03:28 | jmontleon | Exaga, rabeeh fwiw even with the obscene ssd temps it still works great |
03:34 | jmontleon | s/ssd/ssd induced/ |
04:19 | jnettlet | cbxbiker61, oh when loadbootscript is run. yes I don't run autodetectfdt there because I assume if you are running a bootscript you want to replace the entire boot command. |
04:20 | jnettlet | I think if you writing your own custom bootscript you can decide if you want to run autodetectfdt or not. |
08:00 | R0nd | what's the best way to take a backup of the running system on cubox? should I just rsync it? |
10:38 | kbouhara | Hi guys, is there a porting of freescale adv7180 driver on linux-fslc v3-14 ? |
16:27 | matoking | I installed network-manager |
16:28 | matoking | but it also installed systemd and it seems that I can no longer boot the system or access it via SSH |
16:28 | matoking | Basically I receive a bunch of "Welcome to emergency mode! Give root password for maintenance or type Ctrl+D to continue" messages during bootup |
16:29 | matoking | I spam the hell out of Control+D and other than that the boot process appears to continue normally until nothing more appears |
16:34 | matoking | Any idea how I can remove systemd or restore normal boot so I can actually login to the terminal and uninstall it? |
16:35 | Humpelstilzchen | you can always put the sd-card in your linux-pc |
16:35 | matoking | I have the root filesystem on an USB HDD |
16:35 | matoking | is there some file I can edit to disable systemd? |
16:35 | Humpelstilzchen | no you have to uninstall it and reinstall sysvinit |
16:37 | matoking | Well I tried logging into maintenance mode but it always gave me an error about the login being incorrect |
16:38 | Humpelstilzchen | well if it can't read the rootfs.. |
16:39 | Humpelstilzchen | well as said I'ld plug the disk in your Linux PC, mount it, chroot and reinstall sysvinit |
16:43 | quitte | Hi. I'm having trouble finding the documentation for the usom so I can tell which are the jtag testpads |
16:44 | quitte | Also: is the jtag voltage supposed to be 3.3,3.0 or 1.8V ? |
16:45 | matoking | I checked and installing network-manager didn't uninstall anything with sysvinit on it |
16:46 | matoking | Still, I don't really have time to mess with my Linux installation right now |
16:47 | matoking | I was just hoping to install network-manager to make network managing less of a hassle, but apparently it thought that would've made things too trivial :P |
16:50 | quitte | found the manual. So now I only need to find a good point to get JTAG VRef from |
16:58 | deniska | Humpelstilzchen: it is not so easy to chroot into cubox drive considering it uses arm |
16:58 | deniska | matoking: probably mount configuration is wrong |
16:59 | deniska | systemd uses something else instead of fstab afair, and when it uses fstab, it may treat some defaults differently |
16:59 | Humpelstilzchen | deniska: it is binfmt to the rescue |
17:03 | Humpelstilzchen | and qemu |
18:56 | Exaga | rabeeh: are you around? |
18:58 | Exaga | malte: o/ wie gehtes ? |
19:33 | rabeeh | Exaga: here |
19:34 | Exaga | hi rabeeh |
19:34 | Exaga | can i pick your brains my friend? :D |
19:34 | rabee | 19:34 * rabeeh opens his brains for service |
19:34 | Exaga | i'm having a problem with "S0 respawning too fast" console error |
19:35 | Exaga | now i know the hummingboard uses console=ttymxc0 |
19:35 | Exaga | and slackware pretty much uses console=tty1 |
19:35 | Exaga | so my question is... can i specify both? |
19:35 | rabeeh | yes |
19:35 | Exaga | excellent |
19:35 | rabeeh | did slackware move to systemd? |
19:36 | rabeeh | or sysv? |
19:36 | Exaga | hmm i'm not sure |
19:36 | Exaga | you'd have to ask the dev's about that one :> |
19:36 | rabeeh | if sysv then look at /etc/inittab; that's the place to define the login per tty |
19:36 | Exaga | all i know it... tty1 gives me a respawn error and ttymxc0 does not |
19:36 | Exaga | but ttymxc0 blanks the screen during boot |
19:37 | Exaga | until you hit the login prompt |
19:37 | Exaga | the truth is rabeeh... i'm a noob with console and i'm trying to find a solution |
19:37 | Exaga | if you have any advice it will be very well accepted right now lol :D |
19:37 | Exaga | i'm currently testing using both console=tty1 and console=ttymxc0 |
19:38 | Exaga | specified in the uEnv.txt |
19:38 | rabeeh | console=ttymxc0,1152oo or console=tty1 are only where to throw the boot message at |
19:38 | rabeeh | it's completely unrelated where you open a login prompt |
19:38 | rabeeh | opening a login prompt is decided in /etc/initab in sysv style and under /etc/systemd/ ... in a systemd based one |
19:39 | Exaga | yes |
19:39 | rabeeh | do you have /etc/inittab file at all? if so please pastebin.com it |
19:39 | Exaga | hang on rabeeh |
19:40 | Exaga | http://rpi.fatdog.eu/files/inittab |
19:40 | Exaga | i'm currently testing with this one |
19:40 | Exaga | my modification is .. |
19:40 | Exaga | # Local serial lines: |
19:41 | Exaga | s0:12345:respawn:/sbin/getty -L ttymxc0 115200 vt100 |
19:41 | Exaga | .. |
19:41 | Exaga | only that line |
19:41 | rabeeh | looks great :) |
19:41 | Exaga | ok. now you know better than I |
19:41 | Exaga | if you say it's good i shall continue to test |
19:41 | Exaga | would you like to see my uEnv.txt? |
19:42 | rabeeh | c1:12345:respawn:/sbin/agetty --noclear 38400 tty1 linux |
19:42 | rabeeh | why do you need --noclear? |
19:42 | rabeeh | just to keep the boot log? |
19:42 | Exaga | i'm not sure rabeeh. i just left that alone |
19:42 | Exaga | as i say... i am a noob with console :D |
19:43 | rabeeh | --noclear |
19:43 | rabeeh | Do not clear the screen before prompting for the login name (the screen is normally cleared). |
19:43 | rabeeh | from man agetty |
19:43 | rabeeh | so it's really once you don't have 'console=ttymxc0,115200' then it defaults to tty1 |
19:43 | rabeeh | and the --noclear wouldn't clear the screen and keep the boot info just before issuing a login prompt |
19:44 | Exaga | ok. this is way over my head |
19:44 | Exaga | do i need to use ttymxc0 ? |
19:44 | Exaga | or can i just use tty1 ? |
19:45 | Exaga | rabeeh: http://rpi.fatdog.eu/files/uEnv.txt |
19:45 | Exaga | let me know what you think :> |
19:46 | Exaga | if i just use console=tty1 i get the s0 respawning too fast error |
19:56 | rabeeh | Exaga: i'm assuming that fatload and bootz being in different lines and not concatenated to mmcargs2 is a cut and paste mistake |
19:57 | rabeeh | console=ttymxc0,115200 means i'm a developer, and know how to hook a serial port and see what's going on |
19:57 | rabeeh | console=tty1 means that i'm just browsing around to see what happens (display oriented) |
19:57 | rabeeh | for simplicity i would go with console=tty1 |
19:57 | Exaga | right |
19:57 | Exaga | i know that much at least hehe |
19:57 | rabeeh | or even better; just remove 'console=tty1' and let it choose the first graphical tty |
19:57 | Exaga | ahhh ok great. i shall try that |
19:58 | Exaga | because using just tty1 gives me the respawn error |
19:59 | Exaga | and i see the file with 5 different lines |
20:00 | Exaga | if there is an error please let me know |
20:00 | Exaga | should there just be 2 lines? |
20:01 | rabeeh | two lines |
20:01 | Exaga | ok then i shall change it |
20:01 | rabeeh | mmcargs2=run autodetectfdt; fatload mmc 0:1 0x10800000 zImage; fatload mmc 0:1 0x19000000 ${fdt_file}; bootz 0x10800000 - 0x19000000 |
20:01 | rabeeh | mmcargs=setenv bootargs root=/dev/mmcblk0p3 rootwait console=ttymxc0,115200n8 earlyprintk=serial,ttymxc0,115200,keep video=mxcfb0:dev=hdmi console=tty1 raid=noautodetect consoleblank=0 ro; run mmcargs2 |
20:01 | Exaga | yes that's how i've changed it |
20:01 | Exaga | thanks rabeeh |
20:01 | rabeeh | please remove console=tty1 too |
20:01 | Exaga | ok |
20:02 | Exaga | done |
20:02 | Exaga | rabeeh: http://rpi.fatdog.eu/files/uEnv.txt |
20:02 | Exaga | refresh it |
20:04 | rabeeh | great |
20:04 | Exaga | cool. thanks a million <3 |
20:04 | rabeeh | no you mentioned ramdisk; is it now initramfs inside the kernel? |
20:04 | Exaga | yes initramfs |
20:04 | rabeeh | i would add a note in uEnv.txt then (before bootz) |
20:04 | Exaga | there is no initrd in the file |
20:04 | rabeeh | a simple '# Notice that the below zImage includes initramfs' |
20:05 | Exaga | you can use #comments in uEnv.txt ? :O |
20:05 | rabeeh | yeah i understand; just a simple note for the people viewing the uEnv.txt file and asking themselves where the hell did this ramdisk come from |
20:06 | Exaga | refresh it rabeeh :> |
20:06 | Exaga | like that? |
20:06 | rabeeh | yup |
20:07 | Exaga | excellent |
20:07 | Exaga | but now when i boot... the screen is blank until i get to the login prompt |
20:07 | Exaga | and that takes approx 20-30 seconds |
20:11 | rabeeh | so probably console=tty1 related? |
20:11 | Exaga | yes |
20:17 | Exaga | so... a bootload wallpaper would be good here |
21:12 | matoking | So, I just realized I can use a microUSB cable to use the terminal in my Cubox-i |
21:12 | matoking | Instead of having to swap my keyboard's USB cable between my computers constantly |
21:12 | matoking | Anyway, I can't get a /dev/ttyUSBx device to pop up |
21:14 | matoking | I'm thinking it's that the kernel isn't getting far enough in the boot process to load usbserial |
21:27 | jmontleon | matoking, nothing has to load on the cubox-i to get serial console |
21:28 | jmontleon | you can view u-boot sequence, etc. before linux even boots |
21:28 | jmontleon | i've had trouble with some micro-USB cables making a good connection (just this weekend actually); if you have another maybe give it a try instead |
21:30 | jmontleon | i think the usb port on your PC actually provides enough power to get the ttyUSB0 interface up - you shouldn't even need the cubox-i to actually even be plugged in |
21:30 | matoking | jmontleon: Yeah, I think the reason why the USB activity light is blinking has something to do with all these "new full-speed USB device" messages I'm getting at dmesg constantly :P |
21:31 | matoking | Wait, not actually |
21:32 | matoking | jmontleon: Wait, don't you need to configure the uBoot to enable serial console? |
21:33 | matoking | Yeah, the USB activity light stays on even if the Cubox-i is powered off |
21:33 | rabeeh | matoking: CuBox-i pro carrier (typically CBi2ex and higher end) comes with a built-in usb to serial (ftdi chip) that is powered from the PC's USB port |
21:34 | matoking | Well the cable I'm using is the data+charger cable for my smartphone |
21:34 | matoking | Do you suppose that should normally work? |
21:36 | jmontleon | 'should', yes. though my phone charger cable won't even fit even though it's a micro-USB cable; the plastic just blocks it out. |
21:36 | jmontleon | as i said i've had bad luck with 1 or 2 cables |
21:36 | jmontleon | in general though it shouldn't be picky |
21:37 | matoking | jmontleon: Yeah, getting it to fit seems kinda flaky |
21:37 | matoking | The USB activity light goes on but it doesn't go all the way in |
21:38 | matoking | No idea if I have another type B microUSB cable somewhere, I'll have to check tomorrow |