20100912-0 build does not boot on beagle xm

I ran apt-get update && apt-get dist-upgrade and it still boots. Although configuring the update-initramfs in the dpkg sense doesn't seem to be working:

Setting up initramfs-tools (0.98.1ubuntu1) ...
update-initramfs: deferring update (trigger activated)
Processing triggers for initramfs-tools ...
update-initramfs: Generating /boot/initrd.img-2.6.35-1005-linaro-omap
Using u-boot partition: /dev/mmcblk0p1
mount: block device /dev/mmcblk0p1 is write-protected, mounting read-only
Creating backups of uImage and uInitrd... cp: cannot create regular file `/tmp/tmp.5pTemEd34J/uImage.bak': Read-only file system
dpkg: error processing initramfs-tools (--configure):
 subprocess installed post-installation script returned error exit status 1
Errors were encountered while processing:
 initramfs-tools

so I don't know how serious a test of my system that is.

Hi Michael,
one data point....if I use linaro-media-create to create an sd image with daily build http://snapshots.linaro.org/10.11-daily/linaro-headless/20100913/0/images/tar/linaro-m-headless-tar-20100913-0.tar.gz, the XM board boots up to the shell prompt with no problems. http://pastebin.ubuntu.com/493305/

Two questions...
1) when you created your sd image of the linaro daily build, what format did you use to create it?

2) when you performed your "apt-get update && apt-get dist-upgrade" sequence above, which image were you starting from (linaro beta 3? )?

Hi,
went back and ran with Linaro image 0912 ( http://snapshots.linaro.org/10.11-daily/linaro-headless/20100912/0/images/tar/linaro-m-headless-tar-20100912-0.tar.gz ) and it also boots to the shell prompt .... http://pastebin.ubuntu.com/493384

tried from another angle...used linaro-media-create to load and install linaro Beta image 20100830-2 followed by a " apt-get update" and then "apt-get dist-upgrade"

During the dist-upgrade kernel image linux-image-2.6.35-1005-linaro-omap was being installed as I expected (kernel image linux-image-2.6.35-1005-linaro-omap is kernel running in linaro build 20100913-0).

A few errors as dist-upgrade executed, followed by boot partition being mounted read-only. Details can be found in.... http://pastebin.ubuntu.com/493417/

the read-only fix (patch) was not added to the linaro kernel by Beta time so its expected there are problems with the boot partition being mounted read-only.

As such, if I place the sd card on my host machine, call mkimage on the new kernel (kernel from linaro build 20100913-0) and re-insert the sd card back into the XM, the XM boots to the shell prompt once more, and this time its running kernel image Linux version 2.6.35-1005-linaro-omap (buildd@satinash) (gcc version 4.4.5 20100902 (prerelease) (Ubuntu/Linaro 4.4.4-13ubuntu2) ) #10-Ubuntu Thu Sep 9 05:49:25 UTC 2010

as expected.

I did the same -- put the beta on a card, did the dist-upgrade, finished the dist-upgrade by running mkImage from my laptop and it works, booting to a prompt (and the funny business with the mounting things as read only is fixed too). So, um. No idea.

I notice the same thing as Tom in this report: https://bugs.edge.launchpad.net/linux-linaro/+bug/630283 -- could that be related? It seems a bit unlikely.

In my case for what I had reported in Bug #637459, the image I was trying to boot was a fresh 9/13 daily using the current linaro-media-create.

I've tried the 9/14 daily this morning on my Rev A2 BB XM and the results are the same as the 9/13 daily.

Texas Instruments X-Loader 1.4.4ss (Sep 6 2010 - 08:19:49)
Beagle xM Rev A
Reading boot sector
Loading u-boot.bin from mmc

U-Boot 2010.03-rc1 (Jun 01 2010 - 09:57:03)

OMAP3630/3730-GP ES1.0, CPU-OPP2, L3-165MHz,
OMAP3 Beagle board + LPDDR/NAND
I2C: ready
DRAM: 512 MB
NAND: 256 MiB
*** Warning - bad CRC or NAND, using default environment

In: serial
Out: serial
Err: serial
timed out in wait_for_pin: I2C_STAT=1000
I2C read: I/O error
Unrecognized expansion board: 0
Beagle xM Rev A
Die ID #505a0000061000000156166b0a026014
Hit any key to stop autoboot: 0
mmc1 is available
reading boot.scr

407 bytes read
Running bootscript from mmc ...
## Executing script at 82000000
mmc1 is available
reading uImage

3689500 bytes read
reading uInitrd

4790757 bytes read
## Booting kernel from Legacy Image at 80000000 ...
   Image Name: Linux
   Image Type: ARM Linux Kernel Image (uncompressed)
   Data Size: 3689436 Bytes = 3.5 MB
   Load Address: 80008000
   Entry Point: 80008000
   Verifying Checksum ... OK
## Loading init Ramdisk from Legacy Image at 81600000 ...
   Image Name: initramfs
   Image Type: ARM Linux RAMDisk Image (uncompressed)
   Data Size: 4790693 Bytes = 4.6 MB
   Load Address: 00000000
   Entry Point: 00000000
   Verifying Checksum ... OK
   Loading Kernel Image ... OK
OK

Starting kernel ...

Uncompressing Linux... done, booting the kernel.
[ 0.000000] Initializing cgroup subsys cpuset
[ 0.000000] Initializing cgroup subsys cpu
[ 0.000000] Linux version 2.6.35-1005-linaro-omap (buildd@satinash) (gcc version 4.4.5 20100902 (prerelease) (Ubuntu)
[ 0.000000] CPU: ARMv7 Processor [413fc082] revision 2 (ARMv7), cr=10c53c7f
[ 0.000000] CPU: VIPT nonaliasing data cache, VIPT nonaliasing instruction cache
[ 0.000000] Machine: OMAP3 Beagle Board
[ 0.000000] bootconsole [earlycon0] enabled
[ 0.000000] Reserving 12582912 bytes SDRAM for VRAM
[ 0.000000] Memory policy: ECC disabled, Data cache writeback
[ 0.000000] OMAP3630 ES1.2 (l2cache iva sgx neon isp 192mhz_clk )
[ 0.000000] SRAM: Mapped pa 0x40200000 to va 0xfe400000 size: 0x100000

Tom, how are you creating the image? Is this a fresh image via linaro-media-create or an upgrade?

I've installed 9/12 and 9/13 builds with linaro-media-create and it boots to the shell with no problems. I run into trouble if I dist-upgrade from Beta to 9/13 (for example) because the missing patch in Beta kernel supporting the rw file system.

ah-ha...bingo...I pulled down the most recent linaro-media-create tool and am finally able to re-create this bug.

The diff between the linaro-media-create tool I was using and the one I just pulled down is....

--- orig_linaro-media-create 2010-09-13 15:35:37.000000000 -0500
+++ new_orig_linaro-media-create 2010-09-14 11:10:50.000000000 -0500
@@ -57,11 +57,11 @@
 if [ ! ${IMAGE_FILE} ]; then
   for device in /dev/disk/by-id/*; do
     if [ `realpath $device` = $MMC ]; then
- if echo $device | grep -q -- "-part[0-9]*$"; then
+ if echo "$device" | grep -q -- "-part[0-9]*$"; then
         echo "device $MMC must not be a partition part ($device)" 1>&2
         exit 1
       fi
- for part_id in `ls $device-part*`; do
+ for part_id in `ls "$device-part"*`; do
         part=`realpath $part_id`
         part_no=`echo $part_id | sed -e 's/.*-part//g'`
         # echo "part $part_no found: $part_id" 1>&2
@@ -198,10 +198,10 @@
  sudo mount ${MMC1} ${DIR}/disk
  case "$DEVIMAGE" in
    beagle|igep)
- if [ "$DEVIMAGE" == "beagle" ]; then
- if test -e binary/${parts_dir}/${MLO_FILE} -a -e binary/${parts_dir}/${UBOOT_FILE}; then
- sudo cp -v binary/${parts_dir}/${MLO_FILE} ${DIR}/disk/MLO
- sudo cp -v binary/${parts_dir}/${UBOOT_FILE} ${DIR}/disk/u-boot.bin
+ if [ "$DEVIMAGE" = "beagle" ]; then
+ if [ -e binary/${parts_dir}/${MLO_FILE} ] && [ -e binary/${parts_dir}/${UBOOT_FILE} ]; then
+ sudo cp -v binary/${parts_dir}/${MLO_FILE} ${DIR}/disk/MLO
+ sudo cp -v binary/${parts_dir}/${UBOOT_FILE} ${DIR}/disk/u-boot.bin
           fi
      fi
      sync

with the daily 9/14 build

and the MLO and u-boot from http://www.sakoman.com/feeds/omap3/glibc/images/beagleboard/201009091145/

I am able to boot my Rev A2 BB XM and it also show the full 512Meg RAM is being used.

after much investigation, it appears there are fixes added to uboot to help solve this problem. A new version of uboot is forth coming and with that version we should be able to boot up with no problems?

This problem should be fixed in our next Linaro build.

I can confirm that the 20100915/1 build of linaro-headless, flashed onto a sd card with r92 of linaro-media-create, boots successfully to a prompt. Marking this bug fixed, feel free to undo if that's not the right thing to do.