#linuxcnc-devel Logs

Mar 22 2019

12:03 AM linuxcnc-build: build #5879 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/5879 blamelist: Dewey Garrett <dgarrett@panix.com>
12:06 AM linuxcnc-build: build forced [ETA 54h34m44s]
12:06 AM linuxcnc-build: I'll give a shout when the build finishes
12:06 AM linuxcnc-build: build #5880 of 0000.checkin is complete: Failure [4failed fetch branch to local git repo] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/5880
06:24 PM linuxcnc-build: build #3012 of 1903.clang-wheezy-amd64 is complete: Failure [4failed apt-get-update install-missing-build-dependencies configuring] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1903.clang-wheezy-amd64/builds/3012 blamelist: Chris Morley <chrisinnanaimo@hotmail.com>
06:24 PM linuxcnc-build: build #4019 of 1400.rip-wheezy-i386 is complete: Failure [4failed apt-get-update install-missing-build-dependencies configuring] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1400.rip-wheezy-i386/builds/4019 blamelist: Chris Morley <chrisinnanaimo@hotmail.com>
06:24 PM linuxcnc-build: build #3535 of 1402.rip-wheezy-rtpreempt-i386 is complete: Failure [4failed apt-get-update install-missing-build-dependencies configuring] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1402.rip-wheezy-rtpreempt-i386/builds/3535 blamelist: Chris Morley <chrisinnanaimo@hotmail.com>
06:24 PM linuxcnc-build: build #4020 of 1403.rip-wheezy-amd64 is complete: Failure [4failed apt-get-update install-missing-build-dependencies configuring] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1403.rip-wheezy-amd64/builds/4020 blamelist: Chris Morley <chrisinnanaimo@hotmail.com>
06:42 PM skunkworks: pcw_home: what needs to be set to get DPLL working on 7i92
06:59 PM linuxcnc-build: build #5881 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/5881 blamelist: Chris Morley <chrisinnanaimo@hotmail.com>
09:00 PM pcw_home: for the stepgen, something like
09:00 PM pcw_home: # latch 50 usec before nominal read time:
09:00 PM pcw_home: setp hm2_[HOSTMOT2](BOARD).0.dpll.01.timer-us -50
09:00 PM pcw_home: setp hm2_[HOSTMOT2](BOARD).0.stepgen.timer-number 1
09:27 PM skunkworks: pcw_home: thanks!
09:44 PM pcw_home: Ack.. new bug in the TP
09:46 PM skunkworks: uh oh.. what?
09:47 PM pcw_home: user on forum found it
09:47 PM skunkworks: (it really is amazing that there really has only been a handfull of bugs found...)
09:48 PM skunkworks: wow - I have to close some browser tabs when I cannot really figure out what I have up..
09:48 PM pcw_home: looks like another of those tiny move things (I think Rob E has been working on these but I dont think he's merged any of the work)
09:50 PM skunkworks: I could poke him.. he has been really responsive lately..
09:51 PM skunkworks: This laptop is at the edge with a 1khz thread.. Darn - other than that it is my perfect laptop
09:51 PM pcw_home: I suspect this may have been fixed but not merged in master
09:51 PM pcw_home: I have a picture on the forum
09:51 PM pcw_home: very short simple gcode triggers it
09:52 PM skunkworks: I sent him an email with the forum post..
09:52 PM pcw_home: If you dont have high acceleration, a 500 hz servo thread may be OK
09:56 PM pcw_home: or set the packet timeout to about 1/2 the period and use the packet error workaround and just let it drop occasional packets if it times out
09:57 PM skunkworks: I was letting linuxcnc run 3dchips for a few hours.. It got a realtime delay error - but no following error.. Might be just fine for testing..
09:57 PM pcw_home: dpll for the encoder is basically the same, just set the encoder timer number to 1 also
09:58 PM skunkworks: no matter what settings I tweek - it seems to have +100us latency.
09:58 PM skunkworks: compared to this laptop which peaks around 80us
09:59 PM pcw_home: What is your packet timeout setting?
10:00 PM skunkworks: 80
10:00 PM skunkworks: it looks out
10:00 PM skunkworks: *it looks like
10:00 PM pcw_home: try 30...
10:00 PM skunkworks: thanks
10:02 PM pcw_home: you may get following errors unless you add the packet drop workaround
10:04 PM pcw_home: This dropping slow to respond packet scheme will only work if the latency is caused the Ethernet access (it often is)
10:06 PM pcw_home: I sometimes wish LinuxCNC would have the option to just log a count of latency errors
10:06 PM skunkworks: Cool - I will play with it
10:06 PM skunkworks: rob says he has a pull request..
10:07 PM pcw_home: also following error for warning, and following error for stop
10:08 PM skunkworks: that would be cool...
10:08 PM skunkworks: your following error is above x.xxx
10:09 PM skunkworks: That could be done in hal though.. the stop would be the normal ini setting - the warning could be just a hal trigger
10:09 PM pcw_home: also memorize peak following errors
10:10 PM pcw_home: yeah most of this could be done in hal but it would be nice if it was always there for debugging
10:10 PM skunkworks: sure
10:13 PM pcw_home: a pull request means Rob needs a LinuxCNC developer to merge it?
10:26 PM skunkworks: yes