#linuxcnc-devel Logs

Jun 24 2017

07:55 AM jepler: skunkworks: (added a note on how to enter an isolcpus value when booting the live image at https://github.com/jepler/stretch-live-build)
07:58 AM jepler: well that's weird: ssh: connect to host afk.unpythonic.net port 22: Operation not permitted
10:21 AM seb_kuzminsky: nice, zsync can use a local v0.iso as the seed for downloading the -r1.iso.zsync
10:28 AM jepler: how much did it help to do so?
10:29 AM seb_kuzminsky: used 395419648 local, fetched 782799881
10:29 AM seb_kuzminsky: "i'd like to buy a comma"
10:31 AM jepler: 1/3?
10:41 AM seb_kuzminsky: used 1, downloaded 2, how i read it
10:55 AM seb_kuzminsky: we're done with Hardy, right? there will never be another linuxcnc release on hardy?
10:59 AM jepler: http://wiki.linuxcnc.org/cgi-bin/wiki.pl?MinimumSoftwareVersions doesn't even bother listing Hardy
11:03 AM seb_kuzminsky: the most recent "non-no-op" build on Hardy was in May 2016: http://buildbot.linuxcnc.org/buildbot-admin/buildslaves/hardy-rtai-i386?numbuilds=1000
11:03 AM seb_kuzminsky: i'm going to shut down the three hardy buildslaves and replace them with two stretch buildslaves (i686-pae and amd64)
11:04 AM jepler: sounds good to me
11:04 AM jepler: you're going to install them from this .iso image?
11:07 AM seb_kuzminsky: i was planning to use the upstream d.o installer, and switch to the -rt kernel by hand
11:21 AM jepler: oh that's fine too
11:27 AM jepler: join #linuxcnc-meet for this little meeting
11:28 AM jepler: beep seb_kuzminsky cradek in particular
11:51 AM jepler: seb_kuzminsky: if there's to be much buildbot downtime, let's have it after 2.7.10
11:51 AM seb_kuzminsky: agreed
11:55 AM KGB-linuxcnc: 03Jeff Epler 05master e7bfd26 06linuxcnc 10src/emc/usr_intf/axis/scripts/axis.py Merge branch 'axis-numeric-key-shortcuts' of https://github.com/phillc54/linuxcnc * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=e7bfd26
11:55 AM cradek: drat
11:55 AM jepler: hi cradek
11:55 AM cradek: hi
11:55 AM jepler: off by DST bug or something?
11:55 AM cradek: sorry, was driving and missed it
11:55 AM jepler: oh well
11:55 AM jepler: it was uneventful, a room full of yes-persons
11:55 AM cradek: can I have a belated vote?
11:56 AM jepler: to thank yourself? no, you've already been thanked
11:56 AM mozmck: my wife had a baby thursday so I've been a little busy myself.
11:56 AM jepler: mozmck: congratulations! I hope they are doing well.
11:56 AM cradek: yay!
11:56 AM mozmck: doing well!
11:56 AM cradek: that's much better than my excuse
11:56 AM mozmck: But I vote yes as well fwis.
11:56 AM mozmck: fwiw
11:59 AM jepler: I wonder when to mention http://www.linuxcnc.org/testing-stretch-amd64-rtpreempt/ on the mailing list
12:04 PM seb_kuzminsky: jepler: that mention should come with an explanation of the switch frmo rtai to rt-preempt, and the implications for current rtai users
12:45 PM jepler: seb_kuzminsky: yes
12:52 PM jepler: curious, I don't think I did anything with the shared memory limit on the live image..
01:01 PM jepler: something like http://media.unpythonic.net/stretch-announce-draft.txt
01:03 PM jepler: seb_kuzminsky: truetype-tracer and mesaflash are not in the jessie repo on wlo
01:03 PM jepler: I'd like to see that corrected, and also make sure they get into stretch
01:04 PM jepler: is that something that is best to just let you do?
01:07 PM jepler: (I heavily cribbed from http://article.gmane.org/gmane.linux.distributions.emc.user/52401 which is why it occurred to me)
01:27 PM seb_kuzminsky: jepler: yes, i'll do that part
01:27 PM seb_kuzminsky: maybe open a ticket for me so i dont space it
01:27 PM seb_kuzminsky: today is fix-the-dryer day, not hack-on-linuxcnc day unfortunately
01:46 PM jepler: hah first things first
01:47 PM jepler: can you fix it by buying a length of twine and stringing it in the garage?
01:56 PM pcw_home: Has image-to-gcode been tested with stretch? that's one of the things that doesn't work with Ubuntu 16.04/Mint18 (numpy issue?)
02:17 PM jepler: not by me
02:18 PM jepler: NotImplementedError: tostring() has been removed. Please call tobytes() instead.
02:18 PM jepler: stupid brokenness for the sake of brokenness
02:30 PM jepler: hm now I'm making all kinds of work for seb
02:43 PM TurBoss: can someone point me to the bits of AXIS-GUI where the progress bar(when loading a path) gets filled? Thnaks
03:30 PM cradek: jepler: oh my, those changes
03:51 PM jepler: turboss: the 'progress' class in axis.py does that
03:52 PM jepler: er, 'Progress'
04:00 PM cradek: ok I shut off the git!
04:00 PM cradek: well not really. what do we need to do?
04:17 PM seb_kuzminsky: i need to finish the cgi script on the buildmaster that github pokes
04:18 PM jepler: cradek: well the main thing is, we need to set a timeline and notify about what it is
04:18 PM jepler: I think the main things are: get buildbot ready, figure out how to do emc-commit or say it won't work out, then announce & wait a week, then do.
04:19 PM jepler: the day of, we turn glo off, adding some users on github, and actually make the buildbot config live
04:20 PM jepler: maybe there should be a period of time that glo is readonly or something
04:20 PM jepler: write docs for users (change clone instructions) etc
04:20 PM jepler: delete the SOB instructions
04:20 PM jepler: there's a lot, somebody should keep a list
04:23 PM hazzy: jepler: Thanks! So it looks like Progress.update is called in 'open_file_guts', but there is a lot going on in there. Much more than just opening a file!
04:23 PM hazzy: What TurBoss and I are trying to do is get the loading progress of a file opened using the python interface's 'command.program_open'
04:23 PM hazzy: Is there already a way to get the file load progress in the python interface? If not, would the best place to add this be in the python interface or in the GUI?
04:24 PM jepler: to "get" the progress? no, there's not an API for programs to ask AXIS about progess loading a part program.
04:27 PM hazzy: But I assume that the python interface must load the file in a similar way to 'open_file_guts' in axis?
04:27 PM hazzy: Or is it completely different?
04:28 PM jepler: the progress bar in AXIS is all about the preview generation. that step doesn't exist for command.program_open.
04:29 PM hazzy: Oh, so the time consuming part of loading the file is generating the preview.
04:29 PM jepler: yes
04:31 PM hazzy: So if I were to try adding something to track the file loading status the best place would probably be gremlin?
04:31 PM jepler: I don't know offhand if gremlin does a progress indication for preview generation. I think it might not.
04:32 PM hazzy: OK, Ill have a look. Thank you!
04:36 PM jepler: looks like AxisCanon is where the progress of the gcode interpreting phase happens
04:36 PM jepler: it takes a Progress instance in its constructor; then calls it from the next_line method to keep the progress bar going
06:53 PM skunkworks: is there a place to point sources to stretch for rt-preempt? or not yet
06:54 PM skunkworks: sources.list
06:54 PM skunkworks: ie W: Failed to fetch http://buildbot.linuxcnc.org/dists/stretch/2.7-rtpreempt/binary-amd64/Packages 404 Not Found
06:57 PM seb_kuzminsky: not yet
06:57 PM seb_kuzminsky: i'm working on it
06:57 PM seb_kuzminsky: i just put up the first stretch package, truetype-tracer
06:57 PM seb_kuzminsky: it's in base
07:00 PM skunkworks: I was going to try to upgrade but maybe I will wait a bit
07:03 PM seb_kuzminsky: we dont have linuxcnc debs for stretch yet
07:03 PM seb_kuzminsky: in a couple of days or so
07:09 PM seb_kuzminsky: linuxcnc-build: force build --branch=master 0000.checkin
07:09 PM linuxcnc-build: build #5004 forced
07:09 PM linuxcnc-build: I'll give a shout when the build finishes
07:15 PM linuxcnc-build: build #0 of 1630.rip-stretch-rtpreempt-amd64 is complete: Failure [4failed apt-get-update install-missing-build-dependencies configuring] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1630.rip-stretch-rtpreempt-amd64/builds/0
07:16 PM linuxcnc-build: build #0 of 1610.rip-stretch-rtpreempt-i386 is complete: Failure [4failed apt-get-update install-missing-build-dependencies configuring] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1610.rip-stretch-rtpreempt-i386/builds/0
07:22 PM seb_kuzminsky: heh oops
07:32 PM seb_kuzminsky: linuxcnc-build: force build --branch=master 0000.checkin
07:32 PM linuxcnc-build: The build has been queued, I'll give a shout when it starts
07:36 PM linuxcnc-build: build #1 of 1610.rip-stretch-rtpreempt-i386 is complete: Failure [4failed install-missing-build-dependencies configuring] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1610.rip-stretch-rtpreempt-i386/builds/1
07:36 PM linuxcnc-build: build #1 of 1630.rip-stretch-rtpreempt-amd64 is complete: Failure [4failed install-missing-build-dependencies configuring] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1630.rip-stretch-rtpreempt-amd64/builds/1
08:07 PM linuxcnc-build: build #5004 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/5004
08:07 PM linuxcnc-build: build #5005 forced
08:07 PM linuxcnc-build: I'll give a shout when the build finishes
08:33 PM linuxcnc-build: build #2 of 1610.rip-stretch-rtpreempt-i386 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1610.rip-stretch-rtpreempt-i386/builds/2
08:43 PM linuxcnc-build: build #2 of 1630.rip-stretch-rtpreempt-amd64 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1630.rip-stretch-rtpreempt-amd64/builds/2
08:46 PM linuxcnc-build: build #3 of 1610.rip-stretch-rtpreempt-i386 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1610.rip-stretch-rtpreempt-i386/builds/3
08:50 PM linuxcnc-build: build #3 of 1630.rip-stretch-rtpreempt-amd64 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1630.rip-stretch-rtpreempt-amd64/builds/3
09:27 PM linuxcnc-build: build #5005 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/5005
10:54 PM jepler: a2x: ERROR: missing configuration file: /etc/asciidoc/dblatex/asciidoc-dblatex.xsl
10:54 PM jepler: well that's a heck of an error
10:54 PM jepler: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=851018
10:55 PM jepler: >>> Anyway, for a solution, changing the dependency on "asciidoc" to
10:55 PM jepler: "asciidoc, asciidoc-dblatex (>= 8.6.9-5) | asciidoc (<= 8.6.9-3.1)"
10:55 PM jepler: will probably suffice to still be able to build the package on Debian
10:55 PM jepler: 8 Jessie, too.<<<
10:56 PM jepler: I must have that file left because .. it is a conffile of the old asciidoc package and therefore not removed on upgrade?
10:56 PM jepler: 'night
10:59 PM KGB-linuxcnc: 03Sebastian Kuzminsky 052.7 f5e0a5b 06linuxcnc 10debian/configure build: build-depend on asciidoc-dblatex on debian stretch * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=f5e0a5b
11:00 PM seb_kuzminsky: luckily(?) we build different debian/control files on different architectures, so we can just say what we mean
11:38 PM linuxcnc-build: build #4 of 1630.rip-stretch-rtpreempt-amd64 is complete: Failure [4failed compile get-dmesg] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1630.rip-stretch-rtpreempt-amd64/builds/4 blamelist: Sebastian Kuzminsky <seb@highlab.com>
11:45 PM linuxcnc-build: build #4 of 1610.rip-stretch-rtpreempt-i386 is complete: Failure [4failed compile get-dmesg] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1610.rip-stretch-rtpreempt-i386/builds/4 blamelist: Sebastian Kuzminsky <seb@highlab.com>
11:53 PM linuxcnc-build: build #5006 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/5006 blamelist: Sebastian Kuzminsky <seb@highlab.com>