Open sstickel opened 9 years ago
Odd, can you do me a favor and pastebin.com your full "terminal" log after running that "mk_mmc.sh" command..
Sure, here you go: http://pastebin.com/dtKJHdsk
Thanks! Btw, also what version of sfdisk? ( i also just added it to the mk_mmc.sh script for debugging )
My es is at home, so i'll try it over lunch. but on my non-es, it's working. So MLO might just be having issues finding u-boot.img.
U-Boot SPL 2014.07-00014-g2efdfb0 (Jul 22 2014 - 10:58:52)
OMAP4430 ES2.1
SPL: Please implement spl_start_uboot() for your board
SPL: Direct Linux boot not active!
reading u-boot.img
reading u-boot.img
U-Boot 2014.07-00014-g2efdfb0 (Jul 22 2014 - 10:58:52), Build: jenkins-github_Bootloader-Builder-371
CPU : OMAP4430 ES2.1
Board: OMAP4 Panda
I2C: ready
DRAM: 1 GiB
MMC: OMAP SD/MMC: 0
Using default environment
cosmo@rollmops2: ~/netinstall $ sfdisk --version sfdisk von util-linux 2.20.1
Generally u-boot works in combination with my ES. I was playing with u-boot v2014.07 this morning and was able to boot trusty. (Almost precisely doing what is documented here: http://eewiki.net/display/linuxonarm/PandaBoard)
Very odd:
voodoo@poseidon:~$ sudo sfdisk --version
sfdisk from util-linux 2.20.1
U-Boot SPL 2014.07-00014-g2efdfb0 (Jul 22 2014 - 10:58:52)
OMAP4460 ES1.1
SPL: Please implement spl_start_uboot() for your board
SPL: Direct Linux boot not active!
reading u-boot.img
reading u-boot.img
U-Boot 2014.07-00014-g2efdfb0 (Jul 22 2014 - 10:58:52), Build: jenkins-github_Bootloader-Builder-371
CPU : OMAP4460 ES1.1
Board: OMAP4 Panda
I2C: ready
DRAM: 1 GiB
MMC: OMAP SD/MMC: 0
Using default environment
btw, it still doesn't get too far: Although my es might be terminal...
[ 5.683990] twl6030_uv_to_vsel:OUT OF RANGE! non mapped vsel for 1375000 Vs max 1316660
[ 5.692413] twl6030_uv_to_vsel:OUT OF RANGE! non mapped vsel for 1375000 Vs max 1316660
[ 5.700836] twl6030_uv_to_vsel:OUT OF RANGE! non mapped vsel for 1375000 Vs max 1316660
[ 5.709228] twl6030_uv_to_vsel:OUT OF RANGE! non mapped vsel for 1375000 Vs max 1316660
[ 5.717651] twl6030_uv_to_vsel:OUT OF RANGE! non mapped vsel for 1375000 Vs max 1316660
[ 5.726074] twl6030_uv_to_vsel:OUT OF RANGE! non mapped vsel for 1375000 Vs max 1316660
[ 5.734497] twl6030_uv_to_vsel:OUT OF RANGE! non mapped vsel for 1410000 Vs max 1316660
[ 5.742919] omap2_set_init_voltage: unable to find boot up OPP for vdd_mpu
[ 5.750122] omap2_set_init_voltage: unable to set vdd_mpu
[ 5.755828] omap2_set_init_voltage: unable to find boot up OPP for vdd_core
[ 5.763153] omap2_set_init_voltage: unable to set vdd_core
[ 5.768890] omap2_set_init_voltage: unable to find boot up OPP for vdd_iva
[ 5.776123] omap2_set_init_voltage: unable to set vdd_iva
[ 5.782775] cpufreq_cpu0: failed to get cpu0 regulator: -19
[ 5.789184] Power Management for TI OMAP4+ devices.
[ 5.794342] Power Management for TI OMAP4.
[ 5.798645] OMAP4 PM: u-boot >= v2012.07 is required for full PM s
U-Boot SPL 2014.07-00014-g2efdfb0 (Jul 22 2014 - 10:58:52)
OMAP4460 ES1.1
SPL: Please implement spl_start_uboot() for your board
SPL: Direct Linux boot not active!
reading u-boot.img
reading u-boot.img
Okay, this definetly has something to do with the partition table. I had put this (http://eewiki.net/display/linuxonarm/PandaBoard) into a script, so it is definitely reproducible what I did now and what I did this morning. This morning everything booted up fine, but now I ran my script again and it didn't boot. My script does not partition the sd card or create a fresh table. I reused your partition table on the sd card, formatted the first patition as fat32 and added the rest as ext4. Ended up exactly the same as with your netinstall script, only welcome string comes up.
I'll be back :)
Uhm, now I am confused.
sudo fdisk -l /dev/sdj
tells me that this is a gpt table???
gparted reports it as msdos table as it should
That is weird, as sfdisk isn't designed for gpt table's. We also zero out the partition table, so sfdisk should see an empty disk: https://github.com/RobertCNelson/netinstall/blob/master/mk_mmc.sh#L845
from: http://linux.die.net/man/8/sfdisk
sfdisk doesn't understand GUID Partition Table (GPT) and it is not designed for large partitions. In particular case use more advanced GNU parted(8).
I'm pretty sure something with the partitioning/file writing was broken by some recent commits.
I was trying to image a couple of beaglebone blacks today.
64GB sd card -> was trying to overwrite debian jessie with ubuntu trusty, kept booting into debian jessie
32GB sd card -> deleted all partitions and flashed with trusty serial installer, got couldn't read uEnv.txt
I tried an old checkout of netinstall from May 30th and that works fine.
Jolan bbb =/= panda.. use a new "issue"... but since your have a bbb, the old "eMMC" bootloader sounds like it's getting in the way.. remove that first..
SD Card does not boot here.
Host System: Ubuntu 14.04 Target: Pandaboard ES B1
Tried using jessie wheezy and trusty, all the same
Output I get on serial console is only the greeting string of u-boot:
Left it alone for another 20 minutes, nothing else happens