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/
mcopik has quit [Ping timeout: 240 seconds]
jaafar has quit [Ping timeout: 246 seconds]
ajaivgeorge has quit [Ping timeout: 248 seconds]
ajaivgeorge has joined #ste||ar
EverYoung has joined #ste||ar
eschnett has joined #ste||ar
EverYoung has quit [Ping timeout: 246 seconds]
K-ballo has quit [Quit: K-ballo]
hkaiser has quit [Quit: bye]
bikineev has quit [Remote host closed the connection]
bikineev has joined #ste||ar
bikineev has quit [Remote host closed the connection]
vamatya has joined #ste||ar
eschnett has quit [Quit: eschnett]
eschnett has joined #ste||ar
bikineev has joined #ste||ar
ajaivgeorge has quit [Ping timeout: 240 seconds]
ajaivgeorge has joined #ste||ar
vamatya_ has joined #ste||ar
bikineev has quit [Remote host closed the connection]
vamatya has quit [Ping timeout: 248 seconds]
bikineev has joined #ste||ar
ajaivgeorge has quit [Ping timeout: 255 seconds]
ajaivgeorge has joined #ste||ar
heller has quit [*.net *.split]
heller__ has joined #ste||ar
vamatya__ has joined #ste||ar
vamatya_ has quit [Ping timeout: 248 seconds]
EverYoung has joined #ste||ar
vamatya__ has quit [Quit: Leaving]
EverYoung has quit [Ping timeout: 246 seconds]
jaafar has joined #ste||ar
jaafar has quit [Client Quit]
ajaivgeorge has quit [Remote host closed the connection]
ajaivgeorge has joined #ste||ar
ajaivgeorge_ has joined #ste||ar
ajaivgeorge has quit [Ping timeout: 252 seconds]
david_pfander has joined #ste||ar
david_pfander has quit [Remote host closed the connection]
david_pfander has joined #ste||ar
bikineev has quit [Remote host closed the connection]
quaz0r has quit [Ping timeout: 248 seconds]
bikineev has joined #ste||ar
quaz0r has joined #ste||ar
bikineev has quit [Ping timeout: 240 seconds]
bikineev has joined #ste||ar
EverYoung has joined #ste||ar
EverYoung has quit [Ping timeout: 255 seconds]
bikineev has quit [Remote host closed the connection]
bikineev has joined #ste||ar
mcopik has joined #ste||ar
bikineev has quit [Remote host closed the connection]
<github> [hpx] StellarBot pushed 1 new commit to gh-pages: https://git.io/v5IgF
<github> hpx/gh-pages 405fdbc StellarBot: Updating docs
bikineev has joined #ste||ar
hkaiser has joined #ste||ar
K-ballo has joined #ste||ar
pree has joined #ste||ar
eschnett has quit [Quit: eschnett]
<github> [hpx] hkaiser pushed 1 new commit to master: https://git.io/v5IP4
<github> hpx/master 7b6a69d Hartmut Kaiser: Circumventing VS2015 warnings
<hkaiser> heller__: would you mind fixing your PR so it can be merged?
eschnett has joined #ste||ar
pree has quit [Read error: Connection reset by peer]
<github> [hpx] hkaiser closed pull request #2854: Support ARM64 in timestamps (master...master) https://git.io/v5U2O
pree has joined #ste||ar
bikineev has quit [Ping timeout: 240 seconds]
hkaiser has quit [Quit: bye]
aserio has joined #ste||ar
bikineev has joined #ste||ar
bikineev_ has joined #ste||ar
bikineev has quit [Ping timeout: 248 seconds]
pree has quit [Read error: Connection reset by peer]
hkaiser has joined #ste||ar
hkaiser has quit [Read error: Connection reset by peer]
hkaiser has joined #ste||ar
pree has joined #ste||ar
pree has quit [Ping timeout: 248 seconds]
bikineev_ has quit [Ping timeout: 240 seconds]
bikineev has joined #ste||ar
hkaiser has quit [Quit: bye]
aserio has quit [Ping timeout: 246 seconds]
aserio has joined #ste||ar
hkaiser has joined #ste||ar
jfbastien has joined #ste||ar
pree has joined #ste||ar
wash has joined #ste||ar
<wash> aserio: can't do today, sorry, got a meeting
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
david_pfander has quit [Ping timeout: 246 seconds]
bikineev has quit [Ping timeout: 260 seconds]
ajaivgeorge_ has quit [Quit: ajaivgeorge_]
jaafar has joined #ste||ar
hkaiser has quit [Quit: bye]
<aserio> wash: Ok
aserio has quit [Ping timeout: 240 seconds]
bikineev has joined #ste||ar
EverYoung has quit [Ping timeout: 246 seconds]
EverYoung has joined #ste||ar
mcopik has quit [Ping timeout: 260 seconds]
bikineev has quit [Ping timeout: 246 seconds]
bikineev has joined #ste||ar
<thundergroudon[m> Hi guys
<thundergroudon[m> can I know how to get multi node gpu cluster on Rostam?
<thundergroudon[m> cuda compatible
<thundergroudon[m> Bahram, reno and tycho don't seem to have multiple node configuration available
<thundergroudon[m> [mseshadri@bob1 cuda]$ srun -p bahram -N 2 ./buffer_cuda_test
<thundergroudon[m> srun: error: Unable to allocate resources: Requested node configuration is not available
bikineev has quit [Read error: Connection reset by peer]
<zbyerly> -p cuda
<zbyerly> ?
<zbyerly> the partition "bahram" is only the node bahram
<zbyerly> thundergroudon[m, ^
bikineev has joined #ste||ar
<thundergroudon[m> oh!
<thundergroudon[m> I understand
<thundergroudon[m> I should have called generically
<thundergroudon[m> my bad
<thundergroudon[m> zbyerly: Thanks
<diehlpk_work> thundergroudon[m, Cuda: Three nodes (Bahram, Reno, Tycho)
<thundergroudon[m> I was calling it in the wrong way diehlpk_work
<thundergroudon[m> Got the mistake now
<zbyerly> it feels good to be helpful
Matombo has joined #ste||ar
hkaiser has joined #ste||ar
hkaiser has quit [Read error: Connection reset by peer]
hkaiser has joined #ste||ar
bikineev has quit [Ping timeout: 260 seconds]
<github> [hpx] hkaiser force-pushed inspect_assert from feed278 to 04ad2dc: https://git.io/v5LEL
<github> hpx/inspect_assert 04ad2dc Hartmut Kaiser: Deprecate use of BOOST_ASSERT and ensure HPX_ASSERT has corresponding #include present
bikineev has joined #ste||ar
mcopik has joined #ste||ar
stevenrbrandt has joined #ste||ar
<stevenrbrandt> Hello, I'm wondering if parallel::task has been replaced.
<stevenrbrandt> My code is: auto f = parallel::transform ( parallel::par (parallel::task), begin(v), end(v), begin(v), [](int i) -> int { return i+1; });
<pree> What's the difference between fixed components and simple components ?
<stevenrbrandt> my error is In file included from input_line_12:1: /home/jup/.cmd_jup_9.cpp:7:28: error: reference to 'task' is ambiguous parallel::par (parallel::task), ~~~~~~~~~~^ /usr/include/hpx/parallel/executors/executor_traits.hpp:45:63: note: candidate found by name lookup is 'hpx::parallel::v3::task' HPX_STATIC_CONSTEXPR parallel::execution::task_policy_tag task{};
<pree> stevenrbrandt : let me wait : )
<pree> Did fixed component cannot migratable ?
<hkaiser> stevenrbrandt: parallel::execution::task
<pree> ^^
<stevenrbrandt> thanks
<hkaiser> pree: fixed components are nothing you need to worry about, they are used by deep hpx innards only
<pree> hkaiser : Thanks just asked
hkaiser has quit [Quit: bye]
<github> [hpx] sithhell force-pushed fix_ubsan from 1ba33f7 to 83bcd87: https://git.io/v5Tol
<github> hpx/fix_ubsan 83bcd87 Thomas Heller: Switching simple_component to be component...
aserio has joined #ste||ar
denis_blank has joined #ste||ar
<github> [hpx] hkaiser pushed 1 new commit to master: https://git.io/v5LKO
<github> hpx/master 6c214b2 Hartmut Kaiser: Merge pull request #2789 from STEllAR-GROUP/resource_partitioner...
<heller__> how much will break?
heller__ is now known as heller
hkaiser has joined #ste||ar
<hkaiser> K-ballo: just a heads-up: I just merged the resource_partitioner branch
hkaiser has quit [Client Quit]
aserio has quit [Ping timeout: 246 seconds]
<K-ballo> cool
bikineev has quit [Remote host closed the connection]
denis_blank has quit [Quit: denis_blank]
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 246 seconds]
bikineev has joined #ste||ar
aserio has joined #ste||ar
<aserio> diehlpk_work: yt?
eschnett has quit [Quit: eschnett]
bikineev has quit [Remote host closed the connection]
<diehlpk_work> aserio, Yes
bikineev has joined #ste||ar
<aserio> diehlpk_work: Have we talked about how we want the GSoC students to present their final work?
hkaiser has joined #ste||ar
<diehlpk_work> aserio, See pm
denis_blank has joined #ste||ar
<hkaiser> aserio: yt?
<aserio> hkaiser: yep
<hkaiser> see pm
pree has quit [Ping timeout: 260 seconds]
<heller> hkaiser: the component stuff should be fixed as well now
<hkaiser> heller: perfect1
<hkaiser> !
<heller> so let's see
<heller> first, lets see how much the RP stuff breaks
<hkaiser> heller: shouldn't be too bad
<hkaiser> the only change which could cause problems I'm aware of is that -t1 is not the default anymore
<hkaiser> the default is now -tall
<heller> right
<heller> which should be ok with the test runner
<hkaiser> good
<heller> but you never know
eschnett has joined #ste||ar
<hkaiser> heller: btw
<hkaiser> I think we can replace the casts in the serialization flagged by adsan with memcpy calls specifying a fixed length
<hkaiser> afaics, those get optimized properly
<heller> ah, that's perfect
<hkaiser> heller: well for me it's still a small diference, but godbolt.org shows that memcpy(dest, src, 8) get's optimized to an uint64 load/store
<hkaiser> I'll prepare a PR for that
<heller> how do you measure the difference now?
<hkaiser> serialization_overhead
<heller> could you also compare it with a consecutive char loads/stores?
<heller> that test has quite a few parameters ;)
<hkaiser> heller: can try
<hkaiser> I ran serialization_overhead --data_size=100000 --iteratons=10000
<zao> Is the typo intended?
<hkaiser> iterations*
<hkaiser> heller: see https://godbolt.org/g/Q4dqsA
EverYoun_ has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
<heller> hkaiser: you should run with 1 thread to avoid the thread stealing contention ;)
<hkaiser> I ran with thread
<hkaiser> 1 thread
<heller> ok
<heller> yeah, I am aware of that memcpy optimization
aserio has quit [Quit: aserio]
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 246 seconds]
stevenrbrandt has quit [Quit: Page closed]
<diehlpk_work> hkaiser, Do you know if any student failed in 2014 or 2015?
<hkaiser> I don't think anybody failed that year
<hkaiser> but I could be wrong
<diehlpk_work> Ok, need it for my statistics
<hkaiser> heller: when did the websockets guy failed?
<hkaiser> last year?
<hkaiser> diehlpk_work: pls look at the blog post, there should be one
<diehlpk_work> hkaiser, Not for 2014
<diehlpk_work> I found 2015 and there no one failed
EverYoun_ has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
<heller> hkaiser: i think it was last year
<diehlpk_work> 2014 we had just two students
<hkaiser> ok
<heller> diehlpk_work: how did you get the information for the failed students?
<diehlpk_work> 2016, and 2016, I was attending.
<diehlpk_work> 2015 from the blog
<github> [hpx] hkaiser created serialization (+1 new commit): https://git.io/v5Lxm
<github> hpx/serialization 83445a1 Hartmut Kaiser: Replace casts/assignments with hard-coded memcpy operations
<hkaiser> heller: ^^
<heller> hkaiser: whats the performance impact of that for you know?
<diehlpk_work> As long we trust the blog and google archive we are fine :)
<hkaiser> ~10%
<hkaiser> varying for different scenarios, obviously
EverYoun_ has joined #ste||ar
EverYoun_ has quit [Remote host closed the connection]
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 276 seconds]
<hkaiser> heller: btw, your PR still has conflicts and inspect errors
<hkaiser> and you wanted to revert the formatting change ;)
<heller> hkaiser: I didn't want to revert the formating change :P
<heller> conflicts?
<heller> oh
<hkaiser> sure you do
<hkaiser> you said you wouldn't have any problems doing so
<heller> stupid thing
<heller> if you insist ;)
<heller> let's see
<heller> I am seeing a appveyor failure
<hkaiser> that's because of the conflict
<heller> ahh
<hkaiser> appveyor runs test on the merged PR
<heller> ok
<heller> hkaiser: so you insist on reverting the formatting change?
<heller> why?
<github> [hpx] sithhell pushed 1 new commit to fix_ubsan: https://git.io/v5LpF
<github> hpx/fix_ubsan 76db2d9 Thomas Heller: Merge branch 'master' into fix_ubsan...
<hkaiser> heller: because we agreed not to reformat whole files just because
<heller> it wasn't reformatted just because
<hkaiser> sure
<heller> ahh, I see the problem now
<heller> which version of clang-format do you use, btw?
<hkaiser> nsec
<hkaiser> sec
<hkaiser> 6.0.0.310580
<heller> ok
<heller> I am using 4.0.1
<heller> no wonder I am getting errors about unknown keys
<hkaiser> the .clang-format might not work
<heller> yes :/
<hkaiser> we should add a comment in that file about the required version
<heller> and for the record: the formatting was triggered because the content of the file changed
<hkaiser> all of it?
<heller> I am not sure how git clang-format decides what to actually reformat
<heller> it normally only affects the diff
<hkaiser> I don't see a change (except formatting, but I might miss something
<heller> from that diff you complain about, reformatting the diff affected the whole file because the indentation changed
<heller> simple_component_tag -> component_tag
<hkaiser> k
<hkaiser> heller: the best would be not to automatically reformat things
<hkaiser> to avoid such massive unrelated changes
<heller> how do you use clang-format?
<hkaiser> manually
<hkaiser> through it's integration with VS
<heller> I think whenever indentation changes at this scope, it is unavoidable that the changeset in the end is massive
<hkaiser> so don't do the change ;)
<heller> so have an exception where to use clang-format?
<hkaiser> shrug
<hkaiser> we can add a clause for what you're doing
<hkaiser> i.e. let git decide what to reformat
<heller> I'd say having the stuff run through the tool that you just worked on will bring us to not having those formatting changes anymore sooner than later
<heller> in that specific case, the formatting is wrong anyway
<github> [hpx] sithhell force-pushed fix_ubsan from 76db2d9 to 9fb4e43: https://git.io/v5Tol
<github> hpx/fix_ubsan e7764e3 Thomas Heller: Switching simple_component to be component...
<github> hpx/fix_ubsan 9fb4e43 Thomas Heller: Merge branch 'master' into fix_ubsan3...
<heller> hkaiser: alrighty. please look at the diff now. This is how it was supposed to be :/
<hkaiser> ok
<github> [hpx] K-ballo force-pushed std-atomic-lite from 7a1ff1a to a08615a: https://git.io/v7tLl
<github> hpx/std-atomic-lite e40a003 Agustin K-ballo Berge: Replace boost::atomic with std::atomic (where possible)
<github> hpx/std-atomic-lite a08615a Agustin K-ballo Berge: Add deprecated checks for Boost.Atomic
EverYoung has joined #ste||ar
EverYoun_ has quit [Ping timeout: 276 seconds]
EverYoung has quit [Ping timeout: 276 seconds]
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
bikineev has quit [Remote host closed the connection]