**** BEGIN LOGGING AT Wed Mar 19 02:59:58 2014 Mar 19 03:16:57 jkridner, mranostay: edited my application and timelines . RFC if possible Mar 19 03:17:13 (on melange) Mar 19 03:47:42 vvu|Log: a late pong. Sorry. Mar 19 03:48:03 Just woke up. That's great news ! Mar 19 03:49:50 So, is it from mode 5 of adk or a different approach ? Mar 19 03:54:37 hello Mar 19 03:59:25 av500: what kind of event do you want to capture from keyboard? I don't have very clear how to pass through Adk and Displaylink a keypress event Mar 19 03:59:33 jkridner, are there any other minor bugs with bonescript? Mar 19 04:00:44 jkridner: hello Mar 19 04:01:44 jkridner: I sent my proposal to google-melang if you have some time please leave feedbacks :) Mar 19 05:35:35 Once I submit my proposal. can I modifiy it? Mar 19 05:48:27 TonyF: yes. till 21st. Mar 19 05:53:41 anujdeshpande:Thank you very much! Mar 19 07:05:46 A rather strange problem with my usb driver code. It's a very small piece of code. Anyone with related knowledge who are willing to help ? Mar 19 07:07:43 https://github.com/praveendath92/hello-kernel-drivers/blob/master/testusb.c Mar 19 07:09:10 Issue explained: https://github.com/praveendath92/hello-kernel-drivers/issues/1 Mar 19 07:55:14 Hello guys Mar 19 07:55:31 I am interested in participating in GSoC for beagle Mar 19 08:18:51 hiiiii Mar 19 08:19:28 any GSoC menters available here? Mar 19 08:19:46 I want to discuss about my GSoC project Mar 19 08:42:32 Hi. I'm currently interning at a web development startup, though work there is not very hectic. It is expected to last until mid July. Would you recommend me to apply GSoC? I was planning to apply for the Getting Started nodekit app. Mar 19 09:48:57 wow BeagleBoard is boiling this year! impressed by the number of projects this year http://elinux.org/BeagleBoard/GSoC/Ideas Mar 19 10:01:22 praveendath92_: ping Mar 19 10:01:33 hohi Mar 19 10:01:51 hello av500 Mar 19 10:02:58 did last night some hacks and adb is still available via usb when adk runs Mar 19 10:03:01 hi vvu|Log Mar 19 10:03:09 vvu|Log: nice Mar 19 10:03:17 how did you test that? Mar 19 10:03:36 just switched my nexus 5 into adk mode then adb shell :) Mar 19 10:04:03 there are some problems compiling latest adb for arm Mar 19 10:04:29 i could cross-compile it but openssl which is linked into the final binary is not made as static Mar 19 10:04:42 so i need libcrypto.so on my BBB which cannot really fix Mar 19 10:04:58 but praveendath92_ has the means to debug at least Mar 19 10:05:09 i was a bit skeptical about that Mar 19 10:06:48 vvu|Log: ping Mar 19 10:07:18 hey Mar 19 10:07:28 That's a great news ! Mar 19 10:07:56 praveendath92|PC: are yu sure your vid/pid are right ? Mar 19 10:08:14 and also when u connect your phone on the BBB can you do a 'lsusb' ? Mar 19 10:08:20 to see what you get there Mar 19 10:08:21 Yes. Mar 19 10:08:50 I tested it with lsusb Mar 19 10:09:17 That's how I got those numbers Mar 19 10:09:53 av500: can you take a look over the cross-platform boot proposal ? Mar 19 10:10:02 Did you see the log I posted in issue ? Mar 19 10:10:18 yep Mar 19 10:11:01 I got some strage messages in the log. Wait let me see the line numbers Mar 19 10:12:19 https://github.com/praveendath92/hello-kernel-drivers/issues/1 Mar 19 10:12:52 When I insert the module with insmod, the probe callback is being called. Without plugging the device Mar 19 10:12:52 praveendath92|PC: for me it works on my Ubuntu machine Mar 19 10:12:58 the testusb_probe gets called Mar 19 10:13:14 Bus 001 Device 005: ID 04e8:6860 Samsung Electronics Co., Ltd Mar 19 10:13:28 yes, that should be there automatically Mar 19 10:13:33 That's what I see for lsusb Mar 19 10:13:42 ok so they match Mar 19 10:14:11 testusb_probe should be called when we insert the device right ? Not on inserting the module. Mar 19 10:14:16 Hmm Mar 19 10:14:47 Another strage message in log: Mar 19 10:14:48 Mar 5 01:12:49 beaglebone kernel: [ 5772.775156] usbcore: deregistering interface driver testusb Mar 5 01:12:49 beaglebone kernel: [ 5772.775368] testusb: disconnect module Mar 5 01:12:49 beaglebone kernel: [ 5703.933939] testusb: driver registered successfully Mar 19 10:14:58 praveendath92|PC: use pastebin/gist Mar 19 10:15:11 https://gist.github.com/ungureanuvladvictor/9638921 Mar 19 10:15:14 this is what i get Mar 19 10:15:16 let me try on BBB Mar 19 10:15:52 Did you try the code I used ? Mar 19 10:15:56 yes Mar 19 10:15:58 On your machine ? Mar 19 10:16:06 ah no more space on the BBB Mar 19 10:16:08 I'm glad my code isn't wrong ;) Mar 19 10:16:17 That's a basic one ! Mar 19 10:17:48 https://gist.github.com/praveendath92/9638956 Mar 19 10:18:43 something does not seem right there Mar 19 10:19:20 Yeah. The last lines is confusing. Mar 19 10:19:22 https://gist.github.com/praveendath92/9638970 Mar 19 10:19:37 When I inmod. The first line also seems wrong. Mar 19 10:19:56 let me flash my BBB Mar 19 10:20:05 cuz i filled it up and anyway wanted to switch to uSD card Mar 19 10:20:12 and then i can test there Mar 19 10:20:42 Okiee. I will wait. Mar 19 10:21:13 vvu|Log: how did you switch into adk mode? Mar 19 10:21:16 from BBB? Mar 19 10:21:33 yep Mar 19 10:21:41 used some libusb code, not from kernel though Mar 19 10:21:46 ah Mar 19 10:21:47 np Mar 19 10:21:55 and then used adb on BBB? Mar 19 10:21:58 so it compiled? Mar 19 10:22:24 it did not compile, took a binary from xda-developers forum Mar 19 10:22:29 ah Mar 19 10:22:31 :) Mar 19 10:22:34 send me a link Mar 19 10:22:36 tried to compile it staically but still have some problems with shared libs Mar 19 10:22:52 av500: http://forum.xda-developers.com/showthread.php?t=1924492 Mar 19 10:22:58 ignore title of the thread :) Mar 19 10:23:08 so its armv6 ):) Mar 19 10:23:29 i think this is th one i have, downloaded multiple ones Mar 19 10:23:37 but anyway tried to cross-compile it Mar 19 10:25:00 av500: https://gist.github.com/ungureanuvladvictor/9639057 Mar 19 10:25:19 and cannot find libcrypto.so.1.0.0 even if i put it into /usr/lib and all the paths Mar 19 10:25:32 run ldconfig Mar 19 10:26:01 the one i have on the BBB is https://gist.github.com/ungureanuvladvictor/9639071 Mar 19 10:26:11 i did ldconfig -p to show all of crypto Mar 19 10:26:21 and? Mar 19 10:26:32 does not work, still cannot locate it Mar 19 10:26:39 ok Mar 19 10:26:43 angstrom or debian? Mar 19 10:26:47 debian Mar 19 10:26:52 *latest-image* Mar 19 10:26:55 ja Mar 19 10:27:01 ill try to install that today Mar 19 10:27:07 need to find place on my desk :) Mar 19 10:28:22 but adk+adb worked from my ubuntu also so won't need to worry about that Mar 19 10:28:36 right Mar 19 10:28:39 we can test from PC Mar 19 10:28:54 what worries me is that if i disconnect my phone *remove usb cable from the port* the BBB freezes :) Mar 19 10:28:56 like a BIG accessory :) Mar 19 10:32:05 vvu|Log: https://gist.github.com/praveendath92/9636437 is this what you used for activating adb mode ? Mar 19 10:32:15 *file you sent me last night* Mar 19 10:32:19 yeah this code Mar 19 10:33:22 Looks like it can be translated to kernel code without much struggle. Mar 19 10:34:02 *Curious to that after this code works* Mar 19 10:46:05 vvu|Log: It worked on my Ubuntu too. With galaxy s2 Mar 19 10:46:30 Problem running with BB. I think a few calls are being delayed. Dun't know why :/ Mar 19 10:48:21 vvu|Log: I shall try the adb mode requesting on my Ubuntu for now. Mar 19 10:50:03 mhm "uevncmd was not defined in uEnv.txt...halting..." Mar 19 11:07:18 av500: https://vdev.ro/adb Mar 19 11:07:23 here is my adb cross-compiled Mar 19 11:07:32 if you can fix the libcrypto stuff it would be nice :) Mar 19 11:08:44 praveendath92|PC: how did u install the headers ? Mar 19 11:09:01 For BB ? Mar 19 11:09:21 yep Mar 19 11:09:24 Let me check where I got them from. Mar 19 11:12:23 https://github.com/RobertCNelson/linux-dev/releases Mar 19 11:12:33 I coul Mar 19 11:12:59 I can't find the exact link I got the headers from. Mar 19 11:13:08 I will send you. Searching.. Mar 19 11:13:46 got it i think Mar 19 11:15:58 It worked ? Mar 19 11:16:14 It took me very long for the process to complete. Mar 19 11:17:46 build_deb.sh? Mar 19 11:18:56 Yep. Mar 19 11:31:01 vvu|Log: https://gist.github.com/praveendath92/9639831 Mar 19 11:31:07 Did I get it right ? Mar 19 11:35:37 Error setting up accessory Mar 19 11:35:45 does this sound right ? Mar 19 11:37:10 praveendath92|PC: can you upload the .deb somewhere ? Mar 19 11:37:15 cuz it takes too long to generate it Mar 19 11:46:25 It doesn't sound right Mar 19 11:46:37 Which .deb file ? Mar 19 11:46:58 with the headers Mar 19 11:48:44 Sure. Let me see. Mar 19 11:49:13 morning Mar 19 11:49:37 It is a .tar that I have Mar 19 11:50:32 linux-dev-3.8.13-bone41.tar.gz Mar 19 11:51:19 yep Mar 19 11:51:22 upload it somewhere Mar 19 11:51:24 Should upload this ? Mar 19 11:53:46 http://home.iitb.ac.in/~praveendath92/linux-dev-3.8.13-bone41.tar.gz Mar 19 11:59:39 vvu|Log: I think that error is because I don't have the corresponding app to handle on the phone. Mar 19 12:15:02 mdp, Hey Matt! Mar 19 12:15:22 I'm working on the proposal to package language support libraries for Debian Mar 19 12:15:57 And I think I would be able to work on a few other bugs/feature requests which are not big enough for a GSoC project by itself Mar 19 12:16:03 Would you have any suggestions? Mar 19 12:16:23 I talking with Kridner and he pointed out the gpio-keys for BoneScript Mar 19 12:20:03 praveendath92|PC: should not care Mar 19 12:20:08 check lsusb again Mar 19 12:20:14 to see if u have changed the pid/vid Mar 19 12:21:14 praveendath92|PC: after running the build_deb script Mar 19 12:21:17 trying with the app installed. Mar 19 12:21:29 don't u get a .deb file ? Mar 19 12:23:33 The script ran for a while and the it worked. Mar 19 12:24:04 Does it say "YOu can now write kernel modules " ? Mar 19 12:24:14 Something like that. Mar 19 12:28:40 vvu|Log: Now I got the exact one that worked for me. Mar 19 12:28:42 wget https://raw.github.com/gkaindl/beaglebone-ubuntu-scripts/master/bb-get-rcn-kernel-source.sh Mar 19 12:28:54 chmod +x bb-get-rcn-kernel-source.sh Mar 19 12:30:07 https://gist.github.com/praveendath92/9640684 Mar 19 12:30:31 Almost forgot gist. This worked for me. Mar 19 12:50:16 praveendath92|PC: ok got it working Mar 19 12:50:46 What worked finally ? Mar 19 12:53:10 to install the headers Mar 19 12:55:30 Yes Mar 19 12:57:29 vvu|Log: Did you do any fixed to the C code or Android app ? http://stackoverflow.com/questions/6524891/android-adk-with-pc-as-usb-host-with-libusb-bulk-transfer-error Mar 19 12:58:58 For ADK connection. Even after installing the app, when I run the script on my PC I get a popup on my phone saying "No installed apps work with this accessory" Mar 19 12:59:05 did u change the vid/pid in the C code ? Mar 19 12:59:10 ah so it changed Mar 19 12:59:11 to adk mode Mar 19 12:59:13 :) Mar 19 13:00:04 anyway i'm off to class Mar 19 13:00:38 Alright. I will try to fix this. Mar 19 13:00:44 Or we can discuss later. Mar 19 13:00:58 that app does not work with this C code Mar 19 13:01:13 they do not match becuse of the manufacturer in the res/xml/filter file Mar 19 13:01:16 in your androdi app Mar 19 13:01:22 read how matching is done Mar 19 13:01:27 Android app with accessory Mar 19 13:01:28 based on what Mar 19 13:01:31 https://gist.github.com/praveendath92/9641145 Mar 19 13:02:09 Did you have to do these changes ? Mar 19 13:03:02 I will try changing them. Thanks :) Mar 19 13:45:21 wiznerd, I don't have any particular suggestions off the top of my head..I'm not familiar with bonescript. Mar 19 14:46:41 ds2: ping Mar 19 15:08:25 mdp: I request you to look into my proposal and comment Mar 19 15:33:36 jkridner, online? Mar 19 15:33:42 yes Mar 19 15:34:36 * jkridner diverts attention away from channel devoid of new posts Mar 19 15:36:03 jkridner, could you have a look at my proposal? Mar 19 15:37:24 There is no compilation of feature-requests on the google group. I skimmed through the posts Mar 19 15:37:29 * jkridner doesn't see any proposals from "wiznerd". Also, I can give greater feedback (line-by-line) via the google group. Mar 19 15:37:41 Feroze Mar 19 15:37:53 My name is feroze Mar 19 15:37:57 * jkridner doesn't understand what "compilation of feature-requests" is. Mar 19 15:38:12 lol third person Mar 19 15:38:24 I mean a list of feature small feature requests Mar 19 15:38:39 like migrating to gpio-keys for bonescript Mar 19 15:38:48 I went through the bonescript issues page Mar 19 15:39:04 can you bring rrittman/vagrantc into the discussion. I'd like to see his feedback. Mar 19 15:39:32 Can you include documentation for your efforts in bone101? Mar 19 15:39:37 I messaged Rcn-eee on gtalk Mar 19 15:40:14 rseethamraju: is melange working for you yet? Mar 19 15:40:23 I could, but I could comfortably fit that in 2-3 weeks. Also, I was hungry for some more actual coding Mar 19 15:41:03 jkridner, I haven't run into vagrantc Mar 19 15:42:35 wiznerd: I sent you his e-mail address via a private message Mar 19 15:44:59 jkridner, got it Mar 19 15:45:22 also, is there any scale of importance/priority for the projects? Mar 19 15:47:11 mranostay: ping Mar 19 15:48:00 wiznerd: Nothing that is obvious. The degree to which you can gain mentor interest is the best metric for the priority. Mar 19 15:48:38 okay. I'll ask if nelson would be interested Mar 19 15:49:24 I will provide proper documentation for the libraries included Mar 19 15:49:33 give more example source code Mar 19 15:49:35 wiznerd: can you move discussion back here and not the private channel? only private things should be on a private channel. Mar 19 15:50:05 I understand. That was a gtalk conversation. That's why I pasted it in the private channel Mar 19 15:51:28 vvu|Log: Ping Mar 19 15:52:16 * jkridner didn't see anything in there that needed to be private. Mar 19 15:52:34 praveendath92|PC: pong Mar 19 15:52:46 https://www.kernel.org/doc/Documentation/fb/framebuffer.txt Mar 19 15:52:58 okay Mar 19 15:53:07 jkridner, https://github.com/jadonk/bonescript/issues/14 and https://github.com/jadonk/bonescript/issues/12 Mar 19 15:53:16 Can you tell me how you took screen shots when you were implementing using screenshots ? Mar 19 15:53:24 These are pretty small bugs Mar 19 15:53:32 are they still there? Mar 19 15:53:34 had some program that dumped th framebuffer Mar 19 15:53:39 they are all over the net Mar 19 15:53:43 CLI tools Mar 19 15:53:55 The link I just gave. Mar 19 15:54:00 do you have any other bonescript issues in mind? Mar 19 15:54:11 It looks like a good option. Mar 19 15:54:17 praveendath92|PC: /dev/fb has a defined API Mar 19 15:54:28 as vvu|Log said, many examples Mar 19 15:54:58 wiznerd: yes, I haven't fixed them yet as I haven't gotten back to EEPROM authoring with BoneScript. Mar 19 15:55:12 jkridner, and could you tell me more about "Implementing and testing core libraries in Userspace Arduino" Mar 19 15:55:16 So displayLink uses frams from this buffer and will be transmitting over usb ? Mar 19 15:55:32 jkridner, So can I add that to my proposal? (EEPROM) Mar 19 15:56:02 sure, but you'd want to look at the current state of EEPROM authoring tools and where the issues currently are. Mar 19 15:56:16 alexanderhiam: not really but it opens from the link in the automatically generated mail about your comment. Thanks! working on it. Mar 19 15:56:44 wiznerd: http://azkeller.com/blog/?p=62 is useful as is http://inspire.logicsupply.com/p/building-custom-cape.html Mar 19 15:56:47 praveendath92|PC: have you looked at the displaylink driver? Mar 19 15:58:21 Not completely. I landed here when I was looking for some codes and documentation Mar 19 15:58:21 http://libdlo.freedesktop.org/wiki/ Mar 19 15:58:40 jkridner, and who is the mentor for "Implementing and testing core libraries in Userspace Arduino" ? Mar 19 15:59:33 I don't think anyone. :( probably a good question to see if there are interested parties on beagleboard and beagleboard-gsoc Google Groups. Mar 19 15:59:38 and on #beagle and here. Mar 19 15:59:53 praveendath92|PC: speak 2night if you are here, now i have another quiz :) Mar 19 16:00:05 so far, I haven't seen anyone jumping to mentor updating Userspace Arduino. Mar 19 16:00:16 wiznerd: try mdp , bradfa Baxsie on #userspace-arduino Mar 19 16:00:23 jkridner: ^^ Mar 19 16:00:36 much more interest in developing PRU code, autopilot stuff and (from me) updating the bone101 content with interactive tutorials. Mar 19 16:00:57 wiznerd: good point... checking out #userspace-arduino is definitely the hang-out. Mar 19 16:01:09 vvu|Log: I will be here. Good luck for the quiz :) Mar 19 16:01:33 * jkridner is surprised the Galileo crowd still hasn't been informed that they are using Userspace Arduino and to collaborate there. Mar 19 16:02:00 * jkridner is surprised no one has reported them to gpl-violations. Mar 19 16:02:01 anujdeshpande, hey! you on gsoc 2013 right :) Mar 19 16:03:08 jkridner: maybe prpplague did. should ask him.. sad that he isn't participating this year :( Mar 19 16:03:24 wiznerd: yepp. i did #userspace-arduino with hatguy__ last year :_ Mar 19 16:03:26 *:) Mar 19 16:03:30 oh! Mar 19 16:03:37 anujdeshpande, applying again this year? Mar 19 16:03:49 wiznerd: yepp. can't get enough of beagle ;) Mar 19 16:03:58 I've read your blog! Mar 19 16:04:08 sweet Mar 19 16:04:13 wiznerd: :D Mar 19 16:04:22 Which project, if I may ask? Mar 19 16:04:45 wiznerd: beagle pilot hopefully Mar 19 16:05:19 whoa Mar 19 16:05:25 all the best mate! Mar 19 16:06:19 av500: libdlo is the name of their open sourced library. Mar 19 16:07:27 praveendath92|PC: ah Mar 19 16:07:29 hmm Mar 19 16:07:41 ok, we need to be careful Mar 19 16:07:54 it seesm there are two approaches, one in-kernel and one as an X11 backend Mar 19 16:08:10 alexanderhiam : hello :) Mar 19 16:08:11 libdlo seems to imply user space Mar 19 16:08:21 yeah it looks like. Mar 19 16:08:59 libdlo uses the kernel driver or interface I mentioned above to get video frames Mar 19 16:09:42 Like you mentioned they seem to have written APIs to access that data and send to displaylink devices. Mar 19 16:09:53 wiznerd: danke Mar 19 16:09:59 praveendath92|PC: there is a user guide Mar 19 16:10:03 alexanderhiam : I believe I have a decent solution to the size problem (of ascii botspeak instructions) :D Mar 19 16:10:15 Sending is done with userspace code I suppose. Since they are using libusb Mar 19 16:10:21 http://people.freedesktop.org/~berniet/Guide-v104.pdf Mar 19 16:10:25 anujdeshpande : Hello! Mar 19 16:10:27 yes Mar 19 16:10:34 karki: hola Mar 19 16:10:46 I haven't gone through it completely. Should I do that now ? Mar 19 16:10:56 it would make sense to read it Mar 19 16:11:02 anujdeshpande : your using UIO for PRU control if I am right? Mar 19 16:11:04 but note that we dont have to do it like that 100% Mar 19 16:11:07 *you are Mar 19 16:11:10 BTW, I was able to detect my mobile using kernel driver. Mar 19 16:11:33 nice Mar 19 16:11:42 With some issues also the setting to adk mode. Mar 19 16:12:27 have a C code which is to be translated to kernel space. May take time but an easy task, as it seems Mar 19 16:12:34 karki: actually If selected I won't be looking after the loading + comm.. that would be sidbh Mar 19 16:13:12 I shall be through this document and update the displaylink parts of proposal Mar 19 16:13:19 just asking in general! what do you use for your PRU control? prussdev ? Mar 19 16:13:49 av500: Is that your final comment ? 'No need buy a hdmi cable' ? Mar 19 16:14:11 :) Mar 19 16:14:12 no Mar 19 16:14:19 I will add as comment Mar 19 16:14:35 That would be great ! Thanks :) Mar 19 16:19:34 jkridner, are you looking for more bone101 demos? Mar 19 16:20:34 jkridner, I think we could port some of the arduino code for commonly used sensors to bonescript and show them as demos? Mar 19 16:21:37 I thought someone else is already working on that? Mar 19 16:21:45 Could we split up tasks then? Mar 19 16:22:26 karki: hello Mar 19 16:23:24 what are you thinking for code size? Mar 19 16:23:47 praveendath92|PC: http://libdlo.freedesktop.org/wiki/displaylink-mod/ Mar 19 16:24:28 http://lists.freedesktop.org/archives/libdlo/2009-July/000258.html Mar 19 16:24:33 this one is in-kernel Mar 19 16:24:56 Nice ! Mar 19 16:25:08 I will have a look at it :) Mar 19 16:31:29 mranostay: I have edited the timeline. Could you review and comment? Mar 19 16:36:37 mdp, could you tell me about the "Implementing and testing core libraries in Userspace Arduino" project Mar 19 16:36:48 That doesn't have any detail about the goal and deliverables Mar 19 16:37:54 anujdeshpande, could I pm you? Mar 19 16:38:21 wiznerd: IRC is kewl. i am falling behind on my mails actually. Mar 19 16:44:38 anujdeshpande, I'm going through parav's blog posts Mar 19 16:45:02 what part is still left? the wiki page hasn't listed specifics Mar 19 16:45:14 it appears like you guys worked on SPI, I2C and all Mar 19 16:46:03 wiznerd: if you are thinking of applying to userspace-arduino, i suggest you contact one of the mentors. because unless and until you have a few mentors who are willing to guide, chances are close to nil of being selected. Mar 19 16:46:10 wiznerd, Serial is left Mar 19 16:46:15 but that not a huge project Mar 19 16:46:22 *that's Mar 19 16:46:24 hey parav! Mar 19 16:46:27 wiznerd, hi Mar 19 16:46:47 panto: I edited the timeline. Does it answer your question? Mar 19 16:46:56 I'm actually looking for bits of work. My main would be packaging and bonescript issues Mar 19 16:46:59 I thought that project was deleted a while back from the ideas page Mar 19 16:47:48 hatguy__: i thought so too. since prpplague isn't interested. Mar 19 16:48:32 checking history... Mar 19 16:50:52 praveendath92|PC: i will try during the weekend to see how to switch stuff to adk mode from kernel, if that test works from there is just your part of coding :) Mar 19 16:51:01 even if i am super interested in the project :) Mar 19 16:51:11 vvu|Log: pls leave some work for praveendath92_ Mar 19 16:51:21 vvu|Log: we will find something else for you Mar 19 16:51:25 Yes vvu. Please do. Mar 19 16:51:53 my work ends there, was just curious how to speak some usb Mar 19 16:52:06 av500: can you take a look over the cross-platform boot proposal Mar 19 16:53:06 i gave my input for it but would like to see some other opinion Mar 19 16:53:24 vvu|Log: I'm super tempted to work on the ADK mode from kernel but need to look at the displaylink for now. Mar 19 16:53:49 my main concern was adb over adk but now it's gone Mar 19 16:54:54 Now I have one less thing to worry about. A big thanks :) Mar 19 17:00:06 alexanderhiam : So, here it is. Since I have two PRU's I plan to split the work wherein PRU0 does the talking with the ARM and PRU1 does the IO. now..... Mar 19 17:00:06 Hello all, pleasure to be here! Mar 19 17:01:05 any GSoC mentors not invited to the invite-only #beagle-mentors channel, please ping me. Mar 19 17:01:33 alexanderhiam : I need the botspeak format, but at the same time I need to save space. [Space shouldn't be a problem if I could map every botspeak instruction to a 8 bit value] Mar 19 17:01:40 karki: wouldn't the inter-PRU communication add a lot more complexity? Mar 19 17:01:51 alexanderhiam, Tartarus: I'm looking at you Mar 19 17:02:07 here'ish sorry Mar 19 17:02:12 mranostay: ^^^ Mar 19 17:02:53 jkridner: ping! Mar 19 17:03:18 jkridner: yes, ping Mar 19 17:03:46 alexanderhiam : I doubt so, they have common scratch pads, data area and also one PRU can write into the other PRU's data ram. And I'm looking at panto's and mranostay's code on github right now. I think they have done inter PRU sync. Mar 19 17:04:05 praveendath92|PC, you have some TBD items on your proposal, best to finish it soon. Mar 19 17:04:13 alexanderhiam : and thats the best way to split work between PRU's Mar 19 17:04:26 VoltVisionFrench: I'm around, though definitely multitasking. Mar 19 17:04:45 jkridner, meeting time is now right? Mar 19 17:04:58 Tartarus, alexanderhiam: I've sent the invites. You not able to join #beagle-mentors? Mar 19 17:05:04 alexanderhiam : I don't see other ways to exploit the existance of 2 PRU's (atleast as of now! ) Mar 19 17:05:23 jkridner, for the bone101 project, what additional documentation is needed? Mar 19 17:05:30 and I'm talking to Robert Nelson now Mar 19 17:06:36 is there any particular demos which are needed? Mar 19 17:06:43 wiznerd: did you look at the stuff Diego has been discussing on the beagleboard-gsoc list? he seems to have the best grasp on what needs to be improved in bone101 right now. for execution, I'll certainly be looking at best abilities to engage mentors and clearly communicate. Mar 19 17:07:06 I'll have a look Mar 19 17:07:13 karki: try to figure out roughly how many of these botspeak scripts you could have running at once with that scheme Mar 19 17:09:00 alexanderhiam : won't running more than one script require some kind of scheduling policies ? (or did I get you wrong? ) Mar 19 17:09:08 jkridner: I have edited the timeline of my proposal; I'm yet to receive comments on the impact of my proposal on the community. May I post just the timeline and the impact part on the group for a detailed comment from your side? Mar 19 17:10:06 karki: yeah, I'm not sure if that's implemented in botspeak, but I remember that being one of the ideas on the mailing list. Is that not something you're planning? Mar 19 17:11:44 hellooo,I want to contact a mentor Mar 19 17:11:57 for discussing my project Mar 19 17:12:46 Jaseel_: there are potential mentors here and in #beagle. suggest you intice them with your amazing proposal idea. Mar 19 17:13:29 #beagle IRC Channel? Mar 19 17:13:35 yes Mar 19 17:13:42 ok Mar 19 17:13:47 probably want to start here Mar 19 17:14:14 Jaseel_: please wait ca. 45 min, there is a meeting going on right now Mar 19 17:14:18 thx Mar 19 17:14:25 ok Mar 19 17:14:30 i will wait Mar 19 17:14:44 jkridner: Pardon my ignorance, but will the invite be in an email or here? Mar 19 17:20:23 alexanderhiam : I was thinking about it, it is definitely on the TODO list. But I thought it may get a bit too hairy for the summer. ( I realized it will be complex cause I spent quite some time trying to sketch out the possibilities) Mar 19 17:23:16 VoltVisionFrench: invite? are you a mentor? Mar 19 17:23:29 * jkridner doesn't know VoltVisionFrench's human name Mar 19 17:24:05 jkridner: I signed up earlier today on the link you provided .... Human Name = Steve French Mar 19 17:24:22 ah! Mar 19 17:25:51 VoltVisionFrench: join #beagle-mentors Mar 19 17:26:18 karki: we don't use the scratch pad in our pru code currently Mar 19 17:26:28 just had no real need to Mar 19 17:29:51 mranostay: Which part of the PRU-based LA do you consider to be the most difficult to implement and would take the most time to? Mar 19 17:31:31 praveendath92: done Mar 19 17:31:56 Abhishek_: the PRU assembly and interface to userspace Mar 19 17:32:04 that could be a bottleneck Mar 19 17:32:11 Thanks av500 Mar 19 17:32:27 mranostay: some guy has written parts of it already Mar 19 17:32:37 I shall update the remaining parts of my proposal today. Mar 19 17:33:25 joel_: thanks for the reminder. I updated moat of them last night. 2 left. Will be doing now :) Mar 19 17:33:25 mranostay: I have made changes to the timeline, considering your comments. Please have a look Mar 19 17:33:46 I did ping you a few hours ago Mar 19 17:34:57 jkridner: I've edited the timetable according to the feedback I received from you yesterday, could you please look, I posted the answer on the mailing list and edited the proposal Mar 19 17:36:18 mranostay : I see ( still have to go through all of your code :p )! but anyway I was just listing it as a possible means of communication. Mar 19 17:38:41 any menters available? Mar 19 17:41:58 abarbarosie: thanks Mar 19 17:45:11 praveendath92, generally proactive candidates get more points ;-) Mar 19 17:45:17 but you're doing fine. Mar 19 17:46:15 I was playing it safe ;) Mar 19 17:47:35 Must admit. Was a complete noob when i first started my proposal. Now i see through every corner of my project. Mar 19 17:50:09 joel_: could you leave some comments on my timeline or any aspect of my proposal ? Mar 19 17:51:29 praveendath92|PC, go ahead and complete the proposal first Mar 19 17:51:43 there's time for editing it, I will certainly go over it when I next get a chance Mar 19 17:52:09 Al right then. I will ping you once I'm done :) Mar 19 17:53:34 joel_: Would like to have your comments on my proposal too. Mar 19 17:55:44 Abhishek_, I don't have any further comments outside of whats already on it. (also panto had a comment there I think you need to respond to that) Mar 19 17:58:13 panto's comments were made before I edited my proposal today. I am replying to the comment right now. Mar 19 17:59:33 cool Mar 19 18:01:50 hiiii,iwant to discuss my GSoc Proposal Mar 19 18:02:35 I want mentor Mar 19 18:03:30 which one is your proposal? Mar 19 18:03:45 BB Embedded Server For Wireless Sensor Network It allows authorized Internet users to establish two-way communication with the sensor network. This allows the user to monitor the operation of the WSN remotely, to periodically download the sensed data, and to change the operation mode of the network. In addition to providing monitoring and data collection services, Mainly Focused In Agriculure sector, Soli moisture,humidity, Mar 19 18:04:09 did you enter that on Melange? Mar 19 18:04:22 no Mar 19 18:04:54 i will submit proposal today Mar 19 18:05:10 av500,r u mentor? Mar 19 18:05:25 I am one of them Mar 19 18:05:43 ok, Mar 19 18:06:01 how was my project? u have any suggestions? Mar 19 18:06:51 I cannot say much Mar 19 18:07:00 how is it beagleboard specific? Mar 19 18:07:12 how does it benefit the beagleboard community? Mar 19 18:07:21 how does it tie into exsiting open source projects? Mar 19 18:07:35 these are all questions you need to be prepared to answer Mar 19 18:07:45 I think Beagleboard is good for Embedded Linux projects Mar 19 18:07:48 sure Mar 19 18:07:51 thats why i selected Mar 19 18:08:31 whats the next step? Mar 19 18:08:50 write up your proposal Mar 19 18:08:55 can i directly submit into melange? Mar 19 18:08:56 answer the questions above Mar 19 18:08:57 yes Mar 19 18:09:02 you can update until the deadline Mar 19 18:09:19 since it is not one of the proposed projects, you need to be convincing Mar 19 18:09:20 ok,Can i submit 2 proposals? Mar 19 18:09:23 yes Mar 19 18:09:31 not saying that is wise Mar 19 18:09:52 what is the proposed projects? Mar 19 18:13:10 https://www.google.com/search?q=beagleboard+summer+of+code Mar 19 18:13:42 panto: I just replied to your comment Mar 19 18:15:03 i got it Mar 19 18:15:17 Sir,Iam interested in BeaglePilot Mar 19 18:15:26 many are Mar 19 18:16:02 because I have experience in flightcontrollers,RC airplanes,quardcopters etc Mar 19 18:16:10 Abhishek_, ok Mar 19 18:16:45 Aeromodelling is m y hobby Mar 19 18:17:19 av500,can i select Beaglepilot Mar 19 18:17:45 you can, but there are already some students engaged Mar 19 18:18:01 ohh, Mar 19 18:18:30 iam also interested in android display Mar 19 18:18:47 we have 2 proposals there too already Mar 19 18:19:01 I'm sorry, I need to go home Mar 19 18:19:05 jkridner: ^^ can you :) Mar 19 18:19:09 ok Mar 19 18:19:13 see u Mar 19 18:19:45 ah Mar 19 18:19:47 well Mar 19 18:19:51 * jkridner feels there is always room for better proposals, but coming in late is a disadvantage given relationships already formed with mentors. Mar 19 18:20:46 looking at discussion on the beagleboard-gsoc list that seems to indicate the plan for a project doesn't look well-formed makes for ripe territory to swoop in with a better proposal. Mar 19 18:21:45 jkridner: still waiting for your response on the library using starterware project. :( Mar 19 18:22:22 electrobuz: what is your name again? Mar 19 18:22:36 naman Mar 19 18:23:33 I don't see it on the melange proposal list. I'll look at the mailing list. Mar 19 18:23:38 ds2: I like your PRU exploration project idea Mar 19 18:23:57 PRUsploitation Mar 19 18:25:18 i havent uploaded the proposal yet. since i was waiting for your response on taking up the energia support part. Mar 19 18:25:24 PRUSSplotation :P Mar 19 18:28:24 jkridner: response on this thread: https://groups.google.com/forum/#!topic/beagleboard-gsoc/OdRCMUyQYGA Mar 19 18:34:28 jkridner: thanks for the reply. the plan is to start out with arduino functions only. after that, adding library support for accessing cards Mar 19 18:35:36 Abhishek_: need to do something now...catch you later Mar 19 18:35:52 Abhishek_: talk to mdp if you see him and he is free Mar 19 18:36:56 jkridner: my goal after creation of the library is to integrate it with energia. seems good enough to pursue? Mar 19 18:36:57 ds2: how many hours after can I expect you to be back? Mar 19 19:06:22 vmayoral, have you played with Xenomai previously? Mar 19 19:33:44 panto: Do you have any follow-up comments? Mar 19 19:40:20 tridge: ping Mar 19 19:40:45 wdnch: hi Mar 19 19:41:27 tridge: hi. i have a quick question Mar 19 19:41:31 sure Mar 19 19:42:33 tridge: is ardupilot using on beaglepilot project? i Mar 19 19:42:55 tridge: i cant sure what is this Mar 19 19:43:17 wdnch: i hope BBB will become a supported board for ardupilot. http://diydrones.com/profiles/blogs/a-peek-into-the-future-of-ardupilot Mar 19 19:43:45 basically I think ardupilot needs to move towards embedded Linux, and BBB seems to be the ideal first target board Mar 19 19:44:03 so I've done an initial port using cheap 10DoF I2C sensors, which is running now Mar 19 19:44:24 ardupilot works with that setup on BBB, but its not flyable Mar 19 19:45:04 next step is building a really good sensor cape (which we hope to have prototypes of in the next few weeks), and start using PRU for PPM/PWM Mar 19 19:45:41 i send a proposal last night end jkridner told me to look beaglepilot project. Mar 19 19:45:51 but i'm still confusing Mar 19 19:45:56 the sensor cape (called the PXF or 'Fire' cape) has a good set of SPI sensors Mar 19 19:46:35 could you please look at my project and tell your idea? Mar 19 19:46:44 can you give me a link to it? Mar 19 19:46:46 are we aim same thing? Mar 19 19:47:09 http://www.google-melange.com/gsoc/proposal/review/student/google/gsoc2014/bbuyukguzel/5629499534213120 Mar 19 19:47:22 Abhishek_, let me take a quick look Mar 19 19:47:32 unfortunately melange doesn't allow me to view that URL, as it has your username embedded Mar 19 19:47:47 ok wait a second to gdocs link Mar 19 19:48:03 what is your username in melange ? Mar 19 19:48:08 bbuyukguzel Mar 19 19:48:11 or project name? Mar 19 19:48:35 Project IMFS Mar 19 19:48:44 ok, found it, reading now Mar 19 19:48:48 ok than Mar 19 19:49:56 wdnch: ok, you want comments here or on the melange site? Mar 19 19:50:46 your choice Mar 19 19:50:46 so I want to implement openCV functions/algorithms like canny's edge detection in PyBBIO there's a 2013 GSoC project that has some algorithms converted to Python including it into PyBBIO will entail just testing right? Mar 19 19:51:22 i can say future features on my projects on here :) Mar 19 19:51:31 Abhishek_, look good Mar 19 19:52:18 ok, fist off, I2C for gyros/accels is not good. To really fly well you need to sample at a high rate, and I2C is too slow. You could create something that flies, but it won't fly really well (ie. won't be as good as an APM2) Mar 19 19:52:26 you need SPI for higher speed sensors Mar 19 19:52:30 panto: could you provide inputs on the "benefits to community" part? Mar 19 19:53:01 I will, but probably tomorrow Mar 19 19:53:08 is it urgent? Mar 19 19:53:40 wdnch: you also seem to be proposing to build the flight code from scratch? Building a full autopilot stack is a huge task. Much better to use an existing one. I've ported ardupilot to BBB, so that is the obvious one to use, but there are other open source autopilot stacks that could also be used. Mar 19 19:55:15 panto: by what time will you be able to give feedback tomorrow? Mar 19 19:55:29 probably will do it in the morning Mar 19 19:56:03 wdnch: why NixOS btw? Mar 19 19:56:21 tridge: yes my goal is writing flight control from scratch. and i'll write all steps in my project website for other users Mar 19 19:57:01 tridge: i seached some rtos and i though nixos would be good. but i have to say, i have no experience with nix before Mar 19 19:57:14 hmm, sure. will be looking forward to it :) Mar 19 19:57:28 wdnch: a lot of people try to write an autopilot from scratch. Many get to the point of very basic flight, but getting all the details right to create a fully featured autopilot is a huge task. ardupilot is around 0.5M lines of code. Mar 19 19:57:56 much better to use an existing one and port it to BBB Mar 19 19:59:33 wdnch: maybe you should watch the video I pointed you at above and look at the AP_HAL_Linux port of ardupilot? Mar 19 20:09:28 tridge: thank you for your helps Mar 19 20:10:52 wdnch: no problem, good luck with it! Mar 19 20:12:51 tridge: btw, can i ask one more question? :) Mar 19 20:12:57 sure Mar 19 20:14:55 how can i appeal to beaglepilot project? propasal text should be different Mar 19 20:18:21 i'm not sure really. I don't even know how many GSoC slots BeagleBoard has asked for and is likely to get this year. Mar 19 20:18:49 I haven't set on the final number, but I'm feeling it'll be in the 6-10 range. Mar 19 20:19:39 and, of course, Google makes the decision. Mar 19 20:20:02 jkridner: ok, we have 3 reasonably well advanced proposals for beaglepilot at the moment. How many do you think we should aim for? I'm guessing you wouldn't want all of the slots to be taken by beaglepilot :) Mar 19 20:21:10 indeed. 3 feels a bit like a maximum to me for BeaglePilot, but we can discuss more on #beagle-mentors. I'm looking to maintain a 2:1 mentor-to-student ratio or greater. Mar 19 20:21:21 3:1 is ideal. Mar 19 20:21:43 hmm, ok, thats a lot of mentors! Mar 19 20:21:59 right, but I've found it helps the projects move smoothly. Mar 19 20:22:05 and avoids mentor burn-out. Mar 19 20:22:18 mentors have many more distractions than students, typically. Mar 19 20:22:24 yep Mar 19 20:22:47 I want to be sure students aren't sitting idle waiting on mentor feedback. Mar 19 20:22:57 shouldn't take too much time of any one mentor. Mar 19 20:24:45 jkridner: sorry about this question, but what should be on application if i want to work on one of these ideas? (not new project) Mar 19 20:25:43 jkridner: it should be different from default propasal text Mar 19 20:25:49 each proposal needs to be independent. It is possible that the combination of multiple proposals helps to build up a bigger project, but I'm looking for the proposals to stand on their own. Mar 19 20:26:17 For example, say I was writing a PWM/PPM generator to run on the PRU for BeaglePilot. That would be one proposal. Mar 19 20:26:40 is implementing opencv on the board too computationally expensive? Mar 19 20:27:01 I don't like that several proposals are written right now to simply say "BeaglePilot". They should be updated to say what *part* of BeaglePilot that particular student will be implementing. Mar 19 20:27:06 Updated my proposal. Mar 19 20:27:22 rseethamraju: there are many running examples of OpenCV on BeagleBone Black. I suggest you google them. Mar 19 20:27:28 praveendath92|PC: received the update, will look tonight or 2morrow morning. Mar 19 20:27:45 vvu|Log: That would be great ! Mar 19 20:28:22 I wanted to discuss a few details related to the ADK connection. Let me know when we can. Mar 19 20:28:33 jkridner: ok thank you so much Mar 19 20:28:59 leave your questions here and ping me in them vvu|Log 'question' :) Mar 19 20:29:04 s they can stay in my log Mar 19 20:29:36 wdnch: I think the mentors looking at the BeaglePilot project will know who actually understands and wrote a project proposal vs. who copied someone else's proposal. working off of existing proposals and re-writing them such that they are *better* seems like acceptable practice to me. The re-write should be associated with engaging mentors and should reflect a greater understanding of the coding challenge and abili Mar 19 20:29:37 ty to succeed if it is to replace an existing proposal.... Mar 19 20:29:55 if it is meant to augment existing proposals, it should be clear how it will interrelated. Mar 19 20:31:06 mentors will have their favorites and I'll work with them to select as many projects for which we have slots and for which we believe we have students with proposals we can successfully mentor. Mar 19 20:32:52 * jkridner recommends mentors look at *actual code* to justify student selections! Mar 19 20:33:06 http://en.flossmanuals.net/GSoCMentoring/notes-for-first-year-organizations/ is a nice intro for new mentors. Mar 19 20:33:24 * vvu|Log needs to read that :) Mar 19 20:34:54 jkrinder: yes. I've used Derek Molly's code and he has an edge detection example. But can a part of my project be providing this in PyBBIO. There's Python code available already so including it wouldn't really be a hard or time consuming. Mar 19 20:36:05 will it? Mar 19 20:36:21 adding opencv to python is a no-op... it already exists. Mar 19 20:36:58 vvu|Log: Will do that. I'm sleepy already so, we discuss them tomorrow. Mar 19 20:37:11 ok thanks. I really needed that cleared! Mar 19 20:38:55 rseethamraju: here's my example of that: https://github.com/jadonk/LEDscape/blob/fc5611735c62d1a72e3aefe379617e307c0f0e5d/colors.py Mar 19 20:39:31 vvu|Log: 'question' : I changed the xml values in the android app and when I connect my phone and run the executable on my PC the android app opens but still it gives the same old error. Any ideas ? Mar 19 20:40:58 rseethamraju: and the vine of it running: https://vine.co/v/htzHbDziV2Q Mar 19 20:41:05 vvu|Log: When you tried to run, did you have to do the mentioned in the answer here: http://stackoverflow.com/questions/6524891/android-adk-with-pc-as-usb-host-with-libusb-bulk-transfer-error Mar 19 20:41:31 jkrinder: Ok. But can I do this? Add a camera library so that any camera supported by the v4l2 driver can be directed to take a picture or record a video with a simple command like capture() or record() with BBIOserver integration to be able to access the output over the internet. Mar 19 20:42:17 vvu|Log: I did the C code correction with no success. Let me know your comments. Good night :) Mar 19 20:42:33 good night :) Mar 19 20:43:18 cool. Alexander Hiam mentioned opencv in my proposal so I just started looking into it. Mar 19 20:44:12 * praveendath92|PC is signing off Mar 19 20:44:35 mranostay: Do you have any follow-up comments? Mar 19 20:44:38 praveendath92|PC: check the android code if it sends data when it oes online Mar 19 20:45:34 When it goes online ? Mar 19 20:45:40 when it starts Mar 19 20:45:48 as i know that android app has a slider right ? Mar 19 20:46:22 Slider ? Mine had a text box. Mar 19 20:46:26 does it automatically get the write stream for the accessory and starts sending stuff Mar 19 20:46:30 i dunno which one u are using Mar 19 20:46:43 check it it starts sending data because the C app expects to read something. Mar 19 20:46:44 Can you send me your app code ? Mar 19 20:47:01 try and make it work urself here :) Mar 19 20:47:25 From what i found, the buffer sizes are not matched in these two codes. Mar 19 20:47:50 ADk anyway returns only 512 chunks if i remember right Mar 19 20:48:08 * praveendath92 says challenge accepted :P Mar 19 20:48:57 Not infront of the code right now but I will use this as a lead. Mar 19 20:49:05 read the specs Mar 19 20:49:07 it says there Mar 19 20:49:25 It is probably with response from java code. Mar 19 20:49:48 Specs ? Mentioned where ? Mar 19 20:50:42 google adk specs Mar 19 20:52:06 Oh, there. I will do that. Mar 19 20:54:14 I will do tomorrow morning. Thanks for the lead. Mar 19 20:54:20 Nite :) Mar 19 21:11:01 hmmmm Mar 19 21:11:11 mmmmh Mar 19 21:12:41 hmmmmh Mar 19 21:14:30 Hi everyone, my name is Kiran Mathew Koshy, and I will be applying for GSoC under Beagle. Mar 19 21:14:49 Abhishek_: sorry, had to run to a lunch appt. Mar 19 21:15:38 np Mar 19 21:16:20 Abhishek_: any thoughts on a timeline for the alt project? Mar 19 21:16:43 Abhishek_: if you don't mind saying... how many GSoC2014 proposals do you have pending at the moment? Mar 19 21:17:39 Just one - the LA Mar 19 21:18:25 <_av500_> kiran_mathew: hi kiran Mar 19 21:18:31 <_av500_> any project in mind? Mar 19 21:18:33 av500: hi Mar 19 21:18:39 <_av500_> note that it's close to the deadline Mar 19 21:18:46 yes, several, actually Mar 19 21:18:52 <_av500_> and quite a few proposals have been entered Mar 19 21:19:21 Abhishek_: Do you think it makes sense to put together a second proposal? Mar 19 21:19:26 av500: i had a project going on which just got over today- i'll submit an application for gsoc tomorrow Mar 19 21:19:46 <_av500_> one of the proposed projects? Mar 19 21:20:23 av500: i have some experience in robotics , c++, some python, etc, so i thought the PyBBIO wasinteresting Mar 19 21:21:30 av500: the CNC and the stepper motor looks good too, but why does it have to be node.js ? Mar 19 21:23:39 av500: what is IIO ? I've never heard that abbreviation Mar 19 21:23:55 <_av500_> https://www.google.com/search?q=linux+iio Mar 19 21:24:00 trouble ;) Mar 19 21:24:20 av500: right :) Mar 19 21:24:49 pwm seems easy too, i've implemented pwm in FPGAs before Mar 19 21:25:11 ds2: I could, however given less than 45 hours remain, I will have to see how it has to be done. I am not very sure about the timeline also. Mar 19 21:25:17 av500: when i submit the application, can i combine a few of these ideas into one project ? Mar 19 21:25:24 Abhishek_: *nod* understood Mar 19 21:26:39 ds2: I was wondering if it could somehow fit in my LA proposal, as I observed considerable overlap in the things that have to be tried on the PRU, and things that have actually to be done to implement the best possible core for the LA Mar 19 21:26:57 <_av500_> kiran_mathew: I would not recommend that Mar 19 21:27:05 av500: why ? Mar 19 21:27:15 <_av500_> I would recommend you try to meet one of the proposed mentors here or on the mailing list and discuss Mar 19 21:27:25 <_av500_> and time is short Mar 19 21:27:39 <_av500_> from experiecne, projects proposed outside of the list have a hard time Mar 19 21:28:05 Abhishek_: what i was thinking of is you clone the LA proposal and make changes to the deliverables... cuz in the context of the LA proposal, it would be incomplete if all you got done is the PRU details. the LA proposal has a big UI component to it Mar 19 21:28:27 _av500_: ping Mar 19 21:28:45 <_av500_> kiran_mathew: which is not meant to discourage you Mar 19 21:28:49 Abhishek_: what I would hate to have happen is if you edit the LA proposal and it turns out to be half assed and rejected Mar 19 21:28:57 can you take a look at my proposal: http://www.google-melange.com/gsoc/proposal/public/google/gsoc2014/abarbarosie/5629499534213120 Mar 19 21:29:01 indeed Mar 19 21:29:17 I believe I'm almost done and I have taken in consideration every feedback Mar 19 21:29:32 <_av500_> abarbarosie: there was some talk about it today in the mentors meeting Mar 19 21:30:25 _av500_ what about exactly? Mar 19 21:30:35 av500: thanks. i'll build a proposal right away Mar 19 21:30:42 it is too bad the rules of GSoC don't allow for more pooling of resources as the PRU explorer part is needed by several proposals Mar 19 21:31:00 <_av500_> abarbarosie: about the scope Mar 19 21:31:03 <_av500_> vvu|Log: you here? Mar 19 21:31:11 <_av500_> I dont have the logs here and I did not get all of it Mar 19 21:31:14 in an ideal world, everyone would spend the initial part working toward that goal then split off and do their own thing Mar 19 21:31:15 yep roaming around here Mar 19 21:31:42 <_av500_> vvu|Log: can you sum up the talk about multiboot? Mar 19 21:31:45 <_av500_> for abarbarosie Mar 19 21:31:49 <_av500_> kiran_mathew: good luck Mar 19 21:32:04 i gave him some feedback on what to include in the proposal Mar 19 21:32:04 av500: thanks Mar 19 21:32:18 mainly the full thingie should be something really easy to use and cross-platform Mar 19 21:32:37 last year i had a lot of newbies that wanted to use my code and was just too tricky with all those binaries floating around there Mar 19 21:32:59 so now the shift would be DFU, use DFU to flash things. and incorporate DFU in a GUI that can be shifted to Linux/Win Mar 19 21:33:14 also jkridner and me are thinking about Mac support Mar 19 21:33:32 ds2: if you don't mind, was there any discussion on the LA idea too? Mar 19 21:33:34 well I tried to focus on building the GUI as simple as possible Mar 19 21:33:37 which is kinda tricky because so far as i know abarbarosie does not have a Mac...i can provide support here but still will be questionable. Mar 19 21:34:10 Abhishek_:not specifically... Mar 19 21:34:33 Abhishek_: there is concern that the PRU projects may get stuck in the wild wild west of the PRU Mar 19 21:35:01 yes, I remember PRU being mentioned as "unexplored territory" Mar 19 21:36:02 there is a subset of things that we should be able to get to work but there is so much more Mar 19 21:36:19 a basic LA can skirt some if not most of it Mar 19 21:36:37 the Beagle Pilot folks might have a little harder time Mar 19 21:37:46 you can avoid DMA and the memory issues by sampling just to local memory and halting when memory is full Mar 19 21:37:52 _av500_: I posted the final version of the proposal on the mailing list as well Mar 19 21:37:56 have the ARM side do the rest of the work. not ideal Mar 19 21:38:09 jkridner: ping Mar 19 21:44:05 ds2: how about in consultation with the mentors involved in PRU-based projects, draft a second proposal titled something like "Developing a LA core with libsigrok and PRU", which would put, say, 8 weeks into exploration of the PRU territory and 3 weeks into building a LA core with libsigrok, and hence get a backend for the LA ready. It could then be used Mar 19 21:44:05 with PulseView over a remote desktop; and then the front-end being considered as post-GSoC goal. Mar 19 21:46:01 Abhishek_: there are GSoC rules that restrict some sharing so I don't know if that is allowed Mar 19 21:46:25 sharing, as in? Mar 19 21:46:34 colaboration Mar 19 21:47:04 there is a midterm thing and then there is a final deliverable Mar 19 21:49:03 But I didn't exactly get what you said about the sharing part and the violation of rules. Mar 19 21:51:09 I remember this being asked in the google mentor list Mar 19 21:51:35 something about having clear deliverables for each student... I suppose if the task can be broken up appropriately, that might fly Mar 19 21:53:05 would any other mentor be able to further comment on this? Mar 19 21:53:20 mdp didn't come online after you were away Mar 19 21:54:24 maybe Mar 19 21:54:31 so we're supposed to mention timeline with dates? Mar 19 21:54:50 rseethamraju: the more plausible your time line is, the better Mar 19 21:55:52 oh ok thanks. I'll change it now. Mar 19 21:56:13 jkridner: are you in touch with Laine-Walker Avina? Mar 19 22:07:30 ds2: Could you give some inputs on the impact of my proposal to the BB community? Mar 19 22:07:53 I have that part pending Mar 19 22:13:40 Abhishek_: to some degree. Mar 19 22:13:50 abarbarosie: pong Mar 19 22:14:34 jkridner: I'v update the proposal and reposted in on mailind list and melange, could you check it? Mar 19 22:14:42 k Mar 19 22:17:55 jkridner: I did mail him regarding long back regarding my proposal, didn't hear back Mar 19 22:42:56 hello mdp Mar 19 23:05:30 Hey so I would really like to the Debian interation with Ruby and am working on my application currently. Anyone have any sound advice? Mar 19 23:07:25 debian interaction? Mar 19 23:08:59 The project calls for a script to be written so the support libraries for Ruby can be integrated into Debian. Sorry I should have been more specific. Mar 19 23:14:31 I have plenty of Ruby experience and I know a good amount about Debian so I think this project meshes two existing skills while still forcing me to build beagle board experience. In essence this is the ideal project for this summer. Mar 19 23:25:17 mranostay: When would you be able to give your inputs on the "impact on community" on my proposal? Mar 19 23:28:13 look good but the self debugging is about impossible :) Mar 19 23:28:34 you know the PRU pins on are almost every useful subsystem as well :) Mar 19 23:38:07 That part of the proposal was suggested by Hunyue as an illustration of how it may be used Mar 19 23:51:09 mranostay: BTW I just came across one of the presentations you gave about libsigrok. You had touched upon the PRU as a possible candidate :) Mar 19 23:52:03 Abhishek_: if your project gets executed successfully, it could provide an all in one learning solution for the next step beyond blinking an LED Mar 20 00:01:21 ds2: would this part elaborated serve as answer 2 / 3 for that question? Mar 20 00:01:50 I am guessing yes... in the middle of assembling boards so I can't easily look it up Mar 20 00:03:03 assembling: soldering? Mar 20 00:03:11 that's part of it Mar 20 00:09:20 Could you leave your comment on my proposal for answer 2 / 3 so that I can add it to my proposal when I check back again? Mar 20 00:22:32 jkridner: I have requested your feedback on my proposal "community impact" section. Could you please check the thread? Mar 20 00:22:43 yeah, I see that. Mar 20 00:28:05 I just mailed you a copy of the complete proposal; the PDF link is outdated Mar 20 00:40:51 * jkridner isn't happy with re-reading entire documents rather than reading in-line follow-ups with feedback Mar 20 00:44:35 jkridner: Comments on my proposal currently are all on melange, I'm sorry, there isn't any inline version of the proposal I can share with you. Mar 20 00:47:07 k, I will reply there. Mar 20 00:57:26 Abhishek_: I think the application is looking pretty good and don't have much specific feedback. I'd like to hear from others on benefits to the community. Mar 20 01:04:47 jkridner: May I submit this snapshot of an online VCD viewer (I mentioned about it the other day with WaveDrom) as something we should try to emulate? : http://s22.postimg.org/g1o96clb3/Capture.png Mar 20 01:06:11 If that is what you think it should look like, yeah, that'd be fine. You might want to replace the timing diagram with an image from the web timing diagram. Mar 20 01:06:50 you mean, the WaveDrom one? Mar 20 01:07:05 http://balsamiq.com/ is an example mock-up tool you could use. Mar 20 01:07:32 yes, the WaveDrom one... just so it looks closer to how it'd likely look in the end. Mar 20 01:07:35 Abhishek_: do you still need anything from me? I saw your mail... Mar 20 01:08:10 http://mashable.com/2012/06/07/mockup-tools/ Mar 20 01:09:39 ds2: Yes. I referred it to feedback from jkridner and he said that he'd like to hear about benefits to community from others. Mar 20 01:10:56 ok... let me look up the question Mar 20 01:12:15 Does anyone have any comments about how adding ruby support libraries to the next debian distro will benefit the Beagle Bone community? Mar 20 01:12:49 Abhishek_: yes, the earlier comments I made was for that Mar 20 01:13:13 EnvelopeHolder: not sure of the real value of Ruby. I have read http://www.ryanfaerman.com/notebook/2012/12/08/beaglebone-in-ruby.html Mar 20 01:13:22 the value is you can create a self contain educational lab that will go from blinking an LED to more complex protocols like SPI/I2C and beyond. Mar 20 01:14:08 Thanks guys! I appreciate any and all comments. Mar 20 01:17:40 ds2: I recall mentioning something similar I had mentioned in my proposal (but lost in the edits): Creating *a portable instrumentation toolbox*. Mar 20 01:21:16 Okay, Is there any other point I can incorporate for Answer 2 / 3 in my proposal? Mar 20 01:34:08 there are many angles on that Mar 20 01:34:18 I was just pointing the educational angle Mar 20 01:36:38 jkridner: I had yesterday noticed that there is a limit to the waveform scaling that is done by WaveDrom. See the example: http://jsfiddle.net/H7nBn/17/ ; I set hscale to less than 1 but it is ignored Mar 20 01:43:00 ds2: not to bother you too much, but could you give a view from a non-educational angle? Mar 20 01:43:16 *another Mar 20 01:47:55 Abhishek_: yeah, I see what you mean. Mar 20 01:50:40 Abhishek_: you're looking for more feedback? Mar 20 01:51:13 yes, on the community impact part. Mar 20 01:52:53 well as one who manages an IO library, it would be awesome to have for "hardware unit testing" code, without having to have specific target devices Mar 20 01:55:41 Abhishek_: this project can provide remote debugging capabilities due to the onboard networking features of the BBB. Other low cost logic analyzers built around FPGAs would require an additional processor to reduce the data to a network capable format. Mar 20 01:55:53 (please fix the spelling) Mar 20 01:56:31 exactly, that's why a web-based client Mar 20 01:57:02 or not even web based Mar 20 01:57:17 I was thinking of a headless instrument feeding data back Mar 20 01:58:16 noted. Mar 20 01:59:11 jkridner: Do you have any suggestion for a workaround? This could limit zooming abilities. Mar 20 02:05:54 alexanderhiam: Do you mean: automated process of sending hardware some commands, and then receive response and compare with expected response? Mar 20 02:08:14 I just meant you could potentially use it to quickly test code that, for example, talks to a particular I2C sensor, without actually having to solder/breadboard/power the sensor Mar 20 02:09:16 okay I get it Mar 20 02:12:09 Abhishek_: You can update the skin to get greater zooming. Mar 20 02:13:42 Abhishek_: can you see http://jsfiddle.net/jkridner/k7Bev/2/ ? Mar 20 02:15:13 yes, I saw it Mar 20 02:17:13 I see, they seem to store compiled templates to draw the waveforms Mar 20 02:18:33 but what if say, I zoom out to a level, say, 10sec (and it contains a 1ms waveform, for example) Mar 20 02:23:29 I am currently browsing the unpacked sources: https://code.google.com/p/wavedrom/source/browse/trunk/unpacked/WaveDrom.js for hscale Mar 20 02:23:34 At some point, you'd need to replace the images rather than just make them smaller. Mar 20 02:23:42 indeed Mar 20 02:24:12 if you see the sources, you would see that they round up hscale to a number Mar 20 02:25:55 and it makes senses it is an integer, since it is a multiple of the width of the SVG blocks provided by the skin. Mar 20 02:26:48 would letting it be fractional break the rendering? Mar 20 02:27:16 * Abhishek_ was looking at https://code.google.com/p/wavedrom/source/browse/trunk/unpacked/WaveDrom.js#1054 Mar 20 02:28:52 Abhishek_: of course it would, it would mean the elements overlapping each other. Mar 20 02:29:20 I see Mar 20 02:30:59 I take leave now, will be back again with the mockup tomm. Mar 20 02:31:07 'night Mar 20 02:32:23 panto: Do leave your community impact feedback on melange. I will check it when I am back. **** ENDING LOGGING AT Thu Mar 20 02:59:58 2014