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/
K-ballo has quit [Quit: K-ballo]
hkaiser_ has quit [Quit: bye]
shahrzad has quit [Quit: Leaving]
diehlpk_work_ has joined #ste||ar
diehlpk_work has quit [Ping timeout: 240 seconds]
hkaiser has joined #ste||ar
K-ballo has joined #ste||ar
jehelset has joined #ste||ar
hkaiser has quit [Read error: Connection reset by peer]
hkaiser has joined #ste||ar
<heller1>
diehlpk_work you should unload darshan before building/running hpx. Just causing unnecessary problems without real gain for us
<ms[m]>
hkaiser: yt? do you understand what's triggered the bugprone-branch-clone warnings/errors? surely we don't want to mark all uses of the logging macros with that... :/
<hkaiser>
ms[m]: clang-tidy checks only the touched sources...
<hkaiser>
but yah, those warnings are silly
<ms[m]>
really? we have it enabled through cmake...
<hkaiser>
no worries, I'll mark everything as needed
<hkaiser>
ms[m]: that was my assumption as otherwise I wouldn't be able to explain why it starts reporting those things occasionally
<ms[m]>
hmmm...
<hkaiser>
ms[m]: #5099 looks ok now - it's good to go, I think
<ms[m]>
hkaiser: thanks for catching my silly mistakes!
<hkaiser>
no worries - you don't care about Windows, I do
<ms[m]>
hkaiser: btw, I've merged everything that I planned on including in the release, so don't be afraid to merge post-1.6.0 prs
<hkaiser>
ok, thanks - I have the perf counter PR and the components module PR on my list
<hkaiser>
#5152 and #5153
<ms[m]>
>thum
<hkaiser>
#5150 is ok as well
<ms[m]>
👍️
<ms[m]>
yeah, I think 5150 still has some deprecation warnings that need to be ignored or fixed, otherwise ready
<hkaiser>
btw, for #4994: I think I will need to generate new a vcpkg repository compiled with c++17, still trying to figure that out
<hkaiser>
ok, I'll look into the deprecation warnings, didnt see them
<ms[m]>
5150 as in the remove PR? maybe gonidelis[m] can have a look? ;)
<hkaiser>
nod, that's an alternative option ;-)
<hkaiser>
I'll talk to him later today anyways
<ms[m]>
for 4994 I was wondering if we can just build the vcpkg dependencies on ci? and then I'll finally look into setting up that caching on gh actions...?
<hkaiser>
ms[m]: that takes too long and makes the builder time out - been there...
<ms[m]>
ugh, that bad? :/
<ms[m]>
boost?
<hkaiser>
builds all of boost...
<ms[m]>
and there's no header-only option for it?
<hkaiser>
or I need to try narrowing the boost part down to a minimum
<hkaiser>
every boost library os a separate target in vcpkg, I was too lazy to select what we need only...
<hkaiser>
is*
<hkaiser>
but sure, this would be a viable option as well
<ms[m]>
ah, at least they have that option, which is nice :D
<hkaiser>
btw, any idea why the asio pr fails on circleci claiming it can't find the asio/* headers?
<srinivasyadav227>
hey hkaiser: I was following Recorded lectures/sessions provided in Docs, John Biddiscombe mentioned in 2016 HPX workshop at cscs talk that you have also given a session on HPX, so if yes could you please share ?
<ms[m]>
link? which builder fails?
<ms[m]>
sorry, circleci... reading too quickly
<hkaiser>
that's in addition to the clang-tidy problems
<ms[m]>
I'll have a look
<hkaiser>
thanks
<ms[m]>
hkaiser: pushed to the asio branch, I think the problem was that the checked out asio wasn't copied over to the next workflow step
<ms[m]>
should we reuse the now removed external directory to checkout apex and asio in there?
<srinivasyadav227>
hkaiser: I got It anyway 2019 c++ conference talk..Thanks :)
<hkaiser>
ms[m]: yes, that is a good idea, alternatively we could switch to using the Fetch* cmake functionality
<hkaiser>
that would put things under the build dir
<hkaiser>
srinivasyadav227: what did you get?
<srinivasyadav227>
hkaiser: your talk on YouTube related to HPX
<hkaiser>
ahh
<ms[m]>
hkaiser: true, that'd probably be even better
<hkaiser>
ms[m]: but for now, sure
<gonidelis[m]>
ms[m]: hkaiser i can have #5150 ready by today... although i think some of the deprecation warnings come from the use of lambda
<ms[m]>
gonidelis[m]: no rush
bita has joined #ste||ar
hkaiser has quit [Ping timeout: 264 seconds]
hkaiser has joined #ste||ar
<k-ballo[m]>
please don't clone stuff into the source directory
<gonidelis[m]>
k-ballo: ?
<hkaiser>
K-ballo: we already fixed that
<ms[m]>
thanks hkaiser for changing apex to use fetchcontent
<hkaiser>
ms[m]: I just copied your asio code -)
<ms[m]>
for those wondering about clang-format failing on their prs, the annotated function pr wasn't properly formatted
<ms[m]>
I've fixed that on master
<ms[m]>
hkaiser: :P
<k-ballo[m]>
if you specify a tag github will let you do a shallow clone
bita_ has joined #ste||ar
bita has quit [Ping timeout: 264 seconds]
<ms[m]>
k-ballo[m]: do you mean fetchcontent will automatically do a shallow clone in that case or there's an option to do it?
<k-ballo[m]>
there's an option, GIT_SHALLOW ON, that only works with a tag (not hash)
<gonidelis[m]>
if I have pushed commit A on my PR and then pushed commit B on the same PR, can I ammend some changes on commit A?
bita_ has quit [Read error: Connection reset by peer]
weilewei has joined #ste||ar
bita has joined #ste||ar
diehlpk_work_ has quit [Remote host closed the connection]
diehlpk_work has joined #ste||ar
<ms[m]>
gonidelis[m]: look up git fixup
<diehlpk_work>
gonidelis[m], gnikunj[m] How is the GSoC writing going?
<gonidelis[m]>
ms[m]: thanks
<gonidelis[m]>
diehlpk_work: nikunj will be out until monday. i am working on a catalog of "10 hints for getting acquainted with HPX" he could add to that by tuesday i reckon
<diehlpk_work>
gonidelis[m], Thanks for the update
<gonidelis[m]>
diehlpk_work: ;)
weilewei has quit [Quit: Connection closed]
jehelset has quit [Remote host closed the connection]