Back
[00:10:39] <skunksleep> Hey rob. I lightly played with your improvement branch and couldn't see much difference between that and the spiral arc branch
[00:11:10] <skunksleep> I will play more tomorrow
[00:14:08] <rob_ellenberg> It only really makes a difference on arcs with large radii
[00:14:20] <rob_ellenberg> and even then only a few percent
[00:14:59] <skunksleep> I will see if I have gcode that would help test that
[00:15:54] <skunksleep> You seem to be digging in again :)
[00:16:32] <rob_ellenberg> Yeah, I've got some more hours to put towards this stuff
[00:17:48] <skunksleep> All I can say is 'YAY!'
[00:22:12] <rob_ellenberg> haha, thanks! It's a great feeling to squeeze another few seconds out of the TP
[11:02:21] <skunkworks> Yay!
[11:02:38] <seb_kuzminsky> hi skunkworks :-)
[11:03:06] <skunkworks> :)
[11:03:31] <skunkworks> I see 2.7 pre - 3? in the near future.
[11:03:37] <skunkworks> 4?
[11:06:41] <skunkworks> seb_kuzminsky, what was the command to show the processor/cache layout
[11:17:59] <KGB-linuxcnc> 03Robert W. Ellenberg 05feature/spiral-arc-handling-2.7 f545aa4 06linuxcnc 10src/emc/rs274ngc/interp_arc.cc interp: reverted to stock 2.7 tolerance checks on spiral error * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=f545aa4
[11:21:37] <KGB-linuxcnc> 03Robert W. Ellenberg 05feature/spiral-arc-handling-2.7 88a47e8 06linuxcnc 10src/emc/tp/blendmath.c tp: better minimum spiral fit coefficient to reduce acceleration ripple on near-circular cases * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=88a47e8
[11:24:17] <seb_kuzminsky> sweet!
[11:24:40] <seb_kuzminsky> skunkworks: lstopo
[11:24:59] <seb_kuzminsky> from the hwloc debian package, available in wheezy
[11:26:11] <skunkworks> Heh - I just found it. Thanks!
[11:26:13] <seb_kuzminsky> and yeah, i'll need to look at these new commits but i bet rob did the right thing, then we'll be ready for ~pre3 in the next few days sometime
[11:27:20] <seb_kuzminsky> francis tisserant has been working on the french translations under po4a, rob's doing tp work and wants to help with the standalone-motion tests
[11:27:26] <seb_kuzminsky> lots of activity, it's pretty exciting
[11:37:50] <linuxcnc-build> build #1117 of 1403.rip-wheezy-amd64 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1403.rip-wheezy-amd64/builds/1117 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[11:38:26] <skunkworks> uh oh
[11:38:34] <linuxcnc-build> build #1117 of 1400.rip-wheezy-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1400.rip-wheezy-i386/builds/1117 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[11:38:53] <linuxcnc-build> build #1308 of 1404.rip-wheezy-rtpreempt-amd64 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1404.rip-wheezy-rtpreempt-amd64/builds/1308 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[11:39:03] <seb_kuzminsky> probably the arc radius tolerance tests need tweaking for the changed acceptance tests
[11:40:38] <linuxcnc-build> build #628 of 1402.rip-wheezy-rtpreempt-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1402.rip-wheezy-rtpreempt-i386/builds/628 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[11:41:51] <linuxcnc-build> build #776 of 1401.rip-wheezy-rtai-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1401.rip-wheezy-rtai-i386/builds/776 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[11:45:06] <seb_kuzminsky> yeah
[11:45:26] <seb_kuzminsky> those failures are expected, i'll fix it tonight if rob doesn't beat me to it
[11:46:48] <linuxcnc-build> build #1147 of 1405.rip-wheezy-armhf is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1405.rip-wheezy-armhf/builds/1147 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[11:49:02] <skunkworks> darn - can't disable hyperthreading in the bios
[11:49:51] <linuxcnc-build> build #2961 of 1306.rip-precise-amd64 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1306.rip-precise-amd64/builds/2961 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[11:49:59] <linuxcnc-build> build #2959 of 1300.rip-precise-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1300.rip-precise-i386/builds/2959 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[11:50:24] <linuxcnc-build> build #2960 of 1200.rip-lucid-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1200.rip-lucid-i386/builds/2960 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[11:51:02] <seb_kuzminsky> skunkworks: you should be able to get the same performance improvement by isolcpu'sing both the vcpus that share the last l2 (as identified by lstopo)
[11:51:21] <skunkworks> ok - I did and it didn't help.
[11:51:40] <linuxcnc-build> build #2960 of 1202.rip-lucid-amd64 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1202.rip-lucid-amd64/builds/2960 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[11:51:51] <skunkworks> Odd situation. (laptop - so really not a big deal) but rtai seems to have decent latency - rt doesn't
[11:51:57] <linuxcnc-build> build #2161 of 1301.rip-precise-rtai-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1301.rip-precise-rtai-i386/builds/2161 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[11:51:59] <seb_kuzminsky> huh, strange
[11:52:33] <seb_kuzminsky> no change in rtai latency when you isolcpu the last vcpu vs the last two vcpus?
[11:55:42] <linuxcnc-build> build #2960 of 1201.rip-lucid-rtai-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1201.rip-lucid-rtai-i386/builds/2960 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[11:55:42] <linuxcnc-build> build #2970 of 0000.checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/2970 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[11:56:19] <CaptHindsight> seb_kuzminsky: isolcpus was broken for quite a while it RTAI
[11:56:46] <CaptHindsight> it/in
[12:13:31] <seb_kuzminsky> hmm
[12:13:44] <seb_kuzminsky> but it's working in our 3.4-9 kernel, right?
[12:15:34] <linuxcnc-build> build #1118 of 1403.rip-wheezy-amd64 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1403.rip-wheezy-amd64/builds/1118 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[12:16:12] <linuxcnc-build> build #1118 of 1400.rip-wheezy-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1400.rip-wheezy-i386/builds/1118 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[12:16:22] <linuxcnc-build> build #1309 of 1404.rip-wheezy-rtpreempt-amd64 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1404.rip-wheezy-rtpreempt-amd64/builds/1309 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[12:17:49] <linuxcnc-build> build #629 of 1402.rip-wheezy-rtpreempt-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1402.rip-wheezy-rtpreempt-i386/builds/629 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[12:19:11] <linuxcnc-build> build #777 of 1401.rip-wheezy-rtai-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1401.rip-wheezy-rtai-i386/builds/777 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[12:24:29] <linuxcnc-build> build #1148 of 1405.rip-wheezy-armhf is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1405.rip-wheezy-armhf/builds/1148 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[12:26:19] <linuxcnc-build> build #2962 of 1306.rip-precise-amd64 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1306.rip-precise-amd64/builds/2962 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[12:26:22] <linuxcnc-build> build #2960 of 1300.rip-precise-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1300.rip-precise-i386/builds/2960 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[12:26:46] <linuxcnc-build> build #2961 of 1200.rip-lucid-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1200.rip-lucid-i386/builds/2961 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[12:27:47] <linuxcnc-build> build #2961 of 1202.rip-lucid-amd64 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1202.rip-lucid-amd64/builds/2961 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[12:28:25] <linuxcnc-build> build #2162 of 1301.rip-precise-rtai-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1301.rip-precise-rtai-i386/builds/2162 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[12:31:57] <linuxcnc-build> build #2961 of 1201.rip-lucid-rtai-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1201.rip-lucid-rtai-i386/builds/2961 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[12:31:58] <linuxcnc-build> build #2971 of 0000.checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/2971 blamelist: Robert W. Ellenberg <rwe24g@gmail.com>
[12:35:33] <skunkworks> CaptHindsight, this was -rt
[12:46:20] <zeeshan|2> http://pastebin.com/0kmgJ2tb
[12:46:24] <zeeshan|2> loadusr -W vfd.py
[12:46:37] <zeeshan|2> keeps waiting forever to let the component load even though have 3 ready calls made in there?
[12:49:55] <skunkworks> I don't really have a clue - but what happens if you put the ready calls at the end of all pin assignments?
[12:50:13] <zeeshan|2> let me try that
[12:54:21] <skunkworks> (although I didn't know that you could create more than one hal componant in a given program..)
[13:06:15] <zeeshan-mill> nope
[13:06:19] <zeeshan-mill> still tries to wait forever
[13:06:44] <zeeshan-mill> instead of making 3 components in there
[13:06:49] <zeeshan-mill> can i just do this:
[13:08:12] <zeeshan-mill> vfd = hal.component("vfd") ;; vfd.newpin("spdvfd.drivestatus", hal.HAL_U32, hal.HAL_OUT)
[13:08:13] <zeeshan-mill> ?
[13:08:20] <zeeshan-mill> and so on
[13:08:43] <seb_kuzminsky> zeeshan-mill: one component per process, i'm pretty sure
[13:08:50] <seb_kuzminsky> but yes, you can name your pins what you like
[13:09:02] <zeeshan-mill> ah
[13:09:59] <skunkworks> I would think this would be a situationi that you call the same componant 3 times.. (how ever you do that :) )
[13:10:26] <skunkworks> * load 3 instances.
[13:10:44] <skunkworks> but maybe you can't do that with python componants.
[13:12:13] <seb_kuzminsky> you can, as long as they take different names each time
[13:12:23] <seb_kuzminsky> not sure how the serial port sharing would work in that case
[13:12:48] <seb_kuzminsky> as long as the writes to the serial port are "packet at a time" it should work i'd think
[13:14:29] <zeeshan-mill> it was giving me issues
[13:14:46] <zeeshan-mill> when i tried to make 3 components cause i think there was no scheduling for the packets
[13:19:31] <zeeshan-mill> vfd.newpin("spdvfd.run", hal.HAL_BIT, hal.HAL_IN)
[13:19:33] <zeeshan-mill> it doe snot like that
[13:19:41] <zeeshan-mill> it guess doesn't like the dot
[13:19:58] <zeeshan-mill> AttributeError: Pin 'spdvfd' does not exist
[13:21:14] <seb_kuzminsky> "." is allowed in pin names, something else must be wrong
[13:22:53] <skunkworks> http://linuxcnc.org/docs/2.6/html/hal/halmodule.html
[13:23:03] <skunkworks> h.newpin("in", hal.HAL_FLOAT, hal.HAL_IN)
[13:23:33] <zeeshan-mill> channging it to . to xx as the spacer
[13:23:34] <zeeshan-mill> worked lol
[13:24:13] <skunkworks> never mind. thought your order was wrong
[13:24:21] <zeeshan-mill> and now it works.
[13:24:27] <zeeshan-mill> you guys were right. one component per program
[13:25:41] <zeeshan-mill> not sure why the dot issue :)
[13:25:43] <zeeshan-mill> oh well
[13:33:03] <skunkworks> those argon drives don't log analog friendly.. atleast the directions suck as to how to set it up
[13:33:53] <skunkworks> *look
[13:38:28] <pcw_home> at least they have bipolar differential inputs
[13:38:58] <pcw_home> better than some DMM with 0..5V
[13:38:58] <skunkworks> and I don't see an enable..
[13:39:15] <skunkworks> but like I say - the documentation sucks
[13:40:03] <pcw_home> I'm pretty sure you need to program that function on the GPIs
[13:40:51] <pcw_home> (lots of drives/VFDs are like that though default usually includes an enable)
[13:43:24] <zeeshan-mill> what pin is used to set the speed when you call like a m3 s1000?
[13:43:35] <zeeshan-mill> i need to scale it before i pass it to vfd
[13:43:42] <zeeshan-mill> whoops wrong chan :D
[13:47:25] <seb_kuzminsky> zeeshan-mill: here's a python component i wrote that uses "." in the pin names, it works fine:
http://git.linuxcnc.org/gitweb?p=linuxcnc.git;a=blob;f=src/hal/user_comps/scorbot-er-3.py;h=dfb48fb18c9e1701fcd406f576e9524f7018a64a;hb=refs/heads/scorbot-er-3#l62
[13:49:44] <zeeshan-mill> duh
[13:49:48] <zeeshan-mill> i was missing a %d
[13:49:49] <zeeshan-mill> lol
[13:50:06] <zeeshan-mill> thank you
[14:17:31] <zeeshan-mill> wanted to update, it wasn't cause of the %d
[14:17:36] <zeeshan-mill> was a silly mistake. wasn't using quotes
[15:15:11] <skunkworks> pcw_home, the drives don't look like they have pass through encoder signals. can you hook them up to 2 things at once?
[15:15:48] <PCW> Thats akward most drives have a simulated encoder output
[15:16:32] <PCW> Y0u can connect to 2 things at once (or use something like our ENCY)
[15:16:38] <skunkworks> unless I am missing it
[15:16:45] <skunkworks> or you can program it..
[15:17:13] <PCW> I has no high speed outputs that I see
[17:11:14] <seb_kuzminsky> falcon 9 launch in 10 minutes:
http://www.spacex.com/webcast/
[18:33:05] <skunkworks> zlog