**** BEGIN LOGGING AT Mon May 09 02:59:59 2016 May 09 08:54:25 Hi panto ! May 09 08:54:47 hi May 09 08:54:50 do I know you? May 09 08:55:23 I dont think so. I am a GSoC student and will be working on BeagleScope project this year. May 09 08:55:35 oh May 09 08:55:36 nice May 09 08:56:03 And I was working on the latest version of remoteproc, And got into some errors. So I was suggested to ask you for help. May 09 08:57:03 it's been quite a long way back since I've touched pru May 09 08:57:13 and the driver has been rewritten May 09 08:57:38 So, I was trying to get the pruss and pru_rproc working on kernel 4.1.18-ti-r55. They were working when getting automatically loaded on boot, but were causing errors when unloaded and loaded back again. May 09 08:57:53 http://pastebin.com/EyNC5BAN May 09 08:58:01 Oh .. ahh . okay. May 09 08:58:28 but would seeing the oops message, would you like to suggest something ? May 09 08:58:34 maybe a work around ? May 09 08:59:44 iirc loading/unloading the PRU didn't work on the new driver May 09 08:59:51 you should take it up with the author May 09 09:02:08 Oh, okay. The unloading and loading of pruss_remotproc driver, form the PRU-framework project (GSoC 2015) worked. May 09 09:02:28 But since that is not the latest version, I cant go on with it. May 09 09:03:59 I know it worked cause I spend a lot of time getting it to work May 09 09:04:15 Oh ! :) Cool ! May 09 09:04:26 take it with the new driver's author May 09 09:04:39 Okay, So I should take it with the author:Summan Anna. May 09 10:13:55 Is there any documentation on the beagleboard x15 similar to the system reference manual for Black? May 09 10:17:05 hmmm, I'd expect so, but given the board isn't on sale it might not be available yet. Have you checked the elinux page on the X15? May 09 10:18:57 Yes, there is a schematic available and a link to the site on TI. The table looks like the one in the Black manual but I guess it will be available as soon as it is realeased then. May 09 10:19:48 There is some information on Github, but only start up and schematics and so on May 09 10:19:58 hmm, you could try checking if there is documentation on the AM55xx EVM May 09 10:20:11 as that's a X15, essentially May 09 10:21:22 http://www.ti.com/tool/TMDSEVM572X May 09 10:22:23 http://processors.wiki.ti.com/index.php/AM572x_General_Purpose_EVM_HW_User_Guide - linked from the bottom of previous page May 09 10:22:48 Thanks very much! May 09 10:23:16 I'll check in on that. And then just wait unitl X15 crawls out of its craddle! May 09 10:23:58 yeah, let's hope that happens, eventually May 09 10:24:13 The one who waits.... May 09 10:25:26 well, I have one, but it would be nice if you and everyone else could get one too May 09 11:43:57 I have problem with serial connection to beaglebone black May 09 11:44:26 I am using Olimex 3 Pin Cable (PL2303) May 09 11:44:46 Board Wire Function Pin 1.....Blue......Ground Pin 4.....Green.....Transmit Pin 5.....Red.......Receive May 09 11:44:55 I did it May 09 11:46:37 but I am getting "�������" May 09 11:46:40 always May 09 11:49:31 I've heard more people having problems with that cable May 09 11:50:00 wrong voltage? May 09 11:50:03 wrong baud rate? May 09 11:50:07 wrong karma? May 09 11:50:30 I am using usb May 09 11:50:47 and I am giving 5V to p1 May 09 11:51:05 baund rate 115200 8n1 May 09 11:51:25 it seems the 3.3V output from the bbb isn't sufficient for it to (reliably) receive data... even though I'd normally expect a 5V uart chip to be okay with a 3.3V signal May 09 11:53:14 ah. PL2303 input specs say Vih(min) is 0.7*Vdd May 09 11:53:17 that's 3.5V May 09 11:53:22 well there you go May 09 11:53:44 ıt is working before May 09 11:54:03 zmatt: not really... 3.3V is just 66% of 5V... if both supply voltages are off by 10% into the wrong direction (which would be still within specs), then it's just 54% May 09 11:54:57 is boundrate true right? May 09 11:55:14 hhe: baudrate is fine, it's the voltage level May 09 11:55:36 hhe: doğru = right May 09 11:56:07 hhe: it can be "true" as well.. depeding on context, but it's usually better translated as "right" May 09 11:58:17 am I use 5V to power or 3.3V for bbb May 09 11:59:50 KotH: ft232r has input switching threshold around 1.0-1.5V regardless of vccio May 09 12:00:05 hhe: the external power supply is 5V, but it's converted to 3.3V internally. I/O voltage is 3.3V May 09 12:00:15 hhe: do _NOT_ use 5V signaling for the BBB May 09 12:00:24 zmatt: ah.. ok May 09 12:00:25 the console port is also 5V tolerant May 09 12:00:40 zmatt: but few chinese usb2ser adapters use ftdi May 09 12:00:43 but no other I/O May 09 12:00:56 KotH: well, presumably the chinese ftdi-clones are similar ;) May 09 12:00:56 zmatt: oh.. right.. there was this thing there... May 09 12:01:01 * KotH forgets stuff easily May 09 12:01:05 * KotH is getting old May 09 12:01:07 * KotH blames av500 May 09 12:02:06 hhe: conclusion: PL2303 operating at 5V is not compatible with BBB May 09 12:02:46 you need a 3.3V cable, or a 5V FTDI-based cable May 09 12:03:36 I have both of them May 09 12:03:39 http://prnt.sc/b1zfaz May 09 12:04:48 now working with 3.3V cable May 09 12:04:50 :D May 09 12:05:09 thanks alot May 09 12:08:40 you can also change most of the PL2303 boards to 3.3V operation May 09 12:08:47 dunno whether the olimex one supports this May 09 12:48:14 Here is my boot log : http://pastebin.com/gznkseWh, as you see theres about 5 seconds delay after message "Starting kernel ..." from uboot. how can i eliminate this huge delay? May 09 12:49:57 i'm using recent rcn kernel 4.4.8-bone10 May 09 13:00:25 Guest43953: use the --boot-very-fast option May 09 13:04:40 laughable May 09 13:29:44 hello May 09 13:30:41 Is there any information if the San Cloud BeagleBone enhanced will be sold? May 09 13:31:35 I mean, for people too late recognizing the IndieGogo campaign... May 09 13:43:04 eaglepiberry, usually the indiegogo has a follow-on page no? May 09 13:43:07 brb gotta reboot May 09 13:59:41 thanks, stt_michael, I will follow them. May 09 14:00:24 certainly was on one of the other crowd-funded pages I looked at recently .. I would imagine the sancloud was similar May 09 15:47:29 Hurray, I can talk about http://octavosystems.com/2016/05/09/octavo-systems-releases-osd3358-system-package-device/ May 09 15:53:27 <_av500_> jkridner: buy why no DIP package???? May 09 15:54:32 I pushed for QFP at first.... May 09 15:55:04 I was overwhelmingly out-voted. May 09 15:55:39 <_av500_> :) May 09 15:55:56 I mean, I agree that the board technology would be more complex for a QFP pitch that would sustain the needed number of pins. May 09 15:56:52 <_av500_> jkridner: at least a bright blue led should have been on the BGA! May 09 15:56:58 :-) May 09 15:57:33 <_av500_> http://i0.wp.com/octavosystems.com/octavosystems.com/wp-content/uploads/2016/05/OSD3358-SBC-Front.jpg May 09 15:57:40 <_av500_> I expected an empty PCB May 09 15:58:11 yeah, flash, HDMI and Ethernet PHY aren't integratd. May 09 15:58:25 also crystals. May 09 15:59:12 but, https://oshpark.com/shared_projects/GWqtFu43 should be all you need to run Linux off of USB or a uSD card and have a USB host. May 09 15:59:18 https://644db4de3505c40a0444-327723bce298e3ff5813fb42baeefbaa.ssl.cf1.rackcdn.com/9d3de29233e7087054ea598d9d84e8ac.png May 09 15:59:55 * jkridner will hear it again soon about the unfilled vias. May 09 16:32:45 jkridner: cool May 09 16:47:07 jkridner: BGA is fine as long as there is a SoM available, and there is :) May 09 17:12:51 blah... May 09 17:22:32 okay, so it's not my custom kernel or dtb... I get the same problem with a stock 4.4-bone kernel and dtb... May 09 17:22:35 [ 282.006424] uio_pdrv_genirq p9.12-irq: could not find pctldev for node /ocp/interrupt-controller@48200000, deferring probe May 09 17:22:44 on loading an overlay May 09 17:22:51 (works fine if part of main dtb) May 09 17:23:11 weird thing is, the overlay doesn't even contain any reference to the interrupt controller May 09 17:27:19 *sigh* May 09 17:28:55 I'm trying to making examples usable by "normal people" (hence the use of an overlay rather than changing the main dtb like I normally do), yet I can't even get anything basic working this way May 09 17:44:40 zmatt, at least you don't have a perl segfault to troubleshoot :/ :P May 09 18:11:01 stt_michael: sounds fun too May 09 18:11:41 zmatt, yeah it is :/ May 09 18:12:38 I've just pushed my non-working example to https://github.com/mvduin/py-uio ... I'd be interested if anyone can get the overlay applied without error, or figure out what the hell the error means anyway May 09 18:13:17 (the steps are in the readme) May 09 18:32:55 found me bug :/ classic! May 09 18:33:14 stt_michael: what was it? May 09 18:35:49 segfaulting perl is usually not easy ... (though I've had my share of them of course, being the original author of Data::Alias :P ) May 09 18:39:56 zmatt: stop messing with stuff under the hood ;) May 09 18:48:00 zmatt, fortunately I knew the likely cause in this case .. it was just a matter of tracking down exactly which bit of code was causing it. May 09 18:48:54 zmatt, in this instance I was referencing a GUI block that I had previously deleted and free'd for which the reference had disappeared. Fixing was simply a case of altering the reference to the correct entry in a reference array May 09 18:49:24 not the nicest bit of code I've ever written, but mercifully it (mostly, normally) works .. :) May 09 18:51:02 ds2: that's not an excuse for such a basic overlay to give inscrutable kernel errors May 09 18:51:26 ds2: seriously, look at it... https://github.com/mvduin/py-uio/blob/master/dts/p9.12-irq.dtsi May 09 18:51:55 that's not what I'd consider "messing with stuff under the hood" May 09 18:53:40 stt_michael: ah, keeping refcounting straight in glue onto C/C++ libraries can be delicate indeed... though often it can be done cleanly May 09 18:55:05 zmatt, this was somewhat lashed code .. but what it does is quite neat May 09 18:56:48 ds2: this example allows gpio-triggered events to be delivered nicely into any kind of event loop, which is quite appealing May 09 18:57:20 (I know sysfs has a notify mechanism too, but it's rather hackish and non-standard) May 09 19:05:39 got a useful crash course in the perl debugger mind. May 09 19:06:32 :) May 09 19:26:54 Hello, to debug a bricked beaglebone I'd like to use an arduino as a uart screen. May 09 19:27:28 I have connected J1 to ground, J4 to RX J5 to TX. Have on the arduino side connected reset to ground May 09 19:27:49 I open the serial monitor on Arduino IDE, set it to 115200 baud. May 09 19:28:16 I now power my beagle bone and only character printed is an 'unprintable' character, a square that I can"t copy paste May 09 19:28:56 Bobox: you can bootload the BeagleBone over that serial port... May 09 19:28:58 This happens in all cases, if I power from DC or USB, if I power with USER button on and a valid SD card. May 09 19:29:09 it prints out characters when it is trying to boot over serial. May 09 19:29:16 hmm May 09 19:29:20 it'll loop trying to boot over serial/USB/etc. May 09 19:29:31 easiest solution is to insert a bootable microSD to reflash it. May 09 19:29:45 Yes but inserting a bootable microSD does not work May 09 19:29:57 I only have the power led on. and nothing else happen May 09 19:30:00 did you hold the BOOT button when applying power? May 09 19:30:11 Yes, the one next to the SD card May 09 19:30:24 Did that. No better results currently May 09 19:30:26 are you confident in your flashing of the microSD card? I happen to like http://etcher.io as a programming tool. May 09 19:31:06 My SD card is one I used 2 days ago to update the eMMc and it worked well at that time May 09 19:31:18 http://linuxkernel51.blogspot.com/2015/08/booting-beagle-bone-black-over-uart.html seems helpful for booting over serial port May 09 19:32:15 I thought I could use serial to 'see' what was happening during boot. May 09 19:32:23 Thanks jkrider I'll look to that link May 09 23:36:52 Really new here . . . I can see the on http://192.168.7.2/bone101/Support/BoneScript/#functions but don't know how to get to the command line May 09 23:43:51 byeeee May 09 23:47:27 zmatt: you been spamming the sancloud BBB-E page?! :P May 10 00:00:09 veremit: in retrospect it would be nice if you could amend your comments there instead of posting additional ones :/ May 10 00:00:32 ah primitive web gateways ftl May 10 00:02:41 ctypes.Structure already makes py-uio a lot nicer to look at :D May 10 00:02:51 could definitely still be better though using python3 metaclasses May 10 00:02:58 ( https://github.com/mvduin/py-uio/blob/master/l3-sn-test.py ) May 10 00:12:51 woo, the l3-sn overlay does seem to work (along with its test script) May 10 00:18:46 getting closer? May 10 00:20:43 I still want this example to work on a stock bbb config though -> https://github.com/mvduin/py-uio/blob/master/p9.12-irq.py May 10 00:21:41 can you check if you also get that weird error if you compile the overlay and try to load it? (follow the first three steps in the example in the README) May 10 00:22:29 the one 10min ago? May 10 00:23:25 https://github.com/mvduin/py-uio May 10 00:23:39 the steps to build and add the overlay May 10 00:25:40 ok hold up May 10 00:25:52 oh and you need the etc/modprobe.d/uio.conf in place if you don't have it already (or just manually "modprobe uio_pdrv_genirq of_id=uio May 10 00:25:55 ") May 10 00:26:52 zmatt: I take it this is arm-based code .. ;p May 10 00:27:33 the examples target the BBB obviously, I also state that explicitly in the README May 10 00:27:53 (the code in general isn't really platform-specific though) May 10 00:27:57 right .. that'll take a bit longer to *arrange* .. :) May 10 00:28:34 I got an imx6 running .. heheh May 10 00:29:23 well then try there, the device tree fragment should be obvious enough to modify to suit another SoC May 10 00:29:30 assuming gpios work vaguely the same May 10 00:30:23 my helper macros for pins and such will be useless obviously May 10 00:30:48 but most people manage to write overlays without them :P May 10 00:32:17 I mean, most people who manage to write overlays at all manage to do so without CPP (since its invocation isn't exactly intuitive) May 10 00:32:20 :P May 10 00:32:51 brb sec... May 10 00:32:58 this will be fun :P May 10 00:37:18 I still don't understand what the kernel is smoking to come up with that bizarre message May 10 00:37:32 "could not find pctldev for node /ocp/interrupt-controller@48200000, deferring probe" May 10 00:37:59 1. this is well after boot, the interrupt controller is obviously there and ready :P May 10 00:38:03 very descriptive :D May 10 00:38:12 2. I don't even reference the interrupt controller May 10 00:38:29 ts a red herring May 10 00:45:29 I mean, normally I'd say fuck overlays if the kernel is this inept at coping with a dynamically added device :P but until there's a good tool to generate a custom main dts (ideally with some pointy-clicky interface) Joe Average is still stuck with using overlays May 10 00:48:58 yeah can't get the staff May 10 00:49:12 pity .. its one of the (many) things holding BBB back as a go-to platform :/ May 10 00:49:50 ok sorry .. got a bit distracted May 10 00:50:34 no hurry, this isn't a one-day project May 10 00:51:12 and yeah, for newbies it's basically insurmountably difficult to actually get at the cool functionality on the BBB May 10 00:57:06 zmatt: this is kinda stupid, but it'll kill two birds with the same stone .. you'll be able to say it works with the imx6, and I'll get easy-peasy python IO on my wandboard ...... May 10 00:57:19 soo.... files copied to /etc May 10 00:59:12 yikes this is ambitious, but hey .. May 10 00:59:27 just checking... does your kernel include the uio_pdrv_genirq driver? :P rcn-ee enabled it in his kernels on my request, it's not a default afaik May 10 00:59:54 most likely not May 10 01:00:01 its a RN kernel .. but .. let me check version May 10 01:00:17 4.1 .. quite possibly then May 10 01:00:35 the version says nothing, he even enabled it in 3.8 kernels May 10 01:00:55 grep PDRV /boot/config-$(uname -r) May 10 01:00:56 w00 loaded May 10 01:01:13 right .. now for a random pin, and the stuff .. so .. May 10 01:01:36 I found the kernel code emitting the message... still no idea what it's doing though, or how it got there May 10 01:03:46 I just had a thought: maybe it's because this is the first irq requested from gpio1, hence gpio1 in turn needs to request its irq from the interrupt controller... but no, I put an irq request for gpio1.28 in my main dts (which works) and then tried to load an overlay requesting one for gpio1.29 .. same error May 10 01:05:25 wait, what May 10 01:05:42 I just realized, this code is about pinctrl May 10 01:05:51 "pctldev" = pin controller device May 10 01:05:52 shit May 10 01:06:03 yeah .. lol May 10 01:06:26 why on god's green earth is it looking for the "pin controller" for /ocp/interrupt-controller@48200000 May 10 01:06:36 that makes absolutely no sense May 10 01:06:59 cos it just Does.. May 10 01:08:16 maybe I should just put a BUG() there, recompile, and see if the stack trace gives any insight May 10 01:08:29 worth a try May 10 01:09:55 it'll probably still mean crawling through this densely opaque neligibly-commented kernel code though... not really looking forward to it :/ May 10 01:13:37 hmm where these pin headers come from then, zmatt? May 10 01:13:45 I'm gonna have to do a bit of crude translation.... May 10 01:14:47 subarctic/pinctrl.h and bone/pins.h ? May 10 01:14:56 they came from my fingers May 10 01:14:58 yip May 10 01:15:08 eh .. oh .. they're .. erm May 10 01:15:35 oh rofl.. right, I see May 10 01:15:38 in the latter case more or less from my spreadsheet, with some regex-fu applied May 10 01:16:07 most overlays just write hex constants May 10 01:16:29 ok this is crude .. make on the totally wrong mach .. haha May 10 01:17:10 the makefile will work on any system with a suitable dtc installed May 10 01:17:21 (and cpp) May 10 01:17:25 uh oh .. apt-get install build-essential needed .. :/ May 10 01:17:48 you don't have cpp on your imx? May 10 01:17:54 I will do shortly .. haha May 10 01:18:08 o buggah its outta date too .. greate May 10 01:18:19 also check which dtc versions are available May 10 01:18:31 yeah probably need RN's "special" one May 10 01:18:45 not hugely special, but it does need the overlays patch May 10 01:18:53 you can also clone and build his git repo, or mine May 10 01:19:38 I don't do enough code-mangling .. but that's all about to change May 10 01:20:49 caution btw: my dtsi-to-overlay perl script doesn't really parse the input remotely properly... to find the beginning and end of a fragment it just assumes you properly indented it :P May 10 01:21:58 I still think there's no good excuse why dtc requires a ridiculous input format for overlays May 10 01:23:35 hmm, I can't unconditionally put a BUG() there since I also get a legit-looking version of that message during early boot May 10 01:23:42 [ 0.242574] omap_i2c 44e0b000.i2c: could not find pctldev for node /ocp/l4@44c00000/scm@210000/pinmux@0800/i2c0, deferring probe May 10 01:26:39 ah .. its probably caused by the same thing .. nothign to do with your code then .. May 10 01:26:45 unless its just aggrivated by it May 10 01:27:00 ok dtc repo? May 10 01:27:46 just apt-get install? May 10 01:28:01 I mean, building from source is an option but not really necessary May 10 01:29:15 I have device-tree-compiler version 1.4.1-0rcnee1~bpo90+20160224+1 May 10 01:29:50 which really is an impressively elaborate version number May 10 01:32:13 oh its there .. and that works? May 10 01:34:31 what's the package called? May 10 01:34:37 bugger.. dtc nor wgetpaste exist lol May 10 01:34:48 device-tree-compiler May 10 01:34:53 just said that :P May 10 01:34:58 oh duh May 10 01:35:02 three-things-at-once May 10 01:35:17 couldn't they find a longer name? May 10 01:36:07 program-which-converts-random-code-format-into-something-the-kernel-understands-about-peripherals-connected-to-it-and-what-to-do-with-them-and-the-appropriate-driver-to-load-and-some-more-stuff-probably ? May 10 01:36:18 -sigh- May 10 01:36:38 lol its one with an rcn revision .. looks good May 10 01:37:38 aha I have a dtbo May 10 01:38:43 well then all you have to do it use my bin/add-overlay May 10 01:38:57 waitinf for gist .. May 10 01:38:59 (first make sure you have no unsaved open documents or whatever on that thing) May 10 01:39:27 https://gist.github.com/mjeveritt/5a446489f226f430434dde0adef6480b May 10 01:39:45 that means it failed, see kernel log May 10 01:40:10 [48904.073199] of_resolve_phandles: Could not find symbol 'am33xx_pinmux' May 10 01:40:11 [48904.079764] create_overlay: Failed to resolve tree May 10 01:40:14 well duh May 10 01:40:16 :P May 10 01:40:19 hehe May 10 01:40:34 do I have to look that up .. or you busy :P May 10 01:41:58 you should either look for an existing overlay intended for your machine and use that as a guide, or check its main dts :P May 10 01:42:13 probably May 10 01:43:29 I'm still trying to figure out how the hell it ends up trying to do a pinctrl lookup on the interrup controller May 10 01:46:53 ok .. so .. location of am35xxxx May 10 01:48:44 assuming the iMX is similarly well-endowed with muxes as the am335x, the pincontroller can't be hard to find... any chunk of device tree that needs pins setup would have a reference May 10 01:49:32 right May 10 01:50:09 hey, I just had an idea... since this problem is apparently somehow triggered by the pinctrl definition, I could just omit it... gpio is the default already on the am335x anyway May 10 01:50:44 #include #include May 10 01:51:12 urgh.. https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/arch/arm/boot/dts/imx6qdl.dtsi May 10 01:51:57 I am missin something :/ May 10 01:52:12 maybe it doesn't have pinmux? May 10 01:52:53 no .. it does .. but possibly not in the same horrible way .. https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/arch/arm/boot/dts/imx6qdl-wandboard.dtsi May 10 01:53:14 I think its done differently, yeah .. May 10 01:53:31 if any of that makes any sense to you :/ May 10 01:53:47 iomuxc is the pin controller's name apparently May 10 01:55:16 https://www.kernel.org/doc/Documentation/devicetree/bindings/pinctrl/ May 10 01:55:37 see fsl,imx-pinctrl.txt May 10 01:55:55 and fsl,imx6dl-pinctrl.txt or fsl,imx6q-pinctrl.txt , whichever is applicable May 10 01:56:07 yeah this board is a 'dl' iirc May 10 01:57:19 it certainly doesn't look nicer May 10 01:57:43 (magic hex constants)-- May 10 01:58:27 -shrug- its all .. horrible :D May 10 01:59:21 hmm, removing the pinctrl did make the overlay load successfully May 10 02:01:07 grr can't find anything gpio for imx in the kernel sources .. yet :/ May 10 02:03:31 sanity check: I assume there are people who use gpio on that thing? can they just export them in sysfs and be done with it? May 10 02:04:16 since if so, you don't need a pinctrl declaration May 10 02:04:17 eh .. I don't want to use sysfs .. if I can do it elsehow .. I want to drive a 2x16 display from the metadata from an icecast source .. (see where I'm going?! ) May 10 02:04:49 shooting for the i2c interface of a port-expander though May 10 02:05:09 so .. might not need gpio per-se .. but I can drive an LED or two .. perhaps through userspace May 10 02:05:29 but.. I'm just extending your concept, really :D May 10 02:05:42 and I'm curious now .. very dangerous! May 10 02:06:23 well my point was, this particular example uses gpio... this doesn't actually need explicit pinmux setup on the am335x and probably not on your board either May 10 02:06:55 right .. ok, so .. May 10 02:06:58 so you could remove the first fragment and the two pinctrl-related properties in the uio device declaration May 10 02:07:17 lets comment those and see .. :D May 10 02:07:23 I'd personally still rather see pinmux working correctly, but at least it'll give you a first test while worrying about pinmux later May 10 02:08:22 @bb938:~/py-uio$ ./p9.12-irq.py May 10 02:08:22 Ping! May 10 02:08:24 :D May 10 02:08:38 ok so without pinctrl the overlay works, and the example works May 10 02:10:25 done the p9 dtsi .. which is the other file? May 10 02:11:15 hold up .. slow down .. more haste less speed. May 10 02:12:52 hmm it loaded my overlay :D May 10 02:13:38 but I doubt its gonna work without some extra -fu .. https://gist.github.com/mjeveritt/aa5730798048408abf1405c64b9dd8a9 May 10 02:14:11 I'm one step behind you, zmatt May 10 02:17:53 or two .. or three .. May 10 02:18:53 python version? May 10 02:18:59 3.4.1 iirc May 10 02:19:07 see readme :) May 10 02:19:12 psh May 10 02:19:15 fix eet :D May 10 02:19:21 that is next on my list May 10 02:19:24 don't make me .. May 10 02:19:24 XD May 10 02:19:38 I might be able to update python .. sec .. May 10 02:23:21 naa I'd have to backports it May 10 02:23:36 good old ancient debian May 10 02:25:47 yeah I really need to reduce the dependency to 3.4 May 10 02:26:06 I run stretch myself, but I know I'm a minority in that regard :P May 10 02:30:03 I run gentoo .. but not *yet* on arm .. but the time is coming .. :D May 10 02:43:31 hey, funky... you're allowed to remove overlays in arbitrary order nowadays? May 10 02:43:46 it used to be a stack May 10 02:43:51 I suppose if the dependencies are right ... May 10 02:44:05 or there is some way of tracking or .. such **** ENDING LOGGING AT Tue May 10 02:59:59 2016