1. Openembedded-Issues
  2. Topics

H1940 Boot Issue

Arrmo
 


Hi,

 

My apologies in advance – as this is a fairly basic operating question (so likely just me doing something stupid … L), and I’m not sure which list to post this to … so for now I’m posting to all three (so sorry for the spam!).

 

I am trying to get OpenEmbedded running on my iPAQ 1940. I have to do this myself, as Angstrom does not seem to have a build setup for this target hardware. However, I have been able to follow the OE wiki, and have the build up and running just fine now – or at least I’m able to generate all the required outputs it seems … I say this, because it’s not quite working on my iPAQ. Let me explain what I see …

 

1)      The kernel does seem to come up fine – and if I turn kernel logging on (using loglevel in the command line passed over by HaRET) I do see expected outputs on the screen (as I’m directing console output to tty0).

2)      I seem to get to the point where the kernel is trying to launch the file system – but that’s where everything just seems to hang. I see a message about loading a RAMDISK (if I use initrd), or mounting the file system if I create a partition and file system on my SD card. The same result in either case – the RAMDISK / mount message, then a console output noting that some initial memory has been freed up … then nothing more. The system is not dead, as I do get messages on the console if I plug / unplug the USB cable, or if I press keys on the iPAQ … but the system does not boot.

3)      I took the same initrd images and tried running them under QEMU (I really did try to debug this myself first! … J). I have to use a different kernel … which I don’t quite understand yet, so if someone could explain this also I would appreciate it, but that’s a side issue. In any case, QEMU boots off its own kernel (for an ARM processor also), and they happily launches my initrd and works just fine. I have tried three images … console-image, x11-image, and opie-image – they all work fine in QEMU (but not on the iPAQ itself).

4)      I tried moving from initrd (RAMDISK) to the file system on the SD card, hoping then to see some information in the Linux logs that would help me out (i.e. /var/log), but after I (forcefully) reboot the iPAQ and pull the SD card back out there is nothing here … so no help unfortunately.

 

Does anyone have any debugging tricks or pointers that they can pass on to me? I’m definitely open to any suggestions! Of course, if you have seen similar issues and know how to resolve this it would be greatly appreciated. Unfortunately I seem to be stuck right now.

 

Thanks in advance for your help!!!

 

 
Arrmo
 

Hi,

One more observation after a bit of debugging here - I am able to use my OE generated kernel with an older (non-OE) initrd was able to find ... so it seems that the OE generated file system / initrd is the issue. I have tried using a ramdisk, and also a SD partition, to debug the issue - but the boot hangs before anything is written to the log files (on the SD partition).

Any suggestions how to debug this?

Thanks!



On Sun, Aug 22, 2010 08:01  AM, <openembedded@...> wrote:

Hi,

 

My apologies in advance – as this is a fairly basic operating question (so likely just me doing something stupid … L), and I’m not sure which list to post this to … so for now I’m posting to all three (so sorry for the spam!).

 

I am trying to get OpenEmbedded running on my iPAQ 1940. I have to do this myself, as Angstrom does not seem to have a build setup for this target hardware. However, I have been able to follow the OE wiki, and have the build up and running just fine now – or at least I’m able to generate all the required outputs it seems … I say this, because it’s not quite working on my iPAQ. Let me explain what I see …

 

1)      The kernel does seem to come up fine – and if I turn kernel logging on (using loglevel in the command line passed over by HaRET) I do see expected outputs on the screen (as I’m directing console output to tty0).

2)      I seem to get to the point where the kernel is trying to launch the file system – but that’s where everything just seems to hang. I see a message about loading a RAMDISK (if I use initrd), or mounting the file system if I create a partition and file system on my SD card. The same result in either case – the RAMDISK / mount message, then a console output noting that some initial memory has been freed up … then nothing more. The system is not dead, as I do get messages on the console if I plug / unplug the USB cable, or if I press keys on the iPAQ … but the system does not boot.

3)      I took the same initrd images and tried running them under QEMU (I really did try to debug this myself first! … J). I have to use a different kernel … which I don’t quite understand yet, so if someone could explain this also I would appreciate it, but that’s a side issue. In any case, QEMU boots off its own kernel (for an ARM processor also), and they happily launches my initrd and works just fine. I have tried three images … console-image, x11-image, and opie-image – they all work fine in QEMU (but not on the iPAQ itself).

4)      I tried moving from initrd (RAMDISK) to the file system on the SD card, hoping then to see some information in the Linux logs that would help me out (i.e. /var/log), but after I (forcefully) reboot the iPAQ and pull the SD card back out there is nothing here … so no help unfortunately.

 

Does anyone have any debugging tricks or pointers that they can pass on to me? I’m definitely open to any suggestions! Of course, if you have seen similar issues and know how to resolve this it would be greatly appreciated. Unfortunately I seem to be stuck right now.

 

Thanks in advance for your help!!!

 

 

1 - 2 of 2