Back
[01:56:08] <mhaberler> Seb: saw your comment - I'm next to ARM hw only monday, will look into it
[01:56:49] <mhaberler> re setcap: yes, one would have to add that step, provided it is important - the commit only shows how it can be done, but it doesnt do it
[01:57:51] <mhaberler> of the actual sim configs, I am not aware of any which have more than one thread, so this would be about the runtest only - the question is: is it worth that
[01:58:16] <mhaberler> if yes, we have a new pkg dep, and a command to add during make setuid
[02:01:18] <mhaberler> it doesnt make terrible sense to work uncoordinated on a particular issue in general - you had pinged me on this, I said I would look into it, then you come up with a patch silently - kinda wasted effort one way or the other
[02:02:34] <mhaberler> btw mail is an improvement over 'let me hope my addressee reads back some IRC log' - this does not make sense to me at all
[10:23:40] <KGB-linuxcnc> 03Norbert Schechner 05master 57a170a 06linuxcnc 10(6 files in 4 dirs) gmoccapy_1_0_1 - solved jog button bug * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=57a170a
[10:29:35] <awallin> what should "./configure --prefix=" be if I want to build a version to install?
[10:31:53] <KGB-linuxcnc> 03Norbert Schechner 05master 60cd047 06linuxcnc 10.gitignore gitignore - added rule to ignore the preference files * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=60cd047
[11:00:58] <linuxcnc-build> build #1830 of precise-amd64-sim is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/precise-amd64-sim/builds/1830 blamelist: Norbert Schechner <nieson@web.de>
[11:11:29] <linuxcnc-build> build #1831 of checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/checkin/builds/1831 blamelist: Norbert Schechner <nieson@web.de>
[11:20:54] <seb_kuzminsky> awallin: do you want to "install by hand" or build debs and install those?
[11:21:51] <awallin> seb_kuzminsky: mostly I just wanted to run a python script that does "import linuxcnc", but doesn't find the module with a rip build
[11:22:40] <seb_kuzminsky> we have tons of scripts that run 'import linuxcnc' in rip builds
[11:23:09] <seb_kuzminsky> did you run 'scripts/rip-environment'? it sets PYTHONPATH correctly, so you can import our python modules without hassle
[11:23:32] <awallin> hm, but that works from the particular shell where rip-envirnoment was run?
[11:23:48] <seb_kuzminsky> awallin: yes, that's true
[11:36:03] <seb_kuzminsky> you can run the same rip-environment script in as many shells as you want
[12:37:50] <linuxcnc-build> build #1421 of deb-precise-sim-binary-amd64 is complete: Failure [4failed shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-precise-sim-binary-amd64/builds/1421 blamelist: Norbert Schechner <nieson@web.de>
[12:38:06] <linuxcnc-build> build #1422 of deb-precise-sim-binary-i386 is complete: Failure [4failed shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-precise-sim-binary-i386/builds/1422 blamelist: Norbert Schechner <nieson@web.de>
[12:39:11] <linuxcnc-build> build #1416 of deb-lucid-sim-binary-i386 is complete: Failure [4failed shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-lucid-sim-binary-i386/builds/1416 blamelist: Norbert Schechner <nieson@web.de>
[12:39:18] <linuxcnc-build> build #251 of deb-precise-rt-binary-i386 is complete: Failure [4failed shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-precise-rt-binary-i386/builds/251 blamelist: Norbert Schechner <nieson@web.de>
[12:40:14] <linuxcnc-build> build #3 of deb-wheezy-sim-binary-armhf is complete: Failure [4failed apt-get-update shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-wheezy-sim-binary-armhf/builds/3 blamelist: Norbert Schechner <nieson@web.de>
[12:55:15] <linuxcnc-build> build #1417 of deb-lucid-rt-binary-i386 is complete: Failure [4failed shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-lucid-rt-binary-i386/builds/1417 blamelist: Norbert Schechner <nieson@web.de>
[12:55:49] <linuxcnc-build> build #1416 of deb-lucid-sim-binary-amd64 is complete: Failure [4failed shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-lucid-sim-binary-amd64/builds/1416 blamelist: Norbert Schechner <nieson@web.de>
[13:56:04] <KGB-linuxcnc> 03Michael Haberler 05master b7a72d4 06linuxcnc 10src/hal/user_comps/Submakefile xhc-hb03: link with -lstdc++ * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=b7a72d4
[14:43:35] <cradek> I wonder if we could devise a runtest that would detect nonatomic assignment to hal pins
[14:44:09] <cradek> we used to worry about that and tweak compile flags to force *halpin=value; to be one instruction
[14:44:43] <cradek> now that we have a variety of platforms and compilers, I wonder how we'd notice if it wasn't
[14:48:06] <pcw_home> are hal pins all 32 bit entities?
[14:48:15] <cradek> no, they are 64
[14:48:58] <cradek> er I'm wrong; hal_u32_t and hal_s32_t are what they say
[14:49:08] <cradek> hal_float_t is a double
[14:49:47] <cradek> volatile double __attribute__((aligned(8)))
[14:50:24] <pcw_home> hal_float_t is public? (didnt know there were 64 pin pins)
[14:50:32] <pcw_home> 64 bit pins
[14:50:41] <cradek> hal_float_t is what almost everything is made of
[14:51:28] <cradek> we changed it years back because positions in degrees were really crunchy with only 32 bits
[14:51:51] <cradek> when people turned their rotary axes around a few times it got bad
[14:52:54] <pcw_home> also hard to overflow
[14:53:13] <cradek> changed in 2008
[14:53:49] <cradek> http://emergent.unpythonic.net/01225680565
[14:54:24] <cradek> http://emergent.unpythonic.net/01189609097
[14:55:45] <cradek> obviously jepler has put some thought into this already
[14:58:00] <pcw_home> So its a good question.... (about how this is handled for ARM for example)
[14:58:28] <cradek> yes seems like it could lead to very subtle and very bad breakage
[14:58:56] <cradek> reading seb's open questions about monotonic scheduling made me think of it again
[15:01:21] <pcw_home> yeah a base thread interrupt at the wrong time could cause some pain
[15:02:54] <cradek> disassembling and looking would be easier than devising a runtest, but a test that we'd always have in the future would sure be superior.
[15:04:30] <cradek> maybe someone who can build on arm can get doubletest from
http://emergent.unpythonic.net/01225549319 to run again
[15:22:46] <linuxcnc-build> build #1422 of deb-precise-sim-binary-amd64 is complete: Failure [4failed shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-precise-sim-binary-amd64/builds/1422 blamelist: Michael Haberler <git@mah.priv.at>
[15:22:52] <linuxcnc-build> build #1423 of deb-precise-sim-binary-i386 is complete: Failure [4failed shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-precise-sim-binary-i386/builds/1423 blamelist: Michael Haberler <git@mah.priv.at>
[15:24:07] <linuxcnc-build> build #252 of deb-precise-rt-binary-i386 is complete: Failure [4failed shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-precise-rt-binary-i386/builds/252 blamelist: Michael Haberler <git@mah.priv.at>
[15:24:25] <linuxcnc-build> build #1417 of deb-lucid-sim-binary-i386 is complete: Failure [4failed shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-lucid-sim-binary-i386/builds/1417 blamelist: Michael Haberler <git@mah.priv.at>
[15:24:30] <linuxcnc-build> build #4 of deb-wheezy-sim-binary-armhf is complete: Failure [4failed apt-get-update shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-wheezy-sim-binary-armhf/builds/4 blamelist: Michael Haberler <git@mah.priv.at>
[15:39:14] <linuxcnc-build> build #1418 of deb-lucid-rt-binary-i386 is complete: Failure [4failed shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-lucid-rt-binary-i386/builds/1418 blamelist: Michael Haberler <git@mah.priv.at>
[15:41:48] <linuxcnc-build> build #1417 of deb-lucid-sim-binary-amd64 is complete: Failure [4failed shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-lucid-sim-binary-amd64/builds/1417 blamelist: Michael Haberler <git@mah.priv.at>
[16:14:38] <seb_kuzminsky> all this deb building breakage is from norbert's recent gmoccapy commits
[16:31:11] <seb_kuzminsky> oops, there's another gmoccapy problem
[16:31:51] <seb_kuzminsky> after building a deb, git status shows this:
[16:31:56] <seb_kuzminsky> # modified: ../lib/python/gmoccapy/dialogs.py
[16:31:56] <seb_kuzminsky> # modified: ../lib/python/gmoccapy/notification.py
[16:31:57] <seb_kuzminsky> # modified: ../lib/python/gmoccapy/player.py
[16:31:57] <seb_kuzminsky> # modified: ../lib/python/gmoccapy/widgets.py
[16:32:29] <KGB-linuxcnc> 03Sebastian Kuzminsky 05master 1fda8e6 06linuxcnc 10src/Makefile fix "make install" for gmoccapy * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=1fda8e6
[17:57:33] <linuxcnc-build> build #5 of deb-wheezy-sim-binary-armhf is complete: Failure [4failed apt-get-update shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-wheezy-sim-binary-armhf/builds/5 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[18:09:17] <KGB-linuxcnc> 03Dewey Garrett 05dgarr/ini_includes da928ed 06linuxcnc 10docs/src/config/ini_config.txt ini_config.txt: update for #INCLUDE directive * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=da928ed
[19:32:24] <linuxcnc-build> build #6 of deb-wheezy-sim-binary-armhf is complete: Failure [4failed apt-get-update shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-wheezy-sim-binary-armhf/builds/6 blamelist: Dewey Garrett <dgarrett@panix.com>
[19:38:16] <dgarr> configure: error: no convert, documentation cannot be built
[20:19:51] <KGB-linuxcnc> 03Dewey Garrett 05dgarr/ini_includes c33fa8f 06linuxcnc 10configs/sim/axis/ini_with_includes/includes_demo.ini 10docs/src/config/ini_config.txt 10scripts/linuxcnc.in linuxcnc.in: cd to ini dir in handle_includes() * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=c33fa8f
[21:43:07] <linuxcnc-build> build #7 of deb-wheezy-sim-binary-armhf is complete: Failure [4failed apt-get-update shell_3] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/deb-wheezy-sim-binary-armhf/builds/7 blamelist: Dewey Garrett <dgarrett@panix.com>
[22:51:22] <KGB-linuxcnc> 03Dewey Garrett 05dgarr/ini_includes a61d885 06linuxcnc 10configs/sim/axis/ini_with_includes/includes_demo.ini 10scripts/linuxcnc.in linuxcnc.in: annotate autogenerated file * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=a61d885
[23:21:33] <seb_kuzminsky> dgarr: to see what build dependencies you're missing, see here:
http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Installing_LinuxCNC#Resolving_outstanding_build_dependencies
[23:23:12] <dgarr> my reference was to :
http://buildbot.linuxcnc.org/dists/precise/scratch-sim/binary-i386/linuxcnc-doc-en_2.6.0~pre~dgarr.ini.includes~c33fa8f_all.deb
[23:23:41] <dgarr> oops I mean reference to:
http://buildbot.linuxcnc.org/buildbot/builders/deb-wheezy-sim-binary-armhf/builds/6/steps/shell_3/logs/stdio
[23:24:01] <dgarr> it seems convert is not there (imagemagick?)
[23:26:30] <dgarr> on the buildbot for deb-wheezy
[23:28:33] <seb_kuzminsky> oh, i didnt understand that, but now i see what you mean
[23:29:03] <dgarr> i think the branch for ini_includes is working ok and has documentation now
[23:29:18] <seb_kuzminsky> i saw that you made that commit but i didnt read it yet - thanks
[23:31:15] <seb_kuzminsky> hmm
[23:31:45] <seb_kuzminsky> all the deb-* builders build in a pbuilder chroot, and install everything the source package says it build-depends on
[23:31:54] <seb_kuzminsky> that includes imagemagick, which supplies convert
[23:33:06] <seb_kuzminsky> the ini-includes branch is based on master, so it won't build on arm, but it shouldnt fail to find convert...