**** BEGIN LOGGING AT Wed Jun 04 02:59:58 2014 Jun 04 03:06:01 jlusby: the schematics for the WL1835MOD_w/_Chip_Antenna cape show unpopulated external antenna connectors, it might be possible to mod it for your purposes Jun 04 03:06:55 jlusby: might be worth sending an email to circuitco asking Jun 04 03:08:37 alright, cool ty smurray Jun 04 03:09:18 no problem, good luck with your project Jun 04 04:08:06 So, is the RevC actually shipping from anywhere yet? Jun 04 04:08:47 Or did they just ship ever last unit they had all at once the first go-around? Jun 04 04:23:19 i have an issue, every time when plug in the beaglebone in the computer whitout electric blinks for like 3 seconds ant then turns off , any solution? Jun 04 08:44:41 What is the status of mainlining the kernel drivers of BBB? Is there somewhere an overview? Jun 04 09:02:17 markusr: you could probably ask on the beagle list Jun 04 09:02:32 but recent kernels have been getting a lot of good things AFAIU Jun 04 09:02:49 someone even mumbled something about working musb hotplug Jun 04 09:03:27 tbr lol Jun 04 09:05:27 ok. I will ask the mailing list Jun 04 09:28:01 http://arxiv.org/pdf/1404.1499v2.pdf Jun 04 12:14:04 hi, I'm having trouble using gdbserver under my BBB running debian. I always get the message that my gdbserver is not compiled with python support. Jun 04 12:14:45 where can I find an explanation to download/cross-compile gdbserver for debian armhf? Jun 04 12:15:30 does it need python? Jun 04 12:16:39 qtcreator debbugger says so... Jun 04 12:17:42 thts the message from qtcreator: "The selected build of GDB does not support Python scripting. It cannot be used in Qt Creator." Jun 04 12:18:28 I already searched alot for this so I decided to compile my own gdbserver Jun 04 12:19:12 but I'm wondering how the configure statement is to crosscompile gdbserver using the linaro toolchain Jun 04 12:20:01 btw. i installed gdbserver on BBB using apt-get install from debian archives Jun 04 12:21:18 so you have gdbsever on the bbb? Jun 04 12:22:14 yes Jun 04 12:22:35 apt-get install gdbserver Jun 04 12:23:23 it tells me it's version 7.4.1 Jun 04 12:23:38 7.4.1-debian Jun 04 12:24:11 so, on the BB side you should be fine, no? Jun 04 12:24:52 i think on bbb side it should work, but it's missing python scripting support Jun 04 12:25:35 this error message is on the PC? Jun 04 12:25:37 no? Jun 04 12:25:39 I downloaded gdb sourcec 7.4.1 and tried to configure: ./configure --host=arm-linux-gnueabi --prefix=/usr/local/gdbserver-python --with-python Jun 04 12:25:55 yess error message from my debian vm Jun 04 12:25:58 so Jun 04 12:26:08 you have a debian problem on your PC Jun 04 12:26:11 not on the BBB Jun 04 12:26:13 no? Jun 04 12:26:25 the configure output says :: checking whether we are cross compiling... no Jun 04 12:26:48 configure: WARNING: unrecognized options: --with-python Jun 04 12:26:53 can you start gdbserver on BBB or not? Jun 04 12:27:27 yes I can start gdbserver on BBB Jun 04 12:28:00 so why do you want to cross compile it? Jun 04 12:28:35 I want to compile it because it seems it has no python scripting support Jun 04 12:28:47 btw. python is installed on bbb too Jun 04 12:28:52 I think scripting is on the client side Jun 04 12:28:55 on your PC Jun 04 12:29:03 where you run your QT foo Jun 04 12:29:39 qtcreator is invoking it, so I cant really tell where exactly its located Jun 04 12:30:14 I'm using a debian wheezy vm running qtcreator Jun 04 12:31:27 http://qt-project.org/forums/viewthread/42848 Jun 04 12:35:45 OK, that read I try to compile gdb with python. Maybe you are right and the problem is on my HOST. -- I'll be back. Jun 04 12:38:45 Question: with Debian, using only the USB connector, is the BBB running at a lower clock speed than 1GHz? Jun 04 12:41:31 Peanut: check with "cpufreq" on the shell Jun 04 12:42:05 If it isn't 1 GHz, you can force it with the same command Jun 04 12:42:14 Abhishek_: thanks - I don't have the machine on my at the moment, trying to decide if I need to buy a 5V PSU or not. Jun 04 12:42:36 Peanut: no Jun 04 12:42:51 bbb will run full speed on usb Jun 04 12:43:10 nyt: thanks Jun 04 12:43:19 Peanut: see http://paste.debian.net/103373/ Jun 04 12:43:26 should be 1GHz on demand Jun 04 12:44:39 Letothe2nd: I'll give that a try as soon as I am home then. Jun 04 12:55:42 jkridner: im still having issues with davinci-evm.c not compiling in module mode... ive tried to add definition for struct it says is missing but getting similar errors Jun 04 12:57:44 jkridner: i also need to know what are the differences between 3.8 and 3.14 please, saying one is better for capes and other has better usb/enet doesnt tell me enough Jun 04 12:59:18 hi dogrocket Jun 04 12:59:27 heyo Jun 04 12:59:35 * jkridner just finished building the kernel Jun 04 12:59:53 was going to look at your defconfig now. Jun 04 13:00:02 what's the main thing you were trying to turn on? Jun 04 13:00:04 ok... whats prognosis doc? Jun 04 13:01:17 i need to get alsa - soc working so i can talk to a codec with i2s, was told by soc guy because we dont know how things are gonna work out yet we should do all alsa-soc as modules in case we need to fix things Jun 04 13:02:10 so i just turned everything soc audio to modules, and its breakin :P Jun 04 13:04:23 i believe i even saw an error with soc in non-module mode but it wasnt enough to break build Jun 04 13:04:34 dogrocket: yep, turning stuff to a module breaks because those #defines are "blank" when the stuff compiled in and are resolved only when compiling as modules Jun 04 13:05:35 yeah, some of that stuff is broken for as-modules. Jun 04 13:07:12 well i at least need to do codec drivers as modules, and plug codec in and see what happens.. Jun 04 13:07:45 I was able to add additional drivers as modules. Jun 04 13:09:07 http://paste.ubuntu.com/7587380 built fine for me Jun 04 13:09:08 ok... will try it Jun 04 13:11:20 * jkridner invokes 'make modules' (silly not-so-kernel-hacker) Jun 04 13:15:02 any way to restart build from where it broke? Jun 04 13:15:18 kernel-hacker :P Jun 04 13:15:45 oh... nvm, duh gotta change config Jun 04 13:16:04 if you change that you won't be able to resume then Jun 04 13:16:37 yepper... fixed file, starting build over Jun 04 13:17:42 shouldn't have to start over for a minor config change Jun 04 13:18:13 the make system should be smart enough to figure out what has changed and rebuild those portions only Jun 04 14:23:56 hi all! :) how are you? :) I have a question. I have to implement my own OS (bare metal). I have configured my MMU and currently I am trying to implement the ethernet driver using Lwip (like in the StarterWare example). My problem is, that I get a dabt-interrupt, when I try to read a variable. But the behavior is very strange, because when I read the value using CCSv5 Expression view there is no problem. I can see the value clearly. Jun 04 14:24:21 Does anybody has a clue about why I can read the variable using the expression view but I cannot access it in code? Jun 04 14:42:36 properly word-aligned? Jun 04 14:52:41 @pjustice do you mean me? Jun 04 14:53:45 both values are of the same type - so it shouldnt be a problem I guess Jun 04 14:59:39 Well, data aborts apparently occur for three reasons. One is a write scenario, so not relevant. The other two are unaligned access, and access to a reserved area. Knowing that CCS EV is kinda useful, in that you know it can be done, but only kinda, because you don't know _how_ it's doing it. Therefore, I'd guess that either you've arranged for an unaligned access, or that the address you _think_ points to ... Jun 04 14:59:45 ... your variable is wrong, or got damaged by some other code bug Jun 04 15:13:05 okay - thanks for the info Jun 04 15:13:09 i will check that Jun 04 15:24:39 hi Jun 04 15:27:17 hi Jun 04 15:28:10 ok so i have a project due this week and just need help starting it. Jun 04 15:28:36 You will want to use cherrypy framework for the beaglebone server. www.cherrypy.org So here are the tasks: 1. Generate an HTML "GUI" to control the VDB (TI board) 2. Implement the CherryPy server 3. Implement the SPI communication protocol Jun 04 15:30:07 * Letothe2nd really loves the sound of deadlines swooshing by! Jun 04 15:30:16 wow, kids of today have so much fun at school! Jun 04 15:33:58 dogrocket: http://paste.ubuntu.com/7588249/ build all modules for me just fine Jun 04 15:34:21 ok.. checking it out Jun 04 15:35:01 Guest81317: honestly - break down the project in parts you can handle. start with the first one, and once you hit a road block, ask again about the specific problem. nobody here will handhold you for your "project" Jun 04 15:36:31 ok, so the part that i would really need help is implementing the SPI Jun 04 15:37:07 Guest81317: there is the spidev driver, that you can poke directly from userspace. little magic there. Jun 04 15:38:21 Guest81317: see http://www.armbedded.eu/node/318, starting from "Using the SPI User Mode Device Driver" Jun 04 15:39:42 I just made the connections with the beaglebone, (BB pins) 28 - 9 (AArdvark I2C/SPI) - (BB pins) 29- 5 (AArdvark I2C/SPI) -(BB pins) - 30-8 (AArdvark I2C/SPI) - (BB pins) 31 -7 (AArdvark I2C/SPI) 1-2 and 3-6 Jun 04 15:42:02 jkridner: man im not having any luck... is this because i disabled all capes but left cape management enabled? Jun 04 15:42:08 drivers/built-in.o: In function `bone_iio_helper_remove': Jun 04 15:42:08 :(.text+0x6c82a): undefined reference to `iio_channel_release' Jun 04 15:42:08 drivers/built-in.o: In function `bone_iio_helper_probe': Jun 04 15:42:08 :(.text+0x6c90a): undefined reference to `iio_channel_get' Jun 04 15:42:08 :(.text+0x6c992): undefined reference to `iio_channel_release' Jun 04 15:42:08 drivers/built-in.o: In function `iio_helper_show_mvolts': Jun 04 15:42:10 :(.text+0x6ca9e): undefined reference to `iio_read_channel_raw' Jun 04 15:43:05 it was at last step... :/ Jun 04 15:43:26 LINK vmlinux Jun 04 15:43:27 LD vmlinux.o Jun 04 15:43:27 MODPOST vmlinux.o Jun 04 15:43:27 WARNING: modpost: Found 1 section mismatch(es). Jun 04 15:43:27 To see full details build your kernel with: Jun 04 15:43:27 'make CONFIG_DEBUG_SECTION_MISMATCH=y' Jun 04 15:43:31 GEN .version Jun 04 15:43:33 CHK include/generated/compile.h Jun 04 15:43:35 UPD include/generated/compile.h Jun 04 15:43:37 CC init/version.o Jun 04 15:43:39 LD init/built-in.o Jun 04 15:43:41 drivers/built-in.o: In function `bone_iio_helper_remove': Jun 04 15:43:47 :(.text+0x6c82a): undefined reference to `iio_channel_release' Jun 04 15:44:20 Guest81317: well then mux/enable spi first, and then be inspired by the above link Jun 04 15:48:22 To be honest , i started working on beagelbone 5 days ago, for the first time, same with python which is the language i have to use. And right now I am really confused Jun 04 15:55:36 Guest81317: then maybe complete everything but the spi part first. Jun 04 15:56:12 Guest81317: besides, http://elinux.org/BeagleBone_Black_Enable_SPIDEV looks pretty good. Jun 04 16:01:08 * zmatt sighs at TI dude on e2e acting all secretive about the crypto accelerators Jun 04 16:05:04 dogrocket: http://jkridner.wordpress.com/2014/06/04/yet-another-set-of-notes-on-building-beaglebone-kernel/ Jun 04 16:06:16 dogrocket: can you start over again and only make 1 defconfig change at a time or minimize the defconfig changes? Jun 04 16:07:03 you might or might not be running into real bugs (likely are), but don't necessarily need to be running into them and reporting them will be easier if you change fewer things at once. Jun 04 16:14:22 CAN YOU GIVE ME YOUR E-MAIL ADDRESS, IF I HAVE ANY OTHER QUESTIONS PLEASE Jun 04 16:15:47 jkridnet: ok.. im going to use your paste, and do i simply copy over defconfig and not .config before build? Jun 04 16:16:18 jkrider: rather Jun 04 16:16:27 lol... aah Jun 04 16:18:07 jkridner: i reloaded every thing from square 1. uboot , device trees... etc gonna try this again Jun 04 16:31:38 jkridner: doing build, i only removed multimedia devices from your config Jun 04 16:50:25 wow its terribly/surprisingly difficult to get hold of a BBB Jun 04 16:50:34 * SiC wishes he bought one a couple of months ago when you could Jun 04 16:50:51 how many units have been sold so far on the BBB? Jun 04 16:52:13 Guest81317: certainly no. come back here and ask. Jun 04 16:52:43 jkridner: ok build went ok, if i need to make some changes can i just do make command? Jun 04 16:53:03 SiC: Eight billion. Jun 04 16:53:18 SiC: And you still can, just order one and wait. Like the topic says. Jun 04 16:54:20 so theres more BBB than 1pp in the world? ;) Jun 04 16:54:32 but yea, I've just put in my order Jun 04 16:54:39 one more to add to the backlog :p Jun 04 16:55:27 dogrocket: you need to include the ARCH= etc. arguments after editing .config. Best to use 'menuconfig' or 'xconfig' or something like that than to edit by hand. Jun 04 16:55:59 ok... i see your notes now... thank you Jun 04 16:56:37 SiC: The backlog is not as deep as people think. People at my hackerspace have been getting theirs in reasonably quickly. Jun 04 16:56:46 yea I did wonder that Jun 04 16:56:47 http://cpc.farnell.com/circuitco/bb-bblk-000/beaglebone-black-cortex-a8-dev/dp/SC13013 || http://cpc.farnell.com/element14/bbone-black/beaglebone-black-cortex-a8-sbc/dp/SC13391 Jun 04 16:56:51 whats the difference between the two? Jun 04 16:56:57 is it just the manufacturer? Jun 04 16:57:09 I saw on the blog that E14 is manu them now? Jun 04 16:57:23 Yeah, neither of those links work. Jun 04 16:57:31 oh ok srory Jun 04 16:57:33 Oh...speaking of hackerspaces. Jun 04 16:57:37 initially I read that the element 14 board was for asian markets only, is that still true? Jun 04 16:57:44 both work for me Jun 04 16:58:15 the e14 board is a AM3358 vs the other which is AM3359 Jun 04 16:58:18 jkridner: Are you still working on that 3D printer project? Do you still need assistance? I might be able to help in the coming weeks if I get a day or so notice. ;-P Jun 04 16:58:42 SiC: Well, I think you just answered your own question. ;-P Jun 04 16:59:04 whats the difference between them? Jun 04 16:59:13 agmlego: yes. Unfortunately, I'm on travel all next week. The week after would be good to sit down at i3 and hack through it, if you have some time. Jun 04 16:59:20 spec looks pretty much identical Jun 04 16:59:31 SiC: YOu just said, one uses an AM3358 and one uses an AM3359. Jun 04 16:59:48 do both support gles? Jun 04 17:00:38 jkridner: So, weekend of the 14th or so? Jun 04 17:01:03 thurgood: I am sure that is in their respective datasheets. Jun 04 17:01:16 agmlego: so whats the difference apart from model numbers? Jun 04 17:01:20 their specs look identical Jun 04 17:01:41 SiC: There must be a difference, or there would not be a change in model. Read the datasheets. Jun 04 17:01:48 unhelpful :/ Jun 04 17:02:17 SiC: Never said I was going to give you an answer, especially when it is one I do not myself know. Jun 04 17:02:26 SiC: THis channel is all about the self-help. Jun 04 17:02:26 don't you think I've had a flick through the datasheets already? Jun 04 17:02:43 SiC: Basedf on 99% of the people who come in this channel and ask for help? No. Jun 04 17:03:04 agmlego: I get back on Father's day... so during that week or the next weekend. Jun 04 17:03:39 well instead of saying go RTM, it's common curtosy to either say don't know, or not respond. as we all know, smart-arse answers only get peoples backs up ): Jun 04 17:03:57 SiC: Has not really been an issue in this channel, to be honest. Jun 04 17:04:08 jkridner: Mmm, OK. Jun 04 17:04:14 jkridner: if following roberts eewiki instructions this works for remake make -j8 ARCH=arm CROSS_COMPILE=${CC} Jun 04 17:04:25 hardly welcoming for newcomers tbh Jun 04 17:04:46 jkridner: So the weekend of the 21th, then, unless I have some free time during the week preceeding. Jun 04 17:04:57 SiC: Is that the goal of this channel? Jun 04 17:05:59 what, to be not friendly, welcoming and helpful to newcomers? Jun 04 17:06:04 anyway this is a silly argument Jun 04 17:06:23 going back to the original point, the datasheets don't make it clear what the differences are Jun 04 17:06:39 thanks agmlego Jun 04 17:06:45 SiC: Then it probably does not matter. Jun 04 17:06:49 jkridner: No problem. Jun 04 17:06:55 the AM3359/AM3358 are all listed on the same datasheet, but with no clear differences Jun 04 17:07:33 surely if TI label their silicon under different part no., there must be something in there Jun 04 17:07:47 SiC: I think the only notable difference is EtherCAT support. Jun 04 17:07:55 ah ok thanks jkridner Jun 04 17:08:02 actual difference is in ability of CAN network usage, new chip has it removed Jun 04 17:08:05 SiC: So, the first result on Google for "AM3359 vs AM3358" is http://e2e.ti.com/support/arm/sitara_arm/f/791/t/221700.aspx Jun 04 17:08:07 And, I don't think there is a 1GHz version of AM3359 Jun 04 17:08:16 no other changes Jun 04 17:08:26 on the MSP430 datasheets, they usually have a nice little table of the differences between parts in the family Jun 04 17:08:30 SiC: That link states that the AM3358 has multiple speed grades, while the AM3359 has one speed grade and EtherCAT. Jun 04 17:09:51 SiC: So, yes, I get a little bristly when I, having no knowledge of the subject, can find the answer in literally 17 keystrokes and a mouse click. Jun 04 17:10:02 agmlego: sorry meant Ethercat, your right.. Jun 04 17:10:17 dogrocket: No worries, I literally just looked it up. Jun 04 17:17:37 http://elinux.org/Beagleboard:BeagleBoneBlack#Board_Revisions_and_Changes Jun 04 17:17:50 thats what I'm looking for Jun 04 17:17:58 "This version moves to the AM3358BZCZ100 processor as we are no longer able to get the limited production version of the AM3359AZCZ100. No changes in features or operation of the board resulted from this change." Jun 04 17:18:32 on a Rev B board Jun 04 17:18:38 (incase anyone else is following along) Jun 04 17:47:49 Hmm, AM3359 is listed as being an 800mhz part... and BBB revC is running it at 1ghz, or am I missing something? Jun 04 17:52:22 that's XAM3359AZCZ , the "X" was a limited edition part and they switched to AM3358 for Rev B from A6A Jun 04 17:54:42 ah Jun 04 17:58:27 Oh, wait, they switched FROM the AM3359 to the AM3358. Jun 04 17:59:26 Either way, one of the two parts doesn't have a 1ghz version, yet it was/is being run at 1ghz? Jun 04 18:02:27 I must be screwing something up, revB on adafruit says 3359, revc 3358, both are 1ghz, but it would make sense if the A6A was 800mhz and a 3359 and both RevB and RevC are 3358 and 1ghz, with the eMMC being the B->C revision... Jun 04 18:03:19 (As the A6A is listed as being a 700mhz board it seems according to google) Jun 04 18:15:58 AE7IK: maybe they qualified the part they are using for 1GHz Jun 04 18:23:25 are you confusing the beaglebone with the beaglebone black ? Jun 04 18:30:26 sometimes my own stupidity makes me wonder how i got as far as i did lol Jun 04 18:31:13 i just sat scratching my head trying to figure out why my beaglebone was not booting of sd card when i suddenly realised i didnt set the boot flag on the boot partition lol Jun 04 18:40:28 partition boot flags aren't really relevent to most operating systems... Jun 04 18:41:21 except that ROMBL checks it IIRC Jun 04 18:41:21 <_av500_> they are to the TI ROM code Jun 04 18:41:29 * tbr high-fives _av500_ Jun 04 18:41:38 huh. surprise. Jun 04 18:41:46 i was just thinking they are to most bios,s and most boot rom code Jun 04 18:41:58 nnnnnobody expects the spanish inquisition! Jun 04 18:42:22 i'm pretty certain i've booted a beagleboneblack from an SD card without the boot flag set on the partition... Jun 04 18:42:37 <_av500_> you did not Jun 04 18:42:50 <_av500_> there is no reasoning with the ROM code Jun 04 18:43:02 yes there is Jun 04 18:43:06 maybe if i talk louder it will listen Jun 04 18:43:07 mr rom, meet mr laser Jun 04 18:43:44 what might give you the impression though: SD card gets ignored, eMMC kicks in, emmc u-boot finds SD, boots from that Jun 04 18:44:01 bah i was just typing that lol Jun 04 18:44:54 tbr: yes, that's what i'm assuming happened... Jun 04 18:48:37 * vagrantc puts broad-sweeping generalizations back under a rock where they belong Jun 04 18:57:35 hi Jun 04 20:02:41 [kernel] RobertCNelson pushed 1 new commit to 3.8: http://git.io/3-SMZQ Jun 04 20:02:41 kernel/3.8 7f2f39d Robert Nelson: 3.8: merge in all cape changes up to bone56... Jun 04 20:09:03 hmm can somebody tell me which clock in the am33xx-clocks.dtsi is the ARM (ARM_FCLK) from TRM? I've no idea how to connect clocks from that file to TRM :( Jun 04 20:27:24 av500, https://github.com/joelagnel/microsd-switch/raw/master/microsd-switch.pdf Jun 04 20:30:51 <_av500_> thx Jun 05 00:00:27 hi, i want to ask a question about bonescript Jun 05 00:00:38 does beagleboard xm support bonescript? Jun 05 00:01:21 i cant find any documentation on the net Jun 05 02:24:29 anyone worked with alsa soc to connect to a codec on bbb? Jun 05 02:29:19 is https://github.com/RobertCNelson/bb-kernel now the canonical place for the kernel? Jun 05 02:29:37 I have been normally using the linux-dev git repo Jun 05 02:32:10 bah **** ENDING LOGGING AT Thu Jun 05 02:59:58 2014