Open in new window / Try shogun cloud
--- Log opened Sat Feb 08 00:00:49 2014
-!- sonne|osx [~sonne@e179077113.adsl.alicedsl.de] has joined #shogun06:38
@sonney2klisitsyn, could you please give us the full error http://buildbot.shogun-toolbox.org/memcheck/20140207-0059.html07:01
@sonney2klisitsyn, otherwise it is too hard to fix issues07:01
shogun-buildbotbuild #44 of osx2 - python is complete: Success [build successful]  Build details are at http://buildbot.shogun-toolbox.org/builders/osx2%20-%20python/builds/4407:31
shogun-buildbotbuild #180 of clang34 - undefined behaviour analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20undefined%20behaviour%20analysis/builds/180  blamelist: Soeren Sonnenburg <sonne@debian.org>07:48
shogun-buildbotbuild #175 of clang34 - thread analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20thread%20analysis/builds/175  blamelist: Soeren Sonnenburg <sonne@debian.org>08:04
shogun-buildbotbuild #185 of debian wheezy - memcheck is complete: Failure [failed memory check]  Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/185  blamelist: Soeren Sonnenburg <sonne@debian.org>08:37
shogun-buildbotbuild #190 of clang34 - static analysis is complete: Failure [failed analyse]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20static%20analysis/builds/190  blamelist: Soeren Sonnenburg <sonne@debian.org>08:46
-!- lambday [67157c4d@gateway/web/freenode/ip.103.21.124.77] has joined #shogun15:05
lambdaysonne|osx: hi!15:11
lambdaysonne|osx: last year I used shogun for subsequent string kernel (for one of my class projects) - which is not in shogun yet.. I never sent a PR for that15:11
lambdaysonne|osx: today I got a mail from someone who queried about that - do you think we can add this?15:12
lambdaylol subsequence*15:18
-!- lambday [67157c4d@gateway/web/freenode/ip.103.21.124.77] has quit []16:17
-!- adrin [~adrin@guest-202.mpi-inf.mpg.de] has quit [Ping timeout: 250 seconds]17:04
-!- adrin [~adrin@guest-90.mpi-inf.mpg.de] has joined #shogun17:12
-!- zxtx [~zv@c-98-223-196-32.hsd1.in.comcast.net] has quit [Ping timeout: 250 seconds]20:20
@sonney2klisitsyn, ping?21:41
lisitsynsonney2k: pong!21:41
@sonney2klisitsyn, can you do anything about the no error shown xslt?21:42
lisitsynsonney2k: yeah just wanted to do that21:42
@sonney2klisitsyn, please do - we really need to get that stuff green before actual gsoc21:42
lisitsynsonney2k: sure will do now21:43
lisitsynsonney2k: the confusing part is that it will be a big error thing21:43
-!- shogun-buildbot [~shogun-bu@7nn.de] has quit [Quit: buildmaster reconfigured: bot disconnecting]21:44
-!- shogun-buildbot [~shogun-bu@7nn.de] has joined #shogun21:44
@sonney2klisitsyn, show me21:45
lisitsynsonney2k: give me half an hour21:45
lisitsyn;)21:45
@sonney2kyou have 5 minutes21:46
@sonney2k(that is what kirk is supposed to say)21:46
@sonney2kok osx bot now works too21:49
@wikingsonney2k: ?21:49
@wikingsonney2k: what bot? :)21:49
@sonney2kwiking, http://buildbot.shogun-toolbox.org/builders/osx2%20-%20python21:49
@wikingah u mean it went green :)21:49
@wikingi thought we've got an osx bot21:49
@wikingkudos for the fix21:50
@wikingsonney2k: what was the problem?21:50
@sonney2kwiking, floating points21:50
@wiking:<21:50
@sonney2kthe way you compute how two floats are similar is far from trivial21:50
@wikingok i owe one fix for the coveralls21:50
@sonney2kwiking, I just reconfigured the buildbot to create warnings not failues for the static analyze stuff21:51
@sonney2kwiking, one question - do you know if the wheezy memcheck will fail if we have only 'potential memory leaks'?21:52
@sonney2kwiking, ohh and good news we have the google cloud grant worth 1k$21:52
@sonney2kwiking, so maybe if you have time you move the stuff from fatbot -> google cloud?21:53
@wikingsonney2k: account?21:54
@wikingand for how long a 1k usd will get us going ?21:54
@sonney2kwiking, no idea21:56
lisitsynsonney2k: who am I supposed to send updated xslt?21:56
@sonney2klisitsyn, could you give us some example output first maybe?21:56
lisitsynyeah21:57
@wikinglisitsyn: just updated the gist21:57
lisitsynwiking: https://gist.github.com/lisitsyn/879300221:58
lisitsynupdated21:58
@sonney2kwiking, just sent you an invitation21:58
@wikingsonney2k: got it, just trying to see what's this :)22:00
@wikinglisitsyn: sonney2k i've just updated the xslt... so the next build should generate the new output22:01
lisitsynI don't mind to filter this output somehow22:01
lisitsynbut it is a trickery to do in xslt :D22:01
@sonney2klisitsyn, could you just create an example please22:01
@sonney2kso we don't need to wait hours for each iteration?22:01
lisitsynsonney2k: yeah let me upload it22:02
lisitsynsonney2k: https://dl.dropboxusercontent.com/u/10139213/share/tmp/test.html22:03
lisitsyndamn22:03
lisitsynit blocks htmls I guess22:03
lisitsynhttps://dl.dropboxusercontent.com/u/10139213/share/tmp/test.html.txt22:04
lisitsynsonney2k: please download and rename then open22:04
@sonney2kwiking, https://cloud.google.com/products/compute-engine/#pricing22:04
@sonney2kthat's the prices22:04
@sonney2kwiking, so it is sth like 1$ per computing hour22:05
@sonney2kwiking, hard to reach if we don't do a cloud x-val project :D22:06
@wikingsonney2k: mmm i think they bill based on the online number of hours22:12
@wikingnot the actual computing time or?22:12
@sonney2klooks like22:19
@sonney2kwiking, so choose sth that can last for ~100 days or so :)22:20
@wikingsonney2k: ok i'm just testing now22:21
@wikingbtw where should the instance be22:21
@wikingus or eu?22:21
@wikingseems us is a bit cheaper22:21
@wikingok cool this works nicely22:23
@wikingi'll shut down the instance now22:24
@wikingand i'll get back to continue hackint it later22:24
shogun-buildbotbuild #181 of clang34 - undefined behaviour analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20undefined%20behaviour%20analysis/builds/181  blamelist: Soeren Sonnenburg <sonne@debian.org>22:33
@sonney2kmulticlass ocas seems to be totally b0rken22:40
@sonney2k66 uninited memory reads. wtf!?22:40
@sonney2kvoila22:43
@sonney2kand fixed22:43
@sonney2kwiking, how can we require the osx test on the buildbot to work?22:44
@sonney2kwiking, it is currently allowing it to fail22:44
lisitsynsonney2k: is html ok?22:45
@sonney2klisitsyn, where is it?22:45
lisitsynhttps://dl.dropboxusercontent.com/u/10139213/share/tmp/test.html.txt22:45
@sonney2kahh sorry22:45
@sonney2kchecking22:45
@wikingsonney2k: check the .travis.yml file ;)22:46
@wikingsonney2k: simply remove the lines that are for osx build from allow_failures part22:46
@wikingnamely remove these two22:46
@wiking- language: objective-c env: OSX=122:46
@wikingwhich is just before22:47
@wikingvirtualenv22:47
@sonney2klisitsyn, would it be possible to add the failures in the short list so we have them at first glance?22:47
@sonney2klisitsyn, the pull down stuff is nice though :)22:47
@sonney2kwiking, doing thanks22:48
lisitsynsonney2k: what do you mean by that?22:48
@sonney2klisitsyn, currently it lists only ./tests/unit/unit-MCLDA22:48
@sonney2klisitsyn, woudl be nice if it said /tests/unit/unit-MCLDA (Potential Memory Leak (3))22:48
lisitsynah okay easy22:49
shogun-buildbotbuild #176 of clang34 - thread analysis is complete: Failure [failed test]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20thread%20analysis/builds/176  blamelist: Soeren Sonnenburg <sonne@debian.org>22:49
@sonney2kwiking, now since python works too on osx - could we not just build it on the osx bot too?22:50
@sonney2kahh no I guess we need swig 2.0.12 / 3.022:51
lisitsynsonney2k: https://dl.dropboxusercontent.com/u/10139213/share/tmp/test.html.txt22:53
lisitsynwait wiat22:53
lisitsyndropbox is having fun22:54
lisitsynwiking: ich updated das xslt https://gist.github.com/lisitsyn/879300222:56
@sonney2klisitsyn, ok fine!22:59
@sonney2klisitsyn, now fix that malsar conditional memory read22:59
@sonney2klisitsyn, then we only have one more critical in xml serialization23:00
lisitsynsonney2k: it is not malsar that is buggy :D23:03
lisitsynbut a test23:03
lisitsynI am seeing the code and wondering if I was drunk or what23:03
@sonney2klisitsyn, then fix it :D23:08
lisitsynyeah sure23:08
lisitsynmakes sense23:08
@sonney2klisitsyn, and of course the answer is yes23:08
lisitsyn:D23:08
@sonney2kyou are Russian23:08
@sonney2kmy guess is that maybe you were not drunk23:08
@sonney2kso no wonder...23:09
lisitsynsonney2k: didn't we have some work on compilation speedup by besser82?23:09
lisitsynhow did it end?23:09
@sonney2klisitsyn, not ready - will be a gsoc project23:09
@sonney2klisitsyn, but you need libshogun only23:09
lisitsynI know23:09
@sonney2kand that is pretty fast anyway23:10
lisitsynjust curious23:10
lisitsynyeah once parallel it is23:10
@sonney2kman what a crap - we need to get rid of this useless use of SG_UNSTABLE23:11
@sonney2kcluttering stuff with useless warnings doesn'thelp23:11
@sonney2kwiking, any idea what this NON-XML-CHAR stuff in serialization xml tests is about [NON-XML-CHAR-0x1B][1;34m[WARN][NON-XML-CHAR-0x1B][0m In file /home/buildslave/debian_wheezy_-_memcheck/build/src/shogun/structure/DisjointSet.cpp line 19: /home/buildslave/debian_wheezy_-_memcheck/build/src/shogun/structure/DisjointSet.cpp:CDisjointSet::CDisjointSet():23:12
shogun-buildbotbuild #186 of debian wheezy - memcheck is complete: Failure [failed memory check]  Build details are at http://buildbot.shogun-toolbox.org/builders/debian%20wheezy%20-%20memcheck/builds/186  blamelist: Soeren Sonnenburg <sonne@debian.org>23:16
shogun-buildbotbuild #191 of clang34 - static analysis is complete: Failure [failed analyse]  Build details are at http://buildbot.shogun-toolbox.org/builders/clang34%20-%20static%20analysis/builds/191  blamelist: Soeren Sonnenburg <sonne@debian.org>23:37
@sonney2kwe have a problem when serializing XML - inf values cause valgrind errors / issues23:43
@sonney2kserializing AS xml23:43
-!- travis-ci [~travis-ci@ec2-50-19-161-94.compute-1.amazonaws.com] has joined #shogun23:46
travis-ci[travis-ci] it's Soeren Sonnenburg's turn to pay the next round of drinks for the massacre he caused in shogun-toolbox/shogun: http://travis-ci.org/shogun-toolbox/shogun/builds/1849612323:46
-!- travis-ci [~travis-ci@ec2-50-19-161-94.compute-1.amazonaws.com] has left #shogun []23:46
--- Log closed Sun Feb 09 00:00:51 2014