#linuxcnc-devel | Logs for 2016-07-27

[07:29:24] <jthornton> I just got the latest master installed from buildbot on wheezy and the Axis calibration and show hal configuration do not work and no errors are reported... hmm let me set DEBUG=0x7fffffff and try again
[07:31:32] <jthornton> no extra errors they just don't do anything
[07:33:41] <jepler> ugh another 10 CVE bugs in php5 with remote DOS capability at a minimum. ubuntu still hasn't gotten around to fixing the last set of 6 from July 14.
[07:33:48] <jepler> :shakes fist at sky:
[07:57:24] <tinkerer> wow, the JA changes for the config files is a mess...
[07:59:50] <tinkerer> oups, high sophisticated dyslexia...
[08:00:16] <tinkerer> config file changes for JA
[08:10:07] <tinkerer> interestingly that [TRAJ]MAX_LINEAR_VELOCITY is obligatory.
[08:14:23] <tinkerer> and the message "Missing <max linear speed> specifier" belongs to this entry. IMHO this message should be altered.
[08:17:30] <skunkworks_> ok - figured out dads problem... when you exit escape out of a program - the fixture offset always get set to 54
[08:18:24] <skunkworks_> even though you have mdi'ed say g57 - when you abort a program or a program stops - it defaults back to g54
[08:18:59] <skunkworks_> I don't know if that is new to 2.7.5 or not. have not had time to play more
[08:19:35] <jepler> skunkworks_: I wouldn't be surprised if it were new to 2.7.5.
[08:21:32] -linuxcnc-github:#linuxcnc-devel- [13linuxcnc] 15jethornton opened issue #125: Axis Calibration and Show Hal Configuration 02https://github.com/LinuxCNC/linuxcnc/issues/125
[08:21:55] <skunkworks_> it makes the preview wrong and confusing. (because the xyz positions are for the G54 even though the program is running g57 or whatever)
[09:01:02] -linuxcnc-github:#linuxcnc-devel- [13linuxcnc] 15gmoccapy commented on issue #121: Am 27.07.2016 um 04:18 schrieb Jeff Epler:
[11:01:31] <mozmck> This is annoying - I can get a regex working like I want on regex101.com or pyregex, but when I go to use it in python it fails.
[11:02:56] <mozmck> or at least I think it does - it's probably my lack of understanding still.
[11:03:55] <mozmck> I'm trying to match something like o<touchoff>, but NOT if there is a / or ; preceeding it in the line.
[11:17:15] <seb_kuzminsky> skunkworks_: yeah i bet i messed that one up too
[11:17:31] <seb_kuzminsky> :shakes fist at seb:
[11:18:06] <seb_kuzminsky> what does you [RS274NGC]RS274NGC_STARTUP_CODE say? anything?
[11:24:47] <skunkworks_> no entry
[11:44:18] <seb_kuzminsky> is it your program that sets g57? or did you set that through mdi before running the program?
[11:49:15] <seb_kuzminsky> oh, i see you said you set it by mdi
[12:25:56] -linuxcnc-github:#linuxcnc-devel- [13linuxcnc] 15jethornton commented on issue #125: As a temporary fix you can run LinuxCNC from a shell with the following commands... 02https://github.com/LinuxCNC/linuxcnc/issues/125#issuecomment-235650184
[12:27:25] <KGB-linuxcnc> 03Sebastian Kuzminsky 052.6 075a445 06linuxcnc 10src/emc/nml_intf/interpl.cc interp list: log calls to clear() when debugging * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=075a445
[12:27:25] <KGB-linuxcnc> 03Sebastian Kuzminsky 052.6 54bcbf8 06linuxcnc 10(7 files) test: add a hard limit test * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=54bcbf8
[12:27:25] <KGB-linuxcnc> 03Sebastian Kuzminsky 052.6 ef5dc9d 06linuxcnc 10src/emc/motion/control.c motion: when motion disables, mark all joints as "in position" * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=ef5dc9d
[12:28:03] <KGB-linuxcnc> 03Sebastian Kuzminsky 052.7 1407654 06linuxcnc 10src/emc/motion/control.c Merge remote-tracking branch 'origin/2.6' into 2.7 * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=1407654
[12:28:14] <JT-Shop> is setting a label for an issue something that seb does?
[12:28:49] <seb_kuzminsky> JT-Shop: i think anyone with enough permissions in our "LinuxCNC" group on github can do it
[12:28:56] <seb_kuzminsky> i think so far only jepler had done it
[12:29:11] <seb_kuzminsky> JT-Shop: try v2.7.5-31-g1407654, i think it fixes the "hang on limit switch" problem you reported
[12:29:21] <seb_kuzminsky> (if the buildbot can stay up long enough to build it)
[12:32:08] <JT-Shop> I didn't see a way to set the label for the issue I posted
[12:33:52] <seb_kuzminsky> JT-Shop: i just invited you to the github linuxcnc team, see if that lets you set labels on your issue
[12:36:21] <seb_kuzminsky> when i look at an issue, on the right side of the page about level with the first comment there's a "Labels" widget i can click on to add labels to the issue
[12:44:17] <JT-Shop> I just see Labels None yet
[12:44:32] <seb_kuzminsky> click on Labels
[12:44:48] <JT-Shop> it just selects the text
[12:44:59] <seb_kuzminsky> it looks like text but it's an Interactive Web 2.0 Widget (tm)
[12:45:10] <seb_kuzminsky> hm
[12:45:12] <JT-Shop> must not have a high enough clearance
[12:45:24] <seb_kuzminsky> does it have a little gear icon to the right of it?
[12:45:35] <JT-Shop> on
[12:45:36] <JT-Shop> no
[12:45:40] <seb_kuzminsky> ko
[12:45:46] <seb_kuzminsky> emal
[12:46:01] <seb_kuzminsky> accept my invitation into the team and see if that give you label-priviledge
[12:46:23] <JT-Shop> ah I see the email
[12:46:52] <JT-Shop> had my real flight simulator up on that box
[12:47:22] <seb_kuzminsky> brb
[12:49:52] <JT-Shop> ah shows me as a Member and the rest are Owners
[13:02:17] <linuxcnc-build> build #3618 of 1301.rip-precise-rtai-i386 is complete: Failure [4failed compile runtests] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1301.rip-precise-rtai-i386/builds/3618 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[13:11:03] <KGB-linuxcnc> 03Norbert Schechner 052.7 252d423 06linuxcnc 10src/emc/usr_intf/gmoccapy/gmoccapy.py 10src/emc/usr_intf/gmoccapy/release_notes.txt gmoccapy_1_5_6_4 - bug in user tabs button * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=252d423
[13:11:03] <KGB-linuxcnc> 03Norbert Schechner 052.7 ffe9b71 06linuxcnc Merge branch '2.7' of ssh://norbert@git.linuxcnc.org/git/linuxcnc.git into 2.7 * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=ffe9b71
[13:12:14] <seb_kuzminsky> i see you're a member now, can you label issues now?
[13:14:08] <linuxcnc-build> build #4418 of 1201.rip-lucid-rtai-i386 is complete: Failure [4failed compile runtests] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1201.rip-lucid-rtai-i386/builds/4418 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[13:14:08] <linuxcnc-build> build #4423 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/4423 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[13:16:30] <seb_kuzminsky> sigh
[13:17:23] <seb_kuzminsky> i think that's caused by lag between Motion and Task, and/or between Task and the UIs
[13:23:40] <KGB-linuxcnc> 03Sebastian Kuzminsky 05master dd498ec 06linuxcnc 10scripts/linuxcnc.in linuxcnc launcher: don't try to run update_ini without a DISPLAY * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=dd498ec
[13:26:14] <JT-Shop> seb_kuzminsky: I see I'm a member and the rest are owners I think that is the factor to adding labels
[13:34:28] <seb_kuzminsky> JT-Shop: i just invited you to be a "collaborator" on the "linuxcnc" repo, i bet that's what's needed
[13:36:12] <KGB-linuxcnc> 03Sebastian Kuzminsky 052.6 afdaadd 06linuxcnc 10tests/hard-limits/test-ui.py hard-limits test: be robust against realtime delay warnings * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=afdaadd
[13:41:37] <JT-Shop> seb_kuzminsky: you did it! works now
[13:44:50] <linuxcnc-build> build #2080 of 1402.rip-wheezy-rtpreempt-i386 is complete: Failure [4failed runtests] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1402.rip-wheezy-rtpreempt-i386/builds/2080 blamelist: Norbert Schechner <nieson@web.de>, Sebastian Kuzminsky <seb@highlab.com>
[13:55:30] <linuxcnc-build> build #4424 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/4424 blamelist: Norbert Schechner <nieson@web.de>, Sebastian Kuzminsky <seb@highlab.com>
[14:39:51] <seb_kuzminsky> the next 2.6 build should fix these failures
[15:44:58] <skunkworks_> The program also had a g57 in it.
[15:48:59] <linuxcnc-build> build #4421 of 1201.rip-lucid-rtai-i386 is complete: Failure [4failed compile runtests] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1201.rip-lucid-rtai-i386/builds/4421 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[15:50:59] <seb_kuzminsky> hey that wasnt supposed to fail any more
[15:59:57] <seb_kuzminsky> wow
[16:00:59] <KGB-linuxcnc> 03Sebastian Kuzminsky 052.6 93f7913 06linuxcnc 10tests/hard-limits/test-ui.py hard-limits test: try again to detect end-of-motion * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=93f7913
[16:18:31] <jepler> hm we should probably switch from C "volatile" to C11 atomics for hal data types
[16:19:23] <jepler> extern atomic_int *i;
[16:19:24] <jepler> int fn() { return *i; }
[16:19:24] <jepler> void gn(int v) { *i = v; }
[16:19:41] <jepler> given this code, it looks like gcc generates appropriate barrier instructions
[16:21:48] <jepler> including, I think, getting doubles right on arm
[16:22:20] <jepler> the downside is, the HAL memory model doesn't exactly want a full barrier before and after every access to an item in the HAL shared memory region
[16:31:44] <linuxcnc-build> build #4426 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/4426 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[16:36:47] <seb_kuzminsky> jepler: sounds good to me
[16:37:03] <seb_kuzminsky> voltailes are so 1979
[16:41:35] <seb_kuzminsky> i'm feeling beaten by task, and i'm considering reverting most or all of my task changes in 2.6 and 2.7, to undo the flood of bugs they caused
[17:14:53] <KGB-linuxcnc> 03Sebastian Kuzminsky 052.7 d4c6c1e 06linuxcnc Merge remote-tracking branch 'origin/2.6' into 2.7 * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=d4c6c1e
[17:47:46] <linuxcnc-build> build #542 of 4017.5.deb-wheezy-armhf is complete: Failure [4failed shell_3] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/4017.5.deb-wheezy-armhf/builds/542 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[18:47:59] -linuxcnc-github:#linuxcnc-devel- [13linuxcnc] 15SebKuzminsky commented on issue #117: I believe this is fixed in v2.6.12-63 and v2.7.5-31. I'd appreciate feedback from @jethornton and @c-morley, let me know if the issue is resolved for you in those versions. 02https://github.com/LinuxCNC/linuxcnc/issues/117#issuecomment-235751036
[18:57:26] <jepler> whee runtests on uspace-plus branch (using preempt-rt kernel) ran without failure all day (bare metal, 4.6 kernel)
[19:08:44] <jepler> fwiw I'm planning to merge uspace-plus to master maybe tonight
[19:15:25] <jepler> #125 must be package only, I didn't reproduce it RIP
[19:18:22] <jepler> hah it was suggested that .. yeah I was probably wrong in that commit
[19:26:48] <KGB-linuxcnc> 03Jeff Epler 05master 795d782 06linuxcnc 10scripts/linuxcnc.in Revert "runscript: remove variables now set by rip-environment" * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=795d782
[19:27:26] -linuxcnc-github:#linuxcnc-devel- [13linuxcnc] 15cradek closed issue #125: Axis Calibration and Show Hal Configuration 02https://github.com/LinuxCNC/linuxcnc/issues/125
[19:43:30] <PCW_> Hmm, making the LBP16 parser recursive for nested stored procedures is a PITA. I think Ill just forbid it
[19:43:31] <PCW_> Either that or I'll have to add a data stack separate from the return stack in the processor
[19:45:47] <jepler> forbidding it sounds like a good first step
[19:47:37] <skunkworks_> lbp16?
[19:50:50] <jepler> skunkworks_: the communication protocol used in some of the hostmot2 cards, including ethernet
[19:50:55] <jepler> "low bandwidth protocol 16" or somesuch
[19:51:01] <jepler> or maybe "larry, bob, peter 16"
[19:53:19] <skunkworks_> cool
[19:53:46] <skunkworks_> I keep missing andy - his carousel componant is working 100%
[19:53:53] <skunkworks_> (in master)
[19:54:23] <skunkworks_> When I was having issues before - I think (and he mentioned it) the fix in 2.7 wasn't in master yet. Now jogging and homing work as expected
[19:54:27] <skunkworks_> in masgter
[19:54:28] <skunkworks_> master
[20:03:08] <skunkworks_> The y axis on the matsuura was clunking - it was either a loose connection, dirty tach or hung up brush.. (we probably should have done one thing at a time)
[20:09:49] <skunkworks_> we thought initally it wat a ballscrew issue... but decoupling the servo - the clunking stayed with the servo
[20:35:09] <skunkworks_> https://www.youtube.com/watch?v=k1ICebF5MTk
[20:36:17] <Tom_itx> air or mist coolant?
[20:36:31] <Tom_itx> can't really tell there
[20:40:58] <seb_kuzminsky> looking good skunkworks_
[20:41:41] <skunkworks_> just coolant
[20:41:44] <skunkworks_> no air or mist
[20:43:09] <Tom_itx> k
[22:01:25] <KGB-linuxcnc> 03Sebastian Kuzminsky 05master 9633f07 06linuxcnc 10configs/by_interface/parport/plasma-thc-sim/simulator.hal 10src/emc/motion/control.c 10src/emc/usr_intf/axis/scripts/axis.py 10src/emc/usr_intf/gmoccapy/gmoccapy.py Merge remote-tracking branch 'origin/2.7' * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=9633f07
[22:03:22] <seb_kuzminsky> oh oops
[22:03:30] -linuxcnc-github:#linuxcnc-devel- [13linuxcnc] 15cradek closed issue #117: UI becomes unresponsive after hitting limit switch 02https://github.com/LinuxCNC/linuxcnc/issues/117
[22:04:04] <seb_kuzminsky> i forgot to update the ini of the new test for ja, that push to master's gonna fail
[22:08:00] <jepler> zzzz loooks like the evening got away from me and uspace-master still unmerged
[22:08:42] <seb_kuzminsky> somebody needs to invent git unpush
[22:18:47] <seb_kuzminsky> huh, in master's linuxcnc python module, axes don't have limits
[22:19:22] <seb_kuzminsky> joints do
[22:34:36] <KGB-linuxcnc> 03Sebastian Kuzminsky 05master 4a36e42 06linuxcnc 10tests/hard-limits/hard-limits.hal 10tests/hard-limits/hard-limits.ini 10tests/hard-limits/test-ui.py hard-limits test: update for joints/axes * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=4a36e42