01:11 <github> [hpx] hkaiser force-pushed client_base_registration from 5741e55 to f38b014: https://git.io/vFJ7a

01:11 <github> hpx/client_base_registration f38b014 Hartmut Kaiser: Making sure client_base instance that registered the component does not unregister it when being destructed

01:18 <github> [hpx] hkaiser pushed 1 new commit to master: https://git.io/vFU9Y

01:18 <github> hpx/master 1139481 Hartmut Kaiser: Fix inspect error

01:19 <github> [hpx] hkaiser pushed 2 new commits to master: https://git.io/vFU93

01:19 <github> hpx/master 55c20cd Mikael Simberg: Fix small typos

01:19 <github> hpx/master b6362c2 Hartmut Kaiser: Merge pull request #2973 from msimberg/typo-fixes...

09:06 <github> [hpx] StellarBot pushed 1 new commit to gh-pages: https://git.io/vFTYu

09:06 <github> hpx/gh-pages 9f557a3 StellarBot: Updating docs

12:01 <heller> hkaiser: jbjnr: did you see Rolfs Mail?

12:02 <hkaiser> no

12:02 <hkaiser> heller: forward, pls

12:03 <heller> Regarding next hpx course in Stuttgart

12:03 <heller> You should have gotten it

12:03 <hkaiser> have not seen it :/

12:03 <heller> Ok

12:03 <heller> Forwarded it

12:04 <hkaiser> got it

12:04 <heller> March 15-16

12:04 <heller> I'd be on board

12:04 <hkaiser> I'm in Japan right before that

12:04 <hkaiser> doubt that I will make it

12:05 <heller> Oh

12:05 <heller> Vacation in Japan?

12:05 <hkaiser> no, conference

12:06 <jbjnr> hkaiser: heller HLRS course. I'll clear it with my boss and get back to you

12:07 <heller> Ok

12:07 <jbjnr> Also - I am prepared to take over as GSoC admin for HPX if diehlpk_work does not want to continue doing it - he suggested at some point we do it together this yeat to ensure I know what to do.

12:07 <heller> Sure

12:07 <hkaiser> flag into your hand - go for it

12:08 <heller> Me and Hartmut used to share the tasks, then Patrick took over

12:08 <heller> Now it's your job :p

12:08 <jbjnr> I like the T-Shirts and stickers :)

12:11 <heller> I only got one sticker!

12:11 <hkaiser> heller: I can give you the one I received

12:11 <hkaiser> ;)

12:11 <jbjnr> very po this year indeed. and it was the same as last year's one

12:11 <jbjnr> ^poor

12:11 <heller> If you're listed on the portal, you get the free stuff :p

12:12 <heller> You don't have to do anything ;)

12:12 <jbjnr> did you not mentor anyone?

12:12 <hkaiser> shhh, don't tell him

12:12 <jbjnr> rotten swizz

12:58 <github> [hpx] biddisco force-pushed alloc_membind from 6524d7c to d9b91d1: https://git.io/vdnl9

12:58 <github> hpx/alloc_membind e10c802 John Biddiscombe: Add hwloc alloc_membind support and nodeset bitmap type...

12:58 <github> hpx/alloc_membind 513a4a4 John Biddiscombe: Add ostream operator for hwloc bitmap printout

12:58 <github> hpx/alloc_membind d51fbe5 John Biddiscombe: Move hwloc_bitmap_ptr to threads namespace

12:59 <jbjnr> dashboard looking much greener at the moment.

13:01 <jbjnr> K-ballo: just fyi - after cleaning up a bit - the only consts I'm touching are here https://github.com/biddisco/hpx/commit/0a8f4bf8871cb4802d3b9a3e99356ccdc821ff9b

13:02 <jbjnr> still not certain if they are the right ones, but code cimpilaes and runs ok now

13:12 <hkaiser> jbjnr: looks good - I'd forward<>() the executor however if it's being passed in using a forwarding ref

13:15 <jbjnr> good point. missed that.

15:49 <zbyerly> I have a question that might shock some people:

15:49 <zbyerly> can I create an arbitrarily large number of localities on a single node?

15:49 <zbyerly> or am I limited by the number of cores

17:01 <hkaiser> zbyerly: you can create as many localities on a node as you wish, it will just run slow as hell

17:02 <hkaiser> in general I'd suggest to make sure that the sum of all -tN for all localities on a node does not exceed the number of available cores

20:52 <jbjnr> hkaiser: here's a stupid question for you : the then_execute function is templated over the predecessor and Ts...&&args - but the .then only ever takes the predecessor as an argument. The Ts...args are never used. Are they there just for completeness - or is it allowed to say .then(future, more args) and add args to the continuation?

20:53 <hkaiser> the args are there for symmetry with all the other executor API

20:53 <jbjnr> ok. I assumed as much