**** BEGIN LOGGING AT Fri Mar 29 02:59:59 2013 Mar 29 03:02:01 ds2, I think I'll call it the Heater Cape... Mar 29 03:02:13 since its gonna be HOT. Mar 29 03:08:07 heheh Mar 29 03:08:14 ka6sox: what kind of drive currents per phase? Mar 29 03:08:26 and how high of a drive voltage are you allowing? Mar 29 03:10:51 25v and 1.9A Mar 29 03:13:56 is the 25V from the chip specs? Mar 29 03:15:59 iirc yes Mar 29 03:16:12 ds2, did you sign up as a mentor and add your name to the project? Mar 29 03:17:10 does someone want to be my mentor Mar 29 03:17:16 i need help :p) Mar 29 03:21:11 ka6sox: I am on the mentor list. do I need to add myself to a project? Mar 29 03:22:19 it would be so nice if we can get whole Linux CNC package on a bone Mar 29 03:22:50 its possible Mar 29 03:23:06 the UI should be something Nodie... Mar 29 03:24:22 Nodie? Mar 29 03:24:30 is that a new touch screen one? Mar 29 03:24:44 using node.js Mar 29 03:24:49 ah..hehe Mar 29 03:24:56 * ka6sox watches mranostay kringe Mar 29 03:25:19 compared to industrial machines using DOS, it might not be that unpalatable Mar 29 03:25:31 but the 3D one is kind of nice though. and we do have a SGX Mar 29 03:25:49 ka6sox: drinking now.. don't care.. :) Mar 29 03:26:05 ka6sox: do you use Altium? Mar 29 03:26:12 not if I can help it! Mar 29 03:26:20 I have used it in the past. Mar 29 03:31:40 mranostay, 1beer, mildly care...2beers, meh....3 beers, you can paint it pink for all I care! Mar 29 03:36:00 uh oh....he's baaaack.... Mar 29 03:36:08 :) Mar 29 03:40:25 isn't the application committed? or is my clock off? Mar 29 03:40:40 heck if I know... Mar 29 03:40:48 jkridner, is it gone? Mar 29 03:41:10 still working on idea page cleanup Mar 29 03:41:13 I thougth it said 1900UTC Mar 29 03:41:35 and its 0241UTC Mar 29 03:43:48 jkridner, if I need to clean up ours just let me know... Mar 29 03:43:52 Oh UTC Mar 29 03:43:55 thought it was PST Mar 29 03:43:56 I'm working on DT stuff... Mar 29 03:44:22 well, make sure you have yourself listed as a possible mentor on anything you are willing to mentor. Mar 29 03:49:11 jkridner, my concern is signing up for something and everyone else dropping out and I trying to carry forward things like...patches to mainline Mar 29 03:49:41 ka6sox: you mean the motor project? Mar 29 03:50:00 jkridner: where do I indicate possible mentor? in the mentor list or? Mar 29 03:50:02 ds2, that one I think either one of us can carry by ourselves. Mar 29 03:50:53 pushing patches upstream is sometimes harder than doing brain surgery. Mar 29 03:51:11 (and more delicate) Mar 29 03:51:36 I mean pushing EVT upstream is going to be a "challenge" Mar 29 03:51:39 ka6sox: heheheh Mar 29 03:52:02 i can do most things but I prefer to be paid for upstreaming work :D Mar 29 03:52:49 the motor thing is fun and I think will attract folks... Mar 29 03:53:18 I just wish we *could* bill it as a "complete replacement" but its not there yet... Mar 29 03:53:25 need to talk to koen first about what he did. Mar 29 03:53:31 ds2: under each project, there is an entry for 'possible mentor' Mar 29 03:53:47 jkridner, is the deadline really today? Mar 29 03:53:54 the idea is to help the students looking to help make an early connection with mentors. Mar 29 03:53:57 yes. Mar 29 03:53:59 well... Mar 29 03:54:05 you can still add ideas later... Mar 29 03:54:06 but... Mar 29 03:54:07 I got that mail at the end of the workday on a holiday weekend ;) Mar 29 03:54:14 today is the last day to let Google see them Mar 29 03:54:19 sorry, I missed that. Mar 29 03:54:29 oh, well Mar 29 03:54:29 can I edit it now or will that conflict with your cleanup? Mar 29 03:54:36 edit away Mar 29 03:54:39 'k Mar 29 03:54:56 ds2, when you are done let me know and i'll get in there Mar 29 03:55:30 Ohhh I see what went wrong Mar 29 03:55:35 I cut and pasted a bad template Mar 29 03:55:42 so it missed that stuff. will fix it now. Mar 29 03:55:47 ka6sox: ok. give me about 10 Mar 29 03:58:22 ds2, np Mar 29 03:58:25 go ahead Mar 29 03:58:31 I'm still fixing patches. Mar 29 04:00:49 ok done. Mar 29 04:01:01 ka6sox: did you want those parts for stepper motor drive? Mar 29 04:01:05 jkridner: I put myself down for the Android one too Mar 29 04:02:30 ds2: great! Mar 29 04:02:35 yes, I could use some Mar 29 04:02:40 want the full PN? Mar 29 04:03:13 this reminds me... I need to get my board with the Allegro stuff working Mar 29 04:03:27 but that board is MSP430 based Mar 29 04:05:29 I think the PRUs are the right way to go...much better coupling. Mar 29 04:05:33 DRV8825PWPR is the only one I can find to sample Mar 29 04:05:48 stepper motor drive, right? Mar 29 04:06:04 thats the ones! Mar 29 04:06:09 the *exact* ones Mar 29 04:07:05 steps even! Mar 29 04:07:06 not a QFN fan... Mar 29 04:07:11 steps are so 1990 Mar 29 04:07:23 yes. the PRU would solve the whole RT patch issue on ARM. Mar 29 04:07:39 Next year, we can add encoder feedback ;) Mar 29 04:08:00 ka6sox: recent experiments with hot air have made me like QFNs Mar 29 04:08:38 jrawson: don't feel bad about asking questions.... or getting trolled for asking what might be considered a dumb question Mar 29 04:08:52 ;) Mar 29 04:08:56 trolls. where Mar 29 04:09:02 we never troll Mar 29 04:09:04 surely not Mar 29 04:09:10 well mostly not Mar 29 04:09:10 ds2, curiosity, what temp do you use? Mar 29 04:09:34 Russ: my latest experiments was at about 280C, IIRC Mar 29 04:09:44 what is that for? Mar 29 04:09:46 pretty low for a hot air gun Mar 29 04:09:51 Russ: these are 6x6mm MEMS sensors Mar 29 04:10:00 rohs? Mar 29 04:10:02 the reflow profile calls out about 270C as peak temp Mar 29 04:10:06 a 28pin QFN seems a bit more "fun" to do. Mar 29 04:10:06 yeah, rohs unfortunately Mar 29 04:10:12 Well, I'm snarky enough to handle trolling, this is just a new community to me, and so I'm trying to figure out norms. Thanks for the gentle prodding though. :) Mar 29 04:10:21 I'd think w/28pin, it is more of a nozzle size thing Mar 29 04:10:24 thermal pad? Mar 29 04:10:31 theres nothing normal here Mar 29 04:10:34 Russ, definately Mar 29 04:10:41 its a Stepper Driver. Mar 29 04:10:43 don't recall, but I think so... it is more of a ground middle pad Mar 29 04:10:45 and the nature of the place changes with the time of day Mar 29 04:10:48 oh you mean that chi Mar 29 04:10:49 same deal Mar 29 04:10:57 surprised you get it hot enough to reflow with 280C Mar 29 04:10:58 when the europeans wake up Mar 29 04:11:05 Russ: my first experiment was at almost 350C Mar 29 04:11:18 the chip "floated" in the solder Mar 29 04:11:22 I assume that was good enough Mar 29 04:11:25 we flowed the allegros in my toaster oven. Mar 29 04:11:25 so far I've been going at 320C Mar 29 04:11:32 still takes a long time on some bga parts Mar 29 04:11:38 that soudns really hot Mar 29 04:11:48 FWIW, I am using RA flux w/nonconductive residue Mar 29 04:11:50 Anyway, long and short of it, working on a biofeedback-ish project using a beagleboard-xM where we want to use the dsp, followed http://elinux.org/BeagleBoardUbuntu and got what I think is the dsp parts installed, wondering if the community has pointers to working with tidspbridge-y things. Mar 29 04:12:12 bridge? hehehehe Mar 29 04:12:23 trolls live under bridges Mar 29 04:12:36 there are lots of pointers, but most of them are pretty difficult to follow... Mar 29 04:12:40 shouldn't it be dsplink-y or syslink-y stuff? :D Mar 29 04:12:43 Seems like some of the dsp pages are old also, so not sure what is current. Mar 29 04:12:54 Russ: what kind of flux are you using? and what kind of solder? Mar 29 04:12:56 I'd like to think there is an easy answer, but their isn't. Mar 29 04:13:10 ds2, quik chip and rohs Mar 29 04:13:27 I thought syslink was omap4 stuff though? Mar 29 04:13:31 Russ: wire or paste? Mar 29 04:13:44 with the bga's, what ever is already on the bga Mar 29 04:13:54 I've done some qfps that I soldered on wrong Mar 29 04:14:07 ahhh Mar 29 04:14:13 for those I just lifted them, put down flux, put them back down, and reapplied hot air Mar 29 04:14:17 I wonder if the BGA balls are higher temp Mar 29 04:14:26 And other documentation I've seen said that dsplink didn't have the features that dspbridge did. Mar 29 04:14:28 I think its harder to heat them Mar 29 04:14:32 oh wait... this is not fully RoHS now that I think of it Mar 29 04:14:52 jrawson: http://elinux.org/BeagleBoard/DSP_Howto is probably as good as any Mar 29 04:14:56 before I do my next one, I need to actually time things Mar 29 04:14:59 jrawson: there are camps that support one over the other...so it can be a mess Mar 29 04:15:14 do a board preheat stage, and then time how long it takes to remove the chip Mar 29 04:15:22 jkridner, done... Mar 29 04:15:28 that way, I can do the reverse properly Mar 29 04:15:32 Russ: I think I been using wire "silver" solder from Ratshack but it isn't the normal RoHS stuff nor is it 60/40 Mar 29 04:15:48 Russ: which heat gun are you using? Mar 29 04:15:54 hmmm... don't see a dspbridge branch anymore. :( Mar 29 04:16:05 ds2, the cheap one Mar 29 04:16:09 t858d+ or whatever Mar 29 04:16:11 jkridner, really compile(+d) PRU < hmmm... might be in staging Mar 29 04:16:32 jkridner: Started following that, the ubuntu builds come with staging drivers for dspbridge that work on 3.7.x kernels, and the DSP_howto page only talks about 2.6.x things. Mar 29 04:16:38 ds2, if I was adding solder though, I'd use paste Mar 29 04:16:41 Russ: ahhh. I am using the Circuit Specialist one Mar 29 04:16:54 Russ: I pretin the pads and use that as solder...but then I am using QFNs Mar 29 04:17:03 maybe I should start using temp probes to see what temps I'm actually getting Mar 29 04:17:16 yeah, I doubt anyone has spent much effort documenting it recently. Mar 29 04:17:16 Isn't bridge for the 3430 or the 3630? Mar 29 04:17:36 Russ: is there an active temperate measurement/display on that? Mar 29 04:17:38 Russ, your temps seem way too high Mar 29 04:17:44 ds2, it has a digital set point Mar 29 04:17:50 Right, which is the DM3530 from what I read, which is what the beagleboard-xM has. Mar 29 04:17:54 ds2: haha... jrawson: omap3630 and dm3730 are essentially the same part Mar 29 04:17:56 ds2^ Mar 29 04:18:05 er, yeah, what jkridner said. Mar 29 04:18:16 * ds2 whistles innocently Mar 29 04:18:16 ds2, for assembly, I would really recommend using a stencil Mar 29 04:18:28 you can still use the hot air gun, but you could also use a hot plate or oven Mar 29 04:18:47 Russ: I am trying things one step at a time. ultimate goal is to have an oven with a thermal couple + a SS relay Mar 29 04:19:12 * jkridner is happy to see http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/drivers/staging/tidspbridge Mar 29 04:19:14 but still, I am very pleased with the hot air reflow Mar 29 04:19:14 ds2, are you going to or shall I register the "project" in accordance with the rules about being a "registered beagleboard.org project"? Mar 29 04:19:27 ka6sox: please go ahead Mar 29 04:19:33 okay will do Mar 29 04:19:38 since we don't have a webpage yet Mar 29 04:19:41 ds2, before I used stencils, I'd use toothpicks to apply paste Mar 29 04:19:41 But since that's omap3 and docs suggested dspbridge was more fully featured than dsplink, I've been chasing that for a week now. Mar 29 04:20:08 jrawson: practically, reading that code and the commit messages is probably the best way to figure out the accurate details. you can check out the old docs for an overview. Mar 29 04:20:08 Russ: why not the syringe the paste comes in, prehaps w/a nozzle? Mar 29 04:20:24 ds2, the syringe has too much delay Mar 29 04:20:33 I'd end up applying way too much paste Mar 29 04:20:35 http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/drivers/staging/tidspbridge/Documentation/README seems kinda limited Mar 29 04:20:37 Russ bpic? Mar 29 04:20:40 ugh. Mar 29 04:20:44 I think you need a pump to do it properly Mar 29 04:20:49 Russ: oh I see. Mar 29 04:20:52 bpic? Mar 29 04:21:04 sorry that meant to be toothpick Mar 29 04:21:07 ah Mar 29 04:21:11 jkridner: I've been using https://github.com/RobertCNelson/stable-kernel Mar 29 04:21:27 you can use a sryinge and dispending needle Mar 29 04:21:32 problem I see with a stencile is that requires me to commit to the board at once Mar 29 04:21:34 mrpackethead, I tried Mar 29 04:21:42 I like to assemble enough to test. then add more Mar 29 04:21:44 for small quanitys on the work bench i recommend ZephPhaste Mar 29 04:21:46 ds2, you can still move vias and traces around Mar 29 04:22:16 the lucky thing with paste is that location accuracy isn't that important, its the quantity that matters Mar 29 04:22:23 a lot of the quality of the pasting comes down to your paste as well Mar 29 04:22:36 the stuff we use in produciton is'nt realy suitable for prototyping Mar 29 04:22:38 as i have learned Mar 29 04:22:54 I really like the tackyness of the stuff I use Mar 29 04:23:05 suppose that makes it harder to apply with a syringe Mar 29 04:23:12 flux gets you the tackiness too Mar 29 04:23:26 this stuff is great Mar 29 04:23:27 http://www.zeph.com/zephpaste.htm Mar 29 04:23:35 and it stores really well Mar 29 04:23:43 mrpackethead: how well? Mar 29 04:23:51 Anyway, sounds like for what I'm doing I'm walking a path that probably won't dead-end. Thanks guys. Mar 29 04:23:54 i have had some 12 months and its still fine Mar 29 04:24:01 non aircondititioned room well? Mar 29 04:24:02 ka6sox: do you mean for GSoC? I think you only need to register the projects when someone actually starts working on them. Mar 29 04:24:03 they say, 6 Mar 29 04:24:09 but i've had some and its been ok. Mar 29 04:24:10 okay... Mar 29 04:24:20 jkridner, did you see my "edit" for you. Mar 29 04:24:23 or should I do it. Mar 29 04:24:26 don't mind storing my stuff in the fridge Mar 29 04:24:31 ka6sox: I'm talking about registering at http://beagleboard.org/project Mar 29 04:24:41 i keep mine in the fridge Mar 29 04:24:43 as well Mar 29 04:24:45 jkridner, understood. Mar 29 04:24:54 the stuff we use for production has a 6 week life Mar 29 04:25:11 ka6sox: thanks Mar 29 04:25:13 and needs to be baked with 2 hours of being pasted Mar 29 04:25:20 but we get excellent results Mar 29 04:25:26 jkridner, really compile(+d) PRU programmes < add the d Mar 29 04:25:32 :) Mar 29 04:25:40 but its not good for hand placing, or on the bench Mar 29 04:26:00 I did qfn's (the allegros) placed by hand... Mar 29 04:26:01 mrpackethead, I probably got 4 hours with my paste, and hand placing Mar 29 04:26:03 it wasn't fun. Mar 29 04:26:11 like I said, very tacky Mar 29 04:26:15 also make sure its properly warm. Mar 29 04:26:15 how do you get that long..I get 2hrs at most! Mar 29 04:26:19 when you apply it. Mar 29 04:26:27 yup, let it "thaw" Mar 29 04:26:31 ka6sox: theres 100's of types of paste Mar 29 04:26:40 all subtly different Mar 29 04:26:44 ka6sox, the rohs chip quik stuff from digikey Mar 29 04:26:48 * jkridner wonders how the PRU firmware isn't firmware Mar 29 04:26:52 the stuff we got was from sparkfun Mar 29 04:27:09 jkridner, it is firmware. Mar 29 04:27:26 its a reprogrammable onboard device Mar 29 04:27:33 http://www.flickr.com/photos/russdill/8543373764/ Mar 29 04:27:34 yeah, I don't know how it gets quotes around it, because it is about as firmware as firmware gets. Mar 29 04:27:46 (paste, not yet solder) Mar 29 04:27:53 if you can find the documentation for the PRU Mar 29 04:28:16 I'm goign up to Shenzhen in a few weeks Mar 29 04:28:17 russ tell me you did NOT hand place the Virtext7 Mar 29 04:28:18 good release toot http://www.flickr.com/photos/russdill/8541847969/ Mar 29 04:28:27 ka6sox, its a 1mm bga, its easy Mar 29 04:28:29 i might go to the market and buy one of those IR reflow ovens Mar 29 04:28:32 and see how it goes Mar 29 04:28:34 676? Mar 29 04:28:38 784 Mar 29 04:28:41 whoa Mar 29 04:28:56 Russ, did you reflow? Mar 29 04:29:11 http://www.flickr.com/photos/russdill/8542289999/ Mar 29 04:29:39 ahh.. "please contact your consultant, yoru booking is too complex" Mar 29 04:30:27 mrpackethead, sounds like an awesome trip Mar 29 04:30:30 jkridner: Can we assume someone will be assisting in providing an official copy of the docs for the PRU(vs the stuff that is floating around)? Mar 29 04:30:34 WLG -> HKG -> BER -> CGN // SXB --> LHR -> MCO -> DFW -> SNA // LAX -> AKL -> WLG Mar 29 04:30:41 mrpackethead, when I came to .nz my booking agent was able to get both of us there for $560/ea Mar 29 04:30:43 its *mostly* work Mar 29 04:30:43 ka6sox: is http://tomkingcommunications.com/ you? Mar 29 04:31:05 hmm...I really need to do my webpage... Mar 29 04:31:07 jas Mar 29 04:31:41 see pm Mar 29 04:31:46 ds2: https://github.com/beagleboard/am335x_pru_package is official, but I expect we can get it put back into the TRM. Mar 29 04:31:48 ka6sox, I used a steel stencil and can control the oven process pretty good, I didn't have any reflow problems Mar 29 04:32:03 who made your stencil? Mar 29 04:32:13 advanced circuits Mar 29 04:32:25 jkridner: okay.. that should suffice. Mar 29 04:32:47 I usually use ohararp, but couldn't imagine using kapton for a board that size Mar 29 04:32:57 it was certainly the right choice Mar 29 04:32:58 I cut proto stencils in mylar plastic Mar 29 04:33:04 on the laser Mar 29 04:33:11 and get good results Mar 29 04:33:31 mrpackethead: mylar cuts okay w/CO2? Mar 29 04:33:37 no problme Mar 29 04:33:39 applying paste with a kapton (or probably mylar) stencil is certainly an art form Mar 29 04:33:46 i use our 50W Mar 29 04:33:53 at 20%, 100mm/s Mar 29 04:33:55 at 100% duty cycle? Mar 29 04:33:57 no.. Mar 29 04:34:00 oh 20%... ok Mar 29 04:34:07 it cuts very easily Mar 29 04:34:09 so a 40W should work fine Mar 29 04:34:17 is the mylar u using the transparency stuff? Mar 29 04:34:37 not sure, i buy it from a book restoration supplier Mar 29 04:34:42 it comes in 1200mm x 12m roll Mar 29 04:34:47 oh.. how thick? Mar 29 04:34:57 75 or 100micron Mar 29 04:34:59 1.2mx12m? wow... big roll Mar 29 04:35:01 i use the 75micro Mar 29 04:35:04 arrggggg microns Mar 29 04:35:13 what's that in mil's or thou's? Mar 29 04:35:14 75micron is about 3mill Mar 29 04:35:29 3/1000ths Mar 29 04:35:36 'k thanks. sorry, thicknesses are always in mils...not good at translating microns Mar 29 04:36:13 i do .4mm TQFP's and QFN's Mar 29 04:36:15 no problem Mar 29 04:36:17 more and more datasheets make stencil recommendations in um Mar 29 04:36:55 3mil is pretty thin, IIRC... transparency is about 6mil, I think Mar 29 04:37:10 6 is too thick Mar 29 04:37:15 most stencils are around 6 mil, no? Mar 29 04:37:20 our stainless stencils are 3 or 4 Mar 29 04:37:37 Russ: do you keep a mic around for this sort of stuff? Mar 29 04:37:52 I have a cheap stereo amscope Mar 29 04:37:55 6 mil is too thick Mar 29 04:38:03 3mil is easier Mar 29 04:38:07 ah Mar 29 04:38:16 ka6sox: but is transaparency material about 6mil? Mar 29 04:38:32 yes, but we were using 304SS Mar 29 04:38:53 how you cutting the 304SS? Mar 29 04:39:03 yag laser? Mar 29 04:39:23 I see jessica is working late :D Mar 29 04:40:01 we bought it from a place in Vista Mar 29 04:40:12 $300 in the frame Mar 29 04:40:18 (not cheap! Mar 29 04:40:25 but we got it next day. Mar 29 04:40:34 I wonder why my guy gets it Mar 29 04:40:44 I get charged like $250 for a framed stencil Mar 29 04:41:00 http://reprap.org/wiki/Melzi Mar 29 04:41:07 those boards...with the QFN drivers Mar 29 04:41:36 (allegros) Mar 29 04:41:55 allegro has SOIC ones Mar 29 04:42:34 ya, those are the pollolu's Mar 29 04:42:38 POJ Mar 29 04:43:04 jkridner, does she not see we have been editing! Mar 29 04:43:20 :-) Mar 29 04:43:36 I think some folks haven't and she's responding to a blanket request I made to her... Mar 29 04:43:46 kk... Mar 29 04:44:06 I asked if she could encourage people who've expressed interest in being mentors to sign up as possible mentors for specific project. Mar 29 04:44:11 thank you for already having done so. Mar 29 04:44:14 I am using the code on https://www.kernel.org/doc/Documentation/spi/spidev_test.c I just edit /dev/spidev1.1 to spidev4.0 on beagleboard-xm revC, but it outputs only FF http://pastebin.com/PQraM8pX when I run the spidev_test.c what is wrong with it? Mar 29 04:44:42 angs: pinmux? Mar 29 04:44:43 angs: moon not aligning? Mar 29 04:45:06 that is a very vague question Mar 29 04:45:07 angs: did you configure the pinmux and does it have RX enabled? Mar 29 04:45:39 pinmuxing seems fine, pin 20 (in), 18 (in), 12(out) are on mode 1 Mar 29 04:45:53 http://pastebin.com/PQraM8pX doesn't give me a lot of information about what you're doing Mar 29 04:45:57 uh... that sounds wrong Mar 29 04:46:05 CLK needs I/O Mar 29 04:47:17 let me check again Mar 29 04:47:30 and you do have the preq. wire, RIGHT? Mar 29 04:47:54 I have not wired miso to mosi Mar 29 04:48:06 expecting this to be wireless SPI? :D Mar 29 04:48:44 I am not adding -l to loopback Mar 29 04:48:54 hi I would like to interface a lcd character display to beaglebone, what would be the best option for suck a display? Mar 29 04:49:21 use the LIDD? Mar 29 04:49:21 such* Mar 29 04:49:30 ds2: http://elinux.org/index.php?title=BeagleBoard%2FGSoC%2FIdeas&diff=235742&oldid=235736 ? Mar 29 04:49:40 let me look Mar 29 04:50:12 jkridner: looks fine. Mar 29 04:50:20 bonenewb: what LCD character display do you have laying around? if you are going to buy one, have you googled for an example? Mar 29 04:50:23 ds2: thanks? Mar 29 04:50:53 jkridner: actually... the existing project is a bit off... this is more of a generic IIO support but that is the closest link Mar 29 04:51:27 * jkridner takes a quick look at https://groups.google.com/forum/?fromgroups=#!topic/beagleboard/IlDGhHQcTwA for bonenewb's question, but then gets back to GSoC application Mar 29 04:52:03 bonenewb: look at the TRM. search for "LIDD" Mar 29 04:52:04 ds2: I think they'd need the patches for Bone for testing, but I agree it should work against mainline w/o Bone dependencies Mar 29 04:52:08 it is a hardware block for that stuff Mar 29 04:52:28 jkridner: okay, I am okay with that. Mar 29 04:53:33 I have the following display lying around Mar 29 04:53:34 http://www.newhavendisplay.com/specs/NHD-C0220BiZ-FSW-FBW-3V3M.pdf Mar 29 04:54:04 but I think its busted, cause I built the circuit as suggested using the beaglebone 3.3 V input Mar 29 04:54:16 jkridner, pin muxing http://pastebin.com/W01H3DA4 matches with http://elinux.org/BeagleBoard/SPI Mar 29 04:54:16 sorry I mean 3.3 V output Mar 29 04:54:27 I wired mosi to miso, but it outputs FF again Mar 29 04:54:36 right back Mar 29 04:54:47 when I did the i2cdetect nothing came up Mar 29 04:54:50 get the LIDD code working Mar 29 04:54:56 there was nothing on the screen either Mar 29 04:54:58 does spi work full duplex or half duplex on beagleboard-xm? Mar 29 04:55:21 actually, if it is I2C... Mar 29 04:55:24 wtf is the question? Mar 29 04:56:02 SPI is FD Mar 29 04:56:10 SPI *must* be FD. Mar 29 04:56:45 you can't get data unless you run the clock. Mar 29 04:57:10 and when you run the clock, you need to pay attention to the data you are sending. Mar 29 04:57:37 you can throw out data you capture, but there's always something sitting at the pin when you clock it. Mar 29 04:57:45 so, it has to be full-duplex Mar 29 04:58:07 ds2: angs is just working on using spidev Mar 29 04:58:40 jkridner: better? http://imagebin.org/252060 Mar 29 04:58:58 bonenewb: that's I2C. what's the challenge? Mar 29 04:59:49 prppplague: looks like a lot more room to put some documentation on the slikscreen. :) Mar 29 04:59:57 prpplague even Mar 29 05:00:10 jkridner: in the context of a loop back, sure. but you can have HD SPI where MISO and MOSI are connected together with a resistor to reduce wire count Mar 29 05:00:11 slikscreen is eggcellent Mar 29 05:00:17 jkridner: yea Mar 29 05:00:18 Bacon cape. Mar 29 05:00:22 what is this. Mar 29 05:00:40 prpplague: I notice the via count doubled ;-) Mar 29 05:00:49 jkridner: yea Mar 29 05:00:50 three fets,a dn a LED Mar 29 05:00:57 jkridner: The problem is that I don't see anything at all when I do i2cdetect Mar 29 05:01:02 jkridner: no real worries there Mar 29 05:01:47 mrpackethead: mutant version of the Beacon Board Mar 29 05:01:54 suggestions. Mar 29 05:01:58 swap teh Through hole parts out Mar 29 05:02:01 for smd parts Mar 29 05:02:12 jkridner: Also shouldn't I see something on the lcd display Mar 29 05:02:15 prpplague: any chance I can get some labels? Mar 29 05:02:21 ?* Mar 29 05:02:27 jkridner: yea, you let me know what you want labeled Mar 29 05:02:41 mrpackethead: the TP parts are less expensive Mar 29 05:02:46 rubbish Mar 29 05:02:55 how much is a TH RGB part? Mar 29 05:03:01 P8_4x next to the switch. Mar 29 05:03:13 jkridner: can you email it to me? Mar 29 05:03:14 P9_x next to the pot Mar 29 05:03:30 mrpackethead: less than a penny Mar 29 05:03:36 k. deadline? Mar 29 05:03:51 RGB leds is less than 0.01? Mar 29 05:03:53 should I give you a rough placement? Mar 29 05:04:02 jkridner: i want to have the gerbers ready by 8am monday Mar 29 05:04:07 k Mar 29 05:04:21 jkridner: yea, general info Mar 29 05:04:32 mrpackethead: hehe when you buy in bulk from china Mar 29 05:04:56 i do buy in bulk, i buy Millions of RGB leds each year Mar 29 05:05:21 mrpackethead: hehe then you should know, hehe Mar 29 05:05:26 mrpackethead: http://www.elinux.org/BeaconBoard Mar 29 05:05:47 mranostay: total cost for the beacon was just under $1 for 1k units Mar 29 05:05:56 and you are paying less than 1cent per led? Mar 29 05:06:07 the Ws2801 are about 0.12c etch Mar 29 05:06:26 better question : is there a lcd character out there that someone has used for the beaglebone? Mar 29 05:06:49 mranostay: yea i purchased 5k of the RGB leds Mar 29 05:07:02 for $50? Mar 29 05:07:23 mranostay: yea Mar 29 05:07:37 got a link? Mar 29 05:07:47 that is very low cost Mar 29 05:07:57 mrpackethead: not handy, that was over a year ago Mar 29 05:08:06 must have been surplus Mar 29 05:08:13 mrpackethead: but i picked it up of alibaba Mar 29 05:08:25 i have seen such crazy pricing on stuff in the market Mar 29 05:08:25 mranostay: i only used 1k units so i have plenty left over Mar 29 05:08:31 at the SEG Mar 29 05:12:21 nice Mar 29 05:15:15 jkridner, do I need to do anything further with the GSoC page tonight? Mar 29 05:15:43 nah. thanks! Mar 29 05:16:33 i now have a scary project to start work on Mar 29 05:16:36 :-( Mar 29 05:16:49 mrpackethead: a 20000A, 20KV switch? Mar 29 05:16:50 :D Mar 29 05:17:08 ethernet --> profifnet bridge Mar 29 05:17:21 the Beagle is a great starting point Mar 29 05:17:34 oh blah... tame stuf Mar 29 05:17:35 f Mar 29 05:17:42 so... any advice lcd character display anyone? Mar 29 05:18:12 what's the I2C bus looking like on a LA? Mar 29 05:19:45 LA? Mar 29 05:19:54 logic analyzer Mar 29 05:20:45 ah interesting question, hmm I don't have a logic analyzer Mar 29 05:20:54 I just bought a multimeter Mar 29 05:21:00 would that work? Mar 29 05:21:11 not really Mar 29 05:21:15 but that is a start Mar 29 05:21:21 set it on AC and see if you get any action Mar 29 05:21:26 is the lines high (@3.3V)? Mar 29 05:21:31 if not, find out why Mar 29 05:22:20 I checked all the lines they were either 0 or 3.3 volts, I was in the lab once and I used a ossciloscope, and nothing came up Mar 29 05:22:23 it was flat signal Mar 29 05:22:35 ie DC at 3.3 V Mar 29 05:22:41 3.3v (both) should be idle Mar 29 05:23:05 (if its the master) Mar 29 05:23:12 should there be something that comes up on the screen? Mar 29 05:24:10 if assuming all the connected are made except the sdl and sda to the beaglebone? Mar 29 05:26:25 which I2C port are you using? Mar 29 05:26:35 I2C1 Mar 29 05:26:44 pin 17 and 18 on p9 header Mar 29 05:26:48 that port should be working Mar 29 05:26:57 IIRC, that is the port used to read the Bone's I2C prom Mar 29 05:27:15 use a scope and look at that while i2cdetect is running Mar 29 05:27:25 it is possible your device is not respondign to i2cdetect but is otherwise fine Mar 29 05:27:36 and make sure you have the right address Mar 29 05:27:51 interesting thats somethign i wanted to ask someone.. Mar 29 05:28:19 if you create a cape, is there a registry to get an ID fro the EEPROM from? Mar 29 05:28:22 okay cool, but is there a possiblity the lcd display might be defective? Mar 29 05:32:18 *shrug* never used those Mar 29 05:32:28 usually use the older hitach controller ones or the SPI ones Mar 29 05:40:52 mrpackethead: get your connectors? Mar 29 05:41:01 they are ordered Mar 29 05:41:13 probably get them next week i hope Mar 29 05:42:40 how can people stomach all that stacking? Mar 29 05:42:57 are we trying to play bone junga? Mar 29 05:44:38 lol. Mar 29 05:45:48 * mranostay orders nixie cape BOM items Mar 29 05:47:00 d00m! Mar 29 05:47:28 i've just been blowing holes alloy cases Mar 29 05:49:43 5.56mm or 7.62mm? Mar 29 05:49:53 hehe Mar 29 05:53:55 uups, the war and weapon channel Mar 29 06:04:37 0.50 Mar 29 06:07:22 0.51 Mar 29 06:08:28 got to find time to work on the M3 Mar 29 06:08:29 0.26+0.26 Mar 29 06:08:44 = 0.54 Mar 29 06:08:56 heh Mar 29 06:09:16 ds2: your stm32 project? Mar 29 06:10:06 dm8tbr, yup! Mar 29 06:10:54 emeb_mac: yes and eventually the efm32 one Mar 29 06:10:57 * mranostay leechs onto dm8tbr and ka6sox Mar 29 06:11:12 ds2: big fun. Mar 29 06:11:24 emeb_mac: just need time. Mar 29 06:11:36 ds2, whats time? Mar 29 06:11:49 apparently this rare thing unless one has a tardis Mar 29 06:11:51 ds2: time is always in short supply Mar 29 06:12:04 emeb_mac: if I can only get a tardis Mar 29 06:12:33 a Tardis is much smaller than the Backstep machine. Mar 29 06:12:52 plus there is extra storage in a tardis Mar 29 06:13:04 a tardis would be cool Mar 29 06:13:16 wouldn't need to dig a basement for more storage... Mar 29 06:13:48 assuming you are in aplace you can dig a basement Mar 29 06:14:10 ds2: probably not. It's an old joke at our house. Mar 29 06:14:34 someone invariably gets a spoon from the silverware drawer and says "Get started" Mar 29 06:15:13 sure it wasn't just a bad leak at one point? ;) Mar 29 06:15:25 kind of like the grand canyon Mar 29 06:15:32 * dm8tbr throws mranostay a black bone Mar 29 06:15:37 i have two Mar 29 06:15:40 ISTR stories about Seymour Cray having a hobby of digging tunnels in his yard. Mar 29 06:15:47 i mean black what are you talking about!?!? Mar 29 06:16:10 emeb_mac, I believe it....he was eccentric Mar 29 06:16:23 his boxes sure ran fast though. Mar 29 06:16:31 emeb_mac: tunnels to what? Mar 29 06:16:37 * dm8tbr meanwhile browses for industrial strengh active and passive attenuating headsets Mar 29 06:16:39 tunneling was probably easier in Minnesota than it would be here in AZ Mar 29 06:16:47 ya Mar 29 06:17:05 mranostay: apparently tunnels to nowhere. Just random mazes of tunnels around his house. Mar 29 06:17:12 the prices are 'interesting' - anywhere between 500 and 900€ Mar 29 06:17:43 * dm8tbr fancies the Sennheiser S1 Digital... Mar 29 06:17:54 mranostay, why can't we use the kernel firmware loader for PRU firmware? Mar 29 06:18:08 dm8tbr: talk to folks working on an aircraft carrier? Mar 29 06:18:33 ka6sox: that is what i want to do Mar 29 06:18:39 ds2: working in an office with two folks behind my back having telcos at noise levels of an starting fighter jet Mar 29 06:18:45 emeb_mac: sure he isn't part mole? Mar 29 06:19:17 why they _refuse_ to use our meeting rooms is beyond me, both have laptops Mar 29 06:19:44 meeting rooms are overrated Mar 29 06:19:52 I prefer a noisy office Mar 29 06:19:56 can't think in silence Mar 29 06:20:16 I can't think if there are 2-3 conversations going on around me Mar 29 06:21:13 I can't function if there are 2-3 conversations going on at once Mar 29 06:21:26 headphones with music FTW. Mar 29 06:22:11 I tried my closed sennheisers. I had to resort to Hard Rock at earpiercing levels to drown them out. Mar 29 06:22:42 if my employer wants me to be aggressive during my work days that ofc would work too Mar 29 06:22:54 yep - when I worked in a cube farm at Intel I had headphones on anytime I wasn't in a conference call. Mar 29 06:23:25 our previous office had a much better noise attenuation and was far more spacious Mar 29 06:23:54 you'd hardly _ever_ hear a conversation and it was largely open space Mar 29 06:24:11 now it's more like open space, low dividers and crammed up Mar 29 06:24:31 and they put me next to the two guys who spend most of their days in telcos Mar 29 06:24:50 who previously had rooms, for that very reason Mar 29 06:25:06 ka6sox: headphones with music get me through the day Mar 29 06:25:38 drowns out the micromanager boss and the GUI talk i hear next door :) Mar 29 06:26:27 OOP talk gives me hives. Mar 29 06:27:00 these help somewhat, there is still something, but with in ear headphones I can drown that out with even classical music at low levels http://catalogue.3m.eu/en_EU/EU-PPE/Ear_Plugs_and_Ear_Muffs/Passive_Ear_Muffs/Peltor%E2%84%A2~Optime%E2%84%A2_III~Ear_Muffs~nocode Mar 29 06:28:07 dm8tbr, when I'm onsite I often wear those. Mar 29 06:28:52 the 100kW TX is very noisy Mar 29 06:32:59 :) Mar 29 06:33:13 forced air coolin, I guess? Mar 29 06:33:18 +g Mar 29 06:33:33 yes, very large Mar 29 06:33:40 that is what she said Mar 29 06:33:43 i mean.. Mar 29 06:34:26 dm8tbr: that url is invalid Mar 29 06:34:31 * dm8tbr visited a site that had 10 of that class. The two largest ones even went to 200kW AM on HF Mar 29 06:34:36 * ka6sox rolls eyes Mar 29 06:34:42 mranostay: WFM Mar 29 06:34:55 I have the prebuild ubuntu image on beagleboard-mx from http://elinux.org/BeagleBoardUbuntu (uname -a> Linux arm 3.7.8-x8 #1 SMP Sat Feb 16 04:15:13 UTC 2013 armv7l armv7l armv7l GNU/Linux). In order to have spidev, I added buddy=spidev on /boot/uboot/uEnv.txt. ls /dev shows that spidev4.0, spidev3.0 and spidev3.1 are available. I use the test code on https://www.kernel.org/doc/Documentation/spi/spidev_test.c and replace /dev/spidev1.0 Mar 29 06:34:58 the source (gcc spidev_test.c -o a). then execute ./a -s 500000), the beagleboard is connected to atmega chipset (slave) but the mcu does not get data from beagleboard, the linux terminal outputs only FF. I also tried the code by connecting mosi to miso but did not change the output. pin_muxing looks as it supposes to be http://pastebin.com/W01H3DA4 , am I using the wrong testcode or what? Mar 29 06:35:04 their transmitter matrix switch was awesome! Mar 29 06:35:10 why it does not work Mar 29 06:35:50 wow Mar 29 06:36:22 ka6sox: I guess you know such stuff: http://www.ruecker.fi/~dm8tbr/ham/dw/pict2244.jpg Mar 29 06:37:36 deusche welle? Mar 29 06:37:59 yeah, Juelich Mar 29 06:38:22 I love Arc Welders... Mar 29 06:38:27 feel free to browse the album, has lots of pix of that site Mar 29 06:38:50 sadly Juelich was abandoned and demolished :( Mar 29 06:39:20 I built one of those curtains...in Alaska. Mar 29 06:39:26 * mranostay dances Mar 29 06:39:28 and a 250kW TX. Mar 29 06:39:40 that was probably quite an experience Mar 29 06:39:47 * ka6sox arrests mranostay ...for excessive dancing. Mar 29 06:40:00 * mranostay dances out of the handcuffs Mar 29 06:40:00 34.5kV@14.5A Mar 29 06:40:36 yup, decent QRO Mar 29 06:40:57 was that for VoA into Russia? Mar 29 06:41:33 something like that.... Mar 29 06:41:34 :) Mar 29 06:41:42 hehehe, *nod* Mar 29 06:48:45 nice album Mar 29 06:50:14 it was a very enjoyable visit Mar 29 06:51:36 we also got to see their offices, but weren't allowed to take pictures. We got the visit and tour as three of our cand PhDs had written DReaM and they were using it for performance evaluation of their transmissions Mar 29 06:52:12 voa? Mar 29 06:52:19 voice of america Mar 29 06:52:30 ah right Mar 29 06:52:46 it's either that or number stations ;) Mar 29 06:53:33 * dm8tbr realizes he hasn't listened to EAMs in probably 5 years Mar 29 06:53:56 morning Mar 29 06:54:17 morning koen Mar 29 06:54:28 hey ka6sox Mar 29 06:54:47 -NAME = Unicycling Gorilla Mar 29 06:54:48 +NAME = Displaced Humerus Anterior Mar 29 06:54:51 3.8.5 is out Mar 29 06:55:04 with a special greg kh codename Mar 29 06:55:38 :) Mar 29 06:55:39 lol Mar 29 06:56:12 koen, what did you do with a Bone and the 3D pringer Mar 29 06:56:18 printer! Mar 29 06:58:02 sorry for shouting...I was having sticky fingers. Mar 29 06:58:09 (ice cream) Mar 29 06:59:52 dm8tbr, I have one of those: http://www.ruecker.fi/~dm8tbr/ham/dw/pict2208.jpg Mar 29 07:00:17 :) Mar 29 07:00:59 I think that card sticking out below was the injection point for the DRM signal into their modulation PWM Mar 29 07:01:58 thats the *good* kind of DRM. Mar 29 07:02:14 there is a schematic on one of the photos, I don't know though where they injected the DRM modulation exactly. The modulation setup was in the stand alone 19" rack in front of one of the TXs Mar 29 07:02:18 indeed Mar 29 07:03:04 ka6sox: I have one working cupcacke close with a bebopr and at the office we are working on a makergear m2 + bebopr cape Mar 29 07:03:36 url for the cape? Mar 29 07:03:40 not seen that one. Mar 29 07:05:21 okay found it Mar 29 07:05:27 so you are using the PRUs? Mar 29 07:05:39 ka6sox: yes Mar 29 07:05:42 [kernel] koenkooi pushed 4 new commits to 3.8: http://git.io/swa2dw Mar 29 07:05:42 kernel/3.8 5daaa39 Koen Kooi: 3.8: add db.txt from https://git.kernel.org/cgit/linux/kernel/git/linville/wireless-regdb.git/plain/db.txt... Mar 29 07:05:43 kernel/3.8 2b062a0 Koen Kooi: 3.8: add SSD130x patches and provide a CRDA db... Mar 29 07:05:43 kernel/3.8 c9bd50e Koen Kooi: 3.8: integrate db.txt into OE recipes... Mar 29 07:05:53 do I need to apply a patch to make spi work on rcc-ee's prebuild image or do I need to use a different spidev_test.c rather than https://www.kernel.org/doc/Documentation/spi/spidev_test.c ? Mar 29 07:06:01 ka6sox: https://github.com/modmaker/BeBoPr/ Mar 29 07:06:10 ka6sox: teacup fork Mar 29 07:06:39 I see..the PRU code is not FOSS i see. Mar 29 07:07:05 ka6sox: Bas said he'd considered open sourcing it if TI won't sue him for using the secret instructions Mar 29 07:07:09 (not a joke) Mar 29 07:07:11 it shows FFs on the screen,but does not output any signal from the pins Mar 29 07:07:42 koen, I am worried about that too..as I'd like to make that work as well. Mar 29 07:07:50 koen: fun fun Mar 29 07:16:00 * mranostay sues koen for a random reason Mar 29 07:27:08 is there anyone who uses spi on beagleboard-xm without problem? Mar 29 07:36:18 angs, the US is gone to bed and europe is *just* waking up...let them at least have their coffee. Mar 29 07:38:42 ka6sox, I am fine with waiting although I have not slept all night Mar 29 07:39:01 then go to bed mon...it will be here later! Mar 29 07:40:21 thank you, I will go to bed in couple hours eventually Mar 29 07:48:58 its 00:48 here....and I am bleary eyed after working on DT patches... Mar 29 08:03:23 ka6sox, before you go to sleep, can you take a look at dmesg outputs for DT http://pastebin.com/7XP2qfg9 , is that why I cannot use spi on beagleboard? Mar 29 08:05:46 did you set the pinmux and pinfunc for mcspi? Mar 29 08:06:24 I have not set them by myself, but they look fine http://pastebin.com/W01H3DA4 Mar 29 08:06:55 is the dmesg output point any problem? Mar 29 08:10:53 I've not spent any time with that subsystem...spending all my time with Irish Snakes... Mar 29 08:11:25 I don't know either of them:) Mar 29 08:11:28 neither Mar 29 08:11:49 (mcasp's) Mar 29 11:31:08 Hi , I need to load an image to one of my Nand partitions with cat ( I have nothing else like dd ) and the system is running ( the whole system loads from a Nand, rootfs is on different partition , I need to change my /usr which is on this nand but on another partition ) Mar 29 11:31:57 is just doing some thing like unmount /usr and cat myimage > /dev/mtd4 gonna work? Mar 29 11:32:19 or it may burn my device? Mar 29 11:34:28 I don't know if programming a Nand partition from a runing linux on the target is ok or not Mar 29 11:35:36 [ It's not a BB ] Mar 29 11:46:15 doubt cat will work like that Mar 29 11:57:53 hmm thanks thurbad, people in rasbperry channle told me that cat and dd and other byte-copy stuff will not check bad blocks , then it's not so good Mar 29 12:13:19 nkh, nandwrite is your friend Mar 29 12:14:29 panto: Ya I was looking for that , part of linux-mtd, right? Mar 29 12:14:50 yep Mar 29 14:29:13 mourning Mar 29 14:30:05 I hate wireless Mar 29 14:30:13 * koen stabs userspace crda app Mar 29 14:30:25 is the usb port on the bone 480ma max? Mar 29 14:30:29 hi ka6sox-away Mar 29 14:30:37 hiya panto Mar 29 14:30:41 my realtek wifi adaptor is asking for 492 :X Mar 29 14:31:05 usb wifi in general doesn't work Mar 29 14:31:16 on the bone, cause it asks for too much juice Mar 29 14:31:16 panto +1 Mar 29 14:31:19 panto: any suggested alternatives? Mar 29 14:31:27 powered usb hub Mar 29 14:31:38 or the TI wifi cape (when it's ready) Mar 29 14:31:49 usb hub it is Mar 29 14:32:03 and the proto box grows in size again Mar 29 14:32:16 "And Leon's getting larger" Mar 29 14:32:58 panto, if the existing firmware loads fragments @0 and @2 I can add my bits @4, @6 or does it really matter what number *after* @2? Mar 29 14:33:17 no, the numbers are meaningless Mar 29 14:33:26 just don't clash Mar 29 14:33:38 ok Mar 29 14:34:13 note that the devices will be registered in the order in which they are defined in the file Mar 29 14:34:35 by @ number or just physical presence in the file? Mar 29 14:34:47 by physical presence Mar 29 14:34:52 kk Mar 29 14:35:37 forgive me if I oversimplify this.... Mar 29 14:36:29 but ~boot/dts is like saying Y in makemenuconfig (builtin) and firmware/capes is more like insmodding a module? Mar 29 14:37:18 yeah Mar 29 14:37:21 it's supposed to be simple Mar 29 14:37:33 HA! Mar 29 14:38:08 is that why I've spend 2 weeks working on it? Mar 29 14:39:26 the patches in ~patches/hdmi are no longer used iirc right? Mar 29 14:39:33 heh Mar 29 14:39:48 not as far as I know Mar 29 14:39:57 hmmm... Mar 29 14:40:11 koen, know anything more about it? Mar 29 14:41:25 however you are using cape-boneblack-hdmi-00A0.dts? Mar 29 14:41:53 (trying to make sure I'm patching the correct file. Mar 29 14:42:03 yes Mar 29 14:42:56 dethridge found the critical name I've been looking for to get sysclk working for the driver. Mar 29 14:45:37 nice Mar 29 14:45:47 lost in a maze of clocks all alike Mar 29 14:46:25 the coldplay or alan parson's project kind? Mar 29 14:46:36 dali ... Mar 29 14:46:53 ya, that was the one I forgot about! Mar 29 14:48:40 [kernel] koenkooi pushed 2 new commits to 3.8: http://git.io/f-YUcQ Mar 29 14:48:40 kernel/3.8 53ce4a0 Koen Kooi: 3.8: delete obsolete hdmi patches... Mar 29 14:48:40 kernel/3.8 58f53aa Koen Kooi: 3.8: enable fanotify... Mar 29 14:48:52 there you go Mar 29 14:53:24 koen, thanks for removing confusion...there is enough in DT currently. Mar 29 14:58:31 good morning beagleboard! Mar 29 14:58:51 jkridner: will you please go look at the gsoc admin and see if I am properly registered? Mar 29 15:07:03 Has anyone seen this when building am335x-bone.dtb from 3.8.5? "ERROR (phandle_references): Reference to non-existent node or label "userleds_pins" Mar 29 15:09:22 georgem, I'm sure I'd run into it later today...but not yet Mar 29 15:10:26 * koen checks Mar 29 15:10:27 DTC arch/arm/boot/dts/am335x-bone.dtb Mar 29 15:10:27 DTC arch/arm/boot/dts/am335x-boneblack.dtb Mar 29 15:10:27 DTC arch/arm/boot/dts/am335x-tester.dtb Mar 29 15:14:53 cwicks_: as mentioned on the phone, the app went through, so you must have done it right. Mar 29 15:14:59 morning Mar 29 15:15:21 cwicks_: I have your Id down as 'wicks'. Mar 29 15:16:37 jkridner: my public name for gsoc on profile is TIUnivProg do i need to change it to 'wicks' Mar 29 15:16:40 ? Mar 29 15:16:52 Looks like maybe my dtc version is old or something. I'll check that. Mar 29 15:17:11 cwicks_: I think I needed to put in the ID, not your public name Mar 29 15:17:35 panto: sd card image: http://dominion.thruhere.net/koen/angstrom/beaglebone/Angstrom-Cloud9-IDE-GNOME-eglibc-ipk-v2012.12-beaglebone-2013.03.29.img.xz Mar 29 15:17:48 with 3.8.5 Mar 29 15:17:50 F-R-I-D-A-Y!! Mar 29 15:17:57 cwicks_: I'm not sure who can see https://google-melange.appspot.com/gsoc/org/application/google/gsoc2013/8484202 Mar 29 15:19:18 koen, thx Mar 29 15:19:34 I'll hold on for a bit cause I'm in the middle of something Mar 29 15:20:58 mranostay we have holiday Mar 29 15:24:07 jkridner: okay now I see where it references my user name. Mar 29 15:24:18 jkridner: also yes I can see the app Mar 29 15:24:41 k, let me know if you have edits that you can't make Mar 29 15:24:52 woglinde: whow is we? Mar 29 15:24:56 *who Mar 29 15:25:27 jkridner, I have a local hackerspace nearby Mar 29 15:25:35 want to setup something over there? Mar 29 15:28:41 koen: if I were to patch the DT to change the pinmux setting on the serial port? Mar 29 15:28:56 panto: sure, but I need a real pattern on what to setup. Mar 29 15:29:12 did you have some specific ideas? Mar 29 15:29:19 err, I mean I can set it up Mar 29 15:29:37 just a presentation, maybe a workshop Mar 29 15:29:47 a couple of bones would be fine Mar 29 15:30:44 jkridner: I must be missing some info, what do you want to accompllish? Mar 29 15:30:57 jkridner: were you able to test the cppi patches? Mar 29 15:30:58 set a pulldown on the serial RX line. :( Mar 29 15:31:27 koen: not yet. probably not getting to it until Monday. :-/ Mar 29 15:32:11 the serial thing is a blocker for Gerald and the flasher. Seems like a hardware issue to me, but if we can work-around it in software, we should. Mar 29 15:34:26 jkridner: you'd have to patch MLO or teach linux about uart1 pins (yes, fortran numbering) Mar 29 15:35:34 k. I'll look to patch u-boot instead. thanks Mar 29 15:36:26 koen: do you know if you're using DTC 1.2.0 that is included in linux-3.8.5/scripts/dtc/dtc, or are you using a newer version? Mar 29 15:36:37 builtin Mar 29 15:41:50 hi emeb Mar 29 15:42:56 fortran FTW Mar 29 15:43:15 no smalltalk Mar 29 15:44:10 cobol makes the money ... Mar 29 15:45:03 fortran too Mar 29 15:45:11 <_av500_> ADA Mar 29 15:45:16 hmmm.... looks like a pull-up is already enabled, :( Mar 29 15:45:27 jkridner: it's a good app, I added some small tweeks. On thing - do we need to have 'software' clearly in the first sentence of the org? Mar 29 15:45:56 Hi woglinde. Mar 29 15:46:21 dunno... as long as it is clear we are advancing the state of open source software, especially for embedded Linux, I think we are good. Mar 29 15:46:50 * jkridner heads to breakfast Mar 29 15:56:41 * mranostay blares some techno at panto's flat Mar 29 15:56:51 meh wrong channel but whatever Mar 29 15:57:18 * panto pleads to to ops - kick the punk out for a bit Mar 29 15:57:23 *to the Mar 29 15:59:31 <_av500_> op? Mar 29 15:59:44 it's you Mar 29 15:59:49 <_av500_> oh Mar 29 15:59:49 kick... Mar 29 16:03:07 oh, I guess -@ is required for it compile now Mar 29 16:04:48 panto: what you do without me here? :) Mar 29 16:05:04 georgem, yes Mar 29 16:36:03 lumidudm Mar 29 16:36:10 gm prpplague Mar 29 16:36:23 woglinde: greetings Mar 29 16:36:28 hi prpplague Mar 29 16:36:42 panto: hey hey hey Mar 29 16:36:49 panto: it's mister tester! Mar 29 16:36:54 meh Mar 29 16:37:22 the crap I have to do to make the thing work Mar 29 16:39:37 hi Rus Mar 29 16:39:57 morning panto Mar 29 16:40:03 *Russ even Mar 29 16:40:15 3h to beer o'clock Mar 29 16:40:19 * panto looks at the clock Mar 29 16:47:15 panto: hey there old dude Mar 29 16:53:15 hi Russ Mar 29 16:53:32 mranostay: you seem bored Mar 29 16:53:53 hey mranostay Mar 29 16:53:55 biot: a little :) Mar 29 16:54:21 work! Mar 29 16:54:38 bah Mar 29 17:25:06 <_av500_> work? today? Mar 29 17:26:23 <_av500_> mranostay: do you still have that CDROM I gave you at ELCE? Mar 29 17:26:52 ah probably Mar 29 17:27:11 <_av500_> I totally forgot Mar 29 17:27:17 <_av500_> need to set you up with Mike Mar 29 17:27:40 heh need to locate first in my apartment Mar 29 17:27:44 <_av500_> yeah Mar 29 17:41:33 jkridner: email for bacon labels Mar 29 17:41:36 * prpplague nags jkridner Mar 29 17:43:14 k Mar 29 18:05:39 bacon labels? Mar 29 18:23:02 mmm labels. Mar 29 18:25:33 koen: I'm trying to get you a patch series with all the latest lowpan stuff (which has been upstreamed) in it Mar 29 18:26:03 koen: did you commit the stuff I gave you yesterday? Mar 29 18:26:50 koen: nevermind, I see it Mar 29 18:27:14 do { rule_3(); } while (!understand); Mar 29 18:31:30 mranostay: do you have a schematic so I know which color LED is tied to which of P9-14, 16 or 29? Mar 29 18:31:33 hehe, I should print that line out and hang it over my desk Mar 29 18:31:39 or perhaps you can just say. Mar 29 18:32:22 jkridner: pr^2 has the latest Mar 29 18:32:47 he pinged me and bailed! Mar 29 18:34:34 * jkridner wants to move one via to make room for silkscreen. :) Mar 29 18:35:18 via this Mar 29 18:36:23 alan_o: only the dtbo, not the hacks.patch Mar 29 18:38:31 jkridner: the silkscreen better not have any nodejs download instructions :) Mar 29 18:41:20 LOL Mar 29 18:42:05 Shadyman: oh i'm not kidding :) Mar 29 18:42:26 koen: if you have the hacks.patch in front of you, which parts in the dtsi do I need, and do I actually need it. It doesn't seem right that I'd have to have anything in there, but that's the only way so far I've been able to make it work. I'm loading the dtbo fragment with echo BB-BONE-MRF24J40 >/sys/devices/bone_capemgr.7/slots on the command line. I think that's the right way, but it seems I need the cape@7 section (in hacks.patch). A Mar 29 18:42:26 m i doing it wrong? Mar 29 18:42:38 jkridner: Good Friday is a TI holidy? Mar 29 18:42:42 *holiday Mar 29 18:42:48 theoretically. Mar 29 18:43:06 alan_o: don't have it in front of me, sorry Mar 29 18:43:39 ok no prob. We can talk about it later. Mar 29 18:57:08 holy spamming alan_o Mar 29 18:57:23 wha??? Mar 29 18:57:32 just asking a question.... Mar 29 18:57:44 do we have an 80-col limit in the channel? Mar 29 18:58:19 no Mar 29 18:58:21 72col Mar 29 18:58:23 alan_o, let me looks at that cape's dts Mar 29 18:58:50 panto: https://github.com/beagleboard/kernel/blob/3.8/patches/not-capebus/0146-firmware-DT-Fragment-for-MRF24J40-BeagleBone-Cape.patch Mar 29 18:59:27 hmmm Mar 29 18:59:49 panto: http://dpaste.com/1039345/ <-- the hacks i was talking about. Mostly concered about cape@7 Mar 29 19:00:03 dts looks OK Mar 29 19:00:23 what is the problem? Mar 29 19:00:50 it uses DTS ;) Mar 29 19:00:53 I get the feeling I shouldn't have to touch the dtsi in order to have a cape dts fragment Mar 29 19:00:55 ds2: :) Mar 29 19:01:09 but I may be wrong Mar 29 19:01:16 alan_o, it's a helper Mar 29 19:01:29 if you name the fragment the same as your part number you don't have to Mar 29 19:01:43 panto: is that the right way? Mar 29 19:01:53 it's fine Mar 29 19:02:03 BB-BONE-MRF24J40-00A0.dts should work without touching the common Mar 29 19:02:37 panto: so I should rename my dts fragment to that then Mar 29 19:02:49 (that wasn't the convention when I started this, I don't think) Mar 29 19:03:19 alan_o, it's a worthwhile convention to start now Mar 29 19:03:27 ok, will do Mar 29 19:03:30 I realize that people don't want to mess with the boot dts at all Mar 29 19:03:52 panto: not just that, the 3rd party capes will want to just add a dtbo into /firmware and be done Mar 29 19:04:16 the capemap section is a helper really Mar 29 19:04:17 * ka6sox-farfarawa agrees with that sentiment Mar 29 19:04:42 I am worried that insane people will put part-numbers that are not valid filenames in there Mar 29 19:04:44 panto: ok, cool. I'll make the adjustments. Mar 29 19:04:49 which will break the firmware loader Mar 29 19:05:01 if you put a sane part number it will just work Mar 29 19:05:07 alan_o: you might notice me being lazy and adding a few BB-BONE-CAPITALIZED-CAPES.dtsi files :) Mar 29 19:05:59 panto, the whole bonelt vs boneblack thing needs to be sorted. Mar 29 19:06:09 sure Mar 29 19:06:13 jkridner, what will it be? Mar 29 19:06:22 lt? black? purple? Mar 29 19:06:27 RED! Mar 29 19:07:25 ka6sox-farfarawa: there's no 'bonelt' anymore :) Mar 29 19:07:34 ka6sox-farfarawa: mdp ran sed with a vengeance Mar 29 19:07:43 alan_o, no! I don't Bono to come after me Mar 29 19:08:02 ka6sox-farfarawa: i agree, its rattling people. Mar 29 19:08:03 panto: he thinks he owns the color red? Sounds about right. Mar 29 19:08:20 * alan_o is unrattled Mar 29 19:08:25 to the point i'm not sure if i can trust CCO Mar 29 19:08:44 "you can just design aroudn teh White bone", it will be fine Mar 29 19:08:48 .. right .. Mar 29 19:09:07 they are entitled to do what they like. Mar 29 19:09:10 so, am i Mar 29 19:09:24 panto: what needs to be sorted? Mar 29 19:09:29 for capes boneblack is a clusterfuck Mar 29 19:09:32 the board name is black. Mar 29 19:09:40 just stay with white for capes Mar 29 19:09:55 unless you don't need anything on P8 Mar 29 19:10:07 or the PMIC header Mar 29 19:10:11 where is the dependencies on bonelt other than on the board ID EEPROM? Mar 29 19:10:24 koen, well if we manage to keep the extra peripherals on reset it should work Mar 29 19:10:30 the bone is a clusterF* :) Mar 29 19:10:30 there were patchnames with bonelt in them Mar 29 19:10:32 unfortunately it's easier said that done Mar 29 19:10:40 back a bit later Mar 29 19:10:58 ds2: really? Mar 29 19:11:13 koen, so does that mean I have a pseudo-cape called boneblack-hdmi and another cape that is bone-hdmi? Mar 29 19:11:36 * jkridner didn't see the need to rename patches from bonelt to boneblack, but the name of the board is black so if that helps with clarification, that's fine. Mar 29 19:11:43 jkridner: the whole eeprom scheme.... Mar 29 19:11:58 ds2: for capes/ Mar 29 19:11:59 ? Mar 29 19:12:03 for the entire system Mar 29 19:12:15 teh main board depends on having a eeprom or it goes kaloooie Mar 29 19:12:18 it isn't super, but it's way more provided to daughterboard makers than what was there before. Mar 29 19:12:35 ds2: simple u-boot patch. Mar 29 19:12:43 jkridner: nope. Mar 29 19:12:47 http://dominion.thruhere.net/koen/angstrom/beaglebone/BBB-eMMC-flasher-2013.03.29.img.xz Mar 29 19:12:49 kernel side has crap interwoven with that Mar 29 19:13:12 jkridner: I started ripping out layers of the onion before I ran out of time Mar 29 19:13:46 but to be fair, I can't quite completely blame the bone Mar 29 19:13:54 it is the EVM that pushed this monstrosity Mar 29 19:14:09 panto: again, what is the real question Mar 29 19:14:18 koen, since we are treating the HDMI as a onboard cape on the black and a regular cape on the white shouldn't we jut drop the boneblack refernces to HDMI? Mar 29 19:14:45 HDMI hehehehe Mar 29 19:14:46 if they are the same, you can drop the reference Mar 29 19:14:55 another pending disaster Mar 29 19:15:22 just get an intern to plug those holes on p8 that are used by the HDMI... Mar 29 19:15:44 P8 is fine as long as you keep it as video Mar 29 19:15:54 other than slot and whether they are invoked by firmware or boot they should be the same. Mar 29 19:16:11 panto, am I all wet here? Mar 29 19:16:38 the eMMC is the bigger problem Mar 29 19:16:42 it's how it should work Mar 29 19:17:10 the problem that I see with eMMC is that it comes up enabled. Mar 29 19:17:16 ds2, I sure hope the emmc tristates everything when at reset Mar 29 19:17:17 (hardware wise) Mar 29 19:17:41 panto: I hope so too or CCO's shipping and receiving folks would be very busy ;) Mar 29 19:18:26 the issue is that it needs to have a hardware enable...not come up by default ALWAYS on unless you add something. Mar 29 19:18:42 jumper? Mar 29 19:19:10 * kkeller add money ? Mar 29 19:19:13 a PFET swtch on the lines Mar 29 19:20:41 panto: what's the status of capebus wrt upstream? Mar 29 19:21:03 alan_o, we finally got a response by the DT maintainer Mar 29 19:21:12 looks promising, but it will take a few iterations Mar 29 19:21:47 we didn't have a 'that's so stupid' answer, so that's good Mar 29 19:21:52 there is a switch...but that doesnt' seem to hold the unit in reset. Mar 29 19:22:07 doesn't hold the eMMC in reset Mar 29 19:22:15 hardware bug Mar 29 19:22:27 or do you mean the boot switch? Mar 29 19:22:32 its connected to sysboot pins Mar 29 19:22:44 but not the reset pin of the emmc. Mar 29 19:22:50 (unless I'm missing that bit) Mar 29 19:23:24 s2 - uSD Boot Mar 29 19:23:40 yes Mar 29 19:23:46 it's nothing like a reset Mar 29 19:23:54 you would need a proper jumper for that Mar 29 19:24:13 isn't reset hooked up to the am335x? Mar 29 19:24:33 koen yes Mar 29 19:24:56 but you would need to make sure u-boot keep it in reset when not booting from it Mar 29 19:25:09 plus the kernel way of handling that it's not obvious Mar 29 19:25:09 REGEN or something like that Mar 29 19:25:10 correct Mar 29 19:25:52 its connected to gpmc_a4.gpio1_20 via a driver Mar 29 19:27:40 so if sys_boot2 is 0 then gpmc_a4.gpio1_20 needs to be 1 Mar 29 19:27:46 but thats acutally too late. Mar 29 19:27:55 ka6sox-farfarawa, it's a mess Mar 29 19:28:12 I have a solution for it, but it's way crazy Mar 29 19:28:15 a ball for everything and everything on its ball Mar 29 19:29:04 u3-pin2 needs a pullup Mar 29 19:29:06 panto: software solution? Mar 29 19:29:13 ds2, there isn't one Mar 29 19:29:26 I can detect button presses Mar 29 19:29:28 <_av500_> is that a thing to have your avatar/icon look like bacon? Mar 29 19:29:32 ka6sox-farfarwa: that is my thought but I figure I'd ask in case someone else figured a way Mar 29 19:29:40 its too late Mar 29 19:30:09 I have an extra set of pinmux entries Mar 29 19:30:20 a latch + an opencollector buffer Mar 29 19:30:24 once system reset comes off then everything becomes active Mar 29 19:30:29 so what s/w wants to detect a press switches to the other pinmux entries Mar 29 19:30:38 the configuration of the pins doesn't change Mar 29 19:30:51 the uSD switch is on or off Mar 29 19:31:04 is it a pb? Mar 29 19:31:06 let me check Mar 29 19:31:15 so you can momentarily add switch the gpio-keys driver in and have it detect Mar 29 19:31:20 it's super hacky Mar 29 19:31:47 the PB needs to be a slide Mar 29 19:31:55 dynamically changing a pinmux is not hacky Mar 29 19:32:26 while the hdmi display is working? Mar 29 19:32:41 cause guess what? omap_device doesn't like to unregistered Mar 29 19:32:54 in fact no-one has ever tried to do that so far Mar 29 19:33:09 all this dynamic crap doesn't work. Mar 29 19:33:17 i been arguing that since the xM days Mar 29 19:33:21 seems to work Mar 29 19:33:28 are we done with eMMC now? Mar 29 19:33:34 it works for some cases but not all cases Mar 29 19:33:44 if you want to capture anything out of the hard core hardware hacker audience Mar 29 19:33:48 sure Mar 29 19:34:09 but you can't expect people coming from an arduino background to figure out how to modify the kernel Mar 29 19:34:17 scary Mar 29 19:34:28 they have trouble understanding how to access gpios via a driver ffs Mar 29 19:34:33 okay is s2 a pb or slide? Mar 29 19:34:42 (it looks like a push) Mar 29 19:34:47 push Mar 29 19:34:51 bzzzzt Mar 29 19:34:55 that is an education thing Mar 29 19:35:04 good luck trying to educate people Mar 29 19:35:04 if we expect people to be morons, they will remain morons Mar 29 19:35:18 most people _are_ morons Mar 29 19:35:21 we need to encourage folks to grow Mar 29 19:35:25 that should be a slide... Mar 29 19:35:28 folks that can't should be left behind Mar 29 19:35:33 otehrwise you have stagnation Mar 29 19:35:40 you can only let the one that can hack it stop being morons Mar 29 19:36:06 ds2, it's not a binary scale Mar 29 19:36:06 ka6sox-farfarawa: a latch would be more consistant Mar 29 19:36:16 it is a gradual scale Mar 29 19:36:27 panto: i agree but we are not encouraging people to grow Mar 29 19:36:31 a PB means you *have* to catch it at the right spot int eh boot Mar 29 19:36:33 I think we are Mar 29 19:36:45 panto +1 Mar 29 19:36:46 we can tell them: this is what you can do easily Mar 29 19:36:48 ka6sox-farfarawa: I think that is qualified by the REGEN signal Mar 29 19:36:59 this is what you can do if you put more effort in Mar 29 19:37:03 you are catering to the lowest folks Mar 29 19:37:13 aim higher, let them work on it and get a real archeivement Mar 29 19:37:16 we all were lowest folks at some time Mar 29 19:37:23 mru: that is a fun thread to read btw :) Mar 29 19:37:28 but we had things to strive toward Mar 29 19:37:31 we grew Mar 29 19:37:32 ds2, the node stuff is catering to the folks who can't/won't hack on a kernel Mar 29 19:38:07 anyways Mar 29 19:38:27 so what are we going to tell Gerald re: s2 and the emmc collisions that are bound to happen. Mar 29 19:38:48 ds2, you can't possible say that the old ways were better all the time Mar 29 19:38:49 it i F'ed and he has 2 weeks to respin board? :D Mar 29 19:38:51 hehehe Mar 29 19:39:20 death to all nodejs users! Mar 29 19:39:22 panto: not all the time... I am saying the new way encourages stupidity whereas we need to encourage growth Mar 29 19:39:28 sorry in a Friday mood Mar 29 19:39:42 flashing an LED is not a major achievement, it is just a checkbox along the way Mar 29 19:39:46 ds2, no, I'd say find a slide switch that fits the footprint and add the doco about which way to set it in the BoneSRM Mar 29 19:39:50 ds2, how can you encourage growth when you bust the user's balls all the time Mar 29 19:40:13 it's one thing for things being hard, because, yeah things are some times hard Mar 29 19:40:27 it's another to go out of your way to make things more cryptic and difficult Mar 29 19:40:27 ka6sox-farfarawa: I think that will still lead to fried boards Mar 29 19:40:52 panto: one grows beyond the ball busting Mar 29 19:41:08 one can find other things to do Mar 29 19:41:15 and we would lose a good engineer Mar 29 19:41:21 * ka6sox-farfarawa checks to see if teh capes are held in reset while the onboard. Mar 29 19:41:22 ka6sox-farfarawa: and since it is a momentary thing, the boards might die much much later Mar 29 19:41:44 the capes i have seen don't qualify I/O with reset :/ Mar 29 19:41:46 pinmux will be the DEATH of this... Mar 29 19:42:23 we could lose A good engineer but at least you can identify the ones that we don't lose Mar 29 19:42:33 no sorting process is perfect Mar 29 19:42:43 but putting everything on a stagnant level is just bad Mar 29 19:42:55 no mru as a mentor? Mar 29 19:43:25 okay cape bus has SYSTEMreset_n Mar 29 19:43:34 so that means we are gonna FRY. Mar 29 19:43:53 it is the "Black" for a good reason ;) Mar 29 19:43:56 * ds2 ducks Mar 29 19:44:25 I'm sorry...I've been focusing on other areas and see a major Stir Fry about to happen. Mar 29 19:44:43 a lot of "this )(&)#%&*)_& won't boot!" Mar 29 19:44:48 followed by Blue Smoke. Mar 29 19:45:23 all are doomed? Mar 29 19:45:31 let grab beer then Mar 29 19:45:36 beer is good Mar 29 19:45:48 and pringles Mar 29 19:46:05 I hope that eMMC is cheap Mar 29 19:46:18 if you buy enough at a time, they are cheap Mar 29 19:46:32 lessee... Mar 29 19:46:45 but you are assuming the eMMC is what frys Mar 29 19:46:50 is the driver an OC? Mar 29 19:46:51 not the random capes Mar 29 19:46:59 some pins are OC Mar 29 19:47:02 but not all Mar 29 19:47:22 the older specs allows you to stack MMC cards onto a single controller. that is current deprecated but is retained Mar 29 19:47:26 no, I'm asking about the driver for holding the eMMC in reset Mar 29 19:48:48 re prpplague Mar 29 19:49:05 its a hardware hack to make that work... Mar 29 19:49:14 but its open drain Mar 29 19:49:40 what driver do you mean in this context? Mar 29 19:49:58 woglinde: greetings Mar 29 19:49:59 driver as in a software or driver as in a totem pole/OC/OD/etc? Mar 29 19:50:00 SN74LVC1G06 Mar 29 19:50:39 the real solution is to separate out systemreset_n and create a bus reset line that is separate Mar 29 19:50:50 that way you enable the bus *after* you set things up. Mar 29 19:51:12 I'd suggest a different scheme Mar 29 19:51:27 can we do it without a hardware hack? Mar 29 19:51:30 put the emmc behind a buffer/switch... have that switch be controlled by the latched sysboot state Mar 29 19:51:39 too late Mar 29 19:51:41 no. all I can think of uses HW Mar 29 19:51:49 me too. Mar 29 19:52:08 seperating out reset and requiring it to be qualified before an expansion board drives is at best causing compatibility problems Mar 29 19:52:12 if the bus reset were separate (like ISA/PCI/etc) then we could Mar 29 19:52:18 at works, a bigger abomination Mar 29 19:52:36 but for bus reset to matter, you still need to ask all expansion boards to qualify their outputs with that reset line Mar 29 19:52:57 I'm saying *replace* the line with a different one Mar 29 19:53:03 not additional Mar 29 19:53:19 you mean have the boards drive that ? Mar 29 19:53:31 or you mean use that to control if the board drives the conflicting lines or not? Mar 29 19:53:39 no have the processor take a GPIO and turn it into a bus_reset Mar 29 19:53:49 what would that do? Mar 29 19:53:52 but again thats not going to happen Mar 29 19:54:31 the problem is the emmc can crash with a cape. Mar 29 19:54:50 or vice versa Mar 29 19:54:55 there is no way for it to be used then disabled before the bus becomes active Mar 29 19:55:12 DNP'ing the eMMC is probally the safest thing to do given the current schedule Mar 29 19:56:04 ka6sox-farfaraw: I think we are looking at 2 variants of hte problem. I am worried about hard outputs fighting and to solve that with a bus reset means the bus needs to not drive unless it is cleared by the reset Mar 29 19:56:20 bus_reset Mar 29 19:56:40 not system_reset Mar 29 19:56:50 yes Mar 29 19:57:05 but most capes I have seen drive the pins pretty much immediately Mar 29 19:57:08 that again brings up 2 issues Mar 29 19:57:20 1 the hardware doesn't support it Mar 29 19:57:26 time to go Mar 29 19:57:31 for that to mean anything - you must buffer and qualify that with the bus_reset Mar 29 19:57:33 2 the software doesn't support it. Mar 29 19:57:39 goodnight fellow nerds/trolls/assorted creatures Mar 29 19:57:46 so conclusion - we are SOL? Mar 29 19:57:51 later panto Mar 29 19:57:59 bbl, I'm getting food Mar 29 20:22:55 koen: so the default pinmuxes for SPI have been removed from bone-common.dtsi? Mar 29 20:24:16 so, did we get to the bottom of the black debarcle Mar 29 20:24:33 mrpackethead: yes Mar 29 20:24:58 mrpackethead: http://www.raspberrypi.org/ Mar 29 20:24:58 so, all exisiting capes will work flawlessly with the blackbone Mar 29 20:25:15 becuase, clearly, the whitebone will get dropped from production Mar 29 20:25:26 its going to cost them a bunch less to build Mar 29 20:26:09 mrpackethead: bonewhite won't get dropped soon Mar 29 20:26:15 so, they say Mar 29 20:26:32 yes, so we say Mar 29 20:26:40 that's where the margins are :) Mar 29 20:27:01 but you'd not want to be developing a new product based on the bone, and not actually know with some certainly whats happening Mar 29 20:27:27 welcome to the real world Mar 29 20:27:31 mrpackethead: you shouldn't be developing a product based on the bone Mar 29 20:27:33 it's a dev board Mar 29 20:27:36 community supported Mar 29 20:27:38 we're stilling building beagleboard classics Mar 29 20:27:44 that's a board from 2008 Mar 29 20:27:49 if you want futureproof, get a white-label from CCo Mar 29 20:28:17 alan_o: white label is just a bone, with no label, thats sematics Mar 29 20:28:25 and yes, that is what i woudl do. Mar 29 20:28:27 mrpackethead: no it's not Mar 29 20:28:42 so, its different? becuase CCO told me it was identical Mar 29 20:28:51 minus the label Mar 29 20:29:08 mrpackethead: it's a design that you control. Want more of _your_ bone at some time in the future? they'll make you _that_ design Mar 29 20:29:17 right now, it's the same minus the label Mar 29 20:29:51 get the difference? It's not about now, it's about the future, and availability at a future time. Mar 29 20:30:09 and in which case, i might as well own the design and build it ourselves. Mar 29 20:30:25 be prepared to $$$ then Mar 29 20:30:28 that's the tradeoff Mar 29 20:30:36 we build lots of electronics Mar 29 20:30:38 need a million of them? Don't use a bone. Mar 29 20:30:49 i know what it costs to build small run electroncis Mar 29 20:30:50 need a few hundred over the next few years? get a white-label. Mar 29 20:30:56 and i know its not cheap Mar 29 20:31:12 so then why the lamentation? Mar 29 20:31:14 the parts and pcb and assmbly are cheap Mar 29 20:31:21 time is expensive Mar 29 20:33:04 beuase theres a supposbly cheaper bone, thats higher powered and more featured thats just aroudn teh corner, but theres no certainity of the compatiblity Mar 29 20:33:43 its just a lot of "unknowns" Mar 29 20:34:00 !identical => !compatible Mar 29 20:34:09 koen, was the idea of the eMMC to be the rootfs after boot? Mar 29 20:39:27 alan_o: as you say, it the real world, and this is just one of the challenges we face every day Mar 29 20:39:39 and we dont' loose too much sleep over it. Mar 29 20:39:49 I'm looking forward to meeting at CCO face to face Mar 29 20:39:56 and then we can probably figure it out Mar 29 20:39:57 :-) Mar 29 20:40:24 cool. I'm glad you're not bitter :) Mar 29 20:40:34 not bitter.. its just a "challenge" Mar 29 20:52:32 koen: still there? Mar 29 20:52:39 koen: how do you want patches submitted? Mar 29 20:53:09 so there are patch files to the kernel, patches to the repo (which contains patch files) Mar 29 20:53:19 koen: put my patch in a patch? Mar 29 20:53:26 koen: or just send the patch Mar 29 20:53:27 :) Mar 29 20:56:15 hmm looks like both ways are in the log Mar 29 20:57:32 alan_o, ya, noticed that...its a bit confusing Mar 29 21:01:58 ok, patch of patch it is Mar 29 21:02:01 didn't take long Mar 29 21:02:54 I just hate those sometimes if hte number of lines change its a bear Mar 29 21:13:45 as I sit here and wait for patch.sh to run, I can't help but wonder if this is getting (or has gotten out of hand). It's starting to bother me that TI doesn't care enough about the am33xx to have upstream support for it. Mar 29 21:14:23 alan_o, I'm not sure that is the case...2/3 of the GSoC stuff is about pushing things upstream. Mar 29 21:14:43 koen works his butt off on this tree, and I feel like he shouldn't have to. Board stuff, sure, but why does he have to maintain patches for basic SoC functionality? Mar 29 21:15:01 ka6sox-farfarawa: yeah, let's hope some of it can happen Mar 29 21:15:18 ka6sox-farfarawa: but seriously, we have to have summer GSoC interns to push am33x stuff into mainline? Mar 29 21:15:32 I don't think its the case that nobody cares..its the same ole tired issue...not enough time/resources. Mar 29 21:22:49 4th of July celeberations do they do that on the evening of the 4th of July, or on teh eve of 4th july? Mar 29 21:23:49 evening Mar 29 21:24:03 (most) Mar 29 21:24:18 looks like i will be in LA on 4th July Mar 29 21:24:25 last year...ours went up within 3.2 seconds of each other. Mar 29 21:24:30 i'm tempted to stay and see the crazynes Mar 29 21:24:36 ah i was wondering what that random question was about Mar 29 21:24:38 do so...I'l be there! Mar 29 21:28:44 * mranostay does the ds2 dance Mar 29 21:40:54 alan_o: github pull requests are the easiest, but patches or patches to patches are fine as well Mar 29 21:43:12 ok, i give up Mar 29 21:43:19 alan_o: omap^H^H^H^Ham335x totally f@#$@ed :) Mar 29 21:43:21 i'll talk to the Travel agent on Tuesday Mar 29 21:43:21 mrpackethead oh bo Mar 29 21:43:26 :-( Mar 29 21:43:27 lol Mar 29 21:44:32 WLG-AKL-HGK-FRA-DRS-SXB-BRU-LHR-IAD-ORL-DFW-SFO-SAN-LAX-AKL-WLG is too hard to book online Mar 29 21:46:29 what? Mar 29 21:46:36 sounds like madness Mar 29 21:47:01 yup, thats a met 50% of my top clients trip Mar 29 21:48:49 how long of a trip? Mar 29 21:49:00 mrpackethead: some of those airport codes are wrong Mar 29 21:49:20 eg john wayne is SNA Mar 29 21:49:20 like? Mar 29 21:49:31 SNA then. Mar 29 21:49:50 what's ORL supposed to be? Mar 29 21:49:51 flying from burbank to lax? Mar 29 21:49:57 orlando probably Mar 29 21:50:04 its MCO Mar 29 21:50:05 that's MCO Mar 29 21:51:19 WLG-AKL-HGK-FRA-DRS-SXB-BRU-LHR-IAD-MCO-DFW-SFO-SNA-LAX-AKL-WLG then Mar 29 21:51:47 Theres some dammed awful routing Mar 29 21:52:05 You can fly LHR-MCO direclty Mar 29 21:52:18 but to get it all on one ticket Mar 29 21:52:22 is hard Mar 29 21:52:39 really? I only know of GTW-MCO Mar 29 21:53:02 Yeah, it might be GTW Mar 29 21:53:11 Virgin Mar 29 21:53:21 i think fly LHR-MCO Mar 29 21:53:25 BA and virgin do that route Mar 29 21:53:31 it all becomes a Blur. Mar 29 21:53:34 gte Mar 29 21:53:37 gtw Mar 29 21:53:43 lol. Mar 29 21:53:45 is that BOAC? Mar 29 21:54:06 in some form, yes Mar 29 21:54:09 Gatwick is half way to the equator! Mar 29 21:54:23 merged with a few others Mar 29 21:54:27 its a mision to get back into the City Mar 29 21:54:33 mrpackethead, take the train from Victoria Mar 29 21:54:38 yes, you can Mar 29 21:54:43 and i do Mar 29 21:54:43 there's a direct train Mar 29 21:54:45 :-) Mar 29 21:55:04 The other ugly one, is Orlando to Dallas Mar 29 21:55:11 not many (any) direct flights Mar 29 21:57:59 mranostay: DFW? Mar 29 21:58:10 Dallas / Fort worth Mar 29 21:58:36 can the bacon cape, book travel Mar 29 21:58:37 ? Mar 29 22:05:51 alan_o: you never fly? :) Mar 29 22:05:58 even i know DFW Mar 29 22:07:55 no Mar 29 22:07:58 I mean... Mar 29 22:08:03 I know what DFW is Mar 29 22:08:12 mrpackethead: said he couldn't get into Dallas Mar 29 22:08:25 I wanted to make sure he was looking at DFW and not DAL (Love Field) Mar 29 22:08:33 you can get there Mar 29 22:08:36 just not direclty Mar 29 22:08:57 yeah, that's what I meant Mar 29 22:09:45 I'm pretty sure I've flown MCO->DFW direct on AirTran Mar 29 22:10:28 yep, just checked, on 12/31/05 I flew nonstop MCO->DFW on AirTran Mar 29 22:10:35 but that's been a long time :) Mar 29 22:15:01 old people in here Mar 29 22:15:44 yeah, you were still scared of girls back in 05 :) Mar 29 22:16:18 were? Mar 29 22:16:23 hehe Mar 29 22:20:24 hmmm Mar 29 22:20:57 mmmh Mar 29 22:28:56 * mranostay pokes mdp Mar 29 22:29:04 * alan_o pokes mdp Mar 29 22:30:25 * Shadyman pokes mdp Mar 29 22:31:19 oh you guys Mar 29 22:32:34 * alan_o pokes mranostay Mar 29 22:32:39 where would one get a 6502 chip? Mar 29 22:32:50 at the chip shop Mar 29 22:32:59 from an old 1541? Mar 29 22:33:03 mranostay: that old dead architecture? Only scrubs and insurance salesmen use 6502... Mar 29 22:34:44 mranostay: (I assume we're trying to get a rise out of mdp, but in case your'e not, he gets his from http://www.westerndesigncenter.com/wdc/chips.cfm) Mar 29 22:35:19 hey i'm serious at times Mar 29 22:35:44 heh this website even has a 1970's feel Mar 29 22:35:49 wikipedia says mranostay was serious in 1974, 1986 and again in 2003 Mar 29 22:35:55 oooh.. they'll even sell you fpga cores Mar 29 22:37:57 only $7 from mouser Mar 29 22:38:31 "only" Mar 29 22:39:09 alan_o, and I will need some 4116's too Mar 29 22:39:17 or 4164's Mar 29 22:40:28 jameco says they have those for $0.89/ea too... Mar 29 22:41:11 so 8 of those puppies Mar 29 22:42:21 mrpackethead: 1974? that wasn't likely Mar 29 22:50:56 alan_o: or i can just steal from ka6sox-farfarawa's garage :P Mar 29 22:51:14 That's what it's sounding like :) Mar 29 22:52:30 ka6sox-farfarawa: so what time you usually go out in the evenings? :) Mar 29 22:52:48 I leave a guard kitten... Mar 29 22:54:34 hmmm soviet's never cloned the 6502..odd Mar 29 22:54:51 lol. Mar 29 22:56:30 no my my Russian Girlfriend still makes Empty State devices! Mar 29 23:05:02 heh Mar 29 23:12:00 mranostay, you haven't met her...you can't say that... Mar 29 23:24:32 what 'heh'? Mar 30 01:24:30 Hi All, sorry the noob question but, I'm trying to see the output of the xm trough the rs232 115200 8N1 but all I can see is junk (yes there is something going out). Any possibilities that the baudrate is diferent than 115200 8n1 ?? Mar 30 01:25:36 nope Mar 30 01:26:21 do you get a continuous strream or does it stop after a second or two? Mar 30 01:29:02 it stops Mar 30 01:30:34 could be your MLO version Mar 30 01:31:21 no problem if I try without SD right ?? it should display something even without SD right ? Mar 30 01:31:40 no, the xM has no NAND Mar 30 01:33:04 withou sd it shows diferent message, but still junk Mar 30 01:33:46 but there is a 60 in the middle. I have read something about it, maybe something about trying to load sd, can remember Mar 30 01:34:26 So do you recommend prepare the SD with a new image ?? MLO is on sd right ? Mar 30 01:37:41 morning Mar 30 01:37:44 Solved. Problem was the SD Mar 30 01:37:48 thanks Mar 30 01:41:11 tis done. I have a ticket booked Mar 30 01:41:15 geepers how hard was that Mar 30 01:44:33 mrpackethead: how long of a trip is that/ Mar 30 01:44:38 mranostay: i'm coming to San Fran, so i'm going to come to troll you Mar 30 01:44:59 leave home 7 June, get back 7 July Mar 30 01:45:44 decided to stay for July 4th fireworks Mar 30 01:46:39 how trouble is turning beagleboard off without shuting down by software. I mean argstrom or ubuntu will corrupt files and something like that... or they are ready for that ?? Mar 30 01:48:26 heh Mar 30 01:48:33 that is a filesystem issue Mar 30 01:49:05 ubuntu is classified in a list with a nother well known name here.. Mar 30 01:49:13 mrpackethead: staying in SF or just the area? Mar 30 01:49:38 not sure yet Mar 30 01:49:43 but probably will stay downtown Mar 30 01:49:55 or fishermans wharf Mar 30 01:50:20 that is a tourist trap Mar 30 01:50:31 STAY OUT OF TENDERLOIN! Mar 30 01:50:47 it is where the real fun is! Mar 30 01:51:41 I would like to know if the embended distributions are "shielded" or better protective in cases of powering down without shutting down the system... Mar 30 01:51:55 noob123: depends on how you configure it Mar 30 01:52:47 noob123: i use debian, and a ext4 filesystem which is jornelled Mar 30 01:52:54 it saves you from most accidents Mar 30 02:00:16 helps if you sync after writing critical data... slows things down though Mar 30 02:01:44 hum, good info. thanks Mar 30 02:02:04 or just keep your filesystem read-only if you can Mar 30 02:02:23 a bunch of parts to make on the printer Mar 30 02:02:23 remount it RW to write, then sync, then remount RO Mar 30 02:19:12 I think I'm ready to send a few boards to the PCB house Mar 30 02:19:45 s/boards/capes Mar 30 02:43:18 3 capes, 10 each, $75 later... Mar 30 02:53:38 what capes are you developing ?? **** ENDING LOGGING AT Sat Mar 30 02:59:59 2013