**** BEGIN LOGGING AT Sat Jun 04 02:59:58 2016 Jun 04 03:04:36 0_0 where's the MLO file? Jun 04 03:08:05 is that mmcblk0boot0 or mmcblk0boot1? Jun 04 03:20:33 neither Jun 04 03:20:41 its written direct to emmc/usd Jun 04 03:20:59 those blocks/addresses aren't used .. much to zmatt's distress Jun 04 03:27:31 it's written direct to emmc/usd? Jun 04 03:28:53 i mean i though the bbb used the fat+mlo scheme to boot Jun 04 03:29:01 so i was checking the fat partition for an mlo but i didn't see it Jun 04 03:30:13 i got all excited when i found a tutorial for writing your own MLO files Jun 04 03:33:49 meh we'll see Jun 04 03:57:24 no there's no fat partition that I'm aware of on emmc Jun 04 04:10:44 oohhh so its doing kind of a raw data deal Jun 04 04:11:01 i get a lot of my info from google soooo Jun 04 05:01:11 ayjay: https://github.com/mvduin/bbb-asm-demo this explains various boot methods Jun 04 05:01:55 including concrete instructions for booting a tiny demo app (basically taking the role of u-boot's SPL) from SD/MMC via both the old FAT-partition method and the new "raw MMC boot" method Jun 04 05:21:51 ah so it is the new raw method! Jun 04 05:22:59 have you looked at fast external booting? Jun 04 05:23:45 I don't have suitable hardware Jun 04 05:24:16 oooh Jun 04 05:25:18 or, well, maybe I could have one BBB boot off the other with PRU simulating the NOR flash :P Jun 04 05:26:02 but I see little point is doing so Jun 04 14:06:49 hi there Jun 04 14:07:01 anyone knows how to read voltages in 4.6 kernel? Jun 04 14:07:12 there is no /sis/bus/iio Jun 04 14:07:30 i meant /sys/bus/iio Jun 04 14:15:26 that was tsupid question Jun 04 14:15:27 :D Jun 04 14:15:38 compiled dtb adding Jun 04 14:15:45 bits needed Jun 04 14:15:47 bye Jun 04 15:46:23 HI at all! i have some problems with my beaglebone black ... i tried to connect it to a raspbarry pi board and let them communicate using the i2c bus Jun 04 15:46:42 I do not really get it, how to configure the beaglebone black! Jun 04 15:50:48 hello? Jun 04 15:51:36 HI at all! i have some problems with my beaglebone black ... i tried to connect it to a raspbarry pi board and let them communicate using the i2c bus Jun 04 15:51:41 I do not really get it, how to configure the beaglebone black! Jun 04 19:11:25 swoo: what have you tried? Jun 04 21:32:45 Hello Jun 04 21:34:10 I have a question regarding my beaglebone black board is this the right chat room? Jun 04 21:34:57 sure is Jun 04 21:35:31 Thanks Jun 04 21:35:34 onio: go ahead and ask whatever your questions are, and wait patiently for a response Jun 04 21:35:44 cool nice one Jun 04 21:36:57 I have a beaglebone black that is BBB-eMMC-flasher-debian-7.5 Jun 04 21:37:49 Should the board boot from SD card if the boot button is no held down when applying power? Jun 04 21:39:42 you need to hold the boot button to get it to boot off of SD Jun 04 21:40:06 I do have a valid kernel and MLO on the SD card but I was under the impression that this should only load from the SD card if I held down the power button Jun 04 21:40:58 this is my question because it is booting of the card without button being held down Jun 04 21:41:26 does it boot without the SD card? Jun 04 21:41:55 If I remove the SD card it would run the debian image Jun 04 21:42:41 aha Jun 04 21:42:47 yep Jun 04 21:42:58 yes the emmc will check to see if there's a uSD present normally .. its a 'feature' Jun 04 21:42:58 might depend on u-boot settings ... might still be loading u-boot from eMMC but is configured to load kernel+stuff off of SD Jun 04 21:43:30 think its a custom uboot option .. you might want to check default environment settings. I recall its set there Jun 04 21:43:39 * vagrantc finds the BBB to behave a little more confusingly than most other boards in the respect of the boot button/u-boot Jun 04 21:44:16 the BOOT button actually forces a uSD boot .. ie. it looks directly at teh uSD MMC interface *before* trying emmc and its tom-foolery Jun 04 21:44:23 according to BeagleBone Black System Reference Manual page 35 this should not be the case Jun 04 21:44:44 onio: that may be ... old ... Jun 04 21:44:54 ahh Jun 04 21:45:21 well, there's a difference from loading u-boot off of eMMC vs. SD, and loading the kernel off of eMMC vs. SD Jun 04 21:45:27 I'm not sure the history .. but it had something to do with interfaces not being named consistently Jun 04 21:45:45 vagrantc: indeeed, and that's why modern stock images use a UUID Jun 04 21:46:01 although that's nt entirely foolproof either Jun 04 21:46:16 there are too many damn variables lol :) Jun 04 21:46:51 unfortunately the Guy Who Knows no longer frequents this channel :( Jun 04 21:46:55 I am new to embedded linux so this is an learning exercise for me Jun 04 21:47:15 onio: lots to learn :) quite a few traps to avoid too! Jun 04 21:47:47 I am trying to gain some understanding of Kernels, u-boot , MLO etc Jun 04 21:48:49 I am so glad I found this chat room because sometimes I feel so isolated and lost :) Jun 04 21:50:18 Is there any one who can confirm this behaviour, if this is an undocumented feature that has silently crept in Jun 04 21:54:32 * veremit scrolls Jun 04 21:54:42 what are you expecting to happen, and what are you observing? Jun 04 21:55:50 I am expecting that with the SD card inserted and power removed and re-applied that the device should boot from eMMC and not SD Jun 04 21:56:24 However when I hold down the boot button it should then boot from the SD card Jun 04 21:56:42 http://dumb-looks-free.blogspot.co.uk/2014/05/beaglebone-black-bbb-boot-process.html - covers some of the basics well Jun 04 21:57:32 the situation is confused when the emmc uboot looks at the usd and 'chooses' that over the emmc :) Jun 04 21:57:47 Thanks for the link Jun 04 21:58:00 but the BOOT button, if held whilst powered-up will fully override the emmc .. now whether the uSD *looks* for the EMMc .. you're screwed lol Jun 04 21:58:10 the description stated is consistent with what I expect Jun 04 21:58:41 what you need to do .. is boot to uboot with a serial debug ..and check what is in the stored environment .. and what the boot commands are trying to do Jun 04 21:59:01 yea I can do that Jun 04 22:00:28 you should then be able to narrow down what's going on .. and/or wrong .. Jun 04 22:02:30 the boot ROM is fixed in hardware .. but beyond that .. you can alternate storage at *any* stage .. lol .. so you could theoretically end up in a crazy loop Jun 04 22:03:55 but since its software .. you should be able to sever the appropriate links if desired Jun 04 22:04:47 I am sure I would get to the bottom of it Jun 04 22:04:48 I used to get quite confused as to which mmc1 was mmc1 .. the uSD or the eMMc .. but I think that's been better pinned down Jun 04 22:05:28 my understanding the that the eMMC is mmc1 and SD card is mmc0 Jun 04 22:06:16 hwardware-wise I think you're correct . but the driver naming hasn't always been so consistent :) Jun 04 22:07:01 okay Jun 04 22:07:39 like I mention earlier. I am pretty new to the embedded linux as a whole but I have been around bare metal programming for decades Jun 04 22:08:33 eg. I think the driver naming has been inconsistent with the hardware interface .. so you can imagine the fun trying to get the right device to boot. This may not necessarily still be the case, but sometimes legacies live on .. lol Jun 04 22:09:43 I am looking at the u-boot environment variables by halting the booting sequence Jun 04 22:11:10 there is a variable named "bootpart=0:1" am I reading this correctly that it is looking at mmc0 partition 1 Jun 04 22:11:34 which is the SD card for booting information Jun 04 22:14:29 you'd have to check syntax .. but quite likely I'd say. Jun 04 22:15:01 I'm not sure there's still an explicit boot partition anywhere Jun 04 22:15:33 I haven't looked for a while .. used to have to be FAT once. Jun 04 22:15:34 I tried changing the variable but its not saving "saveenv" Jun 04 22:15:44 try overriding in uEnv.txt Jun 04 22:15:59 okay I try that now Jun 04 22:16:25 again .. be quite meticuous about your methods .. as you may not be where you think you are :D Jun 04 22:17:53 Thanks for that I am trying my best. I already locked myself one yesterday but luckily recovered with debian flasher image Jun 04 22:19:04 its worth it in the end :) Jun 04 22:19:20 I am enjoying it Jun 05 01:20:36 linaro vs codesourcery? Jun 05 01:21:46 lol i messed up ill be back Jun 05 01:22:06 lol Jun 05 01:26:41 so anyway, linaro or codesourcery, whatsayyou Jun 05 01:28:37 I've used linaro a fair bit .. via RNs eewiki pages .. Jun 05 01:28:48 but I've just rolled my own toolchain via gentoo's crossdev tools Jun 05 01:28:50 they have a channel on freeenode apparently... Jun 05 02:01:32 hmz, so, any DRM/DRI gurus? why do I need to manually write the first (and only) line from fb0/modes to fb0/mode to active a tilcdc-panel output. I'm fairly certain that can't be How You're Supposed to Use It (tm) Jun 05 02:02:54 xorg does apparently init the display somehow, so maybe I should browse its source code.... ew... Jun 05 02:03:59 or wait that might actually be the kernel since CONFIG_VT would have been enabled on that system (it's not on mine) Jun 05 02:04:24 ... Jun 05 02:05:03 what? Jun 05 02:05:43 browsing the xorg source sounds like fun ;) Jun 05 02:06:19 well if there's anything useful there it'd be in the video-linuxfb driver presumably Jun 05 02:07:02 dunno what would be easier to hunt down, the kernel doing it (if I remember correctly that it does when CONFIG_VT is set) or xorg doing it Jun 05 02:07:10 or just examine the code directly Jun 05 02:07:14 bisect ... Jun 05 02:07:19 or take the plunge lol Jun 05 02:07:22 it might just be a bogosity of tilcdc Jun 05 02:08:24 huighly likely Jun 05 02:09:27 I've noticed e.g. that drivers/gpu/drm/tilcdc/tilcdc_panel.c seems quite different from drivers/gpu/drm/imx/parallel-display.c Jun 05 02:10:02 and that surprises you!? Jun 05 02:10:11 TI is .. *special* .. lol Jun 05 02:10:23 do I *look* surprised? :P Jun 05 02:10:33 tbh I'm not even sure I can blame TI here Jun 05 02:11:04 the graphics infrastructure in linux is very complicated and continuously on the move Jun 05 02:11:47 yes, thats a fair comment Jun 05 02:15:29 I've tried on occasion to understand how it fits together, but it's a maze **** ENDING LOGGING AT Sun Jun 05 02:59:58 2016