**** BEGIN LOGGING AT Fri Mar 27 02:59:59 2015 Mar 27 04:14:39 anyone here? Mar 27 04:15:17 I have a rather benign question about the Proposal form Mar 27 04:19:19 ask away Mar 27 04:19:31 Should I keep the formatting in the proposal content, or can I change the formatting (keeping all the required info, and in exactly the same order, just a bit more readable) Mar 27 04:19:59 readable is good Mar 27 04:20:03 alexanderhiam: I've written it up in vim, and its basicly a text dump, I'd like to copy it and just bold the question parts , probably a bit more readable Mar 27 04:20:33 sure, but you better hurry if you want to get feedback before the deadline! Mar 27 04:20:41 alexanderhiam: alright, I'll do that. Mar 27 04:20:53 ill be done with copying it in about 15 mins Mar 27 04:21:17 alexanderhiam: would be very grateful if you can review it and give some feedback Mar 27 05:03:32 alexanderhiam: Hey, are you here per chance? Mar 27 05:04:25 sort of Mar 27 05:04:31 alexanderhiam: hah, ok Mar 27 05:05:01 alexanderhiam: want me to link the proposal ? Mar 27 05:05:51 I see it Mar 27 05:05:52 alexanderhiam: its a bit longish, but I'm mostly having trouble with scoping, eg. I don't havea feel how much everything will take in the weeks timetable Mar 27 05:07:33 alexanderhiam: alright, thanks ! made it public though, so that's why I redacted the email. you have my email in melange, right? Mar 27 05:08:05 yeah I see it on melange Mar 27 05:48:09 neemo: You should consider getting one of these: https://www.indiegogo.com/projects/pixhawk-fire-cape Mar 27 05:49:05 has 2 magnetometers on board Mar 27 06:04:29 Abhishek_: I was looking at it the other day, seems amazing Mar 27 06:05:35 Abhishek_: and yeah, it's shipping ready (April 2015!!) which makes it a better option than the sensorCape I found while doing research Mar 27 06:06:25 Abhishek_: come to think of it, I should've included the fireCape instead of the sensorCape, but my sleepy mind couldnt make the connection Mar 27 06:07:42 Abhishek_: what's the difference between LSM9DS0 compared to MPU9250? Mar 27 06:08:48 that's something you should be able to figure out :) Mar 27 06:09:19 Abhishek_: looking atm, but google is not in the mood :(, my VPN is crapping out again I'd wager Mar 27 06:09:48 Abhishek_: Will report back with findings! (though I gtg soon, will be back in a couple of hours) Mar 27 06:10:13 alexanderhiam: Did you have a chance to peek at the proposal? Any recommendations on improvements Mar 27 06:14:05 Abhishek_: BTW, what the project for the beaglePilot expansion is the Submarine right? Mar 27 06:15:18 yeah, porting OpenROV Mar 27 06:15:36 Abhishek_: Anybody have any more info on that one, looks really cool, but sadly Im uninitiated in the ways of nodejs Mar 27 06:16:04 Abhishek_: yeah, looking at the github page right now Mar 27 06:17:56 Abhishek_: Any plans on doing something more with ROS and beaglepilot Mar 27 06:18:08 Abhishek_: underwater, or not :P Mar 27 06:19:07 ask vmayoral when he's around Mar 27 06:20:02 will do Mar 27 07:12:41 ds2: thank you for your feed back. Mar 27 07:13:00 can you please help me making a reasonable time line ? Mar 27 07:14:47 you need to seed it more Mar 27 07:16:28 ds2: how does the libprussdrv works ? Mar 27 07:17:05 shouldn't you have researched it? :D Mar 27 07:27:24 so the libprussdrv actually reads and writes to the shared memory, they have some flags there for th host processor and th PRUs as well, indicating start and end of the program . Mar 27 07:27:37 ds2: is it correct ? Mar 27 07:31:42 ds2: can you please help me understanding the statement "Currently the widely used libprussdrv supports exporting of 'interrupts' via sysfs, but no clean way of data transfer." Mar 27 07:37:45 if you look at the examples, the current driver (UIO version) allows waiting on an interrupt Mar 27 07:37:57 data xfer can be done with the preallocated buffer Mar 27 07:38:04 but this is really a userland limitation then anything else Mar 27 07:42:33 ds2: so that "waiting for the interrupt" is what we do not like ? we want the PRUs to listen only when there is a data transfer, and this is what we want to achive using remot procedure calls ? right ? Mar 27 07:43:36 not quite Mar 27 07:43:50 catch me in about 14hrs Mar 27 07:43:55 need to sleep now. sorry Mar 27 07:52:00 sidbh: So can you tell me how to avoid those conflicts especially for some dedicated modules like ADC and PWM ? Mar 27 07:56:41 alexanderhiam: Did you see my revised proposal ? Mar 27 07:56:58 vmayoral: I have sent a reply on the topic. Kindly have a look. Mar 27 07:57:15 rohith: morning Mar 27 07:57:20 sure Mar 27 07:58:12 kiran4399: According to me there is no sensible way of doing so, without the user of the library being aware of what he is doing Mar 27 07:58:38 Abhishek_: I did not understand that ... Mar 27 07:58:50 can you be more specific ? Mar 27 07:59:53 The point is, the user has to be very clear while using this library Mar 27 08:00:46 vmayoral: Also, I need your feedback on the impact of the project on the community to add it to my application. Mar 27 08:01:58 any kind of ad hoc mechanism to enforce conflict resolution will make matters worse Mar 27 08:02:54 Abhishek_: Well, the functions are not so complicated as such.. and also I will provide enough documentation about it... But can you give me an example of any conflict you think will happen by this method ? Mar 27 08:05:45 you configure the ADC under linux, and then start using pruduino's analog input Mar 27 08:06:01 which would mess up the register settings which Linux did Mar 27 08:06:55 rohith: regarding community impact, i'd highlight the size of the DIYDrones community (65.000 users) and the growth both in size and interest on the drones field Mar 27 08:07:35 having a strong software architecture for underwater vehicles is a must if the field wants to advance Mar 27 08:07:45 specially in the autonomous path Mar 27 08:08:37 i'd point out the capabilities that APM provides such as well defined logging system, ground control stations, autonomous mission planning, sensor drivers, ROS integration, ... Mar 27 08:11:04 Abhishek_: No, I will not configure ADC under Linux.. I am configuring with the starterware.. But in any case.. conflicts might come as you said.. I am thinking about that..Do you have anything in mind .. how to avoid the conflicts b/w ARM and PRU ? Mar 27 08:11:16 vmayoral: Thanks! Feedback from both the student and the mentor was required on the application form. I shall add the above as the mentor feedback :) Mar 27 08:11:34 sure Mar 27 08:11:56 rohith: let us know if you need anything else Mar 27 08:13:35 vmayoral: Will definitely do so. Thanks again! Mar 27 08:26:07 Abhishek_: How about saving the present configuration of the registers made by the ARM, changing and using those from PRU, after it is done revert the configuration to the default configuration ? Mar 27 08:28:35 alexanerhiam, sidbh: ping :D Mar 27 08:28:41 alexanderhiam, sidbh: ping :D Mar 27 08:32:50 nodebotanist: ping Mar 27 08:42:34 jkridner: urgently need a small favor, wont take more than 5 minutes Mar 27 08:43:09 jkridner: I would like to take your answer for the 'how will it impact the community?' question Mar 27 08:44:13 webglider: you should try again after 4-5 hrs, it's night time there Mar 27 08:44:47 Abhishek_: oops i forgot about that, thanks for the info Mar 27 08:45:44 Abhishek_: do you know about the NW.js app idea? Mar 27 08:47:50 Abhishek_: How about saving the present configuration of the registers made by the ARM, and reverting the configuration to the default configuration once the process by the pru is done ? Mar 27 08:49:58 kiran4399: it have to be done everytime I read something from the ADC, don't you think it will impact the rate of data reading? Mar 27 08:50:34 webglider: How's your experience with UI/UX design? Mar 27 08:51:12 Abhishek_: Hmm. Yes. you are right.. Mar 27 08:51:24 Abhishek_: Could you please answer the question for my proposal "How this project will impact the community ?" Mar 27 08:52:15 Abhishek_: Nothing specific, but I am well versed with CSS and usage of front-end frameworks like JQuery UI. Mar 27 08:52:54 Abhishek_: most of the UI elements to be used in my project are already readily available in most frontend toolkits Mar 27 08:53:08 webglider: Considering the fact that you are building a beginner-oriented application, you should try and post some wireframes with your proposal as well Mar 27 08:54:00 not sure if there's enough time for that right now though Mar 27 08:54:47 can any1 help me with beaglepilot Mar 27 08:55:21 i have planned making swarm quadcopters that can communicate via wifi Mar 27 08:55:33 can any1 suggest me sumthing for that ? Mar 27 08:55:38 * Abhishek_ posted a wireframe with his proposal last year at the last moment Mar 27 08:56:35 Abhishek_: thanks for the suggestion, I'll try working on it Mar 27 08:57:17 webglider: My question is on the design of the whole of the UX overall, more on the aesthetic part of it, not the implementation Mar 27 08:58:03 If you'd want to use it when you are presented with your design Mar 27 09:00:12 plese help me with the issue Mar 27 09:03:19 Abhishek_: well I was planning to keep it as simple a possible, inspired from the traditional desktop setup wizards Mar 27 09:04:09 Abhishek_: maybe I should draw up a prototype design Mar 27 09:06:04 Abhishek_: could I take your answer for the 'how will it benefit the community?' question? Mar 27 09:06:31 can any1 please help me with beagle pilot? Mar 27 09:08:01 dimonga: The project involves adding autopilot capabilities to underwater vehicles... Mar 27 09:09:03 i am planning to build swarm of auropilot with beagleboards which can communicate with each other Mar 27 09:09:15 @rohith Mar 27 09:10:04 dimonga: Nice! Make sure to fill out the application form before the deadline :) Mar 27 09:10:09 is this a good idea? Mar 27 09:10:17 and how can it help the community? Mar 27 09:23:54 shubhangi: Done. Mar 27 09:26:31 Abhishek_: Thanks alot :) Mar 27 09:27:18 The block diagram looks better than before Mar 27 09:27:51 But you seem to be using sysfs for everything Mar 27 09:28:41 vmayoral: I've submitted the first draft of the proposal. Kindly have a look and let me know if any changes are required. I have yet to fill up the timeline, which I shall do after consulting you Mar 27 09:28:50 shubhangi: Where is the device tree? Mar 27 09:29:55 Abhishek_: Isnt the device tree upto the user ? Mar 27 09:30:41 is the swarm of autopilot a good idea? Mar 27 09:31:05 shubhangi: Isn't the PRU driver supposed to take care of initialization of the PRU using settings in the device tree file? Mar 27 09:31:52 If you look into the device tree for BeagleLogic, there are 3 or 4 segments - one big segment handles the PRU and the remoteproc driver is supposed to take care of that Mar 27 09:31:58 Abhishek_: Yes. Forgot about that. Mar 27 09:32:01 sorry Mar 27 09:32:17 all the host-channel mapping and remaining data Mar 27 09:32:25 yes Mar 27 09:32:30 okay Mar 27 09:32:51 and apart from sysfs, a /dev interface ? Mar 27 09:33:09 or ioctl config files Mar 27 09:33:23 ioctl's are very hacky solutions Mar 27 09:34:28 an ABI will have to be established, and should be looking into kernel subsystems (like the FPGA subsystem, for example) for something similar Mar 27 09:34:51 this part is very crucial as it decides how the kernel driver makes it into the mainline Mar 27 09:42:47 Abhishek_: I'll have to read and learn more about ABI. heard it for the first time.Looked it up and it seems very crucial. Currently working on sysfs wrapper and python api demo. Will include it once i understand how it works. Mar 27 09:43:57 The sysfs wrapper is also a form of ABI Mar 27 09:44:49 * shubhangi so much to learn out there Mar 27 09:44:56 There are two ABIs that you have to deal with - (i) PRU -> kernel and (ii) Kernel -> userspace Mar 27 09:46:25 what would be the PRU->kernel ABI Mar 27 09:49:18 morning! Mar 27 09:49:40 morning vvu Mar 27 09:49:46 how is it going ? Mar 27 10:33:33 anyone available for a checkup on my proposal? Mar 27 11:04:42 alexanderhiam: ping Mar 27 11:04:50 sidbh_: ping Mar 27 11:06:27 vvu: hey, any progress with the sniffer? Mar 27 11:06:41 nop, had no luck Mar 27 11:07:19 Tartarus: spoke with somebody at elc about this issue and they have something on OSX...waiting for reply Mar 27 11:07:42 vvu: :( Mar 27 11:08:13 vvu: could I take your answer for 'how will it benefit the community?' question for the NW.js app Mar 27 11:08:29 vvu: running out of time Mar 27 11:16:36 webglider: i will supply it later Mar 27 11:18:21 vvu: thanks alot, could you just drop it as a comment on my proposal Mar 27 11:19:03 vvu: only a few more hours till the deadline.... Mar 27 12:08:11 Hello everybody Mar 27 12:22:29 ls Mar 27 12:23:04 well, that's not rxvt! Mar 27 12:26:38 . Mar 27 12:26:39 .. Mar 27 12:37:10 av500: can you please review my proposal ? Mar 27 12:39:26 jkridner : Could you review my proposal Mar 27 12:40:24 ds2 : Would you be free? Mar 27 12:43:04 alexanderhiam: ping Mar 27 12:44:13 jkridner: Are you free to share your valuable comments Mar 27 12:48:07 ds2: ping Mar 27 12:50:09 I would be grateful if any mentor could assist me in reviewing my proposal Mar 27 12:51:51 sidbh: ping Mar 27 12:52:19 It is under "Using BeagleBone PRUs to control CNC and 3D printer stepper motor Drivers" Mar 27 13:12:01 ds2: ping Mar 27 13:27:43 alexanderhiam: ping Mar 27 13:29:59 kiran4399: pong Mar 27 13:31:27 alexanderhiam: I have been thinking about the ADC conflicts b/w PRU and ARM.. can you suggest me a method for that ? Mar 27 13:33:51 alexanderhiam: One method I've thought is to let PRU use the ADC .. and after that revert back the original module's configuration.. But I think it slow down the ADC .. Mar 27 13:33:59 I've yet to think of a good method... Mar 27 13:34:57 I'm not convinced there is one, besides just disabling the ADC driver Mar 27 13:35:24 OK.. Mar 27 13:41:42 alexanderhiam: Sorry to bother you, but did you have a chance to take a look at my proposal Mar 27 13:41:54 alexanderhiam: Don't want to annoy, but would like some feedback so I can improve the proposal before the deadline if I've missed something Mar 27 13:42:27 neem0: I'm actually reading it right now, give me a few... Mar 27 13:42:53 alexanderhiam: will do. thanks a lot! Mar 27 13:49:40 alexanderhiam: After the pruspeak has done executing the script sent by the node... should the interpreter sent back a message to the node containing data (if there is any GET command) and an event to tell the node that the program has completed executing ?? Mar 27 13:54:51 neem0: just asked some questions on melange Mar 27 13:57:45 hi Mar 27 13:58:01 i want to communicate with a GSoc mentor Mar 27 13:59:32 Guest97018: ask away Mar 27 14:01:17 alexanderhiam: After the pruspeak has done executing the script sent by the node... should the interpreter sent back a message to the node containing data (if there is any GET command) and an event to tell the node that the program has completed executing ?? Mar 27 14:01:56 kiran4399: that would be good Mar 27 14:02:16 alexanderhiam: and for the pru intc ... instead of polling, Can I map the bit to the ARM_ISR... when the interrupt is triggered, the variable in userspace will read the data in from the FIFO buffer.. what do you think ? Mar 27 14:03:56 you certainly could. That would involve changes to the pruspeak kernel driver though, so you'd want to look into the potential side effects of that Mar 27 14:09:29 If the beaglepilot project is successfully completed, what will its impact be on the BeagleBoard.org community? Mar 27 14:12:54 alexanderhiam:please review my proposal on malange Mar 27 14:13:13 chandy: it's next on my list... Mar 27 14:13:43 thank you Mar 27 14:18:51 alexanderhiam: thanks, I'm reading them and writing answers Mar 27 14:20:29 Guest97018: vmayoral, ahcorde and imuguruza are the beaglepilot folks, they will know the most about it Mar 27 14:20:43 alexanderhiam: I modified a lot in my proposal.. can you just have a look at it ? Mar 27 14:20:50 will do Mar 27 14:21:09 alexanderhiam: I'll be with the answers in a bit Mar 27 14:21:33 av500: do we have a mentor meeting scheduled any time soon? Mar 27 14:22:39 alexanderhiam: thought it's quite hard for me to gauge the amount of preprocessing I could get from the PRU Mar 27 14:23:22 alexanderhiam: from what I understood the PRU is mostly of limits for any amount of mathematically intensive computations Mar 27 14:24:42 alexanderhiam: and one of the algorithms uses matrix multiplication with floats, which makes it a no go for the PRU Mar 27 14:24:43 Guest97018: something like this should do it: The project will enable growth in both size and interest in the field of drones in communities such as DIYDrones, which currently has 65,000+ users. Having a strong software architecture for underwater vehicles is a must in order to further improve the field, especially for autonomous navigation. The addition of the capabilities of the APM such as a well-defined lo Mar 27 14:24:44 gging system, groung control stations, autonomous mission planning, drivers for various sensors, ROS integration etc. to the vehicle will definitely improve functionality and make it easier to control underwater vehicles. Mar 27 14:25:14 neem0: have you considered using fixed point math in the PRU? Mar 27 14:25:29 alexanderhiam: te other ought to be a filter based on a stohastic model, which could be offloaded partially to the PRU Mar 27 14:25:52 alexanderhiam: can the PRU do floating point? I was under the impression that it can't Mar 27 14:26:34 alexanderhiam: am I mistaken? Mar 27 14:27:42 vmayoral: sorry to interupt, but whats on the roadmap for ROS integration into beaglepilot, and beaglesubmarine? Mar 27 14:29:30 neem0: well it can, but it's certainly better not to. That's why I suggested fixed point math Mar 27 14:29:30 neem0: haven't decided yet, we at Erle Robotics are prototyping parts of APM (BeaglePilot) as ROS nodes but it's still undecided which path we will follow since there're already available bridges ROS <-> APM that do the job properly Mar 27 14:30:22 alexanderhiam: hmmm, do you have any links on doing floating point on the PRU I can check up Mar 27 14:30:55 alexanderhiam: the PRU is my weak spot (and maybe a bit of overscoping), so I'm really not sure what can and can't be run on the PRU Mar 27 14:31:17 alexanderhiam: That's why I wrote the proposal in a general way where part of the plan is to test the limits of the PRU, before the coding period starts Mar 27 14:31:36 alexanderhiam: So that I know how much can be done on the PRU and how much will have to go on the ARM Mar 27 14:32:49 It doesn't have a floating point unit, so it'll implemented by the compiler and use up a lot of cycles. Do you know what I mean by fixed point math? Mar 27 14:34:06 alexanderhiam: If you think that the PRU can handle what's essentially 9x9 float matrix multiplicaton in real time on the incoming data, then everything can probably be offloaded into the PRUs Mar 27 14:34:39 jkridner: I have update my proposal.. can you please have a look at it ? Mar 27 14:34:55 alexanderhiam: yeah, I understand what you mean by fixed point math. Using a fixed representation of values from the sensors instead of floats Mar 27 14:35:36 alexanderhiam: and doing the math with ints just scaled to the appropriate decimal value Mar 27 14:35:54 and do you understand the implications of fixed vs floating point in terms of cpu cycles? Mar 27 14:36:36 neem0: real time isn't really the concern, you can always slow down to acheive real time behavior. The question is can you get a high enough sample rate Mar 27 14:38:02 if you mean latency, then yeah, that's also a concern Mar 27 14:38:23 alexanderhiam: understood Mar 27 14:39:53 alexanderhiam: though I have no experience in fixed point math, so I dont have an idea of the amount of cpu cycles recquired to do it compared to floating point with an FPU Mar 27 14:40:44 alexanderhiam: I undestand that it will be more demanding and use up more cycles Mar 27 14:40:51 well it's really int vs floating point math Mar 27 14:41:11 alexanderhiam: seems I have some more reading to do. Mar 27 14:41:49 can you gauge if fixed point math would be a viable option? I was counting on investigating that in the precoding phase anyway Mar 27 14:42:04 though should I put it now as an option in the proposal? Mar 27 14:42:07 yes, it'll be very important to understand how this stuff is done by the processor Mar 27 14:42:28 alexanderhiam: ye Mar 27 14:42:30 do you have any experience with assembly? Mar 27 14:43:00 alexanderhiam: yes, though mostly what I was thought at university Mar 27 14:44:03 alexanderhiam: so basic assembly for computer architecture courses and some derivatives, eg STL for automation Mar 27 14:44:09 the pru compiler can spit out a generated assembly code file. I would suggest writing some C code to do integer and floating point operations, compile it with the assembly output option and compare operations Mar 27 14:45:11 and google is your friend -> "integer math vs floating point math" Mar 27 14:46:32 alexanderhiam: yeah, looking at it atm Mar 27 14:47:50 alexanderhiam: btw, is there a chance the proposals will be unlocked for editing after the deadline? Mar 27 14:48:27 it can be unlocked if we find there's something missing that we need to have added Mar 27 14:49:26 alexanderhiam: Ok, thanks. I'll be responding to the questions and probably editing the proposal in soon Mar 27 14:49:52 alexanderhiam: we wanted to schedule it next WED Mar 27 14:49:59 in the same slot as previous ones Mar 27 14:50:11 ok Mar 27 14:50:14 alexanderhiam: if you have question, you can always ping me Mar 27 14:50:29 thanks Mar 27 14:50:54 it would be good to get #beagle-mentors active again Mar 27 14:55:02 chandy: just left some questions on your proposal Mar 27 14:55:55 alexanderhiam: yes, we can do that Mar 27 14:55:59 for the meeting we will Mar 27 14:58:28 alexanderhian:ok Mar 27 15:01:13 jkridner: urgently need a small favor, wont take more than 5 minutes Mar 27 15:01:18 jkridner: I would like to take your answer for the 'how will it impact the community?' question Mar 27 15:01:30 jkridner: running out of time really need this Mar 27 15:01:51 jkridner: could you please drop a comment on my proposal Mar 27 15:02:09 webglider: no luck getting feedback from people on #beagle or #sparkfun or anywhere on how your project could impact them? Mar 27 15:02:33 an answer from me won't be all that super beneficial for the proposal.... Mar 27 15:02:47 I'd just be like "yup, I said what I said".... Mar 27 15:02:50 not all that helpful. Mar 27 15:03:03 knowing that there are real people out there that would use your work is important. Mar 27 15:03:28 jkridner: yeah but it requires atleast one mentors answer Mar 27 15:05:02 webglider: answer can be from others in the community as well. Mar 27 15:05:06 not just mentors. Mar 27 15:05:49 "at least one of whom should be a BeagleBoard.org GSoC mentor" quote from template Mar 27 15:06:01 jkridner: ^ Mar 27 15:06:44 webglider: did you get any non-mentor quotes? Mar 27 15:07:52 jkridner: nope, actually I didn't pay much attention to this Mar 27 15:08:11 point was to get some bonding with the community talking about your project. Mar 27 15:08:12 jkridner: I should have probably asked on #beagle or the main mailing list Mar 27 15:08:25 yeah. Mar 27 15:09:48 getting an all-in-one package would make getting started for those with doubts to know they are working with a well-tested starting point, common to a large number of other community members Mar 27 15:09:49 jkridner: would you mind giving some extra time to complete this task? Mar 27 15:10:25 webglider: the proposal matters a lot for the mentors to look over, but having relationships built with mentors over the last few weeks will make the biggest difference. Mar 27 15:10:51 * jkridner knows many are frustrated by my traveling, etc. during this time. Mar 27 15:11:20 if proposals are compelling, they can be updated after the deadline. Mar 27 15:12:06 but, you should try to have as much done as you can. Mar 27 15:12:37 webglider: you might note a quote above, but I'd really encourage you to engage on #beagle, #sparkfun, elsewhere to talk about what you want to do and see what they feel about it. Mar 27 15:12:48 it'll help you know what people really want in the community. Mar 27 15:13:06 serving the needs of the community with your code is the primary goal after all. Mar 27 15:14:05 jkridner: yes I will do that, thanks for the motivation :) Mar 27 15:14:48 jkridner: Can you give a mentor quote for pruduino ? Mar 27 15:16:01 alexanderhiam, jkridner: Did you have a look at my updated proposal ? Mar 27 15:20:24 join #beagle Mar 27 15:20:36 ^sorry Mar 27 15:21:13 kiran4399: do you have any quotes from other members of the #beagle community? Mar 27 15:21:36 jkridner: no :-( Mar 27 15:22:52 kiran4399: #sparkfun also has lots of BeagleBone users. Mar 27 15:23:00 as does #i3detroit Mar 27 15:23:06 and #snappy Mar 27 15:24:14 jkridner: So should i ask somebody there ? Mar 27 15:26:32 jkridner: thanks for the boost Mar 27 15:27:37 kiran4399: yeah... #sparkfun might be a good place. or that #nodebots/discussion group on irc.glitter.im Mar 27 15:27:50 * jkridner isn't sure where noduino people hang out. Mar 27 15:28:23 nodebotanist, julianduque: thoughts on where to get some feedback from the community on the concept of pruduino? Mar 27 15:28:31 * jkridner is obviously in favor of it. :-D Mar 27 15:32:19 * jkridner is about to go afk again. :( Mar 27 15:32:43 * jkridner got bogged down yesterday reading about https://thrift.apache.org/ Mar 27 15:34:42 jkridner: I would take to rwaldron about that, also check with any python BBB folks. Mar 27 15:38:09 jkridner: Please have a look at my updated proposal. I included many things.. :) Mar 27 15:40:20 hi Mar 27 15:40:37 alexanderhiam: Please have a look at my updated proposal. I included many things.. :) Mar 27 15:41:23 alexanderhiam: thanks for the support ;) Mar 27 15:41:51 np Mar 27 15:42:55 is there a open list of proposals somewhere ? Mar 27 15:42:57 jkridner: what about the document support during the community bonding period ? Mar 27 15:50:58 alexanderhiam: I was going throught the libpruio library, In that.. ADC is initialized by simple register access... will there be a I/O conflict here ? https://github.com/outer-space-sounds/beaglebone-pruio/blob/master/library/src/pru0_main.c#L259 Mar 27 15:51:27 how about makeing some of the pru code for am437x ;-) Mar 27 15:52:12 kiran4399: that definitely won't be compatible with the ADC kernel driver Mar 27 15:53:35 alexanderhiam: can we disable the kernel driver temporarily ? Mar 27 15:54:20 alexanderhiam: i have gone through the review comments Mar 27 15:54:30 thanks for review Mar 27 15:54:33 kiran4399: it would be the same issue. If the driver initialized while PRU code was using it it would break the PRU code Mar 27 15:55:20 alexanderhiam: instead of PS3 camera the webcam of laptop itself will be more suitable right? Mar 27 15:56:08 also what i woould love to see is u-boot pru code loader and then the reconnect to the booted kernel Mar 27 15:57:07 alexanderhiam: Looks like .. we have to think a lot about it.. in my proposal I just mentioned this challenge and a solution for GPIO... shall I leave it like that.. meanwhile I come up with some good approach ? Mar 27 15:57:35 s/I come / I'll come Mar 27 15:57:49 vmayoral: Added the timeline to the proposal. Let me know if any changes are necessary... Mar 27 15:58:03 chandy: there's no guarantee of UVC support for laptop webcams. I would lean more towards something like the Logitech c270 as a primary target. Lots of people have it, it's relatively inexpensive and it's well supported on GNU/Linux Mar 27 16:00:03 * Abhishek_ has a c270 as well Mar 27 16:00:28 alexanderhiam:ok i thought you are suggesting for making this tutorial by using beaglebone alone Mar 27 16:02:03 chandy: no, you just want to make sure you're targeting a webcam that's readily available and proven working with the BeagleBone Mar 27 16:02:28 ok Mar 27 16:02:32 neemo: Best that you investigate by writing some PRU code yourself so that you know the limits Mar 27 16:03:13 Abhishek_: agreed, that was the plan all along Mar 27 16:04:28 Abhishek_: though I'm counting on doing that in the near future, but I don't think I'll be able to test it and edit the proposal before the deadline Mar 27 16:04:55 alexanderhiam : about mjpg streamer i dnt think there will be much delay ,it depends on the connectivity and hardware ,So i will be testing during GSOC Mar 27 16:04:56 taking feedback from external communities is not as easy as i thought ;) Mar 27 16:05:48 Abhishek_: The PRU testing is actually in my proposal, under Mar 27 16:06:28 Exploring the amount of preprocessing that can be done on the PRUs, and it's planned before the bonding period Mar 27 16:07:00 neem0: It's okay, you can update us later. If so required proposals can be made editable past deadline as well Mar 27 16:07:24 Abhishek_: yeah, I've figured so Mar 27 16:07:35 chandy: I've seen delays upwards of 30 seconds on the beaglebone, so it might be worth testing now if you have a webcam handy Mar 27 16:08:02 neem0: yeah, the sooner you can do that testing the better Mar 27 16:08:18 Abhishek_: Well, I'll do my best to rectify the questions rased in the proposal Mar 27 16:08:36 I'll do my last batch of reading now, so I can edit what I think is the proper idea in the proposal Mar 27 16:09:19 I'm counting on playing with the ARM code in the next few days. probably Sunday/Monday, after I finish up my coursera dues Mar 27 16:11:22 ARM was supposed to be Assembly (sleepyhead) Mar 27 16:12:35 and I'll hopefully have a chance to hook up some hardware in the next week or two. China is a magical land for all the DIY components you could possibly wish for :) Mar 27 16:13:33 > Mar 27 16:14:01 alexanderhiam: now dnt have proper working beaglebone,since 30 sec delay is huge one ,i must must test it Mar 27 16:24:56 alexanderhiam:how about RTMP to rectify delay Mar 27 16:26:12 chandy RTMP isn't open source Mar 27 16:26:33 or parts of it aren't Mar 27 16:26:36 afaik Mar 27 16:27:43 opps ok Mar 27 16:29:31 alexanderhiam : any way from the user experiance in the blog it is found that The delay is almost unnoticeable running at 12fps 640 x 480 Mar 27 16:30:00 i belive it would be good enough Mar 27 16:31:04 chandy: yeah, that would be fine Mar 27 16:31:53 you could always downsample and scale before streaming if you need to do the opencv processing at a higher frame rate or resolution Mar 27 16:33:20 alexanderhiam:ok i will include that in testing phase Mar 27 16:33:59 ok Mar 27 16:49:44 T -2 hours! Mar 27 16:53:07 2 hours? it's not midnight gmt? Mar 27 16:54:56 oh yeah, it is 2 hours Mar 27 16:55:47 hmm, anyone waiting on feedback to make changes?? Mar 27 16:56:13 alexanderhiam: if you get time please comment on my proposal as to how it will impact the community. Mar 27 16:56:36 i am finalizing my proposal right now Mar 27 16:56:48 shubhangi: have you gotten feedback from other beaglebone users? Mar 27 16:56:51 alexanderhiam: Did you see my proposal ? Mar 27 16:57:18 from abhishek and karki Mar 27 16:57:44 shubhangi: e.g. in #beagle Mar 27 16:57:56 or #i3detroit has some beaglebone users Mar 27 16:58:18 can they access the proposal on melange ? Mar 27 16:58:32 kiran4399: yeah, will read through soon Mar 27 16:58:53 hmm Mar 27 16:58:55 i have a google doc too Mar 27 16:59:20 alexanderhiam: is #i3detroit invite only chat ?? Mar 27 16:59:39 shubhangi: no, the idea is to tell potential users briefly what you're proposing and get some feedback Mar 27 16:59:48 geekswine: no Mar 27 16:59:55 alexanderhiam: okay Mar 27 17:05:04 the guys at #i3detroit don't seem to be very happy with beaglers invading their irc :P Mar 27 17:15:47 hi ... i really need a mentor Mar 27 17:16:10 is there mentors of GSoC?! Mar 27 17:16:35 there certainly are Mar 27 17:17:22 ok Mar 27 17:17:55 i want one of the mentors to answer the question following Mar 27 17:18:12 If BeaglePilot project is successfully completed, what will its impact be on the BeagleBoard.org community? Mar 27 17:19:49 Radwan: have you engaged with the mentors for that project? Have you engaged with the BeagleBone community about it? Mar 27 17:24:27 No Mar 27 17:25:08 Hi Mar 27 17:25:28 I do not know for sure what i should do ... but i think i'm in the right place Mar 27 17:25:31 One of the tasks for a user friendly app is programming th board with NW.js app Mar 27 17:25:46 This is to be done via system calls right? Mar 27 17:25:54 And this will depend on OS? Mar 27 17:26:03 I am confused about the best way to do it Mar 27 17:26:04 Hi, I was putting together an application for BeagleRT, I am almost done. I need some help with deciding the timelines. can anybody help out Mar 27 17:26:34 Radwan: sounds like you're cutting it a bit close, there's not much time to get input and feedback. vmayoral, ahcorde and imuguruza are the beaglepilot folks here Mar 27 17:27:20 k29: do you have something to look at? Mar 27 17:27:40 @alexanderhiam ? Mar 27 17:29:12 sgrace_wrk: can you be more specific .. I need feedback from a user in the community which can be kept in my proposal... Mar 27 17:29:14 mr_science: yeah i wrote the basic structure of the work with regards to the preempt Patches and xenomai Mar 27 17:29:24 i need to however concretize the timeline Mar 27 17:29:38 in a draft 0n melange already? Mar 27 17:29:42 prakharsingh95: I don't know much about nw.js, perhaps nodebotanist or julianduque could help? Mar 27 17:29:45 no Mar 27 17:29:47 no Mar 27 17:30:23 mr_science: the main aim of the project is to just test out the various options for RT Mar 27 17:30:50 wait i will put it on melange Mar 27 17:31:05 i'd say repeat the previous test cases/comparison with latest kernel Mar 27 17:31:43 alright Mar 27 17:31:50 i am putting it online Mar 27 17:31:52 pretty much repeat what was done the last time around, unless you find a hole you can fill Mar 27 17:32:00 mind having a look at it? Mar 27 17:32:07 sure Mar 27 17:32:11 :) Mar 27 17:32:20 give me a few minutes Mar 27 17:32:25 will ping you Mar 27 17:32:35 * mr_science in all day dev mtg... Mar 27 17:33:11 alexanderhiam: will the work on adding video streaming directly to clod9ide take much time ? Mar 27 17:34:57 chandy: not really, but I'm not sure where it would physically fit. You could also have a way to serve up a basic HTML page from another port that includes a video stream Mar 27 17:35:26 for the live demos it makes since to have it embedded right in that page though Mar 27 17:39:20 alexanderhiam: I forgot to ask you... are the python bindings implemented in the same way as the node.js ? Mar 27 17:40:30 alexanderhiam: I would like to add dedicated page for video stream ,That seems like more efficient Mar 27 17:41:55 alexanderhiam: i will be using cloud9ide for accessing the opencv library and code . Mar 27 17:42:07 kiran4399: they would certainly talk to pruspeak the same way. There would certainly be some differences based on the differences between Python and JS Mar 27 17:42:56 alexanderhiam :what you think? Mar 27 17:43:03 alexanderhiam: OK !! gotcha.. Mar 27 17:43:25 chandy: there shouldn't be anything that depends on cloud9 Mar 27 17:43:31 1hr 16min more Mar 27 17:43:44 alexanderhiam: can you please give your comment on how pruduino will impact the community ?? Mar 27 17:44:00 ok Mar 27 17:45:06 alexanderhiam : hi can you provide feedback on how orbital imaging cubesat will impact the community ? i have asked beagle group but haven't received any non-trolling response Mar 27 17:45:25 :p Mar 27 17:59:41 alexanderhiam: Also since there is only 4K instruction mem in PRU.. there is a limit to the source code which the interpreter can take right ?? Mar 27 17:59:58 absolutely Mar 27 18:02:45 alexanderhiam: Do you have any idea of how many lines of pruspeak code that would be ? Mar 27 18:05:57 55 minutes remaining !!!! Mar 27 18:06:16 each bytecode is 4bytes. some are 8bytes. you can calculate Mar 27 18:10:29 vmayoral: Is my proposal all right? Mar 27 18:19:17 mr_science: Hey steve Mar 27 18:19:26 i updated the application on melange Mar 27 18:19:38 can you check once please and give me feedback? Mar 27 18:19:47 jkridner: i have tried to answer your queries.. plss see it once. Mar 27 18:21:12 mr_science: i need to concretise the timlines and need help with the community feedback question Mar 27 18:21:47 for the project arduino compatible functions for starterware.. Mar 27 18:21:55 lemme look while i'm not paying attention... Mar 27 18:22:28 jkridner / _av500_ / ds2 feel free to look as well... Mar 27 18:23:28 k29: url? Mar 27 18:23:38 Hi Mar 27 18:23:39 alexanderhiam: ping for the proposal !!! Mar 27 18:23:59 I am submitting a proposal on an idea listed on the wiki Mar 27 18:24:15 mr_science: please help me with the community feedback question. https://docs.google.com/a/itbhu.ac.in/document/d/1OmQnWQ13XTvoSWPhJR5XT-VxP6fhwDHPNU51fH45Xg8/edit Mar 27 18:24:15 Do I need to include a mentor response for the secondlast question Mar 27 18:24:25 Doesn't seem relevant Mar 27 18:24:46 mr_science: just a few lines. if you have time Mar 27 18:28:12 note to students: for future reference, the point of the community feedback on the proposla is to show that you've been engaging with the community to develop your idea. It doesn't really have the same effect if you are putting people on the spot for it feedback right before the deadline Mar 27 18:28:23 k29: just break up the time like an outline, some time for running the tests on each of the three rt options, collecting/analyzing the data, writing up the results Mar 27 18:29:37 mr_science: alright Mar 27 18:29:42 like week wise? Mar 27 18:30:00 yeah, mostl Mar 27 18:30:24 k, you still need the link? i have a gdoc you may refer to as well Mar 27 18:30:27 ? Mar 27 18:30:43 jkridner: got some quality feedback from the i3detroit community Mar 27 18:30:45 kiran4399: speaking of code size, have you tried compiling starterware for the PRU yet? Mar 27 18:31:07 yes... its done long ago !!! Mar 27 18:31:08 mostly analysis probably but you'll need some on the front-end to setup each one, and some tie on the back-end for the writeup Mar 27 18:31:28 k29: got to melange page up Mar 27 18:31:33 jkridner: small suggestion: next time have students do this exercise a bit earlier, it can really help alot Mar 27 18:31:35 s/to/the/ Mar 27 18:31:46 alexanderhiam: yes... its done long ago !!! Mar 27 18:31:53 webglider: it's been in the proposal template the whole time ;) Mar 27 18:31:56 half an hour left.. Mar 27 18:32:48 alexanderhiam: yeah that's true, but it would help if mentors pushed students to do this Mar 27 18:33:02 30 minutes to go! Mar 27 18:33:34 alexanderhiam: until jkridner told me, my impression was that we just had to take feedback from mentors as certification..... Mar 27 18:34:07 I guess we could highlight it a bit more Mar 27 18:34:20 All the best everyone! Mar 27 18:34:21 alexanderhiam: https://github.com/kiran4399/starterware_PRU Mar 27 18:34:58 alexanderhiam: can I have your feedback on the pruduino ? Mar 27 18:35:14 jkridner / _av500_ can i "wish to mentor" more than one at this point? Mar 27 18:35:45 mr_science: yeah Mar 27 18:39:45 kiran4399: I'd say if it succeeded and was well documented it could be a good way to introduce people to the PRU Mar 27 18:40:11 altairpearl: looks pretty good now, think you can get any Mar 27 18:40:20 alexanderhiam: OK.. can I get some more feedback ?? :D Mar 27 18:40:23 other feedback? Mar 27 18:40:45 kiran4399: how much are you looking for?? Mar 27 18:41:10 alexanderhiam: 2 lines to keep it on my proposal.. Mar 27 18:41:32 mr_science: https://docs.google.com/document/d/1OnpoVZHuJe0DMtvDr9PN6rVyc1rEs5xgeyouz0fP8p8/edit?usp=sharing Mar 27 18:41:35 any available mentors like to look at altairpearl's proposal? particularly the PRU bits? Mar 27 18:42:14 k29: i made a comment on the community thing, so get to updating Mar 27 18:42:25 as in now... Mar 27 18:43:11 oh alright Mar 27 18:43:15 on it Mar 27 18:43:54 mr_science: have enough time to look at one more proposal? Mar 27 18:44:23 why not... Mar 27 18:44:43 mr_science: https://www.google-melange.com/gsoc/proposal/review/student/google/gsoc2015/niko_visnjic/5629499534213120?validated=True Mar 27 18:44:47 mr_science: thanks Mar 27 18:44:55 altairpearl: a well optimized orbital positioning system with friendly user space APIs would make the BeagleBone a very approachable platform for educational and hobbyist cubesat development Mar 27 18:44:59 anyone to comment on proposal for Demo android app using BBBAndroid? already got 1 feedback from hendersa ! Mar 27 18:45:10 mr_science: :) Mar 27 18:45:51 mr_science: it's bigish though, but I hope at least researched Mar 27 18:45:55 :) Mar 27 18:46:38 anyone for Library of Arduino-compatible functions for Starterware ?? Mar 27 18:47:00 kiran4399: it would also provide a platform for including PRU demos in the BeagleBone documentation, increasing awareness of the PRUSS Mar 27 18:47:20 jason reviewed it though..but some more feedback ?? Mar 27 18:47:40 mr_science: thank you Mar 27 18:47:51 alexanderhiam : thanks for the feedback :) Mar 27 18:49:45 ah, mr_science == nerdboy Mar 27 18:50:19 10 minutes Mar 27 18:50:29 neem0: melange doesn't want me to see that link... Mar 27 18:50:47 * mr_science still in san jose Mar 27 18:50:50 Are you signed in? Mar 27 18:50:56 if anyone can comment for "Demo app using BBBAndroid" proposal. here it is : https://www.google-melange.com/gsoc/proposal/review/student/google/gsoc2015/iqans/5629499534213120 Mar 27 18:52:20 okay got to it finally Mar 27 18:56:12 3 min Mar 27 18:58:02 * mr_science just hit the comment button Mar 27 18:58:33 alexanderhiam: i can also hit the edit flag, yes? Mar 27 18:59:15 you mean edit past deadline? Mar 27 18:59:20 yup Mar 27 18:59:24 yeah Mar 27 18:59:40 mainly there if we realize there's some information we need added Mar 27 19:00:19 yeah, i just gave the outside comment so he needs to edit that stuff in i think Mar 27 19:00:50 neemo: it does look pretty good... aggressive, but good... Mar 27 19:01:56 mr_science: alexanderhiam Thanks a lot, to both of you for the feedback! saved in the last second! Mar 27 19:02:13 and yeah my apologies if it seemed like I'd vanished from IRC in the worst possible moment Mar 27 19:02:49 no, that was me Mar 27 19:03:07 * mr_science gone at conference and dev mtgs all week Mar 27 19:03:07 what in fact happened is that the lovely love hate relationship I have with Google sudenly got baaaaad, and my vpn snapped as usual Mar 27 19:03:21 mr_science: thanks a lot. updated as you mentioned. cheers Mar 27 19:03:26 :) Mar 27 19:03:41 :-) Mar 27 19:03:43 k29: did you get another comment for the 3? Mar 27 19:03:44 mr_science: well, you were here precisely when was right Mar 27 19:03:52 Frozen Mar 27 19:03:59 nah Mar 27 19:04:20 the edit flag is open... Mar 27 19:04:46 oh yeah Mar 27 19:04:50 and it ends...... Mar 27 19:05:01 edit flag open till? Mar 27 19:05:06 alexanderhiam: is the third comment thing a deal breaker? Mar 27 19:05:07 all the best to all co-students who have applied :) Mar 27 19:05:19 webglider: same to all Mar 27 19:05:26 mr_science: does it ask for 3? Mar 27 19:05:37 alexanderhiam: yes Mar 27 19:05:39 I thought it was 2 Mar 27 19:05:49 2 outside i guess Mar 27 19:06:00 think he only has one... Mar 27 19:06:01 as in 2 from outside, 1 personal Mar 27 19:06:17 I'm not too worried about it, as long as it's clear that the student has engaged with the community Mar 27 19:06:38 yeah, have posted a request to a community user, will add in if the edit flag is open or will simply comment it out? Mar 27 19:06:42 would that be ok? Mar 27 19:07:11 * mr_science still crazy new guy Mar 27 19:07:37 * alexanderhiam mostly just pretends he knows what's going on :P Mar 27 19:08:04 k29: comment would work, or ping a mentor when you get it and ask Mar 27 19:08:13 @alexanderhiam The community bonding period is upcoming right? Do you that into account while reviewing proposals? Mar 27 19:08:33 sure Mar 27 19:08:41 Thanks to everybody who helped me write my proposal !!! I owe you guys a lot !! Mar 27 19:08:57 alexanderhiam: BTW till when is edit proposal option open? Mar 27 19:09:54 prakharsingh95: no, that starts after proposals are chosen, but you should stick around irc and the mailing list until then in case we need to clear anything up Mar 27 19:10:05 k29: until it's turned back off Mar 27 19:10:12 of course thank you Mar 27 19:10:19 :D Mar 27 19:43:48 * vvu has his mailbox full of gsoc mails :) Mar 27 19:44:38 <_av500_> vvu: yup Mar 27 19:46:06 i need to go through them a bit Mar 27 19:46:12 there are some for the usb boot Mar 27 20:47:57 I created a filter to put them in a different mailbox Mar 27 22:13:59 is the whole gsoc thing already planned out ? Mar 27 22:18:03 <_av500_> more or less Mar 27 22:18:13 <_av500_> today is the end of the student application period Mar 27 22:18:21 <_av500_> rob_w: wanted to join as a student? Mar 27 22:18:25 <_av500_> or mentor? Mar 27 22:19:17 i am no student anymore , thats for sure Mar 27 22:31:24 well the pru projects are interesting the least .. and i want a beagle submarine ! Mar 27 22:54:50 hehe Mar 27 22:55:10 the yellow beagle submarine Mar 27 23:02:56 * mr_science pokes jkridner about a psychedelic beagle mascot Mar 27 23:03:24 with flower power Mar 27 23:04:00 there is a beagle already underwater (openrov) Mar 27 23:05:09 jkridner: I should now be in the mentor pool, I had a somewhat emergency business trip to FL this week, and got back yesterday Mar 27 23:05:58 you can run, but you can; Mar 27 23:06:04 hide... Mar 27 23:06:15 <_av500_> calculus: thx Mar 27 23:06:18 +t Mar 27 23:06:28 <_av500_> calculus: can you see if you can still sign up as mentor? Mar 27 23:06:31 <_av500_> in melagne? Mar 27 23:06:53 is that a kind of cheese? Mar 27 23:06:55 well I did this morning before the student app closed, but I can see about trying again Mar 27 23:07:05 <_av500_> calculus: I co.admin bb gsoc with jkridner Mar 27 23:07:38 <_av500_> I dont see pending connections Mar 27 23:07:44 <_av500_> pls try again Mar 27 23:08:34 * calculus trys Mar 27 23:09:35 _av500_: try that Mar 27 23:09:50 <_av500_> jo Mar 27 23:09:52 <_av500_> worked Mar 27 23:09:56 <_av500_> thx again Mar 27 23:10:11 yep, no problem Mar 27 23:10:33 wow, that is a lot of proposals Mar 27 23:11:20 <_av500_> calculus: yeah Mar 27 23:11:20 tim should be signing up today oo Mar 27 23:11:24 too even Mar 27 23:12:10 <_av500_> calculus: we want to hold a mentors meeting next thursday Mar 27 23:12:14 <_av500_> er Mar 27 23:12:16 <_av500_> wednesday Mar 27 23:12:26 <_av500_> same time slot as the previous ones with students Mar 27 23:12:39 <_av500_> there we can discuss in detail how to procees Mar 27 23:13:50 didn'tt even notice he was here already... Mar 27 23:14:10 _av500_: meet moto-timo Mar 27 23:14:24 greetings Mar 27 23:14:26 <_av500_> moto-timo: aloha Mar 27 23:14:40 <_av500_> I need a new irc cheat sheet Mar 27 23:14:52 I need one badly Mar 27 23:14:57 <_av500_> my nick-cache is trashing Mar 27 23:18:48 I just finished registering on melange Mar 27 23:19:59 <_av500_> moto-timo: request a connection with BB Mar 27 23:21:51 <_av500_> moto-timo: acked, thx Mar 27 23:22:01 :) Mar 28 01:08:33 Tartarus: ping Mar 28 01:08:51 vvu: pong Mar 28 01:09:17 i have an u-boot question...have like 500 boards with empty EEPROMS so no maigc/name string Mar 28 01:09:30 BBB boards Mar 28 01:09:32 ouch Mar 28 01:09:43 Yeah, you can kludge a U-Boot to know they're BBB Mar 28 01:09:46 and then flash them Mar 28 01:09:49 i modified the read_eeprom function from board.c to fake a good magic and string Mar 28 01:09:51 I thought the "flasher" images did that, even Mar 28 01:10:41 and after that i am trying to do a i2c_write into the eeprom, the 1st write for the magic number works ok but the 2nd one gives me an error Mar 28 01:10:49 one sec and i'll give a pastebin Mar 28 01:11:58 https://gist.github.com/ungureanuvladvictor/ff55b53d2e849cd73780 Mar 28 01:12:10 i've put a couple of printfs to see which functions are called Mar 28 01:12:35 Tartarus: and i'm writing like this: https://gist.github.com/ungureanuvladvictor/cde1841b2c75259bedb2 Mar 28 01:12:49 ok Mar 28 01:12:56 So, digging my old old notes: Mar 28 01:13:07 it's like 2 writes one after another fail Mar 28 01:13:10 but only one works Mar 28 01:13:19 Yeah Mar 28 01:13:40 I'm kinda hazy on why it doesn't work Mar 28 01:13:48 Try single i2c mw's at the cli Mar 28 01:14:35 or doe it via the 'eeprom' command Mar 28 01:14:38 won't be a good solution for me, i want to modify u-boot so it does all the flashing automatically Mar 28 01:14:39 which splits the writes Mar 28 01:14:59 You could put it into a script to run if it works ;) Mar 28 01:15:06 again, iirc the flasher does it that way Mar 28 01:15:09 ah true that :) Mar 28 01:15:24 "the flasher" ? Mar 28 01:15:41 Hey Tartarus... doesn't the registers get set with the processor type/family on entry to SPL? Mar 28 01:16:08 ds2: nope, that's all in the eeprom normally Mar 28 01:16:18 vvu: yeah, the flasher images from cco or whatever Mar 28 01:16:25 Or do they not make those public anymore? Mar 28 01:16:29 ah, has a custom u-boot maybe Mar 28 01:16:34 Used to be possible to get the one to un-brick boards Mar 28 01:16:35 never really used it Mar 28 01:16:44 And yes, custom u-boot Mar 28 01:16:51 Tartarus: thanks. that explains your comments. Mar 28 01:17:03 i'll see if i2c command works and if so will wrap it up in a script Mar 28 01:17:13 i'm already doing that with MAC address from eeprom :) Mar 28 01:17:50 if i2c mw works Mar 28 01:17:52 try 'eeprom Mar 28 01:18:07 it gets the bulk writes correct too iirc Mar 28 01:18:24 all i2c for value parameter it always expects hex right ? Mar 28 01:18:43 Correct Mar 28 01:18:52 all commands in u-boot are hex Mar 28 01:18:53 ok, perfect Mar 28 01:18:55 except for 'sleep' Mar 28 01:19:07 hehe Mar 28 01:19:15 thx for the help Mar 28 01:19:28 np Mar 28 01:22:25 i2c mw seems to work Mar 28 02:52:59 Evening, all. **** ENDING LOGGING AT Sat Mar 28 02:59:58 2015