**** BEGIN LOGGING AT Wed Mar 07 02:59:58 2012 Mar 07 03:47:17 anyone know how i would ctrl + alt + backspace to quit x on an apple keyboard? Mar 07 03:51:02 ctrl is usually the tourist attraction key. Mar 07 03:53:34 thurbad: MACHINE=beaglebone ./oebb.sh bitbake -f -c deploy virtual/kernel did not do it either :-( Mar 07 03:53:56 :/ Mar 07 03:54:02 it's time fo me to go to bed but I'll come back tomorrow Mar 07 03:54:09 thanks for helping Mar 07 03:55:05 the result was Mar 07 03:55:09 NOTE: Tasks Summary: Attempted 405 tasks of which 405 didn't need to be rerun and all succeeded. Mar 07 03:55:21 and it did Mar 07 03:55:22 Executing: MACHINE=beaglebone bitbake -f -c deploy virtual/kernel Mar 07 03:55:34 gnight Mar 07 05:00:23 hello, I have a problem with devian instaler on beagle bone Mar 07 05:01:32 hello, I have a problem with devian instaler on beagle bone, could some one help me? Mar 07 05:04:59 hello, I have a problem with devian instaler on beagle bone, could some one help me? Mar 07 05:05:48 mario_ - explain your problem instead of repeating yourself Mar 07 05:07:20 ehlo Mar 07 05:08:32 ok thanks Mar 07 05:08:52 I have instaled debian sucsefuly using a 1GB SD card Mar 07 05:09:13 I followed the same porcedure with a 4GB card and is not booting the instaler Mar 07 05:12:32 any clue? Mar 07 05:14:04 are you getting any errors on the serial console? Mar 07 05:14:17 i have a wireless question Mar 07 05:14:17 no Mar 07 05:14:21 just ccccc Mar 07 05:14:35 is not booting Mar 07 05:14:37 i'm using dlink dwa-125 Mar 07 05:14:39 did you try taking out the SD card and putting it back in? Mar 07 05:14:56 im folowing this instructions Mar 07 05:15:13 http://elinux.org/BeagleBoardDebian#Debian_NetInstall Mar 07 05:15:26 is working with 1Gb SD Mar 07 05:15:30 but not with 4GB Mar 07 05:17:52 no error messages but dhclient never gets an ip address Mar 07 05:18:04 any ideas? Mar 07 05:19:13 * mranostay scrolls up Mar 07 05:19:26 god i need another beer :) Mar 07 05:19:50 W1N9Zr0 yes I did try that Mar 07 05:21:28 I also try with two difernet 4GB crds Mar 07 05:25:46 any ideas? Mar 07 05:33:03 hello Mar 07 05:39:05 I have a problem with debian on beagle bone, I am following this instructions http://elinux.org/BeagleBoardDebian#Debian_NetInstall everything works fine I installed debian successfully on a 1GB SD card. My problem is that when I put the installer in a 4GB card the installer is not booting. could some one help me? Mar 07 05:47:05 hello Mar 07 06:02:54 I have a problem with debian on beagle bone, I am following this instructions http://elinux.org/BeagleBoardDebian#Debian_NetInstall everything works fine I installed debian successfully on a 1GB SD card. My problem is that when I put the installer in a 4GB card the installer is not booting. could some one help me? Mar 07 06:11:52 hello Mar 07 06:15:16 mario_: dude we see the question :) Mar 07 06:25:36 so no one can help? Mar 07 06:29:46 mario_: less vague then yes we can Mar 07 06:30:22 Doomed Doomed I say! Mar 07 06:30:33 <_av500_> mario_: maybe ask the #debian people? Mar 07 06:32:00 _av500_: that is right pass the buck :) Mar 07 06:32:37 <_av500_> no, hedge your bets Mar 07 06:47:18 I have a problem with debian on beagle bone, I am following this instructions http://elinux.org/BeagleBoardDebian#Debian_NetInstall everything works fine I installed debian successfully on a 1GB SD card. My problem is that when I put the installer in a 4GB card the installer is not booting. could some one help me? Mar 07 06:47:39 * mranostay smacks head Mar 07 06:49:05 ARRRRRRRRG Mar 07 06:49:07 please stop spamming Mar 07 06:50:16 sleepy time again Mar 07 06:54:19 I'm trying to get Ubuntu up on an OMAP3 device. I installed the SGX libraries according to this: http://elinux.org/BeagleBoardDebian#SGX_Video_Acceleration Mar 07 06:54:52 but when I try to run the run-SGX script I get an error about module bufferclass_ti not found Mar 07 06:55:36 * dm8tbr blinks Mar 07 06:56:53 Is this something I can add to my kernel and build (2.6.37)? I have pvrsrvkm.ko and omaplfb.ko and they are loaded Mar 07 07:03:30 mario_: spamming us will only get us annoyed Mar 07 07:04:17 my last message was 20min ago Mar 07 07:04:50 so? Mar 07 07:05:04 you are going to send it 72 times per day? Mar 07 08:04:44 hi all Mar 07 08:04:53 i m new to beagleboard Mar 07 08:06:01 can anyone guide me Mar 07 08:09:53 http://beagleboard.org/ Mar 07 08:23:27 bonjour Mar 07 08:25:01 hi have probleme to view hyperterm on serial com in beagleboard xm Mar 07 08:28:29 can anyone tell me what the most recent android release for the beaglebone is please Mar 07 08:32:16 patient you are not Mar 07 09:07:43 hi, would like to ask if the autofocus engine in the H3A subsystem is supported by v4l2 driver in beaglebone? Mar 07 09:09:14 and you are sure the bone CPU has H3A? Mar 07 09:16:01 sorry, my mistake -- i meant beagleboard Mar 07 09:17:47 i would like to use the autofocus engine in H3A and i'm researching which board+camera sensor should i use... Mar 07 09:19:32 Hi Mar 07 09:20:08 anything that stops ethernet on Beagleboard-xM Rev C from working when powered via USB? I'm powering via USB because it does not boot from the DC jack (somethign is friend) Mar 07 09:20:11 *fried Mar 07 09:20:19 (definitely not friend) Mar 07 09:21:13 host machine is not reporting overcurrent condition (I've confirmed it can, and will, with other devices that draw even 750mA for too long) Mar 07 09:22:40 fried boards is fried Mar 07 09:22:56 Well, shit. Mar 07 09:23:13 At least it works over serial? :). /alwaysoptimistic Mar 07 09:23:16 (not!) Mar 07 09:23:18 in other words, if the board is already known broken, I would not bet on other stuff to work Mar 07 09:23:22 get it fixed Mar 07 09:23:48 Yeah, that's the issue Mar 07 09:24:09 barely possible to do that, so I was wondering if there was a software herpderp like the thing that used to cause kernel panics when booting from usb Mar 07 09:24:37 where I could insert x line of magical code into kernel code y to make z work fine, etc Mar 07 09:24:53 sure, if you know x, y and z Mar 07 09:25:38 's why I'm here -_-'. I second the 'yep, dodgy board is dodgy' thought, but it's only relevant to an ideal world. Mar 07 09:27:29 [ 4.302917] g_ether gadget: high speed config #1: CDC Ethernet (EEM) Mar 07 09:27:59 beyond that, there is no mention of ethernet, and udev doesn't assign it a device name Mar 07 09:29:17 [ 5.889892] hub 1-0:1.0: unable to enumerate USB device on port 2 Mar 07 09:29:20 few of those, tho Mar 07 09:29:33 leads me to believe, yeah -- fucked. Mar 07 09:30:33 Well, since I'm bored and don't care anymore, I'm gonna abuse the board and try electrocution therapy on it Mar 07 09:30:40 sorry, but I won't put effort into debugging a board that you call "fried" Mar 07 09:30:48 post a video Mar 07 09:31:11 Could you please elaborate? Demonstrating what? Mar 07 09:31:23 "electrocution therapy" Mar 07 09:31:27 oh Mar 07 09:31:33 not 'fried board' Mar 07 09:32:31 by 'fried' I do not mean I saw the board go up in flames. In fact there is zero evidence of that. Just that it does nothing when powered via the DC jack, and the PSU is known-working and can reliably supply 2A of continuous current. didn't try higher. Mar 07 09:32:47 some LEDs are constant-on, I can find out which if anyone cares Mar 07 09:33:42 Though I have a feeling I described it here before and one of you suggested 'power it over usb, inject higher current to usb' Mar 07 09:34:45 if its boots from usb but not from dc-in, it's broken Mar 07 09:55:53 anyone know that connector 'P18' on the Beagleboard-xM is? Can't seem to find it in the SRM Mar 07 09:57:35 Found it Mar 07 09:57:37 ._.' Mar 07 09:58:09 Known as 'P10' in the SRM but the picture shows 'P18' Mar 07 10:22:49 Why can't I completely kill it? :\ Mar 07 10:23:10 shorted the 3v3 regulator, shit got a little warm but Y U NO dead? Mar 07 10:23:51 actually let's see if that magically fixed the USB-connected and therefore theoretically isolated ethernet device. Mar 07 10:24:22 * lunra bangs head against wall. It did more! Mar 07 10:37:51 well, I configured OTG networking Mar 07 10:57:08 I am looking for beagle board get loaded from nfs for this i have set my nfs rootfs folder and ip of device and nfs server. so what should rootfs file structure to get devices booted from nfs Mar 07 11:14:33 Hey Mar 07 11:14:37 anybody here? Mar 07 11:15:41 no Mar 07 11:15:48 I have problems booting the current Angstrom Kernel 3.0.23+ :It's stuck at udevd[72]: starting version 181 Mar 07 11:15:50 we are closed for business Mar 07 11:16:48 I bitbaked my own but have the same stopping with the official releases after a opkg upgrade Mar 07 11:17:10 Hey av500 Mar 07 11:17:10 pastebin a full boot log Mar 07 11:18:20 could it be a problem with missing syscalls (or glibc wrappers for same)? Mar 07 11:18:32 what kernel version was your glibc built against? Mar 07 11:19:43 http://pastebin.com/jRJYAdbY Mar 07 11:21:52 mru what makes me wonder is that the same will happen after I make a opkg upgrade to the core 3.0.17 image Mar 07 11:22:40 just guessing here, missing syscalls is the usual cause of udev hangs Mar 07 11:23:45 i'm not sure if udev actually hangs since I get no error messages during boot Mar 07 12:16:29 mru, av500 did you get a chance to read the pastebin? Mar 07 12:59:10 Hi, on beaglebone I wrote a program to decode an IR signal, using sleep + polling on GPIO. Now, I made interrupts work in another program and therefore I'd like to rewrite the IR decoder to use timers + interrupts. Are setitimer(2) and getitimer(2) the functions I should use for the timer ? Mar 07 12:59:54 and sleep and polling worked? Mar 07 13:00:24 i personally wouldn't do it that way Mar 07 13:00:36 av500: it did, yes Mar 07 13:00:38 I think I'd use GPIO interrupts and record a timestamp when they occur Mar 07 13:00:39 impressive Mar 07 13:00:41 but it's ugly Mar 07 13:00:44 then figure it out based on that Mar 07 13:00:52 so gettimeofday ? Mar 07 13:00:57 and lirc driver does not help? Mar 07 13:01:33 I did not use lirc, I wrote a basic program from scratch that deals with raw data Mar 07 13:01:58 I guessed that much :) Mar 07 13:04:17 okay, gettimeofday is µs precision so it should be fine, thx Mar 07 13:07:31 spow: no Mar 07 13:07:43 it returns time in that precision Mar 07 13:07:53 but that does not mean you get us precision Mar 07 13:08:17 your kernel could be clocked from 32khz, then that is all you get Mar 07 13:09:35 I measured ~25 state changes in ~25ms with variable steps, so something that is expressed in µs is ~1/100 of the actual precision needed Mar 07 13:10:07 but I understand the objection Mar 07 13:12:47 spow: it's amazing that worked. Mar 07 13:13:15 running as a realtime task, you can basically hog the cpu from user space Mar 07 13:13:26 dwery: why is that ? i'm polling a state change Mar 07 13:13:55 spow: I didn't know that you can poll so fast ;) Mar 07 13:13:55 reminds me of a students try i saw once to write a serial driver. he used nonblocking writes and usleeps to bang out bytes one after the other. Mar 07 13:14:09 LetoThe2nd: well done Mar 07 13:14:09 strange it only worked on his particular laptop. no idea why ;) Mar 07 13:14:11 LetoThe2nd: which grade did they get Mar 07 13:14:14 ? Mar 07 13:14:28 no idea. i told them tp get rid of it completely. Mar 07 13:14:35 rid of the laptop? Mar 07 13:14:58 get_rid_of({student, laptop, driver}) Mar 07 13:14:59 of the brain? Mar 07 13:15:20 LetoThe2nd: we had a student once that placed dsp and xtal 15cm apart on the PCB Mar 07 13:15:27 spow: you might want to investigate the PRUSS.. it's a bit hard but it will get you some real satisfaction Mar 07 13:15:37 we told him to move the xtal close to the dsp, he did that by running 15cm more of wires back to the dsp :) Mar 07 13:15:48 av500, hrhrhr Mar 07 13:16:08 av500: kill. Mar 07 13:16:26 dwery: no, laugh really hard for weeks :) Mar 07 13:16:52 dwery: i'll take a look at it for general culture, but i'm doing a really simple project and i'm afraid I won't have the time to integrate that (~50 work hours left including 10 pages latex documentation) Mar 07 13:17:05 spow: why not lorc? Mar 07 13:17:09 lirc even Mar 07 13:17:14 but I know it's important, you guys mentionned it quite a bit in the last days Mar 07 13:17:16 lorc=latex orc? Mar 07 13:17:26 leather Mar 07 13:17:31 linux orc. Mar 07 13:17:45 spow: I'm trying to get more people on it ;) Mar 07 13:17:49 i just wanted to demonstrate basic embedded functions, so I wrote my own programs Mar 07 13:18:03 gettin' high on PRUSS Mar 07 13:18:05 for pwm, IR, sensors, LCD ... Mar 07 13:18:32 dwery: will you work on r-pruss too? Mar 07 13:18:39 sooner or later Mar 07 13:19:01 dwery: maybe bcm was more clever hiding it :) Mar 07 13:19:34 but it was an ambitious project, I have quite a bunch of 'undeclared' hours to the uni Mar 07 13:19:45 :D Mar 07 13:20:19 "undeclared hours" just sound like real life to me. Mar 07 13:20:47 * av500 is reminded to fill out time sheets today Mar 07 13:21:20 how fitting my current music is. "and then there was silence" Mar 07 13:21:21 where I come from only the result matters, but here at rwth it's more about the time spent Mar 07 13:21:27 apparently Mar 07 13:21:36 av500: "3 hours spent showing people how to reformat SD cards" Mar 07 13:22:02 there, done Mar 07 13:22:17 I enter the same thing for 3ys now "multimedia stuff" Mar 07 14:08:39 okay, it's overall much more precise now, but sometimes my results are way of, whereas with poll + sleep I had 100% recognition Mar 07 14:10:27 "now" meaning? Mar 07 14:10:56 I STILL :) have problems booting the current Angstrom Kernel 3.0.23+ :It's stuck at udevd[72]: starting version 181 Mar 07 14:11:03 interrupt + timestamp ... should I give myself scheduler privileges? set_sched (FIFO) ? Mar 07 14:11:25 * LetoThe2nd is pretty certain that it ahven't been 100%. just maybe the failure rate was lower than the total number of values received altogether. Mar 07 14:11:31 Martron: the current is 3.2 if i'm not mistaken Mar 07 14:11:38 no idea if that helps with interrupts in user space Mar 07 14:12:21 spow: Not that I know of Mar 07 14:13:06 spow: beagleboard != beaglebone Mar 07 14:13:39 koen: spoiler! Mar 07 14:13:51 koen: Martron : my bad, need moar coffee Mar 07 14:14:27 no i didn't specify Mar 07 14:14:35 (actually, I need to do less assumptions, but whatever) Mar 07 14:14:41 i mean BeagleBoard of course :) Mar 07 14:15:33 As I wrote before I have the same issues with the official image when doing a opkg upgrade Mar 07 14:17:02 hum, might be irrelevant, but you just do a plain opkg upgrade ? what happens then ? Mar 07 14:17:23 LetoThe2nd: with any failure rate <100%, the probability of a run of successes of any finite length is >0 Mar 07 14:17:55 mru, certainly. finity can be a bitch. Mar 07 14:18:52 well i created a complete new sd-card with the current console image from angström its the 3.0.17 as far as i remember...alll i did with it was: opkg update followed by: opgk upgrade and did a rebbot.....system stuck Mar 07 14:19:29 but the upgrade succedded ? Mar 07 14:19:40 yes Mar 07 14:19:53 i could upgrade to 3.0.23+ Mar 07 14:20:13 but couldn't sucessfully rebooot Mar 07 14:20:26 http://pastebin.com/jRJYAdbY Mar 07 14:21:02 ok then, sorry, no idea Mar 07 14:21:50 no prob just wondering if i'm the only one having problems with 3.0.23+ yet Mar 07 14:25:10 FWIW: Mar 07 14:25:10 root@beagleboard:~# uname -a Mar 07 14:25:11 Linux beagleboard 3.0.23+ #1 Thu Mar 1 10:13:49 CET 2012 armv7l GNU/Linux Mar 07 14:25:40 boring, no 3.3 :p Mar 07 14:26:01 omap is totally fucked in 3.3, no? Mar 07 14:26:17 only omap? Mar 07 14:26:22 s/in 3.3.// Mar 07 14:27:08 koen: since you did the changes i figured it works for you :) Mar 07 14:28:49 did you bitcake console-image for your board? Mar 07 14:29:34 it's an old systemd-gnome-image fro m a few months back that I have been upgrading Mar 07 14:29:34 and how come it won't boot when upgrading the image from the angstrom page Mar 07 14:30:23 yummy, bitcake from chef dotkoen Mar 07 14:30:56 koen did you have a look in the pastebin? Mar 07 14:31:29 Martron: yes, no clue why it's failing for you, sorry Mar 07 14:32:11 ok but just to be sure: it SHOULD work with an BBxM with uLCD7 correct? Mar 07 14:35:03 yes Mar 07 14:35:16 I'm using it on an xm with a ulcd7 right now :) Mar 07 14:36:56 well actually thats good news :) Mar 07 14:49:00 * ynezz laughs at 'Subject: [Angstrom-devel] UNSUBSCRIBE' email :D Mar 07 14:50:11 http://www.mail-archive.com/angstrom-distro-devel@linuxtogo.org/msg05621.html Mar 07 14:50:16 http://www.mail-archive.com/angstrom-distro-devel@linuxtogo.org/msg05620.html Mar 07 14:50:33 that Jane must really hate him :D Mar 07 15:05:14 koen++ Mar 07 15:05:35 reading your mailing list posting about bitbake. You are a helpful person. Mar 07 15:06:18 don't spoil his reputation! Mar 07 15:09:00 Wow all this PRU activity on the mailing list Mar 07 15:09:02 * SilicaGel beams Mar 07 15:09:44 SilicaGel: and remember: there's no PRU on the beaglebone Mar 07 15:10:01 NOT SUPPORTED! Mar 07 15:10:17 wait, are you trying to hypnotize me? Mar 07 15:12:43 jkridner: do you have a reserved EC2 instance for the log bot ? Mar 07 15:29:32 I have a problem with debian on beagle bone, I am following this instructions http://elinux.org/BeagleBoardDebian#Debian_NetInstall everything works fine I installed debian successfully on a 1GB SD card. My problem is that when I put the installer in a 4GB card the installer is not booting. could some one help me? Mar 07 15:30:01 try #debian Mar 07 15:37:59 ok so I have a question. Mar 07 15:38:21 struct omap_dm_timer *t = omap_dm_timer_request(); Mar 07 15:38:37 I get back a timer, this timer has t->id = 3, t->irq = 69 Mar 07 15:38:40 that's all in agreement with the TRM Mar 07 15:38:47 but t->io_base = fa042000 Mar 07 15:39:14 What the hell is that? It's nto the address of timer3 in global address space. And the address returned looks kinda/sorta like a virtual memory address rather than a physical one Mar 07 15:43:27 SilicaGel: use the source :) Mar 07 15:44:04 Hmmmmmmmmmmmmmmmmmmmmmmmm yeah let me look at where the hell that comes from Mar 07 15:50:21 timer->io_base = ioremap(mem->start, resource_size(mem)); Mar 07 15:50:40 I have no idea what the hell that means haha Mar 07 15:51:18 When access passes though page tables, the kernel must first arrange for the physical address to be visible from your driver, and this usually means that you must call ioremap before doing any I/O. If no page tables are needed, I/O memory locations look pretty much like I/O ports, and you can just read and write to them using proper wrapper functions. Mar 07 15:51:21 AHH HAH!!! Mar 07 15:51:23 unhelpful. Mar 07 15:51:36 that address it gives me is not helpful to the PRU which doesn't go through page mapping. Mar 07 15:51:51 yes, but your access to the PRU does Mar 07 15:52:32 In fact, that description is talking about I/O remapping, why does that even apply here when there is no separate I/O address space in an arm Mar 07 15:52:42 hmm Mar 07 15:53:19 what is the issue with the address? Mar 07 15:53:26 you cant access it? Mar 07 15:54:19 oh i was just expecting to return something like 0x48042000 Mar 07 15:54:23 do you see what I mean? Mar 07 15:54:40 so that I could pass that address to the PRU Mar 07 15:54:42 yes Mar 07 15:54:57 well, linux kernel does not know about the PRU Mar 07 15:56:12 so I guess you need to map timer id to phys address yourself Mar 07 15:56:25 make a small table with 11 entries :) Mar 07 15:56:31 not 12! Mar 07 15:56:37 haha Mar 07 15:56:46 i tried omap_dm_timer_request_specific(12) Mar 07 15:56:48 it failed!!!! Mar 07 15:56:56 of course, that timer does not exist Mar 07 15:57:09 That's exactly what they want us to think!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Mar 07 15:57:33 * av500 swaps tin foil hat with SilicaGel to confuse THEM Mar 07 15:57:42 mates I have try debian channel but no one could help me, can some one help me here Mar 07 15:57:43 ? Mar 07 15:57:57 mario_: mate, few people use debian here Mar 07 15:58:19 why? Mar 07 15:58:22 make sure you have a correctly formatted card with MLO and uboot Mar 07 15:58:44 av500 how can I bee sure? Mar 07 15:58:48 i was going to download a picture of the kool-ade man, and use gimp to edit in "THERE IS NO TIMER 12" Mar 07 15:58:52 but i decided it wasn't worth the effort Mar 07 15:58:56 mario_: http://groups.google.com/group/beagleboard/browse_thread/thread/ae8e64e6be02baae Mar 07 15:59:49 * mranostay yawns Mar 07 16:00:31 SilicaGel: http://cdn.memegenerator.net/instances/400x/15783222.jpg Mar 07 16:01:12 HAHAHAHAHAAHAHAHAHAHAHAHAHAHAHAHA Mar 07 16:01:21 * SilicaGel dies laughing Mar 07 16:01:59 * dwery rolls on the floor Mar 07 16:01:59 s/dm/gp/ ? Mar 07 16:02:07 ah frak Mar 07 16:02:09 what ever Mar 07 16:02:16 the TRM really does refer to them as dual mode timers Mar 07 16:02:23 for the am335x Mar 07 16:02:24 haha Mar 07 16:03:50 fixed: http://cdn.memegenerator.net/instances/400x/15783356.jpg Mar 07 16:04:28 +1 Mar 07 16:09:08 has someone built a tool to just generate those? Upload a picture, type your text, and go Mar 07 16:09:19 http://memegenerator.net/ Mar 07 16:09:23 someone could make an appspot app or something and put ads there and win Mar 07 16:09:45 Yeah just like that site Mar 07 16:10:25 there are a lot of sites doing just that Mar 07 16:10:53 and i'm not sure you can make a business plan based on nerd thingies Mar 07 16:11:03 jwinnebeck: http://cdn.memegenerator.net/instances/400x/15783693.jpg Mar 07 16:11:23 Ha I'm cracking up here Mar 07 16:11:23 wow av500 Mar 07 16:11:26 that one amde me sad :( Mar 07 16:11:26 I'm going to get fired! Mar 07 16:11:39 my typing also makes me sad lately. Mar 07 16:12:41 * av500 needs to plug memegenerator into kernel oops backtrace info and make that report to our bugtracker Mar 07 16:12:54 that sounds like a good idea Mar 07 16:13:20 I kind of always wanted a big LED marquee that would scroll the failures (and who to blame) that our continuous integration server sees Mar 07 16:13:27 I don't always crash in function foo_bar, but when I do, I do it at address 0x1234abcd Mar 07 16:14:05 deadbeef... Mar 07 16:14:06 come on Mar 07 16:14:20 * dm8tbr is going to have some dead beef soon Mar 07 16:14:31 roadkill? Mar 07 16:14:31 tasty black angus Mar 07 16:14:44 0xB00B135 Mar 07 16:19:52 this might be a dumb question but, if im running X11 by itself and i launch wahlcade with .xinitrc then that launches another app (emulator) how can i bring it above wahlcade? or do i need some sort of minimal window manager? Mar 07 16:21:13 basically, can i stack windows without a window manager? Mar 07 16:22:49 SilicaGel: that address can be used to access the mapped timer from linux Mar 07 16:23:36 from ... from userspace? Mar 07 16:23:46 or from the driver itself? Mar 07 16:23:58 oh i think i get it Mar 07 16:24:02 from the driver Mar 07 16:25:21 yeah ok Mar 07 16:25:29 beacuse the driver is still in virtual space Mar 07 16:25:34 that's waht I temporarily forgot Mar 07 16:30:43 smplman: you need a window manager Mar 07 16:30:55 jay6981: no way around it? Mar 07 16:31:19 i guess i kinda wanted whalcade to act as my window manager Mar 07 16:31:30 load emulators on top of itself Mar 07 16:31:54 doesn't x11 come with a basic windowing system? Mar 07 16:32:08 it doesn't Mar 07 16:32:59 not even a screen saver? Mar 07 16:35:33 SilicaGel: sorry for the delay, I was on a phone call Mar 07 16:35:40 but you got your answer Mar 07 16:37:25 smplman: if you use xlib, you can probably create and place windows manually wherever you want Mar 07 16:37:54 smplman: maybe your program has a way to specify the window geometry when it loads up Mar 07 16:38:30 jay6981: i believe both wahcade and mednefen (NES Emulator) have this Mar 07 16:38:41 bit it would be x and y, not depth Mar 07 16:39:44 smplman: http://tronche.com/gui/x/xlib-tutorial/ Mar 07 16:41:15 jay6981: gracias Mar 07 16:41:32 hopefully i will get my beagleCade up and running tonight Mar 07 16:41:59 oh, and NFS's are love when it comes to having hundreds of roms Mar 07 16:42:52 SilicaGel: all this PRU activity is impossible, it's too complicated for customers to understand Mar 07 16:57:05 how do I go about building console-image with linux-ti33x-psp_3.2 instead of the default (linux_3.0 i think)? Mar 07 17:03:49 W1N9Zr0: I think 3.2 is the default Mar 07 17:03:55 When I build systemd-image, I get 3.2 Mar 07 17:04:01 also when I build virtual/linux Mar 07 17:04:08 or virtual/kernel, whatever it is Mar 07 17:04:43 mine comes out 3.0, i'm building for beagleboard-xm if that matters Mar 07 17:05:03 oh yeah sorry, I get self-centered, I'm on beaglebone Mar 07 17:05:16 The 3.2 kernel is critical for bone Mar 07 17:05:28 to be honest I don't know about the xM Mar 07 17:05:40 it could be possible there's not a working 3.2 for it yet Mar 07 17:09:16 maybe i'm confusing google groups messages, there's one about 3.2 working on beagleboard xm, and one asking how to build ti33x-psp_3.2 with the answer being use bitbake Mar 07 17:10:43 the ti33x kernel is only for ti33x cpus Mar 07 17:12:15 mdp: really? Mar 07 17:12:21 mdp: do you really mean that? or are you being a wise guy? Mar 07 17:13:07 I am not wise, that's for sure Mar 07 17:13:37 Once I got it actually running, I find the PRU to be butt simple Mar 07 17:13:39 wise guy, in the slang sense..ok..yeah, I needed sartalics there Mar 07 17:14:25 SilicaGel: you might be surprised at justifications used to obscure things in a part…just saying Mar 07 17:14:32 i see Mar 07 17:18:13 I don't understand the logic of not allowing more than irq 0-63 into the PRU on the am335x though. That seems to me to be a mistake. Mar 07 17:18:52 i mean, i'm not trying to be narcissistic here, but you can route USB events/interrupts into the PRU, but you can't route GP/DM timer events into it. Which one are you more likely to need ? Mar 07 17:18:58 I don't understand how many design decisions are made Mar 07 17:19:01 Seems painfully obvious to me. Mar 07 17:19:34 it may involve a bristle board Mar 07 17:20:10 0-63 might have to do with something being 7 bit only :) Mar 07 17:21:47 I'm sure Mar 07 17:21:55 but Mar 07 17:21:58 they love muxes Mar 07 17:22:04 so i think this would have called for another mux! Mar 07 17:45:16 do'h! I mistakenly closed the window.. how many timers did I miss? Mar 07 17:49:19 more than 11, less than 13 Mar 07 17:53:04 ouch! Mar 07 17:53:36 any PRUSS breakthrough? Mar 07 18:40:59 Has anyone had success with the network and audio on a BeagleXM with stock 3.2 kernel? Mar 07 18:47:39 GrueMaster: network, yes…Rev B Mar 07 18:47:42 3.2.5 Mar 07 18:48:09 Do you have the driver compiled in or as a module? Mar 07 18:57:10 compiled in, I don't do modules :) Mar 07 18:57:46 mostly because I'm nfsrooting off of the smsc95xx Mar 07 19:04:20 mdp: Yea, I can get it to work if it is compiled in, but not if it is modular. Having an argument with my kernel engineer about it. Mar 07 19:05:40 smsc95xx as a module or that and ehci? Mar 07 19:07:52 GrueMaster: kernel people were born to argue Mar 07 19:08:19 heh. Mar 07 19:09:05 I think most everything is a module. iirc, the kernel config is closely mirrored to the x86 config (which has had other issues). Mar 07 19:10:59 it's not uncommon to have bugs like this when compiling stuff as a module Mar 07 19:11:39 I'll add it to my growing TODO list upstream Mar 07 19:12:08 need to fix it in mainline first to pull back to stable though and there's regressions on xM to deal with first Mar 07 19:17:13 Is there a tool I can run to verify what rev board I have? It is labeled rev B. Mar 07 19:24:01 I just got an e-mail from somebody who says the PRUSS is not fun. Mar 07 19:24:05 I wonder what's wrong with that guy! Mar 07 19:24:22 this is not the PRUSS you're looking for Mar 07 19:24:36 PRUSS is more fun than a clown on fire! Mar 07 19:26:01 hi mr pruss Mar 07 19:26:13 hi Mar 07 19:26:32 so pru can't see all the system interrupts? Mar 07 19:27:00 nope Mar 07 19:27:04 It can see IRQ0-63 Mar 07 19:27:09 the others aren't mapped Mar 07 19:27:19