**** BEGIN LOGGING AT Thu Jan 24 02:59:56 2019
Jan 24 04:32:15 <mohan__> hi
Jan 24 04:33:32 <mohan__> beagle black board can i use web service ?
Jan 24 04:33:45 <mohan__> any one reply ?
Jan 24 04:43:37 <zmatt> mohan__: your question is really vague and doesn't sound hardware-dependent
Jan 24 04:43:53 <zmatt> it runs linux, hence you can do anything that runs on linux
Jan 24 04:51:39 <kremlin> mohan__: what is web service
Jan 24 12:04:39 <saghir> Hi.
Jan 24 12:04:56 <saghir> I am trying to connect my beaglebone with xfinity wifi router
Jan 24 12:05:16 <saghir> it is connected but when i google i am redirected to xfinity login
Jan 24 12:05:38 <saghir> i am submiting my comcast credentials but it redirects to error page saying link expired
Jan 24 12:05:55 <saghir> how can i register my device to my xfinity account
Jan 24 12:05:57 <saghir> ?
Jan 24 13:17:43 <fred__tv> xinput_calibrator suggested snippets to make calibration permanent still doesn't work !!! any suggestion ?
Jan 24 13:30:06 <fred__tv> beagle is stuck with penguin on display for about 20-230 seconds before login prompt , probably because of this : https://pastebin.com/fSVnrBvL ,  any idea ?
Jan 24 13:37:52 <zmatt> it is weird that udev is running these actions so late (as well as that they're failing), but I don't see any indication here about what might be the reason
Jan 24 13:38:24 <zmatt> does the /dev/pwm directory itself exist?
Jan 24 13:39:48 <zmatt> also, when inspecting logs it may be sure to use monotonic timestamps ( journalctl -o short-monotonic -b ), i.e. time since startup began
Jan 24 13:40:11 <zmatt> since the system clock could also jump during boot due to network time sync
Jan 24 13:40:53 <zmatt> uhh, sentence failed... "it may be a good idea" or "be sure" .. somehow ended up mashed  together
Jan 24 13:41:53 <zmatt> can you pastebin the entire bootup maybe? ( journalctl -o short-monotonic --no-pager -b )
Jan 24 13:42:49 <zmatt> if pastebinit is installed you can directly create a paste from the beaglebone:  journalctl -o short-monotonic -b | pastebinit
Jan 24 13:43:41 <fred__tv> https://pastebin.com/itcTcTxY
Jan 24 13:46:38 <zmatt> o.O
Jan 24 13:49:07 <zmatt> oh, huh, these timestamps can't be right
Jan 24 13:49:32 <fred__tv> Somethin more tricky than transformation matrix for touchscreen ? no ??
Jan 24 13:49:35 <fred__tv> :-((
Jan 24 13:50:04 <zmatt> oh right, nm I understand why the timestamps look like this, it's due to initramfs of course (which I don't use)
Jan 24 13:50:49 <zmatt> so it kinda looks like it's hanging in initramfs
Jan 24 13:51:28 <zmatt> since by the time systemd starts, it's already 37 seconds on the monotonic time
Jan 24 13:52:34 <fred__tv> do you mean jump from 2 to 37 ?
Jan 24 13:53:00 <zmatt> try removing your initramfs ( /boot/initrd.img-$VERSION ) file.  it's not needed to boot the beaglebone, and removing it should help getting more useful log output
Jan 24 13:53:43 <fred__tv> renaming ?
Jan 24 13:53:58 <zmatt> yeah... whatever is happening during the jump to 2 to 37 doesn't seem to be producing log output
Jan 24 13:54:01 <zmatt> yeah
Jan 24 13:54:06 <zmatt> remove or rename, doesn't matter
Jan 24 13:57:21 <zmatt> this is so weird... what the hell is it doing
Jan 24 13:57:29 <fred__tv> https://pastebin.com/FnEJFem9
Jan 24 14:00:11 <zmatt> hey, the errors are gone
Jan 24 14:00:22 <zmatt> it's still slow as fuck though
Jan 24 14:00:43 <zmatt> and I really can't tell why from this
Jan 24 14:00:55 <zmatt> time for a boot plot!  systemd-analyze plot >boot.svg
Jan 24 14:01:07 <zmatt> this creates an svg image of your boot activity
Jan 24 14:03:42 <fred__tv>  perhaps just missed ?
Jan 24 14:03:43 <fred__tv> [   21.824544] beaglebone kernel: random: crng init done
Jan 24 14:03:44 <fred__tv> [   21.824575] beaglebone kernel: random: 7 urandom warning(s) missed due to ratelimiting
Jan 24 14:03:44 <fred__tv> [   44.933812] beaglebone systemd[1]: Listening on Load/Save RF Kill Switch Status /dev/rfkill Watch.
Jan 24 14:03:55 <zmatt> that's not important
Jan 24 14:04:23 <fred__tv> what to do with boot.svg ?
Jan 24 14:04:48 <zmatt> hmm, do you have a way to share it?
Jan 24 14:05:51 <zmatt> btw you can probably fix the urandom warnings by adding "rng_core.default_quality=1" to the "cmdline" variable in /boot/uEnv.txt
Jan 24 14:07:40 <fred__tv> https://filebin.net/b4w2fsazl462hgpz
Jan 24 14:08:24 <fred__tv> cool !
Jan 24 14:12:04 <zmatt> so networking.service appears the main culprit... and maybe generic-board-startup.service, although that could be a side effect of the networking issues
Jan 24 14:12:59 <fred__tv> some dns issue ?
Jan 24 14:13:00 <zmatt> you probably don't want generic-board-startup.service anyway, unless you care about the usb networking, but I think you already broke that anyway
Jan 24 14:13:12 <zmatt> so  systemctl disable generic-board-startup.service
Jan 24 14:13:34 <fred__tv> I have apt remove connman
Jan 24 14:13:52 <fred__tv> then configured /etc/networking/interfaces
Jan 24 14:13:58 <zmatt> yes you already said that
Jan 24 14:14:11 <fred__tv> something broken in doing this ?
Jan 24 14:14:20 <zmatt> yup
Jan 24 14:14:50 <zmatt> ifupdown might be waiting for something, dunno what though
Jan 24 14:15:42 <zmatt> in my experience ifupdown just kinda sucks.  I didn't like connman either, but systemd-networkd has been working pretty much like a charm for me
Jan 24 14:15:56 <fred__tv> anyway.... to wait 20 seconds more at boot is not so tremendous...
Jan 24 14:16:06 <zmatt> oh, maybe also   systemctl disable bb-wl18xx-wlan0.service
Jan 24 14:16:19 <zmatt> since that's completely useless on your platform
Jan 24 14:17:15 <fred__tv> now, really , is touchscreen calibration making me crazy....
Jan 24 14:17:23 <zmatt> 20 seconds? you mean 40 seconds
Jan 24 14:17:40 <fred__tv> uh ?
Jan 24 14:17:48 <zmatt> that's what your boot diagram shows
Jan 24 14:17:55 <zmatt> networking.service is taking 40 seconds
Jan 24 14:18:05 <zmatt> 42.737s
Jan 24 14:18:20 <fred__tv> :-)
Jan 24 14:18:23 <fred__tv> seen
Jan 24 14:19:08 <zmatt> this is what boot can look like if you trim away all the bloat and unnecessary services: https://liktaanjeneus.nl/boot.svg ;)
Jan 24 14:19:24 <fred__tv> impressive
Jan 24 14:21:09 <zmatt> anyway, at least try disabling bb-wl18xx-wlan0.service and generic-board-startup.service
Jan 24 14:21:21 <zmatt> these also mess with networking, so maybe they end up confusing ifupdown somehow
Jan 24 14:23:05 <fred__tv> systemctl disable generic-board-startup.service  ?
Jan 24 14:23:12 <zmatt> ye
Jan 24 14:23:15 <zmatt> s
Jan 24 14:24:49 <fred__tv> both disabled....no improvements...
Jan 24 14:25:18 <zmatt> then I guess it must be purely a problem with ifupdown or its config
Jan 24 14:28:11 <fred__tv> probably...
Jan 24 14:30:09 <fred__tv> I should try to burn the original rcn 9.6 console with connman and see if it is faster....
Jan 24 14:30:43 <fred__tv> can it be a white beagle relate issue ?
Jan 24 14:30:55 <zmatt> seems unlikely
Jan 24 14:35:16 <fred__tv> got a new boot.svg , networking.service still the culprit
Jan 24 14:44:56 <fred__tv> definitely a networking issue
Jan 24 14:45:27 <fred__tv> I can't ping beagle in less than 40 sec
Jan 24 15:18:28 <fred__tv> Why any known setting has no impact on touchscreen behavior ?
Jan 24 15:22:56 * charlie5 waves to zmatt
Jan 24 15:23:29 <LetoThe2nd> fred__tv: have you tried good old ts_calibrate?
Jan 24 15:25:06 <zmatt> fred__tv: did you see my last comments yesterday on that topic?  if the libinput driver is used, you need to use Option "CalibrationMatrix"
Jan 24 15:25:25 <zmatt> (or put the calibration matrix in an udev rule, but that's more complicated probably)
Jan 24 15:25:47 <zmatt> the main thing that bothers me is why explicitly setting the driver to evdev in your xorg.conf.d snippet had no effect
Jan 24 15:27:39 <fred__tv> I just need to understand how touchscreen is recognized, first of all it works :-) but x and y values are inverted (sliding up, pointer goes down, slidin right pointer goes left)
Jan 24 15:27:44 <fred__tv> look:
Jan 24 15:28:44 <fred__tv> https://pastebin.com/7rNxvrvE
Jan 24 15:29:56 <fred__tv> so, should I modify  /usr/share/X11/xorg.conf.d/40-libinput.conf  at section :
Jan 24 15:30:16 <zmatt> no
Jan 24 15:30:18 <fred__tv> Section "InputClass"
Jan 24 15:30:18 <fred__tv>         Identifier "libinput touchscreen catchall"
Jan 24 15:30:18 <fred__tv>         MatchIsTouchscreen "on"
Jan 24 15:30:18 <fred__tv>         MatchDevicePath "/dev/input/event*"
Jan 24 15:30:18 <fred__tv>         Driver "libinput"
Jan 24 15:30:18 <fred__tv> EndSection
Jan 24 15:30:40 <zmatt> I already told you you shouldn't modify stuff in /usr (other than in /usr/local)
Jan 24 15:30:57 <zmatt> and I explained how xorg gathers its configuration from the various files
Jan 24 15:31:39 <zmatt> the path you were using in /etc for your config file is correct, and anything you put that takes precedence over the previous config blocks such as 40-libinput.conf
Jan 24 15:32:10 <zmatt> can you show me your Xorg.log again?
Jan 24 15:32:27 <fred__tv> but xorg tells: Using system config directory "/usr/share/X11/xorg.conf.d"
Jan 24 15:32:33 <fred__tv> yes wait
Jan 24 15:33:33 <zmatt> it should also say  Using config directory: "/etc/X11/xorg.conf.d"   on the line immediately above it
Jan 24 15:35:15 <fred__tv> no Using config file "/etc/X11/xorg.conf.d"
Jan 24 15:35:23 <fred__tv> sry
Jan 24 15:35:33 <fred__tv> Using config file "/etc/X11/xorg.conf"
Jan 24 15:35:48 <fred__tv> here's the log  https://pastebin.com/q2JXURzt
Jan 24 15:36:21 <zmatt> oh shit, does the presence of xorg.conf mean it completely ignores xorg.conf.d ?
Jan 24 15:37:15 <zmatt> try moving xorg.conf to xorg.conf.d/99-video.conf
Jan 24 15:37:20 <zmatt> or something like that
Jan 24 15:37:37 <zmatt> what was in it again? something to configure 16-bit color right?
Jan 24 15:39:10 <fred__tv> etc/X11/xorg.conf    https://pastebin.com/PuX3Kw4U
Jan 24 15:39:31 <zmatt> oh no it's really an xorg.conf
Jan 24 15:39:46 <zmatt> also, you have DefaultDepth commented out?
Jan 24 15:39:56 <zmatt> then just remove the whole file
Jan 24 15:40:08 <zmatt> I don't think it's doing anything useful
Jan 24 15:40:44 <fred__tv> yes I can
Jan 24 15:41:24 <fred__tv> why xinput_calibrator says:  copy the snippet below into '/etc/X11/xorg.conf.d/99-calibration.conf' (/usr/share/X11/xorg.conf.d/ in some distro's)
Jan 24 15:41:25 <zmatt> like I said, on modern linux systems you do not have an xorg.conf normally
Jan 24 15:41:42 <fred__tv> and it doesn't work ?
Jan 24 15:41:57 <zmatt> have you removed the xorg.conf already and tried again?
Jan 24 15:42:09 <zmatt> since it sounds like it might be a significant cause of problems
Jan 24 15:43:27 <fred__tv> removed but not yet addedd nothing about touchscreen snippet
Jan 24 15:43:28 <zmatt> [  1218.579] (**) ti-tsc: Applying InputClass "calibration"
Jan 24 15:43:43 <zmatt> it did find and apply your InputClass block apparently
Jan 24 15:44:03 <zmatt> which is odd
Jan 24 15:44:35 <zmatt> since it's implying it didn't look in /etc/X11/xorg.conf.d/
Jan 24 15:44:38 <fred__tv> ah yes sorry
Jan 24 15:45:16 <zmatt> ?
Jan 24 15:45:21 <fred__tv> I have 99-calibration.conf in /usr/share/X11/xorg.conf.d
Jan 24 15:45:34 <fred__tv> looks like suggested by xinput_calibrator :
Jan 24 15:45:49 <zmatt> please don't.  place it in /etc/X11/xorg.conf.d/
Jan 24 15:45:56 <zmatt> and remove /etc/X11/xorg.conf
Jan 24 15:46:08 <fred__tv> https://pastebin.com/FJgVM0Ed
Jan 24 15:46:40 <zmatt> and can you try disabling all of those Option blocks and instead add:  Driver "evdev"
Jan 24 15:46:55 <zmatt> those Option lines I mean
Jan 24 15:47:11 <fred__tv> ok I'll move to /etc/X11/xorg.conf.d (that directory doesn't exists yet)
Jan 24 15:47:20 <zmatt> ???
Jan 24 15:47:36 <zmatt> how.. what... why have you still not created it?
Jan 24 15:47:41 <zmatt> that's one of the first things I mentioned
Jan 24 15:47:59 <zmatt> I mean, that explains why xorg isn't mentioning the directory... it doesn't exist!
Jan 24 15:48:16 <zmatt> :/
Jan 24 15:48:24 <fred__tv> :-(
Jan 24 15:49:07 <zmatt> so maybe the xorg.conf isn't a problem after all... pretty sure it's still unnecessary though
Jan 24 15:50:07 <fred__tv> ok how that 99-calibration.conf should be ?
Jan 24 15:50:34 <ben_> anyone have any information on Beagle Board X15 power consumption?
Jan 24 15:50:55 <fred__tv> do I leave
Jan 24 15:50:56 <fred__tv>         Identifier      "calibration"
Jan 24 15:50:56 <fred__tv>         MatchProduct    "ti-tsc"
Jan 24 15:50:59 <zmatt> yes
Jan 24 15:51:43 <zmatt> (you can use any Identifier you want of course, it's just a name for this snippet of config)
Jan 24 15:51:50 <fred__tv> removed options, now ?
Jan 24 15:52:04 <zmatt> add   Driver "evdev"
Jan 24 15:52:27 <fred__tv> added
Jan 24 15:53:08 <zmatt> now let's see what Xorg.log says, hopefully it will use the evdev driver rather than the libinput driver
Jan 24 15:53:35 <zmatt> (nothing wrong with the libinput driver, but it's just annoying that xinput_calibrator hasn't been updated yet to be able to produce config for it)
Jan 24 15:56:25 <fred__tv> doesn't find driver:
Jan 24 15:56:38 <zmatt> ohh
Jan 24 15:56:55 <zmatt> dpkg --get-selections | grep xorg-input
Jan 24 15:57:38 <fred__tv> https://ghostbin.com/paste/yg3np
Jan 24 15:58:11 <fred__tv> doesn't find module sry
Jan 24 15:59:07 <zmatt> uhh what? presumably it at least finds the xserver-xorg-input-libinput package?
Jan 24 15:59:16 <fred__tv> xserver-xorg-input-all                          install
Jan 24 15:59:16 <fred__tv> xserver-xorg-input-libinput                     install
Jan 24 15:59:16 <fred__tv> xserver-xorg-input-wacom                        install
Jan 24 15:59:50 <zmatt> lol... "xserver-xorg-input-all" only depends on those two
Jan 24 15:59:54 <zmatt> interesting definition of "all"
Jan 24 16:00:15 <fred__tv> shoud I install xserver-xorg-input-evdev ?
Jan 24 16:00:21 <zmatt> I guess they really want to push libinput
Jan 24 16:00:47 <zmatt> well alternatively we could just try to make the appropriate config for the libinput driver
Jan 24 16:01:38 <fred__tv> in 40-libinput.conf paste before ?
Jan 24 16:01:48 <zmatt> so, just to confirm, what does this show:  xinput list-props "ti-tsc" | grep 'Coordinate Transformation'
Jan 24 16:02:17 <fred__tv> not found "xinput"
Jan 24 16:02:35 <zmatt> oof
Jan 24 16:02:38 <zmatt> apt-get install xinput
Jan 24 16:04:21 <zmatt> the main problem was that it was rotated 180 degrees right?
Jan 24 16:05:28 <fred__tv> no rotation OK
Jan 24 16:05:54 <zmatt> "x and y values are inverted (sliding up, pointer goes down, slidin right pointer goes left)"
Jan 24 16:05:58 <fred__tv> (sliding up, pointer goes down, slidin right pointer goes left)
Jan 24 16:06:12 <zmatt> yes, that's exactly the same as being rotated 180 degrees
Jan 24 16:06:14 <fred__tv> Images are ok
Jan 24 16:06:31 <zmatt> yes I understand, I was talking about the touchscreen data
Jan 24 16:07:04 <fred__tv> xinput list-props "ti-tsc" | grep 'Coordinate Transformation'    no returned data
Jan 24 16:07:15 <zmatt> huh
Jan 24 16:07:36 <zmatt> can you pastebin the full output of  xinput list-props "ti-tsc"  ?
Jan 24 16:09:01 <fred__tv> hmmm I'm working via ssh on beagle , how to catch on real console ?
Jan 24 16:09:11 <zmatt> you can use ssh
Jan 24 16:09:21 <zmatt> first do:  export DISPLAY=:0
Jan 24 16:09:28 <zmatt> then xinput should work
Jan 24 16:09:54 <zmatt> (assuming you're logged in as the same user as x11 is)
Jan 24 16:11:25 <fred__tv> https://ghostbin.com/paste/b72ty
Jan 24 16:11:44 <zmatt> okay so it does work
Jan 24 16:12:02 <zmatt> you probably made a typo in the grep earlier?
Jan 24 16:12:19 <zmatt> anyway, I found some instructions, lemme see what they're saying
Jan 24 16:12:19 <fred__tv> very likely...
Jan 24 16:12:46 <zmatt> run xinput_calibrator in verbose mode: xinput_calibrator -v
Jan 24 16:13:43 <zmatt> what we're interested in is the "Adding click N (X=..., Y=...)" messages
Jan 24 16:14:29 <fred__tv> https://ghostbin.com/paste/ev9j6
Jan 24 16:16:13 <zmatt> okay
Jan 24 16:17:49 <zmatt> what was your display resolution?
Jan 24 16:17:57 <fred__tv> 640x480
Jan 24 16:20:16 <zmatt> okay, hold on
Jan 24 16:20:24 <zmatt> math in progress ;)
Jan 24 16:21:15 <fred__tv> Already said, Coordinate Transformation is diabolic....
Jan 24 16:21:39 <fred__tv> With older versions Evdev was immediate....
Jan 24 16:28:32 <zmatt> well, again the issue isn't that libinput's configuration is problematic, the issue is that xinput_calibrator hasn't been updated
Jan 24 16:29:01 <zmatt> I'm trying to understand what the hell these weirdass formulas do I found on a wiki page
Jan 24 16:29:13 <zmatt> they seem so odd
Jan 24 16:29:21 <fred__tv> https://wiki.ubuntu.com/X/InputCoordinateTransformation
Jan 24 16:31:22 <zmatt> yes that all makes perfect sense
Jan 24 16:31:43 <zmatt> but https://wiki.archlinux.org/index.php/Talk:Calibrating_Touchscreen#Libinput_breaks_xinput_calibrator  does not
Jan 24 16:42:07 <zmatt> well, we could just try them
Jan 24 16:46:38 <zmatt> xinput set-prop "ti-tsc" 'Coordinate Transformation Matrix' -1.054945, 0, 1.0282967, 0, -1.081081, 1.02364865, 0, 0, 1
Jan 24 16:47:55 <fred__tv> I've done like suggested in https://wiki.archlinux.org/index.php/Talk:Calibrating_Touchscreen#Comments_on_making_transformation_matrix_permanent
Jan 24 16:48:32 <fred__tv> ok
Jan 24 16:48:38 <fred__tv> pasted
Jan 24 16:49:01 <zmatt> ohhhhhhh I finally get it! I just realized that xinput_calibrator is having you mark the points at 1/8 - 7/8 of the screen
Jan 24 16:49:05 <fred__tv> what do I do to make X to start with new values ?
Jan 24 16:49:09 <zmatt> that's the bit I was completely missing
Jan 24 16:49:22 <zmatt> d'oh I feel like an idiot now
Jan 24 16:49:46 <fred__tv> think about me..... :-(((
Jan 24 16:50:09 <zmatt> Option "CalibrationMatrix" "-1.054945 0 1.0282967 0 -1.081081 1.02364865 0 0 1"
Jan 24 16:50:54 <zmatt> (and remove the Driver "evdev" of course, if that's still there)
Jan 24 16:51:11 <fred__tv> CalibrationMatrix or TansformationMatrix  ???
Jan 24 16:51:17 <fred__tv> removed
Jan 24 16:51:49 <zmatt> CalibrationMatrix
Jan 24 16:51:53 <zmatt> http://manpages.ubuntu.com/manpages/bionic/man4/libinput.4.html
Jan 24 16:52:41 <fred__tv> so https://wiki.archlinux.org/index.php/Talk:Calibrating_Touchscreen#Comments_on_making_transformation_matrix_permanent  is wrong
Jan 24 16:53:15 <fred__tv> BINGO !!!!
Jan 24 16:53:39 <zmatt> maybe either of them works? I dunno
Jan 24 16:54:12 <fred__tv> TransformationMatrix doesn't work
Jan 24 16:54:18 <zmatt> lol
Jan 24 16:54:24 <zmatt> maybe reply on that wiki talk page ;)
Jan 24 16:58:45 <fred__tv> so how do I convert xinput_calibrator points to matrix ?
Jan 24 17:01:57 <zmatt> lemme see what the easy way is
Jan 24 17:34:16 <fred__tv> not that easy...
Jan 24 17:34:24 <zmatt> https://pastebin.com/N3b5xy9j
Jan 24 17:34:33 <zmatt> run this with  gp --quiet calibration.gp
Jan 24 17:35:09 <zmatt> (obtained using: apt-get install pari-gp  ... you can of course do this on your linux host system rather than the beaglebone)
Jan 24 17:37:32 <fred__tv> installing....
Jan 24 17:37:34 <zmatt> this does linear least squares fitting to the points given
Jan 24 17:41:49 <fred__tv> it gaves me back this: https://ghostbin.com/paste/wxxns
Jan 24 17:44:18 <zmatt> oh whoops, I somehow lost the * 1.0 somewhere to force it to use floating-point instead of fractions
Jan 24 17:45:22 <zmatt> updated
Jan 24 17:45:31 <zmatt> I've also made it slightly cleaner
Jan 24 17:47:52 <zmatt> you can now easily see how to add more calibration points (to perhaps improve accuracy): you can simply add more columns to xi, yi, xo, yo, and w
Jan 24 17:53:01 <fred__tv> How many digit after comma are significative ?
Jan 24 17:53:17 <zmatt> no clue :D
Jan 24 17:53:26 <zmatt> perform two calibrations and check how much they differ
Jan 24 17:53:42 <fred__tv> it gives me 37 decimals...
Jan 24 17:53:50 <zmatt> oh lol, it doesn't here
Jan 24 17:53:56 <zmatt> maybe I have some saner defaults configured
Jan 24 17:55:04 <fred__tv> I'll try some calibrations with accuracy tomorrow
Jan 24 17:55:36 <fred__tv> what timezone are you in ?
Jan 24 17:56:08 <zmatt> there, I've added a line to reduce number of digits in output
Jan 24 17:56:16 <zmatt> CET
Jan 24 17:56:18 <fred__tv> check
Jan 24 17:56:49 <fred__tv> ok so no timezone issue,
Jan 24 17:57:01 <zmatt> gp is pretty cool if you need precision, if you want you can just do:
Jan 24 17:57:07 <zmatt> \p 100
Jan 24 17:57:14 <zmatt> and it'll say:    realprecision = 115 significant digits (100 digits displayed)
Jan 24 17:57:22 <zmatt> :D
Jan 24 17:59:15 <zmatt> (by merely setting "format" rather than the realprecision I just change the output precision but it'll still use its default internal precision of 38 significant digits)
Jan 24 17:59:20 <fred__tv> 8-)
Jan 24 17:59:38 <zmatt> it'll just as happily do 1000 digits of precision if you need it
Jan 24 17:59:51 <fred__tv> tried with "g.7" for 6 digit after comma , excellent
Jan 24 18:00:01 <fred__tv> A very GREAT thank you
Jan 24 18:00:28 <zmatt> you're welcome!
Jan 24 18:01:14 <fred__tv> Tomorrow : UART   (but it should be a little bit simpler...)
Jan 24 18:01:31 <fred__tv> already enabled cape
Jan 24 18:02:13 <zmatt> actually it looked to me like the universal overlay is enabled for you
Jan 24 18:02:38 <zmatt> which means you can also experiment with uarts by using config-pin rather than using an overlay for it
Jan 24 18:02:54 <zmatt> but either should be fine
Jan 24 18:03:04 <fred__tv> ah, i've noted that uboot_overlay_addr0=/lib/firmware/my.dtbo  actually doesn't care if a cape is detected with its own eeprom
Jan 24 18:03:28 <zmatt> yeah it overrides the bootloader, mostly
Jan 24 18:03:32 <fred__tv> (I tried to move beagle to another LCD, without erased eeprom,
Jan 24 18:03:42 <fred__tv> and it works anyway(
Jan 24 18:03:52 <zmatt> it's still a good idea to not have false information in the eeprom though
Jan 24 18:03:57 <fred__tv> right
Jan 24 18:04:29 <fred__tv> dinner time !! (I'm still at work)
Jan 24 18:04:42 <fred__tv> See you !  Thanks !!!!
Jan 24 18:04:54 <zmatt> later!
Jan 24 18:05:16 <fred__tv> perhaps after-dinner  !
Jan 24 18:05:45 <fred__tv> are you a gnuplot guru ?
Jan 24 21:44:50 <tbet> Hi guys - has anyone some experince withe the pocket beagle - after having no issues with the beagle bone - I'm having now a tough time with the beagle pocket - the sd card was done with the latest debian image by etcher - board starts up - all three user leds toggling - but the board doesn't show up as a usb drive under several windows 10 laptops that I tested overhere - is there a known issue ?
Jan 24 21:45:24 <jkridner[m]> You are using the 2018-10-07 image?
Jan 24 21:45:38 <jkridner[m]> Does the "heartbeat" go on USR0?
Jan 24 21:46:11 <jkridner[m]> There was an issue before 2018-10-07.
Jan 24 21:46:20 <jkridner[m]> (MSFT breakage)
Jan 24 21:46:51 <tbet> yes
Jan 24 21:47:27 <tbet> my sdcard is 16GB - couldn't be an issue, right ?
Jan 24 21:48:41 <tbet> May it be worthwhile to checkout another iot - image ?
Jan 24 21:50:59 <tbet> after booting fully (about 10s) user2 led is in a state where it's somehow dimmed due to ongoing cycling - is that normal ?
Jan 24 21:53:01 <zmatt> sure, if it's just showing sporadic activity
Jan 24 21:53:13 <zmatt> it doesn't actually dim of course, it just looks dim because it's flashing briefly
Jan 24 21:53:33 <zmatt> and sdcard size is not a problem
Jan 24 21:54:13 <zmatt> as long as it's big enough to hold the image, it's fine. (and if it isn't big enough then etcher will complain obviously)
Jan 24 21:54:39 <tbet> etcher was fine - validation done without a problem ...
Jan 24 21:55:24 <tbet> is there maybe an issue specific to windows 10 - as a usb host - sorry - no other machines currently available here ;)
Jan 24 21:55:28 <zmatt> you're saying it doesn't show up as usb drive... does it show up as any other sort of device?  I mean, the usb drive part isn't actually important, it's the networking interface that matters
Jan 24 21:58:56 <tbet> no actually there is no kind of recognition as a usb device ...
Jan 24 21:59:03 <zmatt> that's odd
Jan 24 21:59:53 <zmatt> unfortunately with the pocketbeagle I'm not sure how one would approach debugging this
Jan 24 22:00:16 <tbet> ... like the usb power connection is OK - but the datalines are not correctly connected ...
Jan 24 22:00:31 <zmatt> have you tried a different usb cable?
Jan 24 22:00:53 <tbet> ... will do so ...
Jan 24 22:03:19 <zmatt> and you said you have a beaglebone too? can you try booting the beaglebone from that same sd card and see if it shows up as a device in Windows?
Jan 24 22:04:55 <zmatt> those are basically the ideas I have... I can't do a test for you since I have neither a pocketbeagle nor Windows
Jan 24 22:05:58 <zmatt> and I'm not sure how you'd debug issues with the pocketbeagle if you can't reach it via usb... does it still have a debug serial port? I'm not that familiar with the pocket tbh
Jan 24 22:16:50 <tbet> sorry - took a few "seconds" to find another micro usb cable - BUT : sorry for bothering you guys - it was realy the damn cable - even without SD card the device was recognized and the usb drivers installed - after booting the drive came up and now I can go on ...
Jan 24 22:17:02 <zmatt> lol
Jan 24 22:17:13 <tbet> thx :)
Jan 24 22:17:14 <zmatt> maybe a charge-only cable? ;)
Jan 24 22:17:32 <zmatt> you know, because adding two extra wires was too expensive
Jan 24 22:17:34 <tbet> at least now it's one ;)
Jan 24 22:17:56 <tbet> btw. - yes there is still uart0 as a serial debug port - https://github.com/beagleboard/pocketbeagle/wiki/System-Reference-Manual#56_Serial_Debug_Port
Jan 24 22:19:06 <tbet> well - maybe it's that - or it was just me : bending it one time too much ;)
**** ENDING LOGGING AT Fri Jan 25 02:59:58 2019