Re: [Emc-developers] requesting help debugging ubc

2014-01-19 Thread Chris Radek
On Mon, Jan 20, 2014 at 02:11:16AM +0100, Michael Haberler wrote: > > Assuming the .so's are built, it comes down to where they are > installed (/usr/lib/linuxcnc?) and tell ldconfig about it (maybe a > linuxcnc entry in /etc/ld.conf.d) currently we use -Wl,-rpath to do this. --

Re: [Emc-developers] requesting help debugging ubc

2014-01-19 Thread Sebastian Kuzminsky
On 01/19/2014 06:47 PM, Michael Haberler wrote: > > reading back on IRC: > >> [17:20:41] zultron: i'm trying to debug that hal_init >> problem i'm seeing in packaged ubc, and i'm confused as to why the >> rtapi_print_msg() calls in hal_init_mode() aren't showing up in the logs > > This is be

Re: [Emc-developers] requesting help debugging ubc

2014-01-19 Thread Sebastian Kuzminsky
On 01/19/2014 06:11 PM, Michael Haberler wrote: > > Am 19.01.2014 um 16:42 schrieb Sebastian Kuzminsky : > >> I'm trying to get ubc ready to merge, and i'm running into a problem >> that i'd like help debugging. >> >> I have a machine (Precise, x86, with memleak's RTAI kernel) that runs >> the ub

Re: [Emc-developers] requesting help debugging ubc

2014-01-19 Thread Michael Haberler
reading back on IRC: > [17:20:41] zultron: i'm trying to debug that hal_init problem > i'm seeing in packaged ubc, and i'm confused as to why the rtapi_print_msg() > calls in hal_init_mode() aren't showing up in the logs This is because it's a chicken-and-egg problem - RTAPI logging needs a

Re: [Emc-developers] requesting help debugging ubc

2014-01-19 Thread Michael Haberler
Hi Seb, Am 19.01.2014 um 16:42 schrieb Sebastian Kuzminsky : > I'm trying to get ubc ready to merge, and i'm running into a problem > that i'd like help debugging. > > I have a machine (Precise, x86, with memleak's RTAI kernel) that runs > the ubc branch fine when configured for rip, but throws

Re: [Emc-developers] nanomsg

2014-01-19 Thread Michael Haberler
Am 19.01.2014 um 09:27 schrieb John Morris : > On 01/16/2014 03:06 PM, Alexey Starikovskiy wrote: >> Hi, >> It appears that ZeroMQ has a derivative, which is positioned to be >> "done right"... >> here is their story http://nanomsg.org/documentation-zeromq.html There are actual several libraries

Re: [Emc-developers] nanomsg

2014-01-19 Thread Jon Elson
On 01/19/2014 02:27 AM, John Morris wrote: > On 01/16/2014 03:06 PM, Alexey Starikovskiy wrote: >> Hi, >> It appears that ZeroMQ has a derivative, which is positioned to be >> "done right"... >> here is their story http://nanomsg.org/documentation-zeromq.html > Oh, wow. Some scary bits about 0MQ i

[Emc-developers] requesting help debugging ubc

2014-01-19 Thread Sebastian Kuzminsky
I'm trying to get ubc ready to merge, and i'm running into a problem that i'd like help debugging. I have a machine (Precise, x86, with memleak's RTAI kernel) that runs the ubc branch fine when configured for rip, but throws errors when installed from the debian packages i've made (with the seb/ub

Re: [Emc-developers] Udoo Board Problems

2014-01-19 Thread John Morris
To those with UBC3 problem and wishing for more to go on: Enable full debugging by setting the environment variable "export DEBUG=5". (Works for all 2.x branches, maybe?) Send debugging messages to stderr with "export MSGD_OPTS=-s". These break 'runtests', but if run on a known problem runtest,

Re: [Emc-developers] nanomsg

2014-01-19 Thread John Morris
On 01/16/2014 03:06 PM, Alexey Starikovskiy wrote: > Hi, > It appears that ZeroMQ has a derivative, which is positioned to be > "done right"... > here is their story http://nanomsg.org/documentation-zeromq.html Oh, wow. Some scary bits about 0MQ in there, such as that described under 'Asynchronou