**** BEGIN LOGGING AT Tue Apr 29 02:59:58 2014 Apr 29 07:18:24 morning Apr 29 07:27:12 greetings everyone! Apr 29 08:44:49 hello vmayoral ! Apr 29 12:25:08 DiegoTc: can we talk here rather than on G+? Apr 29 12:25:12 or GTalk. Apr 29 12:25:29 yes, better for me. Do I do copy paste? Apr 29 12:25:29 this is logged and visible by other the other mentors (hopefully). Apr 29 12:25:42 1. I'm planning to leave the main page like the one of the template: diegotc.github.io/bone101/Support/GSOC/template.html except for the pink colors. Change CSS Apr 29 12:25:43 you said: 1. I'm planning to leave the main page like the one of the template: diegotc.github.io/bone101/Support/GSOC/template.html except for the pink colors. Change CSS Apr 29 12:26:13 Why didn't you change the shape of the entries to be fixed-sized cards? Apr 29 12:27:30 there's an issue with that. I have a 13 inches laptop, it looks fine, but right now I'm using a 17 inches monitor, horrible, test it on a 15 inches it doesn't looks so good Apr 29 12:27:53 do you have screen captures to show the issue? Apr 29 12:29:06 not right now, I will show them, give me some minutes. But I remember VoltiVision telling that it will be excellente if you could see them on a tablet Apr 29 12:29:21 So that's why I thought I will be better responsive Apr 29 12:31:27 jkridner : ping! Apr 29 12:31:55 2. I'm assuming that for running bone101 and this detecting the board you have to have the project running in node? I'm correct? Apr 29 12:31:56 karki: please try to go straight to questions rather than just pinging. I toggle between a lot of stuff and it wastes a lot of time. Apr 29 12:32:38 DiegoTc: working on a tablet should indeed be a high priority. Apr 29 12:33:13 It need to be responsive them if we want a good looking Apr 29 12:33:30 DiegoTc: what do you mean my 'project' here? you would have to have the bonescript server running to interact with the board from the web page. Apr 29 12:33:48 jkridner : I had a problem understanding how the PRU signals ARM. How does writing r31 with 19+16 signal map to PRU_EVTOUT_0 on the ARM side. Apr 29 12:34:32 karki: you are interacting with panto as well, right? Apr 29 12:34:50 PRU_EVTOUT_0 should be received when we signal out to channel 2. yes, I'am! Apr 29 12:36:07 jkridner: yes, I was trying to validate that. You say it on nice and clear words. 3. Have you talk with David Scheltema. I show him the logs about your idea, but not sure if he gets the feeling Apr 29 12:36:12 karki: is this an example of in code where the PRU signals the ARM? https://github.com/jadonk/pruduino/blob/master/src/ws28xx-pru0.c#L654 Apr 29 12:37:46 jkridner : here https://github.com/beagleboard/am335x_pru_package/blob/master/pru_sw/example_apps/PRU_memAccessPRUDataRam/PRU_memAccessPRUDataRam.p#L100 Apr 29 12:38:20 karki: https://github.com/jadonk/pruduino/blob/master/src/include/pru_defs.h seems to have the constant defs. Apr 29 12:38:57 karki: you want to move quickly beyond that UIO approach and use the remoteproc approach, I believe. panto should be able to confirm. Apr 29 12:39:40 DiegoTc: I will try to spend some more time chatting with him. Apr 29 12:39:48 wish he was on the irc more. Apr 29 12:40:04 DiegoTc: I find him on GTalk a lot. Apr 29 12:40:13 4. jkridner: Do I go responsive or no? For me its better responsive. Apr 29 12:40:26 DiegoTc: I don't understand that question. Apr 29 12:40:54 the lightbox, going to your actual question why not fixed sized on the lightboxes Apr 29 12:42:37 jkridner : I was discussing with panto about remote proc whole of yesterday. But this return value of 19+16 to signal the ARM cpu that it is done is present even in the remote proc code. So I was curious as to why that number! Apr 29 12:43:24 jkridner : My main focus is remote proc, I'm just using UIO material for some general PRU stuff cause it's more easily available. Apr 29 12:43:42 karki: https://github.com/jadonk/pruduino/blob/master/src/include/pru_defs.h#L535 has the operation to trigger the interrupt. Apr 29 12:44:24 thanks ! Apr 29 12:45:18 karki: are you looking through the operation of the mranostay/panto code? Apr 29 12:45:31 the examples from TI are a bit behind in their style. Apr 29 12:45:43 * jkridner has to drop off for a while. Apr 29 12:45:51 * jkridner is currently on a plane. Apr 29 12:46:06 jkridner : yes. Apr 29 12:46:14 I have had a good look at both Apr 29 13:03:35 mranostay : Hey, had a bit of a query regarding your code here : https://github.com/jadonk/pruduino/blob/master/src/include/pru_defs.h#L578 Apr 29 13:27:58 jkridner: I just updated: diegotc.github.io/bone101/Support/GSOC/template.html fixed sized cards Apr 29 13:28:10 I'm going to send screenshots in some minutes Apr 29 13:29:04 they aren't the right layout in my view Apr 29 13:29:50 oh, you are talking about one-level-deep, not the overview. Apr 29 13:30:20 the overview has portrait-shaped cards, rather than landscape... and they don't have a BeagleBone-shaped card outline. Apr 29 13:34:25 hi dschelt Apr 29 13:34:32 hello jkridner Apr 29 13:34:32 DiegoTc: what I see on the overview page: http://imagebin.org/308137 Apr 29 13:35:16 dschelt: DiegoTc was wondering if I'd really pitched the card concept to you and the overall goals for the tutorial-display/creation environment. Apr 29 13:35:20 i noticed gist had some color issues with one of the cards, DiegoTc. Apr 29 13:35:38 jkridner: no, I don't think you'd told me your idea Apr 29 13:35:39 dschelt: +1 on that. the JS highlight CSS is wrong. Apr 29 13:36:14 have you seen http://beagleboard.github.io/bone101/Support/bone101/ ? Apr 29 13:36:25 Note how it has the "fork me on github" link? Apr 29 13:36:33 lots of projects have that... Apr 29 13:37:11 jkridner: that question goes to me? Apr 29 13:37:23 the idea is that on these tutorials, there will be a similar link where you can fork and extend the tutorials yourself, potentially providing updates/improvements/extensions to the existing library of tutorials. Apr 29 13:37:30 DiegoTc: no, to dschelt Apr 29 13:38:38 jkridner: that makes sense Apr 29 13:38:56 so, each of the BoneScript tutorials will exist as tiny git repos and the main bone101 git repo will refer to individual tutorials as an aggregation point, but people are able to use the tool to create their own websites with these interactive tutorials. Apr 29 13:39:15 nice Apr 29 13:39:19 the tutorials would typically include code that can be run from directly within the page. Apr 29 13:40:10 The listings of tutorials would show up a bit like learn.adafruit.com/category/beaglebone, but each tutorial would be shaped like a landscape-oriented BeagleBone and be made up of a deck-of-cards. Apr 29 13:40:31 each card is a step and are also BeagleBone-shaped, but blown up to take up more of the screen. Apr 29 13:40:49 so, you'd click on one of the tutorials and it would zoom-in to the first step. Apr 29 13:41:17 it should be easy to place collections of these "decks" on any website. Apr 29 13:45:21 jkridner: I was wondering about the sizing of the cards. DiegoTc has the card taking up the entire page, but I wasn't sure if it was because I'm on a laptop or if it were by design and dynamic to screen size. Apr 29 13:45:23 dschelt: does that make sense? Apr 29 13:45:28 jkridner: yup Apr 29 13:45:46 and faster than reading old irc logs :) Apr 29 13:45:49 My preference is to set it to a fixed number of pixels that is nice for a typical tablet or small laptop. Apr 29 13:46:31 I say that, because I'd like for it to be very reliable/predictable results in going to a printed page or otherwise having nice graphics. Apr 29 13:47:09 much like mobile phone apps push you to support a set number of resolutions, I'd want us to do the same and make sure it looks good on iPhones, typical Android tablets, etc. Apr 29 13:47:14 ...and print. Apr 29 13:47:45 having it expand to full-screen seems like a bad idea to me. by fixing the resolution, we make sure that people authoring content always have it display the way they intended. Apr 29 13:50:21 yeah, imposing some constraints is a good idea Apr 29 13:50:48 jkrinder I misunderstood you. I thought you where talking about the cards sizes, no the overview.Talking about the beginning. Apr 29 13:51:05 I updated the cards overview layout http://diegotc.github.io/bone101/Support/GSOC/template.html Apr 29 13:51:39 they don't have the same size because, I assume we are giving the freedom to write the introduction to users Apr 29 13:51:48 not a specific amount or words Apr 29 14:09:13 panto : here in mranostay 's code https://github.com/mranostay/ws28xx-lighting-pru, once the overlay was uploaded, the PRU could be talked to like a serial device, right? which driver gave the PRU a serial interface? [or am I getting confused somewhere? ] Apr 29 14:09:50 the driver is the virtio-serial driver Apr 29 14:10:06 but, don't go into that just yet :) Apr 29 14:11:44 panto : okay :) Apr 29 14:12:21 but your answer freed up my mind! I thought I got utterly confused somewhere! Apr 29 14:12:59 that's to be expected Apr 29 14:25:37 morning Apr 29 14:27:27 end semesters finally over *yipeee* Apr 29 14:30:46 Abhishek_: passed well ? Apr 29 14:31:12 * vvu|Log still has 7 final exams :( Apr 29 14:31:24 hopefully. There was too much cramming involved, far more in the last two papers than in this entire semester :P Apr 29 14:32:12 vvu|Log: German's up for us next sem hopefully Apr 29 14:32:30 i am studying in germany but running from german as far as i can :)) Apr 29 14:32:39 lucky that my university is in english Apr 29 14:44:44 * Abhishek_ gave 6 final exams Apr 29 14:56:30 * karki still has 8 final exams to give Apr 29 14:57:11 Praveen must be finished with his exams very soon now Apr 29 14:58:16 lucky IITians ;) Apr 29 14:59:44 mine starts next week : 6th 8th 10th 13th and 15th of May Apr 29 15:00:08 no worries peeps, i start on the 20 and until 27th :) Apr 29 15:00:33 Ours was : 21st 22nd 23rd 28th 29th Apr 29 15:05:42 I think I will barely make it to tomorrow's meeting, probably from the airport Apr 29 15:28:04 hi DiegoTc, sorry to run off. Today is really busy! I'm moving to California tomorrow. Did Jason get back to you regarding the card sizes? Apr 29 15:28:36 hi dschelt, he left. Ahh ok I understand that. Apr 29 15:28:56 ah, he was online in a plane, so maybe that's why he left Apr 29 15:29:09 No he left, but we were not in the same topic, I was talking about the lightboxes, the actual card size, and he was tallking to me about the overview Apr 29 15:29:20 that's why I was quite confuse. Apr 29 15:30:00 I just updated again, still mising the overview way according to Jason requirements: http://diegotc.github.io/bone101/Support/GSOC/template.html Apr 29 15:30:43 ah Apr 29 15:30:50 I thought he was talking with me about the card size Apr 29 15:31:28 he was later on with you Apr 29 15:31:33 ah Apr 29 15:31:44 what's the overview issue? Apr 29 15:32:27 until that momemnt, I knew I was misunderstanding the topics Apr 29 15:33:15 he wants this: the overview has portrait-shaped cards, rather than landscape... and they don't have a BeagleBone-shaped card outline. Apr 29 15:33:35 to have the zoom efect Apr 29 15:40:34 oh, yeah Apr 29 15:46:06 DiegoTc: check your gist syntax highlighting on the cards... still seems to be an issue. Apr 29 15:49:01 yes, I have that as a todo Apr 29 15:49:07 it has to be a css issue Apr 29 15:49:54 OK, wanted to make sure that didn't get lost in the conversation re: card size Apr 29 15:49:55 :) Apr 29 16:05:14 alexanderhiam: ping =) Apr 29 16:05:21 hey Apr 29 16:05:35 so what should the module be called? Apr 29 16:08:04 hmm... _sysfs_util maybe? Apr 29 16:08:49 starting names with an underbar in Python imply that thing shouldn't really be used directly by user code Apr 29 16:09:21 alexanderhiam, rseethamraju: Hello! is there anything I should review in prep for our meeting today? Apr 29 16:09:30 ok Apr 29 16:11:37 * karki wonders what alexanderhiam is upto with PyBBIO Apr 29 16:12:19 VoltVisionSteve: I don't think so, rseethamraju has been figuring out Python C extensions by converting the code that opens/reads/write kernel driver files to C to speed it up a bit Apr 29 16:14:38 alexanderhiam, cool. I will be here until the meeting, so ping me as necessary. Talk soon! Apr 29 16:15:17 alexanderhiam, have you done much with audio on the BBB? I am now teaching myself about ALSA and maybe JACK... Apr 29 16:16:50 not on the bbb, but I've had fun with jack and alsa before Apr 29 16:21:36 hi all from ELC Apr 29 16:22:42 hello mranostay ! Wish I was there! Apr 29 16:26:36 ELC? Apr 29 16:35:28 panto : why is it that the PRU interrupts are polling based and not truly asynchronous? Any design specific reason? [It's been bugging me for a while] Apr 29 16:38:05 the whole reason of the PRU existence is real-time predictability at the ns range Apr 29 16:38:27 interrupts by their definition throw in jitter Apr 29 16:38:38 so no interrupts Apr 29 16:42:32 panto : but doesn't polling reduce performance in quite a few applications? I'd end up wasting a decent bit of time polling! Apr 29 16:43:04 oh Apr 29 16:43:15 okay, never mind. I get it Apr 29 16:44:00 but we are sacrificing performance for no jitter. Apr 29 16:44:15 real time is not about performance Apr 29 16:44:20 if you want performance use the arm Apr 29 16:44:27 real time is about predictability Apr 29 16:49:12 Got it! Thanks panto! Apr 29 17:21:15 jkridner, VoltVisionSteve I think we should define if the card has to have 100% width or we should leave it a fixed size. Right now the template http://diegotc.github.io/bone101/Support/GSOC/template.html has a static size. Apr 29 17:21:34 jkridner I'm working for having the cards on ladscape position. (Talking about the overview). I have some small doubts that this will be completely align. Images will be the issue. But I will try and show you what I have Apr 29 18:02:14 panto : I was reading the PRU INTC section from the TRM, I understood how the PRU signals the ARM, but couldn't figure out how the ARM is supposed to send PRU a signal. None of the 64 system interrupts listed seemed to have ARM MPU as a interrupt source! Apr 29 18:03:14 where as PRU 8 interrupts were listed in the 128 interrupts the ARM INTC handles. Apr 29 18:03:44 so those 8 are for PRU to ARM, but what about ARM to PRU Apr 29 18:04:02 use the source :) Apr 29 18:05:49 you write the event # in the pintc reg Apr 29 18:05:55 look at pru_downcall Apr 29 18:07:08 * karki runs of to read the source Apr 29 18:11:13 panto : got it :) It's just that I have been reading 4 different pieces of source code and 4 different docs; so I'm in quite a mess! But yeah I found out how to do it. Thanks :D Apr 29 18:14:08 * karki thanks panto and crashes into the bed after a looong day Apr 29 18:15:58 karki: btw virtio serial is slow so don't expect to transfer data with it Apr 29 18:16:08 that is why i have a faux spidev interface Apr 29 18:16:52 panto: ping Apr 29 18:17:21 mranostay : but your firmware on the PRU side still is as per virtio, right? Apr 29 18:18:04 karki: yeah slow interface and mostly for configuring Apr 29 18:18:16 aka setting string size for the LEDS Apr 29 18:19:27 mranostay : which part of the code (on the PRU) have you handled the SPI specific part? (or does it not matter.....) Apr 29 18:19:43 karki: look in the upcalls Apr 29 18:20:07 it is pretty clear if you look at the commits to the pru code Apr 29 18:23:16 oh! okay. I didn't look too close into the as yet. I'll be doing it next week! This week I was busy getting acquainted to remote proc and the over all systems design. So I have just looked at snippets of your code till now. Apr 29 18:32:48 mranostay: Hello, and all the best for ELC :) Apr 29 18:37:36 hi jkridner, are you there at ELC too? Apr 29 18:37:47 nope Apr 29 18:48:29 jkridner: i thought it seemed nodeless in here :) Apr 29 18:54:33 mranostay: biot 's there too? Apr 29 18:56:46 yeah Apr 29 18:57:14 When is his presentation on sigrok? **** ENDING LOGGING AT Wed Apr 30 02:59:58 2014