**** BEGIN LOGGING AT Tue Mar 17 02:59:58 2015 Mar 17 03:04:25 hmmmm Mar 17 03:27:31 s/ua/au/ Mar 17 09:12:23 morning ! Mar 17 09:45:15 morning all! Mar 17 09:59:29 sidbh_: ping :) Mar 17 10:10:48 kiran4399, pong Mar 17 10:12:27 sidbh_: Hi.. so I did whatever you said.. now I am not getting the whole ring.. I am only getting the entry which is very good.. but I am still getting garbage string .. Mar 17 10:13:38 sidbh_: did you get the actual string on your BBB ? Mar 17 10:13:48 yes Mar 17 10:14:07 I'm getting "Initial 0 0" as expected Mar 17 10:14:29 your current code, gist it Mar 17 10:15:15 only the firmware part Mar 17 10:17:38 sidbh_: https://gist.github.com/kiran4399/f0a93d9f7136ba8c8087 Mar 17 10:23:23 kiran4399, I just copy pasted your code, compiled it and ran the test, I got: initial 0 0 Mar 17 10:24:01 sidbh_: are you running in the pruspeak ?? Mar 17 10:25:26 I'm just running exporting the DTB, which uploads the firmware Mar 17 10:26:13 sidbh_: I am running in pruspeak !!! Mar 17 10:30:55 see what happens when you just export the BB-PRUSPEAK dtb Mar 17 10:31:06 which is what I'm doing Mar 17 10:33:12 sidbh_: export to where ?? Mar 17 10:33:20 slots Mar 17 10:34:10 sorry, echo Mar 17 10:34:11 yes.. I exported it ... Mar 17 10:34:27 ok then run the test Mar 17 10:34:45 without doing anything else Mar 17 10:35:26 sidbh_: this is from my slots 7: ff:P-O-L Override Board Name,00A0,Override Manuf,BB-PRUSPEAK Mar 17 10:39:38 kiran4399, ok forget everything I just told you, stop all the pruspeak stuff you are running on userspace Mar 17 10:39:44 do this : echo 0:pru0_firmware > /sys/devices/ocp.3/4a300000.prurproc/load Mar 17 10:39:51 run : ./test Mar 17 10:39:58 and tell me the result Mar 17 10:48:42 kiran4399, got my last points? Mar 17 10:56:53 sidbh_: -bash: /sys/devices/ocp.3/4a300000.prurproc/load: No such file or directory ?? Mar 17 10:58:26 kiran4399, do: ls /sys/devices/ Mar 17 10:58:43 ocp.3 is there Mar 17 10:58:57 sidbh_: but 4a300000.prurproc is not there Mar 17 10:59:09 any thing with prurproc Mar 17 10:59:38 sidbh_: which dtb should I load ? Mar 17 10:59:45 if not reboot the BBB and then echo BB-PRUSPEAK Mar 17 11:00:15 I'm using the one inside pruspeak/src/dts Mar 17 11:01:01 if all fails, do: dmesg, and gist the output Mar 17 11:02:27 sidbh_: yes I got it.. ./test is my userspace executable right ?? Mar 17 11:02:51 yes Mar 17 11:03:11 you got what? /sys/devices/ocp.3/4a300000.prurproc/load now exists Mar 17 11:04:24 ��9d� 0 1 0 Mar 17 11:04:44 sidbh_: ^^^ Mar 17 11:05:44 ok, do: echo 0:pru0_firmware > /sys/devices/ocp.3/4a300000.prurproc/load Mar 17 11:05:51 and run test again Mar 17 11:09:46 sidbh_: I ran the executable in one terminal and ran echo 0:pru0_firmware > /sys/devices/ocp.3/4a300000.prurproc/load in another terminal ... Mar 17 11:09:54 sidbh_: but the output is same.. Mar 17 11:13:02 copy paste this into your test.cpp, the userspace: https://gist.github.com/bugobliterator/8bd73e0727f2337c3643 Mar 17 11:15:36 sidbh_: ��9d� 0 0 Mar 17 11:18:48 kiran4399, just reboot, compile everything and upload, because as of now I'm using the same code as you and it runs just fine, so only answer I've right now is probably build is getting something wrong, else I don't see any problem Mar 17 11:19:48 sidbh_: wait a minute.. do you want me to compile test.cpp again or something else ? Mar 17 11:20:30 everything, the dts, the firmware, the test code Mar 17 11:21:58 and reboot BBB too Mar 17 11:26:44 sidbh_: Hey.. BTW I am getting a clock-skew.. in many builds .. does it matter ? Mar 17 11:27:46 nahhh..., probably not, just correct time using ntp and you should be fine Mar 17 11:28:14 where are you compiling your firmware, onboard or offboard Mar 17 11:38:20 sidbh_: onboard :) Mar 17 11:38:50 hmmm... there we are different, I do it offboard :) Mar 17 11:39:22 anyway does the rebooting and recompiling helped Mar 17 11:47:49 sidbh_: From what are you generating pru0_firmware ? Mar 17 11:48:14 the linux x86 pru compiler Mar 17 11:48:25 running on debian Mar 17 11:50:31 sidbh_: I mean what are your source files from which you got pru0_firmware ? Mar 17 11:51:02 ohhh the one you pointed me towards, from your git repo Mar 17 11:51:39 the pru0_firmware.c, I updated with what you just pasted on gist Mar 17 11:54:11 sidbh_: Hmm.. this is really wierd.. How come you are getting it and I am not getting it ? Mar 17 11:54:19 sidbh_: do you mean this repo https://github.com/kiran4399/pruspeak/ Mar 17 11:55:04 yeah that on, do one thing, update this repo with your local changes and push it to a branch and give me a link to that Mar 17 11:55:12 s/on/one/ Mar 17 12:08:55 gm Mar 17 12:09:26 morning Abhishek_ Mar 17 12:09:50 sidbh_: Actually I did the reverse... I recloned the repo to my bbb.... I changed the pru0_firmware.c with your gist... ./install.sh....../run.sh.....went to the userspace directory... echo 0:pru0_firmware > /sys/devices/ocp.3/4a300000.prurproc/load..... g++ test.cpp...../a.out Mar 17 12:10:22 sidbh_: but the output is the same thing... ;,-( Mar 17 12:11:15 vvu: ping Mar 17 12:11:27 left some comments on your proposal Mar 17 12:11:57 you compiled it on board Mar 17 12:12:08 kiran4399, Mar 17 12:12:14 sidbh_: yes.. Mar 17 12:12:25 try off board compilation Mar 17 12:12:32 sidbh_: how do you compile offboard ? Mar 17 12:12:38 use linux pru compiler Mar 17 12:13:19 nothing different everything would be same, only difference you use different compiler, select linux on the PRU compiler download page Mar 17 12:13:26 i.e. install the PRU C Compiler on your PC and use it, however I don't see if it makes a difference Mar 17 12:19:12 sidbh_: you use it make the firmware right ?? Mar 17 12:20:25 kiran4399, download the firmware from here: https://gist.github.com/bugobliterator/8bd73e0727f2337c3643 Mar 17 12:20:43 right click the view raw and save the link as Mar 17 12:21:34 or just click it to download Mar 17 12:21:43 sidbh_: yes.. gotcha. Mar 17 12:23:56 vvu: "just use dd from the host." Mar 17 12:24:04 vvu: but dd will work only on linux and osx hosts Mar 17 12:24:12 vvu: to make it cross platform we will have to call dd from the board right? Mar 17 12:24:48 if the board exposes eMMC to the host as a mass_storage device you can run dd from the host Mar 17 12:24:59 isn't that the whole idea, to expose the eMMC as a mass_storage device ? Mar 17 12:26:22 vvu: but what if the host is a windows system.. then we can't use dd Mar 17 12:26:52 there is WinDD or something Mar 17 12:27:13 webglider: http://www.raspberrypi.org/documentation/installation/installing-images/windows.md Mar 17 12:27:19 like the guys from rpi write their sd cards Mar 17 12:27:21 same utility Mar 17 12:27:30 Win32DiskImager Mar 17 12:28:01 webglider: so in the end i think the bottleneck is BBBlfs not the nw.js dev Mar 17 12:28:10 i suggest you try and look into bbblfs issues Mar 17 12:28:14 do you have a osx computer ? Mar 17 12:31:35 sidbh_: BULL'S EYE !!!! Mar 17 12:31:45 is it working> Mar 17 12:31:48 ? Mar 17 12:31:52 webglider: get a normal IRC client Mar 17 12:32:07 sidbh_: yes !!! Mar 17 12:32:24 so its the build system Mar 17 12:32:53 compile it on x86 machine, don't use ARM compiler Mar 17 12:37:30 vvu: sorry having some connectivity issues. can't use normal IRC client, college network does not allow :( Mar 17 12:38:53 sidbh_: but tell me one thing.. originally pruspeak (without any modification worked right ??) Mar 17 12:39:05 sidbh_: then why is this ?? Mar 17 12:39:20 vvu: so instead of using a different utility why not use dd from the board by talking to it over a network protocol Mar 17 12:40:25 kiran4399, its not messing up with the pruspeak part, just the debug stuff we added, Mar 17 12:40:51 Oh.. OK.. Mar 17 12:41:26 vvu: nope i do not have an OSX computer Mar 17 12:43:04 vvu: I actually thought the project was about building the nw.js app as it was mentioned on the ideas page Mar 17 12:44:04 karki: can you please verify my proposal ? Mar 17 12:44:12 kiran4399, this all PRU compiler stuff is still in beta if I'm not wrong so such behaviours are expected :) Mar 17 12:44:20 link Mar 17 12:44:55 sidbh_: Oh right... so can you please paste the original pru0_firmware.c gist ?? Mar 17 12:45:37 original pru0_firmware.c? Mar 17 12:47:39 webglider: yes but flashing the board is a main part of the nw.js app Mar 17 12:47:44 talk with jkridner also Mar 17 12:48:15 sidbh_: oops sorry.. I got it.. Mar 17 12:48:31 kiran4399, https://gist.github.com/bugobliterator/8bd73e0727f2337c3643 , its what I copied from your gist :) Mar 17 12:49:08 webglider: in order to flash the uSD card you need to boot from eMMC, if u want to flash eMMC you need to boot from uSD, USB Mar 17 12:52:07 webglider: lots of the requirments for the nw.js app are to flash the board Mar 17 12:58:35 praveendath92: ping Mar 17 13:00:21 vvu: Yes ofcourse I understand clearly flashing is a major part as it will simplify the job of updating OS for beginners Mar 17 13:01:08 if you hang around #beagle every day is somebody asking how to update the OS/repair the OS or something related Mar 17 13:02:05 better discuss with jkridner what he expects from the nw.js app to do exactly, start writing the proposal on melange website Mar 17 13:02:08 so we can review it Mar 17 13:02:16 and put there all proposed approaches Mar 17 13:02:28 vvu: yeah sure Mar 17 13:02:37 try to discuss advantages/disadvantages Mar 17 13:03:04 windows i expect you have it right ? Mar 17 13:03:15 vvu: yes ofcourse Mar 17 13:03:49 ok, try and just touch the OS X problem now...without real hardware developing for OSX is useless Mar 17 13:04:45 anyway, gtg now...checkin started Mar 17 13:05:25 webglider: check backlog Mar 17 13:06:25 vvu: see you and thanks for the help Mar 17 13:07:05 karki: the starterware repo in my git is not complete yet... the gpio access is almost done.. but there are bugs creeping ... I haven't yet pushed the complete work onto the repo.. I'll do it as soon I solve them and clean the code.. :) Mar 17 13:07:21 ok Mar 17 13:09:12 karki: back from Singapore? Mar 17 13:09:21 yesterday :) Mar 17 13:09:46 nice Mar 17 13:10:05 how was the trip? Mar 17 13:11:27 cool Mar 17 13:11:31 humid rather :p Mar 17 13:11:42 the talk was nice Mar 17 13:11:53 lots of developers Mar 17 13:12:08 h/w hackers, s/w hackers and even dev ops Mar 17 13:13:40 what event was it? Mar 17 13:15:11 vvu: Pong Mar 17 13:15:31 Sorry. I was in the middle of my lunch Mar 17 13:17:09 FOSS Asia Mar 17 13:23:12 hmm Mar 17 13:26:41 karki: how is it ? Mar 17 13:30:51 jkridner: Hey, I'd like to know your views about this small idea (which i mentioned in one of my previous replies): http://bit.ly/1Evqpf9 Mar 17 13:35:46 kiran4399 : who are you looking for as a prospective mentor? Mar 17 13:37:54 karki: Can I have you and jkridner ? I mean you for pruspeak extension and firmata integration. jkridner for node.js and pruduino front-end Mar 17 13:38:24 hmm.... Mar 17 13:38:26 ok Mar 17 13:41:19 karki: I do not know how mentors are allotted to the projects. Is it that I have to seek for mentors ? Mar 17 13:41:30 nope Mar 17 13:41:38 I was just wondering Mar 17 13:41:47 who you were looking to work with :) Mar 17 13:42:56 karki: I am happy to work with anyone !!! :D Mar 17 13:43:52 karki: I was just wondering if you would be busy with prubridge ?? Mar 17 13:44:04 karki: and alexanderhiam ?? Mar 17 13:44:21 don't worry about the mentors Mar 17 13:44:28 focus on your proposal Mar 17 13:44:51 mentors will want to back a strong proposal, and will back away from a weak one Mar 17 13:45:20 karki: yes.. sure !!! Can you please tell me some changes you would like me to make in the proposal ? Mar 17 13:45:39 later Mar 17 13:45:46 just skimmed thru the proposal Mar 17 13:45:57 I'll take a proper look later Mar 17 13:53:35 karki: ok !! Mar 17 15:00:33 hi Mar 17 15:16:04 hi FREEZX.... Mar 17 15:16:08 hi Mar 17 15:16:12 we don't often answer to 'hi' much.... Mar 17 15:16:23 need simple questions. Mar 17 15:16:55 Do you think it would be a good idea to also have a sort of a minimalistic javascript editor that can be used to write simple scripts by using bonescript and have a button to instantly run the script (this would be very similar to what arduino offers in terms of scripting). Mar 17 15:17:06 in the nw.js idea Mar 17 15:17:33 that would simplify things for people just starting up with beagle Mar 17 15:17:50 FREEZX: that's kinda what is already at http://beagleboard.org/Support/bone101 Mar 17 15:18:04 FREEZX: there isn't a "save" button. Mar 17 15:18:24 FREEZX: DiegoTc's code makes it easy to fork these tutorials using gist.github.com Mar 17 15:18:56 http://beagleboard.org/Support/BoneScript/getPlatform/ is an example with an editor and 'run' button. Mar 17 15:19:14 FREEZX: just need better tutorials. Mar 17 15:19:16 ok, i will look into it Mar 17 15:19:36 http://jadonk.github.io/bone101 has a bit more stuff... Mar 17 15:19:42 to see if they could possibly be integrated, because the functions are a bit similar Mar 17 15:20:02 http://diegotc.github.io/bone101/Support/GSOC/views/ is DeigoTc's code Mar 17 15:20:46 i see Mar 17 15:23:55 i think something like DiegoTc's project could be integrated with the nw.js project Mar 17 15:37:36 jkridner: can you verify my proposal please ? Mar 17 15:37:47 kiran4399: link? Mar 17 15:38:01 kiran4399: I think I replied to your idea. Mar 17 15:39:25 jkridner: i think in the end the nw.js app thingie is going to be 80% trying to fix bbblfs Mar 17 15:39:42 the rest of the nw.js app does not sound too much of work Mar 17 15:39:58 vvu: agreed. :( Mar 17 15:40:16 if we get windows support that would be super nice Mar 17 15:40:45 vvu: originally I had that idea around before BBBlfs and expected to just plug something in and build that type of app in JavaScript, but whatever the easiest way to get BBBlfs working cross-platform should really be the focus. Mar 17 15:40:59 any updates on Mac support? Mar 17 15:41:20 i got somebody on GitHub trying to make a codeless kext Mar 17 15:41:29 need to supply some data about the bbb when plugged into the mac Mar 17 15:42:23 jkridner: https://github.com/ungureanuvladvictor/BBBlfs/issues/16 Mar 17 15:44:22 I am in UK this week so cannot provide anything Mar 17 15:47:03 jkridner: need to go, keep in touch Mar 17 15:56:03 can bbblfs not run on windows with something like cygwin? Mar 17 15:57:55 or be adapted into a javascript library via node-gyp or nw gyp Mar 17 16:02:19 please can anybody give the link to the BBB UI embeded in the card Mar 17 16:02:50 "give me the github rep" Mar 17 16:09:19 tanero, https://jadonk.github.io/bone101/Support/bone101/ Mar 17 16:16:17 *rcn-ee*: I have checked that repository.But could understand what the phrase " One of the main objectives will be to include the Beaglebone-UI embedded on the cards" on the project "BoneScript web pages with live-running examples and documentation" meant.I thought may the UI used on the card was built on open GL.And that what was required here was to reproduce the UI on the webpages using may be Processing Mar 17 16:16:38 "But could not understand" Mar 17 16:17:18 opengl? it's running on top of nodejs as a web-server... Mar 17 16:18:56 rcn-ee, tanero: it just uses canvas, not GL! http://jadonk.github.io/Beaglebone-UI/BBUI.html Mar 17 16:19:22 rcn-ee: If so what can *including the Beaglebone-UI embedded on the cards* mean. Mar 17 16:19:47 Idea is to include that real-time view in Bonecards: http://jadonk.github.io/bone101/Support/bonecard/, http://diegotc.github.io/bone101/Support/GSOC/views/ Mar 17 16:21:32 jkridner: Using an open GL library like Processing? Mar 17 16:22:23 tanero: no, just canvas. Mar 17 16:22:35 tanero: nice thing about web pages is the source is included. :-D Mar 17 16:24:23 jkridner: Did you get the link to my proposal ? Mar 17 16:24:54 kiran4399: where? Mar 17 16:25:37 jkridner: sent you a PM !!! Mar 17 16:25:49 * jkridner mostly ignores PMs Mar 17 16:26:09 jkridner: well it is a writable link !! :-) Mar 17 16:26:32 jkridner: What about the interfacing with potentiometers via twitter,facebook and the rest,will the student be required to come out with bonescript wrappers for these API's or student need to use the API's to demonstrate how these platforms can be used with bbb? Mar 17 16:27:13 Could bbblfs be converted to a node module using node-gyp? Mar 17 16:27:39 kiran4399 (and all prospective students): now that melange is open you can and should submit proposals there asap. That way all the mentors can read them and give feedback, and you can keep editing them until the deadline on the 27th Mar 17 16:28:05 alexanderhiam: sure :D Mar 17 16:28:42 ok Mar 17 16:29:05 we also all get emails when a new proposal is submitted, so it's the quickest way to get feedback Mar 17 16:30:59 jkridner: still there? Mar 17 16:31:15 was looking at kiran4399's doc. Mar 17 16:31:57 tanero: I'd love to see integration with Facebook, Twitter, etc. for sharing Bonecards, but you'd need to share the technical specifics of how that'd be done. Mar 17 16:32:19 alexanderhiam: +1.... sending me a Google Doc is a good way to frustrate me. Mar 17 16:32:43 * jkridner is editing this one, but won't likely do another one. Mar 17 16:32:49 especially by pm! Mar 17 16:34:28 and don't link to docs in your melange proposals Mar 17 16:41:50 Abhishek_: i have a doubt regarding PRU message passing and firmware loading framework rpoject. Does it begin from scratch on the lines of remoteproc or is will the framework work in cohesion with existing linux remoteproc Mar 17 16:42:01 project* Mar 17 16:42:19 or will* Mar 17 16:54:24 shubhangi: it will be based off the curetn rproc driver Mar 17 16:56:04 okay. thanks Mar 17 16:57:33 kiran4399: I provided some responses to help move our conversation along. I'd prefer to continue on the mailing list and Melange (if it needs to be private). Mar 17 16:59:18 jkridner: One last thing... is there anything in my proposal which you did not like ? Mar 17 16:59:28 apart from the comments... Mar 17 16:59:45 kiran4399: you are on the right track, but we have a lot of back-and-forth to go. Mar 17 17:00:24 jkridner: on the node side or pru side ? Mar 17 17:00:28 kiran4399: just want to simplify things down a bit to tasks that can be executed in small chunks of time. Mar 17 17:00:36 kiran4399: both. Mar 17 17:01:20 kiran4399: ability to draw in other mentors beyond me is a huge plus! Mar 17 17:01:44 nerdboy: any cat herding tips for the day? Mar 17 17:02:17 howdy calculus. can we talk you into spending some time mentoring this summer? Mar 17 17:03:07 nerdboy: any ideas for additional simple exercises to test student knowledge/ability? Mar 17 17:06:03 jkridner: umm.. I did not get exactly get what you meant by saying "simplify things down a bit to tasks that can be executed in small chunks of time"... Mar 17 17:06:36 not since last night Mar 17 17:06:47 I mean is it regarding the present context ?? Mar 17 17:06:53 kiran4399: itemize the tasks into small discrete chunks Mar 17 17:07:02 * nerdboy apparently brought home conference bug Mar 17 17:07:06 or during the coding time ? Mar 17 17:08:45 kiran4399: in the proposal. There is a lot of discussion of deliverables that is separate from the project plan. that makes it confusing for me to grok. just walk me step by step through the milestones rather than breaking down the deliverables once and then defining a totally different set of deliverables in the project plan. Mar 17 17:08:48 Oh.. gotcha !! Mar 17 17:10:18 the deliverables and milestones should be the same thing. small discrete chunks of code/docs that can be executed in just a few days. Mar 17 17:11:25 yes, the final deliverables can be specified for clarity, but don't go so far defining them that they conflict with the milestones. Mar 17 17:12:01 jkridner: OK.. and one more thing.. there are some botspeak arguments which are not implemented like servo and tmr, Can I implement them also ?? Mar 17 17:12:05 in fact, best just to define the final deliverables in the last milestone---which always includes documentation and test benches of some kind. :-D Mar 17 17:12:29 kiran4399: as long as they fit into a reasonable project plan. Mar 17 17:13:09 +1 Mar 17 17:13:10 * jkridner just wants progress. measurable progress toward a useful end goal. Mar 17 17:13:19 including documentation testing plans in the proposal is an absolute must Mar 17 17:13:27 +1 :-D Mar 17 17:13:32 documentation *and* testing... Mar 17 17:14:09 * jkridner won't approve proposals w/o documentation and testing plans in the milestones. Mar 17 17:14:21 * jkridner also won't approve proposals with "etc." in them. :-D Mar 17 17:15:13 jkridner: cool !! Mar 17 17:16:22 * nerdboy would approve anything with a giant a tylenol in it... Mar 17 17:16:32 s/a// Mar 17 17:16:39 jkridner: I suggested an idea about ROS And OPEN CV with bone 101 Mar 17 17:17:07 and i need to get stated on this idea Mar 17 17:17:28 which part i should statr working on? Mar 17 17:17:49 start working on? Mar 17 17:19:39 chandy: what is the end result? OpenCV and ROS bath have node.js bindings already, so what would you work on? Mar 17 17:20:24 my plan is to include ROS and OpenCv tutorial in bone101 Mar 17 17:20:31 chandy: cool. Mar 17 17:20:33 node? Mar 17 17:20:40 yes Mar 17 17:20:41 alexanderhiam: i completed adding gpio module support to pruspeak and sent merge request. working on proposal these days. will look into adding uS delay to pruspeak after submitting proposal Mar 17 17:21:13 alexanderhiam: I don't know how well the node bindings actually work. I struggled with them in the 15 minutes I put into them a year ago. Mar 17 17:21:22 chandy: that would be awesome, but would that be enough for a whole GSoC project? Mar 17 17:21:44 I have a doubt on that Mar 17 17:22:07 I guess there's also making sure that ROS implements all the bbb's peripherals I guess Mar 17 17:22:10 chandy: would you use the BoneScript RPC mechanism to add them to bone101? Mar 17 17:22:11 Should i work on one single tutorial or many? Mar 17 17:22:11 jkridner: the other day when I was discussing with alexanderhiam, he told me an important thing that single commands like digitalRead and digitalWrite are never going to be very useful this way since there is so much added latency with all the serializing and parsing... So how would you want me to implement these ? Mar 17 17:22:13 I guess I guess Mar 17 17:22:37 chandy: I suspect a library of tutorials could be fleshed out enough to make a complete project.... Mar 17 17:23:04 edge detection, object detection, face detection, etc. Mar 17 17:23:20 ok Mar 17 17:23:33 it would have to be enough beyond what is already there, but I'd think enough little tutorials would make a full project. Mar 17 17:23:45 alexanderhiam: does that seem right, or not enough fundamental value? Mar 17 17:23:49 * jkridner loves tutorials. Mar 17 17:24:05 tutorials = documentation and RUNNING code. Mar 17 17:24:12 == coding project Mar 17 17:24:54 So what you think ,Shoul I focus a bunch of tutorials or the main coding part? Mar 17 17:25:15 kiran4399: that's exactly right, you don't want to hand off a simple digitalRead/digitalWrite to the PRU alone. yes, it is useful, but not useful enough to be the only mechanism. Mar 17 17:25:36 kiran4399: there needs to be some way to build up more decision making on the PRU. Mar 17 17:26:17 chandy: what else could be done to elaborate on the main coding? Mar 17 17:27:13 chandy: I'd expect you to be done with adding OpenCV and RPC in just a couple weeks, then a couple more weeks for test and basic documentation. Mar 17 17:27:50 Now my plan is as follows 1.interface the camera in OpenCV and and perform maximum 3 tutorials Mar 17 17:28:15 and second one is any good ROS tutorial Mar 17 17:28:29 ok Mar 17 17:29:27 Let me start working on with adding OpenCV and RPC Mar 17 17:30:02 I hope that it will be done in couple of weeks Mar 17 17:31:29 You talking about Beaglebone RPC right ? Mar 17 17:31:40 * jkridner had to take a phone call Mar 17 17:32:06 ok Mar 17 17:33:24 I was digging the documentation of BBB and found this http://elinux.org/BeagleBoard/GSoC/2010_Projects/C6Run Mar 17 17:34:21 but I have no idea about this project . Mar 17 17:34:30 so need some help Mar 17 17:35:57 jkridner: well, for such basic commands like digitalWrite and digitalRead... which are just a straight forward implementations of SET DIO and GET DIO... I think little latency due to the parsing and serialization is inevitable... do you think latency here would really matter a lot ? Mar 17 17:43:46 jkridner, chandy: it that could include some instructions on the hardware side for the ROS examples and some demo videos as well Mar 17 17:45:26 in hardware level i have ultrasonic or laser in my mind Mar 17 17:46:26 shubhangi: saw that. Some tips: make sure to always adhere to the coding style of the repo you're contributing to, meaning comments, indentation, etc. Also, be more specific with your commit messages, when you have three commits with the same message it's not clear what each one actually changed Mar 17 17:47:00 and one more thing ,there was a project in ROS names ardupilot Mar 17 17:47:49 is it possible to adopting such a project Mar 17 17:48:20 alexanderhiam: Thanks. Will clean up code and adhere to repo style. sorry about the commits. still learning git. will get things done correctly asap :) Mar 17 17:50:09 kiran4399: think about actual use cases. For instance, if you want to read the input on one pin and output that value on another pin, you don't want a loop in userspace that calling digitalRead/digitalWrite on the PRU Mar 17 17:51:10 alexanderhiam: that's right !! Mar 17 17:51:12 in hardware level i have ultrasonic or laser in my mind Mar 17 17:51:21 kiran4399: with pruspeak you can write a script that does that digitalRead/digitalWrite, the logic and the looping in the pru Mar 17 17:52:55 alexanderhiam:and one more thing ,there was a project in ROS names ardupilot,is it possible to adopting such a project? Mar 17 17:53:13 chandy: already done, called beaglepilot Mar 17 17:53:36 there's been beaglepilot projects in the past Mar 17 17:54:04 chandy: seems a technical challenge to get video images showing up in the web browser. mjpg-streamer used by openrov doesn't seem to be an option in this case. Mar 17 17:54:09 alexanderhiam: yeah so that's not fair right? Mar 17 17:54:49 chandy: they might have needs to continue improving it Mar 17 17:55:30 jkridner: ok so any alternative ? Mar 17 17:55:47 alexanderhiam: so instead of looping in the userspace.. you loop in the pru .. right ? Mar 17 17:55:56 alternative? Mar 17 17:56:08 kiran4399: right Mar 17 17:57:30 jkridner: I mean is there any another method to get video images showing up webbrowser Mar 17 17:57:55 kiran4399: common application: https://www.npmjs.com/package/pid-controller Mar 17 17:58:07 chandy: yes. Mar 17 17:58:50 jkridner:Could you eloborate Mar 17 17:59:08 chandy: can you google? Mar 17 17:59:25 ok got it :D Mar 17 17:59:28 http://www.webrtc.org/ Mar 17 18:00:25 jkridner; so my challenges will be adding OpenCV and RPC and to get video images showing up webbrowser? Mar 17 18:00:42 * jkridner looks at https://rtc.io/ Mar 17 18:01:18 jkridner: so my challenges will be adding OpenCV and RPC and to get video images showing up webbrowser? Mar 17 18:01:29 I think getting the images to show up in the browser with RPC and a base set of tutorials starts to make it enough for an entire summer project. Mar 17 18:01:56 ok Mar 17 18:02:20 jkridner:So no need of ROS right ? Mar 17 18:02:49 well, if doing ROS, I think that is a different project and should be combined with BeaglePilot. Mar 17 18:03:25 ok fine Mar 17 18:03:34 * jkridner would check with openrov developers if they've looked at https://rtc.io/ Mar 17 18:04:06 jkridner: we talk about mentoring... mainly for me what time commitments are needed, and for you what skillsets of mine would be appropriate for the proposals/ideas Mar 17 18:04:12 we +could* Mar 17 18:04:16 jkridner: I will be sticking on OPEV CV Mar 17 18:04:52 calculus: yeah, I was trying to convince you to be a backup mentor at least. Mar 17 18:05:00 calculus: your familiarity with robotics is very useful. Mar 17 18:05:07 calculus: and general Linux knowledge. Mar 17 18:05:09 kiran4399: cright. That's the point of the scripts. So if you add another layer over pruspeak that doesn't let you run scripts then it's limiting functionality Mar 17 18:06:11 calculus knows more than he lets on... Mar 17 18:06:18 calculus: time input for mentors varies a bit. alexanderhiam might have the best idea. if you pair up with other mentors to share the load, just a few hours a week could make a huge difference. panto has made huge contributions in just a couple hours a week. Mar 17 18:06:53 jkridner: a few hours per week is doable Mar 17 18:07:46 nerdboy: perhaps... I don't want to give away all my secrets Mar 17 18:07:59 but I am starting to work on node for work Mar 17 18:08:10 :-D Mar 17 18:08:22 * jkridner watches ds2 squirm. Mar 17 18:09:00 calculus: julianduque and nodebotanist are node/js people. :-D Mar 17 18:09:33 Yep! Mar 17 18:11:29 panto had been really helpful last year even though we wished he could have engaged more Mar 17 18:13:37 cool, I am just getting started and planning on using express for rest things, and d3 for some charting/graphing Mar 17 18:18:31 alexanderhiam: I don't understand one thing.. in the example which you gave.. once you digitalread the pin... without looping over in the userspace to know what the user has entered after digitalread ... you would'nt know what to do right ?? so why would'nt we not wish to loop in the userspace? Mar 17 18:18:55 atleast for this specific reason ? Mar 17 18:23:42 shubhangi: Do you feel that "microsecond" delay would be hindered by latency between ARM-PRU communication? Mar 17 18:24:09 Which could itself be of the order of microseconds Mar 17 18:29:51 Abhishek_: Not that, the current WAIT command of pruspeak takes values for milis. Mar 17 18:30:49 Abhishek_: adding uSec support would would make the pru wait for tht much uSec using its internal timer, no interaction with linux Mar 17 18:31:37 kiran4399: if you want user input in a pruspeak script you can set a variable from userspace through the pruspeak interpreter and the script can read the value of that variable Mar 17 18:31:38 its an open issue on pruspeak git Mar 17 18:36:53 just to toss a comment into the mix - opencv is not the easiest way to get to algs...there other ones that are much much easier to use Mar 17 18:37:31 been playing with those recently. compilation is next to trivial. and not many, dependencies either Mar 17 18:42:58 Abhishek_: https://gist.github.com/shubhi1407/aad5d39f3f8199584004 Mar 17 18:45:06 *DIO[0] in line 4 and 11 Mar 17 19:12:29 praveendath92: ping Mar 17 19:12:36 pong Mar 17 19:12:49 do you have a beaglebone black near you ? Mar 17 19:13:11 I have one with me. Mar 17 19:13:19 Should be around. Why? Mar 17 19:13:35 can you put it in usb booting mode and hook it up to a linux pc then lsusb -v Mar 17 19:13:43 i need all the descriptors and the interfaces for the ROM Mar 17 19:14:02 lsusb -vvv or how was the command to print with full detail Mar 17 19:15:12 Okay Mar 17 19:15:38 Let me look for that Mar 17 19:15:54 thx a lot! Mar 17 19:20:40 vvu: This will take a while Mar 17 19:20:45 Haven't used it in months Mar 17 19:21:39 lsusb ? Mar 17 19:21:41 or bbb ? Mar 17 19:21:50 It's on. Mar 17 19:22:13 hmm Mar 17 19:22:45