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/
hkaiser[[m]] has quit [Ping timeout: 256 seconds]
hkaiser has joined #ste||ar
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
gedaj has joined #ste||ar
diehlpk has joined #ste||ar
diehlpk has quit [Ping timeout: 268 seconds]
gedaj has quit [Quit: Leaving]
gedaj has joined #ste||ar
hkaiser has quit [Quit: bye]
gedaj has quit [Remote host closed the connection]
gedaj has joined #ste||ar
diehlpk has joined #ste||ar
eschnett has quit [Quit: eschnett]
eschnett has joined #ste||ar
K-ballo has quit [Quit: K-ballo]
EverYoung has joined #ste||ar
EverYoung has quit [Ping timeout: 265 seconds]
daissgr has joined #ste||ar
daissgr has quit [Ping timeout: 256 seconds]
diehlpk has quit [Ping timeout: 248 seconds]
mcopik has quit [Ping timeout: 256 seconds]
mcopik has joined #ste||ar
nanashi55 has quit [Ping timeout: 268 seconds]
Anushi1998 has left #ste||ar ["AndroIRC"]
Anushi1998 has joined #ste||ar
nanashi55 has joined #ste||ar
Nikunj_ has joined #ste||ar
gedaj_ has joined #ste||ar
gedaj has quit [Read error: Connection reset by peer]
parsa has quit [Quit: Zzzzzzzzzzzz]
parsa has joined #ste||ar
parsa has quit [Client Quit]
gedaj__ has joined #ste||ar
gedaj_ has quit [Read error: Connection reset by peer]
Nikunj_ has quit [Quit: Page closed]
hkaiser[m] has joined #ste||ar
Nikunj_ has joined #ste||ar
jakub_golinowski has joined #ste||ar
Nikunj_ has quit [Quit: Page closed]
Nikunj_ has joined #ste||ar
anushi has joined #ste||ar
Nikunj_ has quit [Ping timeout: 260 seconds]
Nikunj_ has joined #ste||ar
david_pfander has joined #ste||ar
anushi has quit [Remote host closed the connection]
anushi has joined #ste||ar
anushi has quit [Remote host closed the connection]
anushi has joined #ste||ar
hkaiser[m] has quit [Read error: Connection reset by peer]
anushi has quit [Remote host closed the connection]
anushi has joined #ste||ar
gedaj_ has joined #ste||ar
gedaj__ has quit [Remote host closed the connection]
Nikunj_ has quit [Ping timeout: 260 seconds]
anushi_ has joined #ste||ar
anushi has quit [Ping timeout: 245 seconds]
anushi_ has quit [Ping timeout: 245 seconds]
anushi has joined #ste||ar
Anushi1998 has quit [Remote host closed the connection]
<hkaiser[m]>
jakub_golinowski feel free to add text to this, please
anushi has quit [Ping timeout: 276 seconds]
<jakub_golinowski>
hkaiser: I can gladly do it, however I am not yet sure what exactly the motivation is. From the boost b2 help message I can see that using build-type=complete means that all the possible variations of library versions will be built (as opposed to only a default library versions for built-type=minimal). So my guess is that HPX is using different version of boost libraries depending on the situation?
anushi has joined #ste||ar
hkaiser[m] has quit [Ping timeout: 256 seconds]
hkaiser[m] has joined #ste||ar
hkaiser[[m]] has joined #ste||ar
hkaiser[m] has quit [Read error: Connection reset by peer]
K-ballo has joined #ste||ar
hkaiser[[m]] has quit [Ping timeout: 265 seconds]
daissgr has joined #ste||ar
jakub_golinowski has quit [Ping timeout: 265 seconds]
jakub_golinowski has joined #ste||ar
quaz0r has joined #ste||ar
daissgr has quit [Ping timeout: 252 seconds]
nikunj has joined #ste||ar
daissgr has joined #ste||ar
Anushi1998 has joined #ste||ar
<simbergm>
jbjnr: another stupid question for cholesky: do you compile with Release or RelWithDebInfo? I seem to get a 50 % performance hit at least with blocksize 64 and RelWithDebInfo
anushi has quit [Read error: Connection reset by peer]
<simbergm>
another thing is the gemm implementation changes when you go below 128 blocksize, do you know that parsec and hpx are calling the same implementation?
EverYoung has joined #ste||ar
EverYoung has quit [Ping timeout: 252 seconds]
hkaiser[[m]] has joined #ste||ar
hkaiser[[m]] has quit [Ping timeout: 240 seconds]
parsa has joined #ste||ar
hkaiser has joined #ste||ar
hkaiser has quit [Ping timeout: 256 seconds]
diehlpk_work has joined #ste||ar
<diehlpk_work>
To all our GSoC students, I have seen few proposals with a time line without any dates or duration of the phases. Please check that your time line has dates before you send your proposal to us
* zao
takes this opportunity to re-read »Falsehoods programmers believe about time»
<diehlpk_work>
Also very important is that you describe what you do in details. For ewxample writing examples it not enough. We like to know which examples and what does each of them show.
<diehlpk_work>
And when sharing the proposal, please allow us to edit it. So we can provide you with comments
eschnett has quit [Quit: eschnett]
<nikunj>
diehlpk_work: Thank you for sharing the shortcomings and providing with your valuable feedback. I will ensure everything is explained in detail. About timeline, I did not include it since I am still not sure of dividing my work according to week and still analysing about the work.
<nikunj>
I will be including the timeline in my actual proposal, which again I will be sharing so as to receive feedback so that I can brush up my proposal.
<diehlpk_work>
nikunj, Good, we need to see the time line to may suggest you that you will need more time for a specific task
<github>
[hpx] msimberg opened pull request #3211: Use preinit_array to set argc/argv in init_globally example (master...msimberg-patch-1) https://git.io/vAQSA
<simbergm>
nikunj: would you mind trying the above please?
<nikunj>
simbergm: Ok, I will make the above changes if it changes stuff.
<simbergm>
nikunj: not sure if I misunderstood but I meant the PR, #3211, it should fix the init_globally example
mbremer has joined #ste||ar
<simbergm>
no need to make changes for that
<nikunj>
simbergm: I meant changing the init_globally file locally ( i.e. to include the changes you made in the file ) and then trying to run it
<simbergm>
nikunj: okay, perfect, no misunderstanding then (you can also checkout the branch, but feel free to do it whatever way is easiest)
twwright_ has joined #ste||ar
hkaiser has joined #ste||ar
eschnett has joined #ste||ar
twwright has quit [Ping timeout: 276 seconds]
twwright_ is now known as twwright
<github>
[hpx] msimberg opened pull request #3212: Fix check in sequenced_executor test (master...msimberg-patch-2) https://git.io/vAQQw
<mbremer>
Hi, would some one mind helping me clear up some confusion about the AGAS layer?
<mbremer>
In particular, I'm confused about what happens to the id_type and gid_type when a component is migrated. I saw in the unit test, that the client for a migrated compnent needs to be reinstantiated, but what does that mean for remote clients of that component?
<hkaiser>
mbremer: what should happen to the id?
<mbremer>
I'm unsure. I was hoping nothing!
<hkaiser>
you're right, nothing
<mbremer>
I guess, I was just confused, why the client gets remade especially if it's just a wrapper for the id_type.
<hkaiser>
you can thin of an id as a global shared_ptr that does not change if the target is moved to another node
<mbremer>
Ah great! That's what I was hoping it would be.
<mbremer>
Kk thanks! That clears that up for me.
<hkaiser>
mbremer: I know migrate returns an id_type, but you can ignore that
<hkaiser>
it's the same as before
<hkaiser>
I meant the returned id is the same as the one passed to migrate
<mbremer>
That was the source of confusion. And then presumably all other clients, will also have their actions rerouted to the correct location, right?
<hkaiser>
right
<mbremer>
Great! Lol I had a mini existential crisis over the point of AGAS over the weekend. But now all's well again!
<zao>
Doubting the fabric of space and time? Sounds like you're well on your way to be a proper HPX developer :)
<hkaiser>
lol
hkaiser_ has joined #ste||ar
aserio has joined #ste||ar
hkaiser has quit [Ping timeout: 256 seconds]
nikunj has quit [Ping timeout: 260 seconds]
hkaiser_ has quit [Client Quit]
<github>
[hpx] msimberg opened pull request #3213: WIP: Relax locks for thread suspension to avoid holding locks when yielding (master...suspension-locks) https://git.io/vAQbx
smrk007 has joined #ste||ar
bibek has joined #ste||ar
Anushi1998 has quit [Remote host closed the connection]
Anushi1998 has joined #ste||ar
aserio1 has joined #ste||ar
aserio has quit [Ping timeout: 255 seconds]
aserio1 is now known as aserio
aserio has quit [Ping timeout: 240 seconds]
simbergm has quit [Ping timeout: 240 seconds]
smrk007 has quit [Ping timeout: 260 seconds]
nikunj has joined #ste||ar
EverYoung has joined #ste||ar
nikunj has quit [Ping timeout: 260 seconds]
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
nikunj has joined #ste||ar
nikunj has quit [Ping timeout: 260 seconds]
mbremer has quit [Ping timeout: 260 seconds]
EverYoung has quit [Ping timeout: 240 seconds]
parsa has joined #ste||ar
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
nikunj has joined #ste||ar
<nikunj>
msimberg: Unfortunately, the output of init_globally does not change and throws exactly the same error as before.
diehlpk has joined #ste||ar
diehlpk has quit [Ping timeout: 245 seconds]
diehlpk has joined #ste||ar
diehlpk has quit [Remote host closed the connection]
diehlpk has joined #ste||ar
jaafar has quit [Quit: Konversation terminated!]
jaafar has joined #ste||ar
jaafar has quit [Client Quit]
jaafar has joined #ste||ar
david_pfander has quit [Remote host closed the connection]
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
diehlpk has quit [Ping timeout: 260 seconds]
diehlpk has joined #ste||ar
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 276 seconds]
victor_ludorum has joined #ste||ar
<nikunj>
msimberg: It does work, it was a bit of misunderstanding. I ran the same file twice :(
david_pfander has joined #ste||ar
aserio has joined #ste||ar
diehlpk has quit [Ping timeout: 276 seconds]
hkaiser[m] has joined #ste||ar
<jakub_golinowski>
I want to do the pull request by executing following steps: 1. cloned the master branch to local. 2. Created local branch 3. want to push local branch to github 4. request merging the branch to master
<jakub_golinowski>
However, I cannot do the step 3. because I get the following error from git:
<diehlpk_work>
You can not work direclty on hpx, you have to fork hpx and work on your local fork and do a pr from there
<jakub_golinowski>
Ok, thank you very much, this what I suspected
<diehlpk_work>
You are welcome, only some people work directly on the hpx repo
hkaiser[m] has quit [Read error: Connection refused]
hkaiser has joined #ste||ar
<hkaiser>
aserio: yt?
<aserio>
hkaiser: yes
<hkaiser>
see pm, pls
<github>
[hpx] Jakub-Golinowski opened pull request #3214: Changed prerequisites.qbk to avoid confusion while building boost. (master...master) https://git.io/vA7zU
eschnett has quit [Quit: eschnett]
Anushi1998 has quit [Ping timeout: 265 seconds]
Anushi1998 has joined #ste||ar
<zao>
I see it as a good thing that you need a separate repo, harder to accidentally force-push master then ;)
<zao>
jakub_golinowski: I strongly recommend that you make PRs from one of your branches.
<zao>
Otherwise, additional work in your 'master' branch will keep advancing the PR.
<jakub_golinowski>
zao: yes I agree that a separate repository is a good thing I could also do another safety check in browser, after doing the push. About doing PRs from branches - I was just thinking about how am I going to keep my fork's master up to date when new commits come before my PR is merged - you provided me with solution
<zao>
Did you fix HPX yet to run on current 2017? :P
<hkaiser>
zao: did it have problems on VS2017?
<hkaiser>
works for me (tm)
<zao>
My workflow with PRs is to fork and keep my master up to speed with upstream master.
<zao>
hkaiser: Note that 15.6 just released, with a lot of changes.
<hkaiser>
will try
<zao>
I've been on its Preview for a while, running into K-ballo-esque horrors with HPX and/or Boost.
<hkaiser>
that's bad
<hkaiser>
I think there is one problem with variadic expansion I ran into yesterday, but that was it - easily fixed...
<zao>
(workflow): Then do my work in a named branch, rebase it against master when it changes meaningfully and I want the current upstream master in there.
<zao>
Works for me as I'm working alone on the branch and as such, can do the rebase stuff. I guess you other people do merges of master in when it feels right.
<zao>
(I found that junk in history when trying to grab commits properly for CI :P)
<zao>
I've been running my hello_world tests with 2015, for example, as I couldn't get a proper build done with 2017.
<zao>
(couldn't figure out how to get CMake to not pick up my secondary install)
<K-ballo>
"K-ballo-esque horrors", not sure if good or bad...
<hkaiser>
stupid pattern matching - they might have deleted part of their database as this worked before...
<hkaiser>
zao: moving the '0,' to the end of the array makes it work
<jakub_golinowski>
zao: (workflows) thank you very much for your advice I think your workflow is clean and makes good use of the github - my next PR will be done in this manner (because I think this one is easy enough that will require very little work (if any) and will quickly be merged)
david_pfander has quit [Ping timeout: 256 seconds]
gablab has joined #ste||ar
<gablab>
hkaiser could I ask you a question about prefetching distance?
<hkaiser>
sure
<gablab>
I send an email to zarha but she didn't answer yet
<zao>
So '0)..., 0};' ?
<hkaiser>
zao: yes
<gablab>
From our last discussion, I concluded that chunk size could not be treated as a continuous variable since you can't have 5.2 iterations
<hkaiser>
right
<gablab>
But what about prefectching distance, I assume it can take any integer value
<gablab>
321,504,2345
<hkaiser>
yah, but large values don't make any sense
<zao>
Bleh... can't get my clang-format to agree with the existing formatting.
gablab has quit [Ping timeout: 240 seconds]
<github>
[hpx] zao opened pull request #3216: Reorder a variadic array to satisfy VS 2017 15.6 (master...fix-vs-15.6-variadic-array) https://git.io/vA7in