hope you don't mind me doing this
<gnikunj[m]> no no. But do let me know if you want me to update it. I don't want you to get busy doing trivial cmake stuff
<gnikunj[m]> hkaiser: everything got deleted on Loni from my /work directory. Do I not have any allocations now? (srun, sbatch etc. are not working either)
<gnikunj[m]> I'm pretty sure that the PR has everything updated, but I wanted to check with my hpx installation in loni. Now I can't :/
<hkaiser> ok
<hkaiser> K-ballo: ?
<hkaiser> ahh, excellent catch
<K-ballo> HPX_CAPTURE_FORWARD moving from an lvalue
<K-ballo> won't make a different for forwarding references anyways, which is hopefully the only thing people are using this on
<hkaiser> thanks, I'll fix it
<hkaiser> actually, we don't need that anymore - can be replaced
<hkaiser> K-ballo: I think this is not used anywhere anymore anyways
<K-ballo> good
<hkaiser> fixed it anyways: #5091
surbhi has quit [Ping timeout: 272 seconds]