Back
[07:28:20] <skunkworks_> zlog
[07:37:46] <jepler> linuxcnc-build: force build --branch=master 1403.rip-wheezy-amd64
[07:37:47] <linuxcnc-build> build forced [ETA 24m20s]
[07:37:47] <linuxcnc-build> I'll give a shout when the build finishes
[07:41:54] <almost_bold> Hello. I am trying to change a small thing in "emc/motion/control.c", and I would like to make a print to the command line for debug.
[07:42:32] <almost_bold> For using printf I need to include the stdio.h
[07:42:37] <almost_bold> as far I know
[07:42:57] <almost_bold> BUT using #include<stdio.h> results in a error:
[07:43:53] <almost_bold> stdio.h: No such file or directory
[07:44:16] <almost_bold> And "obviously" it is installed as I can compile linuxcnc
[07:46:02] <almost_bold> Forcing #include "/usr/include/stdio.h", results in another error: "features.h: No such file or directory", that is a header needed by stdio.h
[07:47:15] <almost_bold> so I am missing some basic setup, that I would be grateful in having a reply.
[07:50:50] <archivist> almost_bold should you be using
http://linuxcnc.org/docs/ja/html/man/man3/rtapi_print.3rtapi.html
[07:55:24] <almost_bold> Thank you, will try. Is the output sent for the command line? I am almost sure I already tested based on others rtapi_prints in the same file, but without any real print.
[07:57:06] <archivist> to stdout I think
[07:57:14] <linuxcnc-build> Hey! build 1403.rip-wheezy-amd64 #2643 is complete: Success [3build successful]
[07:57:15] <linuxcnc-build> Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1403.rip-wheezy-amd64/builds/2643
[08:06:17] <jepler> rtapi_print_msg goes to dmesg for kernel-based realtime and terminal for uspace-based realtime
[08:06:46] <jepler> you are probably using kernel-based realtime if <stdio.h> is not available
[08:07:42] <almost_bold> got it in dmesg
[08:07:47] <jepler> I recommend using uspace-based realtime (sim if necessary) while developing anything but rtai-only device drivers, it's much more civilized. you can attach a debugger to the rtapi_app process and do source-level debugging, for instance; and you don't have to reboot after a crash, just exit linuxcnc/halrun/etc and run again.
[14:10:00] <kirk_wallace> I'm perplexed with G76 threading. In looking at the current paths which are on top here:
http://www.wallacecompany.com/tmp/g76_safe.png
[14:10:34] <kirk_wallace> It seem it would be better to have the Z safe motion as in the lower image.
[14:15:40] <cradek> I agree! If you fix it, be sure not to break tapered entry
[14:17:01] <cradek> hmm, if some of the pullout rapids are shorter than others, the pullouts might not match anymore (might be at different speeds)
[14:17:35] <cradek> or maybe it'd be ok since they're getting longer as you cut deeper
[14:17:39] * cradek squints
[14:18:30] <kirk_wallace> Hmm. I was thinking of getting rid of the tapers.
[14:19:19] <cradek> that's a harder sell
[14:20:40] <kirk_wallace> There would still be the compound angle, but the entry/exit tapers were incompatible with my tapered thread feature.
[14:52:33] <kirk_wallace> Does anyone use the entry/exit tapers? I noticed a FANUC G76 example that mentioned the compound angle and a 45 degree exit taper but didn't give a reason for the exit taper, nor if the exit was synchronized or not.
[15:07:41] <jepler> "does anyone use" -> hard question
[15:08:00] <jepler> I guess we should add telemetry to linuxcnc so that we could know the answer to that, among all internet-connected linuxcnc installations that have opted in
[15:44:18] <kirk_wallace> Another issue with tapered threads might be that the synchronized motion follows the taper and not the thread axis. I suppose I could calculate the difference and tweak the pitch value.
[15:53:35] <cradek> it does, so yes you'd have to do that
[17:32:55] <seb_kuzminsky> jepler: that would be interesting data
[21:41:41] <KGB-linuxcnc> 03Jeff Epler 05master 1b4d4ed 06linuxcnc 10docs/src/Submakefile build: this pipe would hide real build errors * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=1b4d4ed
