Back
[01:56:29] <seb_kuzminsky> jepler: you'll have to rebase your lineardeltakins branch to something that includes the fix for bug #315
[02:00:35] <KGB-linuxcnc> 03seb 05linuxcncrsh-debugging c85d558 06linuxcnc 10src/emc/usr_intf/emcrsh.cc * add a bunch of debug printing to linuxcncrsh
[06:37:02] <jepler> http://emergent.unpythonic.net/files/sandbox/rostock-vismach.png
[06:38:18] <skunkworks> jepler, up late? ;)
[06:38:22] <skunkworks> Very neet
[06:38:22] <jepler> up early
[06:38:33] <skunkworks> neat
[06:38:41] <jepler> siggen is driving the joints, vismach uses the forward kins to get the platform in the right spot
[06:41:11] <KGB-linuxcnc> 03jepler 05jepler/lineardeltakins 55fec86 06linuxcnc 10src/emc/usr_intf/touchy/listing.py * touchy: allow optional counts argument to scroll operations
[06:41:11] <KGB-linuxcnc> 03jepler 05jepler/lineardeltakins 9bf54b3 06linuxcnc 10src/emc/ 10usr_intf/touchy/touchy.glade 10usr_intf/touchy/touchy.py * touchy: allow wheel scrolling of program start point
[06:41:15] <KGB-linuxcnc> 03jepler 05jepler/lineardeltakins e8cd18f 06linuxcnc 10src/emc/usr_intf/touchy/touchy.py * Merge branch 'touchy-mpg-scrolling'
[06:41:22] <KGB-linuxcnc> 03elson 05jepler/lineardeltakins 069894e 06linuxcnc 10docs/src/drivers/pico_ppmc.txt
[06:41:25] <KGB-linuxcnc> 23:08:49 up 15 days, 0 min, 4 users, load average: 0.02, 0.08, 0.09
[06:41:28] <jepler> seb_kuzminsky: oooohhhh (I did)
[06:41:28] <KGB-linuxcnc> USER TTY FROM LOGIN@ IDLE JCPU PCPU WHAT
[06:41:31] <KGB-linuxcnc> elson :-3 - 23Jun13 ?xdm? 4:29m 0.96s /usr/bin/gnome-
[06:41:34] <KGB-linuxcnc> elson pts/1 :0.0 24Jun13 0.00s 55:28 0.29s git commit pico
[06:41:37] <KGB-linuxcnc> elson pts/2 :0.0 26Jun13 22.00s 0.86s 0.03s info vi
[06:41:38] <jepler> (apparently that makes for spew? huh.)
[06:41:39] <KGB-linuxcnc> elson pts/3 :0.0 Sun19 25:30m 0.07s 0.07s bash
[06:41:42] <KGB-linuxcnc> Please enter the commit message for your changes. Lines starting
[06:42:07] <KGB-linuxcnc> 03elson 05jepler/lineardeltakins 01efc2d 06linuxcnc 10docs/src/drivers/pico_ppmc.txt * describe command line options
[06:42:14] <KGB-linuxcnc> 03git 05jepler/lineardeltakins 13273aa 06linuxcnc 10src/emc/rs274ngc/interp_o_word.cc * bug315: terminate skipping state on endsub only, not on return
[06:42:21] <KGB-linuxcnc> 03git 05jepler/lineardeltakins f27301c 06linuxcnc 10src/emc/rs274ngc/interp_read.cc * interp/params: do not require a named param to be defined when parsing a sub
[06:42:28] <KGB-linuxcnc> 03chris 05jepler/lineardeltakins b3bb409 06linuxcnc * Merge remote branch 'origin/bug315-2-fix' into v2.5_branch
[06:42:33] <KGB-linuxcnc> 03chris 05jepler/lineardeltakins f92d100 06linuxcnc 10tests/interp/ 03oword-bug315-p2/README 03oword-bug315-p2/expected 03oword-bug315-p2/test.ngc 03oword-bug315-p2/test.sh * Test for bug 315 part 2
[06:42:42] <KGB-linuxcnc> 03chris 05jepler/lineardeltakins 6233a2a 06linuxcnc 10docs/src/drivers/pico_ppmc.txt 03src/emc/rs274ngc/interp_namedparams.cc * Merge branch 'v2.5_branch'
[06:42:49] <KGB-linuxcnc> 03chris 05jepler/lineardeltakins 7a287c9 06linuxcnc * Merge remote branch 'origin/bug315-fix'
[06:42:55] <KGB-linuxcnc> 03seb 05jepler/lineardeltakins 7c32c4a 06linuxcnc 10tests/ 10(12 files in 11 dirs) * tests: use NO_FORCE_HOMING with the linuxcncrsh tests
[06:43:02] <KGB-linuxcnc> 03dgarrett 05jepler/lineardeltakins 09d6235 06linuxcnc 10src/emc/ 10usr_intf/touchy/filechooser.py 10usr_intf/touchy/touchy.py * touchy: regression fix (loading of file twice)
[06:43:10] <KGB-linuxcnc> 03chris 05jepler/lineardeltakins 13311d6 06linuxcnc 10src/emc/usr_intf/touchy/filechooser.py * Fix traceback when clicking an empty file-load slot
[06:43:18] <KGB-linuxcnc> 03jthornton 05jepler/lineardeltakins 61ebd82 06linuxcnc 10src/hal/components/stepgen.c
[06:43:19] <KGB-linuxcnc> increase stepgen to 16
[06:43:22] <KGB-linuxcnc> Signed-off-by: John Thornton <jthornton@gnipsel.com>
[06:43:32] <KGB-linuxcnc> 03jthornton 05jepler/lineardeltakins d5138ff 06linuxcnc 10docs/man/man9/stepgen.9 * Docs: update man page to 16
[06:43:39] <KGB-linuxcnc> 03seb 05jepler/lineardeltakins c237d02 06linuxcnc * Merge remote-tracking branch 'origin/v2.5_branch'
[06:43:44] <KGB-linuxcnc> 03dgarrett 05jepler/lineardeltakins ab4fa67 06linuxcnc 10(8 files in 3 dirs) * pyngcgui: allow configuring send_to_dir
[06:43:51] <KGB-linuxcnc> 03jepler 05jepler/lineardeltakins cae41ba 06linuxcnc 10(6 files in 3 dirs) * Linear delta kinematics
[06:43:57] <KGB-linuxcnc> 03jepler 05jepler/lineardeltakins dc81550 06linuxcnc 10src/hal/user_comps/ 10vismach/Submakefile 03vismach/rostock.py * Rostock visualization
[07:45:11] <linuxcnc-build> build #1168 of lucid-amd64-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-amd64-sim/builds/1168 blamelist: dummy, Michael Haberler <git@mah.priv.at>, John Thornton <jthornton@gnipsel.com>, Chris Radek <chris@timeguy.com>, Jon Elson <elson@jeuser.pico-systems.com>, Jeff
[07:45:11] <linuxcnc-build> Epler <jepler@unpythonic.net>, Sebastian Kuzminsky <seb@highlab.com>, Dewey Garrett <dgarrett@panix.com>
[07:45:32] <linuxcnc-build> build #1165 of lucid-i386-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-sim/builds/1165 blamelist: dummy, Michael Haberler <git@mah.priv.at>, John Thornton <jthornton@gnipsel.com>, Chris Radek <chris@timeguy.com>, Jon Elson <elson@jeuser.pico-systems.com>, Jeff Epler
[07:45:32] <linuxcnc-build> <jepler@unpythonic.net>, Sebastian Kuzminsky <seb@highlab.com>, Dewey Garrett <dgarrett@panix.com>
[07:49:03] <linuxcnc-build> build #1167 of lucid-i386-realtime-rip is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-realtime-rip/builds/1167 blamelist: dummy, Michael Haberler <git@mah.priv.at>, John Thornton <jthornton@gnipsel.com>, Chris Radek <chris@timeguy.com>, Jon Elson <elson@jeuser.pico-
[07:49:03] <linuxcnc-build> systems.com>, Jeff Epler <jepler@unpythonic.net>, Sebastian Kuzminsky <seb@highlab.com>, Dewey Garrett <dgarrett@panix.com>
[07:49:04] <linuxcnc-build> build #1163 of checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/checkin/builds/1163 blamelist: dummy, Michael Haberler <git@mah.priv.at>, John Thornton <jthornton@gnipsel.com>, Chris Radek <chris@timeguy.com>, Jon Elson <elson@jeuser.pico-systems.com>, Jeff Epler <jepler@unpythonic.net>,
[07:49:04] <linuxcnc-build> Sebastian Kuzminsky <seb@highlab.com>, Dewey Garrett <dgarrett@panix.com>
[08:21:47] <KGB-linuxcnc> 03jepler 05jepler/lineardeltakins abed7dc 06linuxcnc 10src/hal/user_comps/ 10vismach/Submakefile 03vismach/rostock.py * Rostock visualization
[08:50:21] <cradek> I wish kgb was better than it is
[09:01:09] <cradek> jepler: do you want a license in rostock.py?
[09:01:21] <jepler> cradek: yes, I'll add that
[09:12:09] <cradek> http://emergent.unpythonic.net/01373628488
[09:12:33] <linuxcnc-build> build #1166 of lucid-i386-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-sim/builds/1166 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[09:13:15] <KGB-linuxcnc> 03jepler 05jepler/lineardeltakins a04ddb8 06linuxcnc 10(6 files in 3 dirs) * Linear delta kinematics
[09:13:16] <KGB-linuxcnc> 03jepler 05jepler/lineardeltakins 956bd66 06linuxcnc 10src/hal/user_comps/ 10vismach/Submakefile 03vismach/rostock.py * Rostock visualization
[09:13:19] <KGB-linuxcnc> 03jepler 05jepler/lineardeltakins 803969e 06linuxcnc 10configs/sim/axis/ 03axis_rostock.ini 03sim_rostock.hal * sim rostock configuration
[09:13:34] <jepler> cradek: just updated with a screenshot of linuxcnc running the delta!
[09:15:42] <cradek> yay, there's the splash screen
[09:16:08] <seb_kuzminsky> this is a new failure:
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-sim/builds/1166/steps/runtests/logs/stdio
[09:16:30] <cradek> do you have any feel yet for how joint and world acc/vel interact?
[09:16:34] <jepler> cradek: no
[09:16:36] <jepler> seb_kuzminsky: this ? test.hal:1: HAL_LIB: Cannot run 'setexact' after a thread has been created
[09:16:47] <jepler> odd
[09:17:01] <seb_kuzminsky> i guess that's why it failed
[09:17:07] <seb_kuzminsky> yes it's odd
[09:17:15] <jepler> is this another face of the "something running left over" problem?
[09:17:30] <jepler> because of an earlier run that hit the 1200 second timeout and got killed with linuxcnc and hal running
[09:17:40] <seb_kuzminsky> i don't think so, i think i cleaned it up correctly since the last timeout
[09:17:45] <jepler> hm
[09:18:00] <jepler> I sure didn't touch the abs.0 test in my branch
[09:18:07] <seb_kuzminsky> yes :-/
[09:18:08] <jepler> certainly not on purpose
[09:18:12] <seb_kuzminsky> bbl
[09:19:44] <linuxcnc-build> build #1168 of lucid-i386-realtime-rip is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-realtime-rip/builds/1168 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[09:19:44] <linuxcnc-build> build #1164 of checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/checkin/builds/1164 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[09:20:10] <jepler> also I don't know why buildbot calls me "dummy Jeff Epler"
[09:23:54] <KGB-linuxcnc> 03seb 05linuxcncrsh-debugging-master 52c0547 06linuxcnc 10src/emc/usr_intf/emcrsh.cc * add a bunch of debug printing to linuxcncrsh
[09:23:55] <KGB-linuxcnc> 03seb 05linuxcncrsh-debugging-master 0d8ba8f 06linuxcnc 10src/emc/usr_intf/emcrsh.cc * more linuxcncrsh debugging
[09:54:02] <jepler> challenge: make 'daisy' play on a rostock-style machine
[10:04:00] <cradek> ugh
[10:21:24] <linuxcnc-build> build #1170 of lucid-amd64-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-amd64-sim/builds/1170 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[10:33:07] <seb_kuzminsky> when the linuxcncrsh test hangs, the linuxcncrsh process does not have the thread running that's responsible for talking to the client, but the client has sent all the stuff (i saw it go by with tcpdump)
[10:38:10] <linuxcnc-build> build #1169 of lucid-i386-realtime-rip is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-realtime-rip/builds/1169 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[10:38:11] <linuxcnc-build> build #1165 of checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/checkin/builds/1165 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[10:39:03] <seb_kuzminsky> jepler: it's not you, it's me
[10:39:11] <cradek> ha
[10:46:34] <seb_kuzminsky> now the buildbot is building the linuxcncrsh-debugging-master branch, which has a bunch of extra debug printfs in linuxcncrsh
[10:46:47] <seb_kuzminsky> not sure if that will help
[11:34:16] <linuxcnc-build> build #1168 of lucid-i386-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-sim/builds/1168 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[11:37:49] <linuxcnc-build> build #1171 of lucid-amd64-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-amd64-sim/builds/1171 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[11:37:50] <linuxcnc-build> build #1166 of checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/checkin/builds/1166 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[11:39:36] <jepler> f---
[11:40:00] <jepler> in 2011, mah contributed a file to nc_files with the notice (in german) 'Copyright 2010 / All Rights Reserved'
[11:40:28] <seb_kuzminsky> whoops
[11:41:05] <jepler> even more inexplicable, he deleted a file by cradek of the same name
[11:42:54] <jepler> it's cone.ngc
[11:43:23] <cradek> hmmm
[11:44:08] <cradek> so just remove it I guess? at least we've never packaged that up in a release.
[11:44:14] <jepler> yes
[11:44:22] <jepler> is your cone.ngc (still in v2.5_branch) worth reverting to?
[11:44:31] <jepler> or just delete it?
[11:44:58] <cradek> we still have the 5axis sim, so why not put mine back
[11:45:23] <seb_kuzminsky> sometimes i think we should be more careful about reviewing all commits before they go in
[11:45:39] <KGB-linuxcnc> 03jepler 05master 984dc1c 06linuxcnc 10nc_files/cone.ngc * Do not distribute an 'all rights reserved' file
[11:45:44] <linuxcnc-build> build #1169 of lucid-i386-sim is complete: Failure [4failed git] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-sim/builds/1169 blamelist: Jeff Epler <jepler@unpythonic.net>
[11:46:19] <seb_kuzminsky> "failed git" means my network blinked, usually
[11:46:23] <cradek> or, we should just be more careful about our own commits
[11:46:47] <jepler> cradek had to catch me this morning pushing files without copyright statements
[11:47:05] <seb_kuzminsky> yeah, several of us are more careful than any of us
[11:47:11] <cradek> well that's a MUCH smaller sin
[11:47:43] <KGB-linuxcnc> 03jepler 05jepler/lineardeltakins 7a9e3f3 06linuxcnc 10src/hal/user_comps/ 10vismach/Submakefile 03vismach/rostock.py * Rostock visualization
[11:47:43] <KGB-linuxcnc> 03jepler 05jepler/lineardeltakins cfaa854 06linuxcnc 10configs/sim/axis/ 03axis_rostock.ini 03sim_rostock.hal * sim rostock configuration
[11:48:30] <cradek> seb_kuzminsky: could it just wait and try again? not sure what kind of blinks you are getting...?
[11:48:49] <KGB-linuxcnc> 03jepler 05jepler/lineardeltakins 68eabae 06linuxcnc 10src/hal/user_comps/ 10vismach/Submakefile 03vismach/rostock.py * Rostock visualization
[11:48:49] <KGB-linuxcnc> 03jepler 05jepler/lineardeltakins 8d7c29b 06linuxcnc 10configs/sim/axis/ 03axis_rostock.ini 03sim_rostock.hal * sim rostock configuration
[11:49:00] <seb_kuzminsky> dns failure
[11:49:01] <jepler> (yes I just pushed again, found a debug print to remove)
[11:51:22] <jepler> (it's repeating the same commit messages over and over again because I'm rebasing)
[11:52:29] <cradek> yay kgb
[11:52:32] <jepler> it's so fun to feel like I have something to hack on in linuxcnc
[11:52:48] <jepler> unfortunately the fun part is nearly over, and the part that involves hardware is soon to start :-/
[11:52:52] <cradek> haha
[12:08:11] <seb_kuzminsky> jepler: there's a guy at work who's been wanting to buy a kessel(?) and he just got really excited when i shoed him your rostock vismach
[12:09:15] <jepler> seb_kuzminsky: same kins should work on rostock and kessel
[12:09:21] <seb_kuzminsky> yeah
[12:09:27] <jepler> kossel
[12:09:31] <jepler> whatever their darn names are
[12:09:38] <seb_kuzminsky> is that from starwars?
[12:09:43] <jepler> kessel is
[12:10:07] <jepler> 12 parsecs, etc
[12:10:29] <seb_kuzminsky> cool
[12:10:42] <jepler> Kossel may refer to:
[12:10:42] <jepler> Albrecht Kossel (1853–1927), German biochemist, 1910 Nobel laureate
[12:10:45] <jepler> Walther Kossel (1888–1956), son of Albrecht Kossel, German physicist known for his theory of the chemical bond (ionic bond/octet rule), etc.
[12:10:52] <jepler> I assume the machine is named after one of these two
[12:11:16] <jepler> ah yes, the father
[12:12:02] <jepler> wikipedia only knows Rostock as a place, not a person.
[12:12:10] <seb_kuzminsky> when i googled kessel, the first hit was an article on "wookiepedia"
[12:12:20] <seb_kuzminsky> god bless the internet
[12:14:28] <cradek> beware: "loud guy" in the coffee shop says you shouldn't trust wikipedia because anyone can edit it
[12:14:59] <cradek> also, fyi, taxes are bad
[12:39:02] <memleak> Hello everyone! I normally don't idle on IRC, are there always quite a few automated builds that fail?
[12:39:15] <seb_kuzminsky> not normally this bad
[12:39:26] <seb_kuzminsky> i broke something and i've had a really hard time tracking it down :-(
[12:39:33] <seb_kuzminsky> sorry for the noise
[12:39:46] <memleak> Haha, it's ok, I just wasn't sure.
[12:42:32] <linuxcnc-build> build #1172 of lucid-amd64-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-amd64-sim/builds/1172 blamelist: Jeff Epler <jepler@unpythonic.net>
[12:48:24] <seb_kuzminsky> i'm away from my linuxcnc.org git push key... could someone push these two commits to linuxcncrsh-debugging-master?
http://highlab.com/~seb/linuxcnc/linuxcncrsh
[12:48:39] <cradek> sure
[12:48:41] <linuxcnc-build> build #1171 of lucid-i386-realtime-rip is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-realtime-rip/builds/1171 blamelist: Jeff Epler <jepler@unpythonic.net>
[12:48:41] <linuxcnc-build> build #1167 of checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/checkin/builds/1167 blamelist: Jeff Epler <jepler@unpythonic.net>
[12:48:42] <seb_kuzminsky> (the first one should go in mainline, i'll do that later)
[12:48:51] <seb_kuzminsky> thx
[12:51:31] <KGB-linuxcnc> 03seb 05linuxcncrsh-debugging-master 6275c48 06linuxcnc 10bin/.gitignore * don't ignore a file we're tracking
[12:51:32] <KGB-linuxcnc> 03seb 05linuxcncrsh-debugging-master f37c0f4 06linuxcnc 10bin/.gitignore 03bin/traced-linuxcncrsh 10debian/control.in 10tests/linuxcncrsh/linuxcncrsh-test.ini * bogus meta-testing: run linuxcncrsh under strace
[12:51:55] <seb_kuzminsky> thanks
[13:36:59] <jepler> pcw_home: can you tell me the difference between the various uart*.vhd files? In hostmot2-firwmware I have 4: uartr8.vhd uartr.vhd uartx8.vhd uartx.vhd
[13:40:39] <jepler> r and x are receive and transmit ends?
[13:40:55] <pcw_home> the uartr are rx, uartt is tx 8 is 8 bit
[13:41:00] <jepler> and 8 is specialized for only 8-bits?
[13:41:08] <jepler> aha I nearly puzzled it out in time
[13:41:11] <pcw_home> Yes
[13:41:31] <pcw_home> normal (no 8 ) is 32 bit interface UART
[13:42:53] <pcw_home> AFAICR the 8 bit ones are only used by sserial (sserial has a 8 bit processor)
[13:45:13] <linuxcnc-build> build #1173 of lucid-amd64-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-amd64-sim/builds/1173 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[13:47:26] <linuxcnc-build> build #1170 of lucid-i386-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-sim/builds/1170 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[14:17:15] <linuxcnc-build> build #1173 of lucid-i386-realtime-rip is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-realtime-rip/builds/1173 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[14:17:16] <linuxcnc-build> build #1168 of checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/checkin/builds/1168 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[14:17:52] <cradek> % ./a.out
[14:17:52] <cradek> 0 0 0 -> nan nan nan -> nan nan nan
[14:17:52] <cradek> 0 0 1 -> nan nan nan -> nan nan nan
[14:17:56] <cradek> ok, so no
[14:18:57] <cradek> *theta = 180.0*atan(-zj/(y1 - yj))/pi + ((yj>y1)?180.0:0.0);
[14:19:05] <cradek> this has the distinct smell of someone not knowing what they're doing
[14:22:17] <jepler> cradek: rotary tripod kinematics?
[14:22:26] <cradek> yeah
[14:24:11] <cradek> 0 0 -100 -> 1.56074 1.56074 1.56074 -> 0 1.84109e-13 -99.9999
[14:24:12] <cradek> oh hey
[14:26:15] <jepler> (master branch, not ja3) if I am in coordinated mode and jog to an unreachable location, my feedback coordinates go to nan but when I jog back into the volume it repairs itself
[14:26:28] <jepler> but switch to joint mode and everything goes to nan and I don't know if I can get it back
[14:29:36] <jepler> and of course POSITION_FILE will blindly read infs and nans back in
[14:30:13] <cradek> I have a dim memory of hearing that complaint before
[14:42:21] <jepler> yeah me too
[14:43:07] <cradek> oh it's weird because it's translated from Zsombor-Murray's BASIC code
[14:43:25] <cradek> I wasn't far enough in the paper yet
[14:45:41] <cradek> > That is the purpose of this article: to provide a clear kinematic analysis useful to those who may wish to program and employ nice little three legged robots suited to a line and sphere intersection model.
[14:45:57] <cradek> is this, um, a license statement?
[14:51:07] <jepler> how do I cause an estop connection when I detect a kins problem in emc/motion/control.c:get_pos_cmds() (which is where I would detect a problem in inverse kins)
[14:51:21] <jepler> I've got it to spew 'kinematicsInverse failed' until I estop
[14:51:45] <cradek> you could use an amp-fault input? that'll turn the amps off (not estop)
[14:53:18] <andypugh> jlivoria appears to have addressed Steve Blamore's tapered thread problem:
http://www.linuxcnc.org/index.php/english/forum/20-g-code/26760-new-gcode-cycle#36582
[15:08:36] <jepler> Linear move on line 0 has no inverse kinematics solution
[15:08:41] <jepler> kinematicsInverse failed
[15:08:44] <jepler> this is looking more promising
[15:08:47] <cradek> whoah
[15:08:56] <jepler> but I'm probably very silly for working on top of master and not ja3
[15:09:04] <cradek> is that a thing we already had, or did you add it?
[15:09:05] <jepler> either this is already done there, or it's better to do it on top of there..
[15:09:08] <jepler> it's new
[15:09:14] <jepler> kinematicsInverse's return value was never checked!
[15:09:21] <cradek> ha
[15:09:24] <jepler> and joint positions were blindly copied even if they were nans
[15:09:28] <jepler> or inf
[15:09:45] <cradek> yeah you should surely move to ja3 sooner rather than later
[15:09:54] <cradek> or just whistle innocently while you merge it into master
[15:10:02] <linuxcnc-build> build #1174 of lucid-amd64-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-amd64-sim/builds/1174 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[15:10:04] <jepler> I would not want to get on seb's bad side
[15:10:12] <jepler> he'd make me produce 18 million lines of debug output as penance
[15:10:28] <linuxcnc-build> build #1171 of lucid-i386-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-sim/builds/1171 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[15:11:54] <linuxcnc-build> build #1174 of lucid-i386-realtime-rip is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-realtime-rip/builds/1174 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[15:11:54] <linuxcnc-build> build #1169 of checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/checkin/builds/1169 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[15:17:56] <andypugh> Aye, and you would have to remember to keep the line count in a long, or do it all over again.
[15:27:22] <seb_kuzminsky> cradek: could you push 0003 and 0004 for me...?
http://highlab.com/~seb/linuxcnc/linuxcncrsh
[15:27:35] <seb_kuzminsky> on top of linuxcncrsh-debugging-master pls
[15:28:17] <seb_kuzminsky> i should bring my key next time...
[15:41:51] <seb_kuzminsky> jepler: i'm getting close to having the ja branch rebased & cleaned up & ready for review
[15:42:50] <jepler> seb_kuzminsky: great, I'll keep that in mind
[15:43:19] <seb_kuzminsky> do you have your g.l.o push key handy? could you push a couple of test debugging commits ^^^
[15:44:11] <cradek> seb_kuzminsky: do you want to send me another key?
[15:53:24] <KGB-linuxcnc> 03seb 05linuxcncrsh-debugging-master 59bc079 06linuxcnc 10tests/ 10(7 files in 7 dirs) * enable strace on all the tests that use linuxcncrsh
[15:53:24] <KGB-linuxcnc> 03seb 05linuxcncrsh-debugging-master 0d64189 06linuxcnc 10bin/traced-linuxcncrsh * add some helpful info to the linuxcncrsh strace wrapper
[16:03:01] <seb_kuzminsky> i wonder why those tests passed fine for so long, and now have started failing so often
[16:46:56] <seb_kuzminsky> looks like a timing issue with signal delivery
[16:47:11] <seb_kuzminsky> linuxcncrsh uses signal(2) which is marked in the manpages as obsolete, and to use sigaction() instead
[16:50:03] <linuxcnc-build> build #1172 of lucid-i386-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-sim/builds/1172 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[16:51:52] <linuxcnc-build> build #1175 of lucid-amd64-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-amd64-sim/builds/1175 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[16:52:58] <andypugh> <points finger of Blame at Seb>
[17:06:18] <linuxcnc-build> build #1175 of lucid-i386-realtime-rip is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-realtime-rip/builds/1175 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[17:06:18] <linuxcnc-build> build #1170 of checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/checkin/builds/1170 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[17:30:57] <KGB-linuxcnc> 03seb 05linuxcncrsh-debugging-master 45c027d 06linuxcnc 10src/emc/usr_intf/emcrsh.cc * use the non-deprecated way to specify signal handlers
[18:18:47] <linuxcnc-build> build #1173 of lucid-i386-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-sim/builds/1173 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[18:20:43] <linuxcnc-build> build #1176 of lucid-i386-realtime-rip is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-realtime-rip/builds/1176 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[18:20:59] <linuxcnc-build> build #1171 of checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/checkin/builds/1171 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[18:32:10] <seb_kuzminsky> why is linuxcncrsh's client thread getting SIGPIPE when i've told it to ignore SIGPIPE? and why does it only happen on lucid? a bug in lucid's signal handling?
[18:36:37] <seb_kuzminsky> /proc/$(pidof linuxcncrsh)/status shows the expected set of signals being ignored
[18:37:44] <seb_kuzminsky> ugh
[18:38:41] <cradek> does it fork?
[18:38:56] <seb_kuzminsky> it calls pthread_create() on accept()
[18:39:33] <seb_kuzminsky> both the main thread and the newly created thread are ignoring the expected signals, according to /proc/$PID/status
[18:40:42] <seb_kuzminsky> if i sent the process SIGPIPE using kill from the shell, it ignores it correctly
[18:41:09] <seb_kuzminsky> in the strace i see it writing to the closed pipe, write returns -1 (EPIPE), and then the thread gets SIGPIPE and dies
[18:41:12] <seb_kuzminsky> wtf
[18:41:23] <seb_kuzminsky> i'll look at it more later
[19:58:43] <KGB-linuxcnc> 03seb 05linuxcncrsh-debugging-master 4e33653 06linuxcnc 10src/emc/usr_intf/emcrsh.cc * un-race-ify the linuxcncrsh client connection logic
[19:58:43] <KGB-linuxcnc> 03seb 05linuxcncrsh-debugging-master 8538052 06linuxcnc 10tests/ 10(5 files in 5 dirs) * now that connecting to linuxcncrsh is not racy, we dont need sleep
[20:57:19] <linuxcnc-build> build #1174 of lucid-i386-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-sim/builds/1174 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[20:57:44] <linuxcnc-build> build #1177 of lucid-i386-realtime-rip is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-realtime-rip/builds/1177 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[20:58:02] <linuxcnc-build> build #1177 of lucid-amd64-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-amd64-sim/builds/1177 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[20:58:02] <linuxcnc-build> build #1172 of checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/checkin/builds/1172 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[21:00:47] <andypugh> "the blamelist birds go "pointy pointy" "
http://www.youtube.com/watch?v=nU4RD2f2BnY
[21:32:38] <KGB-linuxcnc> 03seb 05linuxcncrsh-debugging-master 540662e 06linuxcnc 10src/emc/usr_intf/emcrsh.cc * make a local variable less global
[21:32:38] <KGB-linuxcnc> 03seb 05linuxcncrsh-debugging-master 9a069e6 06linuxcnc 10src/emc/usr_intf/emcrsh.cc * don't ignore SIGCHLD
[21:32:41] <KGB-linuxcnc> 03seb 05linuxcncrsh-debugging-master 4ec92cf 06linuxcnc 10src/emc/usr_intf/emcrsh.cc * report SIGPIPE action when starting a new thread
[21:35:08] <memleak> seb_kuzminsky, what 3.x kernel again did you say you had working with the master branch?
[21:35:46] <memleak> I just tried 3.8.13 and there was a lot of problems with how linuxcnc was looking at the kernel headers.
[21:36:09] <memleak> a lot has changed from 3.7 to 3.8.
[21:37:13] <memleak> If it was 3.5.7 I understand then, because that was before the monstrous kernel header re-working.
[21:49:26] <andypugh> why is LinuxCNC impossible to find at sourceforge.com? Even searching for "Enhanced Machine Controller" has zero results.
[21:51:59] <Tom_itx> http://sourceforge.net/projects/emc/?source=directory
[21:52:02] <Tom_itx> bingo
[21:52:18] <andypugh> As far as I can tell you can not get to the LinuxCNC bug tracker if your starting point is www.sourceforge.com. Which seems odd given the existence of:
http://sourceforge.net/projects/emc/
[21:53:38] <andypugh> Ah, I am wrong. If you search for "emc" you get it, but not with "emc2" "cnc" "LinuxCNC" "Enhanced" "Enhanced Machine" "Machine Controller"....
[21:54:48] <andypugh> And when you do get there, the one-star worst of all review is "I loved this software" ?
[22:21:19] <CaptHindsight> "Linux precise-rtai-x86 3.5.7-rtai+ #5 SMP Thu Feb 28 00:21:45 MST 2013 i686 i686 i386 GNU/Linux"
[22:21:37] <CaptHindsight> memleak: ^^
[22:23:20] <memleak> I'm diffing the drivers/gpu code from 3.4.52 to 3.7.10 right now figuring out which way to go. they both have the southern islands (HD 7000 series) PCI IDs.
[22:24:58] <memleak> the 3.5 series is basically "bad"
[22:29:14] <linuxcnc-build> build #1175 of lucid-i386-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-sim/builds/1175 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[22:30:07] <linuxcnc-build> build #1178 of lucid-amd64-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-amd64-sim/builds/1178 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[22:30:32] <linuxcnc-build> build #1178 of lucid-i386-realtime-rip is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-realtime-rip/builds/1178 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[22:30:32] <linuxcnc-build> build #1173 of checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/checkin/builds/1173 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[22:30:40] <memleak> 3.4.52 wins! Now to write yet another RTAI patch.
[22:34:41] <memleak> 3.4.6 to 3.4.52 this should be fun :)
[22:40:17] <zultron> logger[mah],
[22:40:17] <logger[mah]> zultron: Log stored at
http://linuxcnc.mah.priv.at/irc/%23linuxcnc-devel/2013-07-13.html
[22:49:37] <memleak> I think I excited someone with a comment of mine :) zultron?
