sb0 changed the topic of #m-labs to: ARTIQ, Migen, MiSoC, Mixxeo & other M-Labs projects :: fka #milkymist :: Logs http://irclog.whitequark.org/m-labs
<bb-m-labs> build #92 of conda-lin64 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-lin64/builds/92
<whitequark> shit
<GitHub134> [artiq] sbourdeauducq pushed 1 new commit to master: https://git.io/v2kwQ
<GitHub134> artiq/master 95dbafd Sebastien Bourdeauducq: gui: add logo to MDI area
<sb0> rjo, ^ you probably will have strong opinions about this. feel free to revert if you dislike it.
<bb-m-labs> build #266 of artiq is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/266
<whitequark> ok hm I completely did not consider the dlls...
<bb-m-labs> build #30 of conda-win32 is complete: Exception [exception interrupted] Build details are at http://buildbot.m-labs.hk/builders/conda-win32/builds/30
<bb-m-labs> build #53 of conda-win64 is complete: Exception [exception interrupted] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/53
<bb-m-labs> build #9 of conda-all is complete: Failure [failed] Build details are at http://buildbot.m-labs.hk/builders/conda-all/builds/9
<whitequark> shit, I completely forgot how library path works on Windows
<whitequark> shouldn't have cancelled this :/ :/
<whitequark> bb-m-labs: force build --props=package=qt5 conda-win64
<bb-m-labs> build forced [ETA 1h57m26s]
<bb-m-labs> I'll give a shout when the build finishes
<whitequark> although on the other hand.
<bb-m-labs> build #54 of conda-win64 is complete: Exception [exception interrupted] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/54
<whitequark> oh, no, I wasn't wrong after all. this needs fixing.
<whitequark> sb0: qt5 installs libGLESv2.dll and libEGL.dll.
<whitequark> for... whatever the hell reason
<sb0> does it build them too?
<sb0> I think the chromium source is because they have a new component that displays html using the chromium engine...
<whitequark> yes, it does
<whitequark> it's ANGLE
<whitequark> opengl-to-direct3d translator
<sb0> there aren't that many ways to display bloated websites like facebook and twitter correctly ...
<whitequark> yes
<whitequark> btw I do not build either qtwebkit or qtwebengine (this chromium thing)
<sb0> ok. and are there problems with that, besides bloat?
<whitequark> no, there aren't, I just checked that qt4 doesn't ship them
<whitequark> all other libraries are prefixed with Qt5
<whitequark> webkit is fun. building webkit with LTO takes ~8 hours on a 12-core i7 at 4.5GHz with 32GB of RAM
<sb0> twitter is remarkably bloated. especially for what it is supposed to do...
<whitequark> you cannot link webkit at all on a 32-bit machine because the symbols alone exhaust the linker's address space
<sb0> did the anaconda guys build webkit in their qt4 package?
<whitequark> yes
<GitHub48> [conda-recipes] whitequark force-pushed master from 501cb32 to 6124bb9: https://github.com/m-labs/conda-recipes/commits/master
<GitHub48> conda-recipes/master 7e98ee6 whitequark: pyqt: rename to pyqt5 again.
<GitHub48> conda-recipes/master 95f09d6 whitequark: qt5: add.
<GitHub48> conda-recipes/master 1fa22e3 whitequark: qt5: disable qtimageformats on Windows....
<whitequark> ok this should be the last qt build (I hope)
<whitequark> bb-m-labs: force build --props=package=qt5 conda-all
<bb-m-labs> The build has been queued, I'll give a shout when it starts
sb0 has quit [Quit: Leaving]
<bb-m-labs> build #10 forced
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #55 of conda-win64 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/55
<bb-m-labs> build #31 of conda-win32 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win32/builds/31
<whitequark> fuuuuuuck
<whitequark> it's always the most idiotic typos
<whitequark> like forgetting a \ or a ^
<whitequark> actually, no, in this case it's just qt not recognizing -verbose but /only on Windows/ ??
<whitequark> bb-m-labs: force build --props=package=qt5 conda-win64
<bb-m-labs> build forced [ETA 1h57m26s]
<bb-m-labs> I'll give a shout when the build finishes
<GitHub155> [conda-recipes] whitequark force-pushed master from 6124bb9 to ea05156: https://github.com/m-labs/conda-recipes/commits/master
<GitHub155> conda-recipes/master 73814aa whitequark: qt5: add.
<GitHub155> conda-recipes/master e63a353 whitequark: qt5: disable qtimageformats on Windows....
<GitHub155> conda-recipes/master bdbbef4 whitequark: pyqt: rename to pyqt5 again.
<bb-m-labs> build #93 of conda-lin64 is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/conda-lin64/builds/93
<bb-m-labs> build #10 of conda-all is complete: Failure [failed] Build details are at http://buildbot.m-labs.hk/builders/conda-all/builds/10
<whitequark> bb-m-labs: force build --props=package=pyqt5 conda-lin64
<bb-m-labs> build forced [ETA 56m19s]
<bb-m-labs> I'll give a shout when the build finishes
<whitequark> do we really need to build qt with bluetooth and accelerometer support...
<bb-m-labs> Hey! build conda-lin64 #94 is complete: Success [build successful]
<whitequark> sb0: ok, linux pyqt5 appears to work from here
<whitequark> ah, hm, haven't tested xkb. but I assume it works since there are no errors.
<whitequark> bb-m-labs: force build --props=package=pyqt5 conda-win32
<bb-m-labs> build #32 forced
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #32 of conda-win32 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win32/builds/32
<whitequark> bb-m-labs: force build --props=package=qt5 conda-win32
<bb-m-labs> build #33 forced
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #56 of conda-win64 is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/56
<whitequark> bb-m-labs: force build --props=package=pyqt5 conda-win64
<bb-m-labs> build forced [ETA 1h55m40s]
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #57 of conda-win64 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/57
<whitequark> hrm
<bb-m-labs> build #33 of conda-win32 is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/conda-win32/builds/33
<whitequark> bb-m-labs: force build --props=package=pyqt5 conda-win64
<bb-m-labs> build forced [ETA 1h55m40s]
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #58 of conda-win64 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/58
<whitequark> sb0: btw I just installed the "GPLPV" drivers on the windows guest
<whitequark> it works much faster now. (I forgot to do it initially for some reasonss)
<whitequark> sigh
<whitequark> my qt5.conf patch works on linux but not windows, somehow.
<whitequark> what sense does this even make
<whitequark> *facepalm*
<whitequark> >patch -p1 0<$RECIPE_DIR/qt5_conf.patch
<whitequark> of COURSE it doesn't work
<whitequark> bb-m-labs: force build --props=package=qt5 conda-win64
<bb-m-labs> build forced [ETA 1h55m40s]
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #59 of conda-win64 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/59
<whitequark> lolwhat
<whitequark> command timed out: 14400 seconds without output
<whitequark> I started it three minutes ago!!
<whitequark> bb-m-labs: force build --props=package=qt5 conda-win64
<bb-m-labs> build forced [ETA 1h55m40s]
<bb-m-labs> I'll give a shout when the build finishes
hozer has quit [Ping timeout: 240 seconds]
hozer has joined #m-labs
hozer has quit [Ping timeout: 240 seconds]
<bb-m-labs> build #60 of conda-win64 is complete: Exception [exception interrupted] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/60
<whitequark> fuck, I forgot to push
<whitequark> bb-m-labs: force build --props=package=qt5 conda-win64
<whitequark> bb-m-labs: force build --props=package=qt5 conda-win32
<bb-m-labs> build forced [ETA 1h55m40s]
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build forced [ETA 1h08m15s]
<bb-m-labs> I'll give a shout when the build finishes
<GitHub132> [conda-recipes] whitequark force-pushed master from ea05156 to 8a664ad: https://github.com/m-labs/conda-recipes/commits/master
<GitHub132> conda-recipes/master b052b5a whitequark: qt5: add.
<GitHub132> conda-recipes/master 064f918 whitequark: qt5: disable qtimageformats on Windows....
<GitHub132> conda-recipes/master 901281c whitequark: pyqt: rename to pyqt5 again.
zoobab has quit [Ping timeout: 240 seconds]
hozer has joined #m-labs
zoobab has joined #m-labs
<whitequark> AAAARGH
<whitequark> what obnoxious asshole was writing the pyqt configure file?!
<whitequark> it specifically disables the --qmake option on windows
<whitequark> fuck you, obnoxious asshole
<whitequark> hm, they never thought of someone renaming qmake to qmake-qt5
<whitequark> and on linux the --qmake option is supposed to only set the path, not filename too
<whitequark> well
<GitHub78> [conda-recipes] whitequark force-pushed master from 8a664ad to da38f64: https://github.com/m-labs/conda-recipes/commits/master
<GitHub78> conda-recipes/master da38f64 whitequark: pyqt5: use prefixed qmake.
<rjo> sb0: why would i object or even have a strong opinion about it? it's like arguing over desktop backgrounds ;) but i can imagine that at some point people will want to have this configurable and want their experiment's logo there...
<whitequark> rjo: I fixed llvm-or1k btw
<whitequark> to run on ubuntu
<rjo> whitequark: i was running on debian stable
<rjo> whitequark: -2 still needs GLIBCXX_3.4.21
<whitequark> hm
<whitequark> rjo: I for some reason decided that it's a good idea to delete the package and reupload with same name and buildnumber
<whitequark> so you aren't actually using my new build (it has 0 downloads https://anaconda.org/m-labs/llvm-or1k/files)
<whitequark> you can delete it: rm -rf miniconda/pkgs/llvm-or1k*
<whitequark> i shouldn't do that, of course. (frankly that anaconda.org even allows to delete a package with >0 downloads is clearly a bug)
<rjo> whitequark: need to do a -3.
<rjo> whitequark: but that new -2 is fixed. yes.
<rjo> whitequark: and its not a bug to have in-use packages deletable.
<whitequark> it is if you have caches.
<whitequark> either-or
<whitequark> well, pinging the server with If-None-Match: <checksum> is also acceptable.
<rjo> whitequark: there are enough reasons why you have to delete a package that dwarf these technical wrinkles.
<whitequark> no, I don't think there are.
<rjo> licensing to name one
<whitequark> mhm, ok.
<whitequark> reuploading under same name should still be forbidden.
<rjo> yes
<bb-m-labs> build #34 of conda-win32 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win32/builds/34
<whitequark> ...
<whitequark> why did it break while installing enginio
<whitequark> enginio isn't even built
<whitequark> why did that happen only while building qt for win32
<whitequark> bb-m-labs: force build --props=package=qt5 conda-win32
<bb-m-labs> build forced [ETA 1h08m15s]
<bb-m-labs> I'll give a shout when the build finishes
<GitHub165> [conda-recipes] whitequark force-pushed master from da38f64 to c899d53: https://github.com/m-labs/conda-recipes/commits/master
<GitHub165> conda-recipes/master 4c33244 whitequark: qt5: add.
<GitHub165> conda-recipes/master 4c54216 whitequark: qt5: disable qtimageformats on Windows....
<GitHub165> conda-recipes/master ddcbf60 whitequark: pyqt: rename to pyqt5 again.
<whitequark> bb-m-labs: force build --props=package=pyqt5 conda-win64
<whitequark> bb-m-labs: force build --props=package=pyqt5 conda-win32
<bb-m-labs> The build has been queued, I'll give a shout when it starts
* whitequark shrugs
<bb-m-labs> The build has been queued, I'll give a shout when it starts
<whitequark> let's hope this will work
<bb-m-labs> build #61 of conda-win64 is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/61
<bb-m-labs> build forced [ETA 1h59m02s]
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #62 of conda-win64 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/62
<whitequark> bb-m-labs: force build --props=package=pyqt5 conda-win64
<bb-m-labs> build forced [ETA 1h59m02s]
<bb-m-labs> I'll give a shout when the build finishes
<GitHub94> [conda-recipes] whitequark force-pushed master from c899d53 to b8745d2: https://github.com/m-labs/conda-recipes/commits/master
<GitHub94> conda-recipes/master b8745d2 whitequark: pyqt5: use prefixed qmake.
sb0 has joined #m-labs
<GitHub131> [artiq] sbourdeauducq pushed 2 new commits to master: https://git.io/v2Ivl
<GitHub131> artiq/master 1a5ccdf Sebastien Bourdeauducq: MANIFEST.in: update
<GitHub131> artiq/master 1299274 Sebastien Bourdeauducq: conda: pyqt -> pyqt5
<bb-m-labs> build #267 of artiq is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/267 blamelist: Sebastien Bourdeauducq <sb@m-labs.hk>
<whitequark> sb0: -win64 package for qt works
<whitequark> -lin64 package for qt should work but I only tried it with xvfb and didn't try any xkb
<whitequark> so test it
<whitequark> -win32 package will work, assuming it compiles
<GitHub191> [conda-recipes] sbourdeauducq pushed 1 new commit to master: https://github.com/m-labs/conda-recipes/commit/f7948bc0471cc94aaf3f80638005ec8306462c39
<GitHub191> conda-recipes/master f7948bc Sebastien Bourdeauducq: quamash: pyqt -> pyqt5
<sb0> bb-m-labs: force build --props=package=quamash conda-lin64
<bb-m-labs> build forced [ETA 38m36s]
<bb-m-labs> I'll give a shout when the build finishes
<whitequark> didn't bump
<whitequark> it'll break.
<bb-m-labs> build #95 of conda-lin64 is complete: Failure [failed anaconda_upload] Build details are at http://buildbot.m-labs.hk/builders/conda-lin64/builds/95
<whitequark> told you
<sb0> hmm we could simply delete the old package, but conda loves its local caches
<whitequark> exactly
<whitequark> so just bump
<whitequark> don't repeat my mistakes
<sb0> bb-m-labs: force build --props=package=quamash conda-lin64
<bb-m-labs> build forced [ETA 38m36s]
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #63 of conda-win64 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/63
<GitHub59> [conda-recipes] sbourdeauducq pushed 1 new commit to master: https://github.com/m-labs/conda-recipes/commit/288f2c7146f7e672d293d43a77ef1ff970bb0c7e
<GitHub59> conda-recipes/master 288f2c7 Sebastien Bourdeauducq: quamash: bump
<bb-m-labs> build #96 of conda-lin64 is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/conda-lin64/builds/96
<whitequark> ugh
<sb0> bb-m-labs: force build artiq
<whitequark> bb-m-labs: force build --props=package=pyqt5 conda-win64
<bb-m-labs> build forced [ETA 4m47s]
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build forced [ETA 1h59m02s]
<bb-m-labs> I'll give a shout when the build finishes
<whitequark> sb0: noooo
<GitHub24> [conda-recipes] whitequark pushed 1 new commit to master: https://github.com/m-labs/conda-recipes/commit/1e876e8b32797b810d174d8b7c3c2a7a6824dfc9
<GitHub24> conda-recipes/master 1e876e8 whitequark: pyqt: fix patch.
<sb0> why?
<whitequark> it's already far overloaded with qt
<whitequark> that pyqt run has been unpacking for 20 minutes
<whitequark> and it's a tiny tar.
<whitequark> and now it's going to spend the next four hours building bitstreams.
<whitequark> well, I'll just go home.
<sb0> let's cancel the bitstream builds
<whitequark> then what's the point of building the artiq package?
<sb0> trying qt5
<whitequark> hrm
<bb-m-labs> build #64 of conda-win64 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/64
<whitequark> configure.py: error: 'c:\slave64\miniconda\envs\_build\Library\bin\qmake-qt5' is not an executable
<whitequark> ARGH
<whitequark> no .exe
<whitequark> this script is too fucking dumb
<whitequark> bb-m-labs: force build --props=package=pyqt5 conda-win64
<bb-m-labs> build forced [ETA 1h59m02s]
<bb-m-labs> I'll give a shout when the build finishes
<GitHub146> [conda-recipes] whitequark pushed 1 new commit to master: https://github.com/m-labs/conda-recipes/commit/8628e3f537421c8e04cfc7bdbbc598856538197e
<GitHub146> conda-recipes/master 8628e3f whitequark: pyqt5: fix --qmake on Windows.
<bb-m-labs> build #65 of conda-win64 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/65
<bb-m-labs> build #268 of artiq is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/268
<whitequark> bb-m-labs: force build --props=package=pyqt5 conda-win64
<bb-m-labs> build forced [ETA 1h59m02s]
<bb-m-labs> I'll give a shout when the build finishes
<GitHub102> [conda-recipes] whitequark pushed 1 new commit to master: https://github.com/m-labs/conda-recipes/commit/16cf2b2e04bf4632be5115dfc10ec8c6e617ce0b
<GitHub102> conda-recipes/master 16cf2b2 whitequark: pyqt5: fix patch.
<bb-m-labs> build #135 of artiq-kc705-nist_qc2 is complete: Exception [exception interrupted] Build details are at http://buildbot.m-labs.hk/builders/artiq-kc705-nist_qc2/builds/135
<bb-m-labs> build #51 of artiq-kc705-nist_clock is complete: Exception [exception interrupted] Build details are at http://buildbot.m-labs.hk/builders/artiq-kc705-nist_clock/builds/51
<bb-m-labs> build #151 of artiq-kc705-nist_qc1 is complete: Exception [exception interrupted] Build details are at http://buildbot.m-labs.hk/builders/artiq-kc705-nist_qc1/builds/151
<bb-m-labs> build #137 of artiq-pipistrello-nist_qc1 is complete: Exception [exception interrupted] Build details are at http://buildbot.m-labs.hk/builders/artiq-pipistrello-nist_qc1/builds/137
<sb0> whitequark, how do I bump when the build number is determined by git? (e.g. pyqtgraph)
<whitequark> you push an empty commit
<whitequark> there's really no better way
<whitequark> ok, qt/win32 has built correctly
<whitequark> pyqt should work too
<sb0> bb-m-labs: force build --props=package=pyqtgraph conda-lin64
<bb-m-labs> build forced [ETA 19m27s]
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #66 of conda-win64 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/66
<whitequark> sb0: haha, pyqt doesn't allow to set the name of moc.
<whitequark> actually, hm.
<whitequark> hmm.
<bb-m-labs> build #97 of conda-lin64 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-lin64/builds/97
<whitequark> oh.
<whitequark> oh i see.
<whitequark> shit
<sb0> and pyqtgraph still installed qt4, for some reason
<whitequark> ok so you need BOTH the prefixed AND the unprefixed version of qt
<whitequark> qt tools
<whitequark> another rebuild :/
<sb0> hm, what?
<whitequark> qmake calls "moc"
<whitequark> not "moc-qt5".
<sb0> oh, ok, you were talking about the other problem
<sb0> also i simply didn't notice my push had failed. pebkac
<whitequark> on linux this succeeds because uh
<whitequark> --bindir is passed to pyqt5
<GitHub49> [conda-recipes] sbourdeauducq pushed 2 new commits to master: https://github.com/m-labs/conda-recipes/compare/16cf2b2e04bf...5d4fa61f3959
<GitHub49> conda-recipes/master c2da5fd Sebastien Bourdeauducq: pyqtgraph: pyqt -> pyqt5
<GitHub49> conda-recipes/master 5d4fa61 Sebastien Bourdeauducq: Merge branch 'master' of github.com:m-labs/conda-recipes
<whitequark> oh, no, that doesn't actually do anything. hm. weird.
<whitequark> ohhhhhh
<sb0> bb-m-labs: force build --props=package=pyqtgraph conda-lin64
<whitequark> oh I see
<bb-m-labs> build forced [ETA 19m27s]
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #98 of conda-lin64 is complete: Failure [failed anaconda_upload] Build details are at http://buildbot.m-labs.hk/builders/conda-lin64/builds/98
<whitequark> qmake will call the moc in lib/qt5/bin if I build qt so
<whitequark> ugh
<whitequark> this coexistence is a real pain in the arse
<whitequark> that's the same old bug
<whitequark> except for some reason that builder isn't attempting to detect it
<whitequark> "sanity check" should've caught
<sb0> and the solution is "git_url: ../.."?
<whitequark> yes
<sb0> do I need to modify anything else? buildbot config?
<whitequark> no, not ../..
<whitequark> actually, I think it's impossible to fix that bug with conda-recipes repo
<sb0> we can easily put the conda files into pyqtgraph, since we had to fork that one already as upstream has been ignoring my trivial 1-line bugfixes patches for 10 months
<whitequark> then yes
<whitequark> I'm sure you'll figure out how to update buildbot config
<whitequark> bb-m-labs: force build --props=package=qt5 conda-win64
<bb-m-labs> build forced [ETA 1h59m02s]
<bb-m-labs> I'll give a shout when the build finishes
<whitequark> bb-m-labs: force build --props=package=pyqt5 conda-win64
<GitHub155> [conda-recipes] whitequark pushed 1 new commit to master: https://github.com/m-labs/conda-recipes/commit/64cd90d5545bc3ec362478008d534d3f25ec74d1
<GitHub155> conda-recipes/master 64cd90d whitequark: qt5: leave binaries in the prefix where they belong.
<whitequark> bb-m-labs: force build --props=package=qt5 conda-win32
<bb-m-labs> The build has been queued, I'll give a shout when it starts
<whitequark> bb-m-labs: force build --props=package=pyqt5 conda-win32
<bb-m-labs> The build has been queued, I'll give a shout when it starts
<bb-m-labs> The build has been queued, I'll give a shout when it starts
<whitequark> this miiiight work
<bb-m-labs> build #35 of conda-win32 is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/conda-win32/builds/35
<bb-m-labs> build forced [ETA 1h21m29s]
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #36 of conda-win32 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win32/builds/36
<sb0> whitequark, how do I/when can I restart buildbot?
<sb0> ah, there's deploy.sh.
<bb-m-labs> build #67 of conda-win64 is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/67
<bb-m-labs> build forced [ETA 1h43m28s]
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> Hey! build conda-win64 #68 is complete: Retry [retry exception slave lost]
<bb-m-labs> build forced [ETA 1h43m28s]
<bb-m-labs> I'll give a shout when the build finishes
<sb0> hmm what happened (I didn't touch it)
<sb0> bb-m-labs: force build --props=package=pyqt5 conda-win64
<bb-m-labs> The build has been queued, I'll give a shout when it starts
<bb-m-labs> build #69 of conda-win64 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/69
<bb-m-labs> build forced [ETA 1h43m28s]
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #70 of conda-win64 is complete: Exception [exception interrupted] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/70
<sb0> bb-m-labs: force build --props=package=pyqt5 conda-win64
<bb-m-labs> build forced [ETA 1h43m28s]
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #71 of conda-win64 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-win64/builds/71
<GitHub139> [buildbot-config] sbourdeauducq pushed 1 new commit to master: https://github.com/m-labs/buildbot-config/commit/3f311c8cf86a457272ba2408c2b3750aaa685fd9
<GitHub139> buildbot-config/master 3f311c8 Sebastien Bourdeauducq: Add dedicated pyqtgraph builder
bb-m-labs has quit [Quit: buildmaster reconfigured: bot disconnecting]
<GitHub28> [buildbot-config] sbourdeauducq pushed 1 new commit to master: https://github.com/m-labs/buildbot-config/commit/edeb5206f6a3fcc1df7f5dfa70d8b5b78ecd54f7
<GitHub28> buildbot-config/master edeb520 Sebastien Bourdeauducq: fix typo
bb-m-labs has joined #m-labs
<sb0> bb-m-labs, force build pyqtgraph
<bb-m-labs> build #0 forced
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #0 of pyqtgraph is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/pyqtgraph/builds/0
<sb0> bb-m-labs, force build pyqtgraph
<bb-m-labs> build #1 forced
<bb-m-labs> I'll give a shout when the build finishes
<bb-m-labs> build #1 of pyqtgraph is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/pyqtgraph/builds/1
<sb0> whitequark, ImportError: libQt5Gui.so.5: cannot open shared object file: No such file or directory
<sb0> for some reason, it works on my machine and I can import your PyQt5 even after renaming my system Qt ...
<sb0> rjo, btw you were right, no one complained about the ion pump magnets in my luggage
<sb0> there was a 'magnets forbidden' sign at warsaw airport, but the same sign also said that radioactive substances are banned, which is not true
<GitHub93> [conda-recipes] sbourdeauducq pushed 2 new commits to master: https://github.com/m-labs/conda-recipes/compare/64cd90d5545b...08d7bb05a4fd
<GitHub93> conda-recipes/master 907f5e8 Sebastien Bourdeauducq: delete pyqgraph recipe (now part of pyqtgraph repository to workaround conda bug)
<GitHub93> conda-recipes/master 08d7bb0 Sebastien Bourdeauducq: Merge branch 'master' of github.com:m-labs/conda-recipes
<sb0> anyway, I uploaded my locally-built pyqtgraph package; now that its bugs and problems are confined inside applets (which use it in a pretty standard way), I hope we won't have to modify it too often
zoobab has quit [Ping timeout: 255 seconds]
zoobab has joined #m-labs
<rjo> sb0: what's "a magnet" anyway?
<sb0> one applet bug less
<sb0> pfui
<sb0> artiq actually works pretty nicely on windows now
<sb0> last applet bug gone. my intuition about window borders was right.
<sb0> there is still this ConnectionAbortedError python bug spamming the logs
<sb0> seems you get it if you try to write on a socket that has been closed a long time ago
<sb0> (closed by the remote end)
<sb0> and the asyncio unittests don't wait long enough to get it
<GitHub121> [artiq] sbourdeauducq pushed 1 new commit to master: https://git.io/v2IK1
<GitHub121> artiq/master e3e4221 Sebastien Bourdeauducq: applets: workaround 'garbage on sides of embedded windows' bug on Windows
<bb-m-labs> build #269 of artiq is complete: Failure [failed make_doc] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/269 blamelist: Sebastien Bourdeauducq <sb@m-labs.hk>
<bb-m-labs> build #52 of artiq-kc705-nist_clock is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-kc705-nist_clock/builds/52
stekern has quit [Ping timeout: 255 seconds]
stekern has joined #m-labs
<sb0> stekern, how good is the mor1kx FPU?
<sb0> new cool conda bug: it freezes/deadlocks while downloading packages
<sb0> seems to only occur since i upgraded glibc
<sb0> (on my laptop)
<rjo> same here. i think that's somewhere else.
<sb0> whitequark, on Linux, it now does this:
<sb0> This application failed to start because it could not find or load the Qt platform plugin "xcb".
<bb-m-labs> build #136 of artiq-kc705-nist_qc2 is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-kc705-nist_qc2/builds/136
<cr1901_modern> What an interesting Qt component I've never heard of. It has the exact same name as an X component
<bb-m-labs> build #138 of artiq-pipistrello-nist_qc1 is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-pipistrello-nist_qc1/builds/138
<bb-m-labs> build #152 of artiq-kc705-nist_qc1 is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-kc705-nist_qc1/builds/152
<GitHub121> [artiq] jordens pushed 3 new commits to master: https://git.io/v2LIW
<GitHub121> artiq/master 294eac1 Robert Jordens: pdq2.mediator: cleanup/style
<GitHub121> artiq/master 7c246b3 Robert Jordens: pdq2: redo program serialization, cleanup
<GitHub121> artiq/master d713c62 Robert Jordens: pdq2 frontends: adapt to new program, cleanup
<bb-m-labs> build #270 of artiq is complete: Failure [failed make_doc] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/270 blamelist: Robert Jordens <jordens@gmail.com>
<sb0> rjo, forget my github comment. i misinterpreted what happened.
<sb0> --target got list in refactoring
<sb0> *lost
<GitHub33> [artiq] sbourdeauducq pushed 1 new commit to master: https://git.io/v2Lt3
<GitHub33> artiq/master ac6e31d Sebastien Bourdeauducq: frontend/rpctool: fix -t/--target. Closes #283
<bb-m-labs> build #137 of artiq-kc705-nist_qc2 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/artiq-kc705-nist_qc2/builds/137 blamelist: Robert Jordens <jordens@gmail.com>
<bb-m-labs> build #153 of artiq-kc705-nist_qc1 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/artiq-kc705-nist_qc1/builds/153 blamelist: Robert Jordens <jordens@gmail.com>
<bb-m-labs> build #271 of artiq is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/271 blamelist: Sebastien Bourdeauducq <sb@m-labs.hk>
<sb0> now I can't install any cygwin package, it says "Download incomplete - try again"
<sb0> this, and the sshd breakage
<sb0> I didn't remember cygwin to be so loaded with bugs
<sb0> it also says so "269%" into the installation
<sb0> any idea how I can get a ssh server on windows that is not overflowing with bugs?
<GitHub199> [artiq] sbourdeauducq pushed 1 new commit to master: https://git.io/v2LmG
<GitHub199> artiq/master e177bbd Sebastien Bourdeauducq: frontend/client: fix screen clear on Windows
<bb-m-labs> build #272 of artiq is complete: Failure [failed make_doc] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/272 blamelist: Sebastien Bourdeauducq <sb@m-labs.hk>
<bb-m-labs> build #154 of artiq-kc705-nist_qc1 is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-kc705-nist_qc1/builds/154
<rjo> sb0: how far are the powershell people with the native openssh port?
<stekern> sb0: tbh, I have only briefly tested it myself
key2 has joined #m-labs
<sb0> apparently, this ConnectionAbortedError comes from the way Qt or whatever quamash is using closes the network connection
<sb0> if I close the connections to the master uncleanly, I don't get it
<sb0> pretty sure this is just another python but
<sb0> *bug
<bb-m-labs> build #138 of artiq-kc705-nist_qc2 is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-kc705-nist_qc2/builds/138
<bb-m-labs> build #273 of artiq is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/273
<sb0> yeah, maybe
<rjo> sb0: the rpcs for the pdqs quickly contain lines that are longer than the max. what should i do?
<rjo> sb0: see the issue report.
key2 has quit [Ping timeout: 252 seconds]
zoobab has quit [Ping timeout: 248 seconds]
zoobab has joined #m-labs
zoobab has quit [Ping timeout: 248 seconds]
zoobab has joined #m-labs
zoobab has quit [Ping timeout: 276 seconds]
zoobab has joined #m-labs
zoobab has quit [Ping timeout: 240 seconds]
zoobab has joined #m-labs
zoobab has quit [Ping timeout: 250 seconds]
zoobab has joined #m-labs
zoobab has quit [Ping timeout: 244 seconds]
zoobab has joined #m-labs
zoobab has quit [Ping timeout: 244 seconds]
zoobab has joined #m-labs
ylamarre has joined #m-labs
ylamarre1 has joined #m-labs
ylamarre has quit [Ping timeout: 244 seconds]
ylamarre1 is now known as ylamarre