#linuxcnc-devel | Logs for 2015-04-17

[10:04:56] <seb_kuzminsky> http://mail.rtai.org/pipermail/rtai/2015-April/026820.html
[10:33:54] <kwallace> Thank you Paul C.
[10:41:42] <skunkworks> zlog
[10:42:57] <skunkworks> wonder what paul is doing now?
[10:43:21] <skunkworks> tuxcnc hasn't been updated for a long time.
[11:01:25] <lair82> Good Morning Gentlemen, I started moving the slides around today on my latest build, and was wondering if there may be a bug with the calibration utility. I am running Debian with RT Preempt, using gscreen.
[14:47:02] <dgarr> seb_kuzminsky: for review (see today's #linuxcnc log) bug fix http://www.panix.com/~dgarrett/stuff/0001-emccalib.tcl-allow-whitespace-on-detected-setp-lines.patch
[14:47:12] <dgarr> should it go to 2.6 or 2.7?
[14:55:31] <jepler> 2.6 would seem reasonable to me fwiw
[14:55:55] <jepler> consider your patch reviewed but not tested by yours truly
[14:56:43] <dgarr> jepler: thanks
[14:59:18] <seb_kuzminsky> dgarr: i assume you tested it? (i read it but did not run it)
[15:13:11] <dgarr> i tested a number of cases after reproducing the problem, seems ok, 2.7 runtests passed, i won't be surprised if there are other places where leading whitespace could be problematic in that file, but i'm not going there
[15:26:15] <seb_kuzminsky> dgarr: great! please push to 2.6
[15:51:24] <KGB-linuxcnc> 03Dewey Garrett 052.6 453533f 06linuxcnc 10tcl/bin/emccalib.tcl emccalib.tcl: allow whitespace on detected setp lines * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=453533f
[15:53:14] <seb_kuzminsky> thanks
[16:52:36] <cradek> dgarr: thanks you! is that what lair82 hinted about earlier?
[19:06:02] <dgarr> cradek: yes, and it led me to a bug fix for 2.7
[19:06:18] <KGB-linuxcnc> 03Dewey Garrett 052.7 d95dd4e 06linuxcnc Merge remote-tracking branch 'origin/2.6' into 2.7 * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=d95dd4e
[19:06:18] <KGB-linuxcnc> 03Dewey Garrett 052.7 2f78543 06linuxcnc 10tcl/bin/emccalib.tcl emccalib.tcl: ignore LIB: hal files * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=2f78543
[19:25:13] <jepler> 13:23:33 < skunkworks> Bring up axis sim -> home -> run splash screen program -> hit estop -> then either hit the refresh or try to open another program.
[19:25:29] <jepler> reproduced. debugging shows axis is hung at the wait_complete call:
[19:25:36] <jepler> File "/home/jepler/src/linuxcnc/bin/axis", line 1096, in open_file_guts
[19:25:39] <jepler> ensure_mode(linuxcnc.MODE_MDI)
[19:25:42] <jepler> File "/home/jepler/src/linuxcnc/bin/axis", line 899, in ensure_mode
[19:25:44] <jepler> c.wait_complete()
[19:31:20] <jepler> emcWaitCommandComplete echo_serial_number=21 serial=21 status=RCS_EXEC
[19:31:51] <jepler> axis expects that for the command to be complete, either the echo_serial_number is bigger than serial, or status is RCS_DONE or RCS_ERROR
[19:40:50] <jepler> in milltask, some of the parts of this long expression are false
[19:40:50] <jepler> 3428 } else if (!taskPlanError && !taskExecuteError &&
[19:41:55] <jepler> (gdb) p emcStatus->task.execState
[19:41:56] <jepler> $12 = EMC_TASK_EXEC_WAITING_FOR_MOTION_AND_IO
[19:42:01] <jepler> (gdb) p emcStatus->motion.status
[19:42:01] <jepler> $13 = 2
[19:42:03] <jepler> [RCS_EXEC]
[19:42:16] <jepler> (gdb) p emcTaskCommand
[19:42:16] <jepler> $10 = (NMLmsg *) 0x256dbc0
[19:42:29] <jepler> (gdb) p emcTaskCommand->type
[19:42:29] <jepler> $16 = 516
[19:43:04] <jepler> [EMC_TASK_PLAN_SYNCH_TYPE]
[19:43:39] <jepler> so IMHO this bug should be investigated as a bug in task, not ui
[19:47:27] <jepler> .. afk
[20:20:59] <cradek> ick
[20:50:44] <skunkworks> I keep meaning to create a tracker
[20:50:48] <skunkworks> *bug