Boot loop

The ppcboot transfered to the box is executed and overtake the control.
The automatic for flashing don't start in your case. Normally if you use flashing with null modem cable the commands are transfered via that null modem cable. You can break off all the attempts of the ppcboot with ctrl and c on your pc keyboard and type then all commands by your own.
 
Talked to a my linux expert at work today.
He confirmed what you have told me in your latest post.
He confirmed that it is possible to replace the file if we can identify the boot loop offender, as long as it isn't the BIOS.
Going to take box to work early next week to test his expertise.
Will post again when I have some progress.

As a slight side issue... do you know how to replace the boot loader image ? I suspect that this is my offender. If I could replace the boot loop box's boot loader it would be interesting to see what it does.
 
There is nothing to replace.
Start with the normal flashing procedure and if the ppcboot is crying for missing files break it off. You get a prompt and the control is at your finger tips.
 
Last edited:
There should be no spaces in the path to your image. No spaces in folder names and no spaces in the image name. Thats maybe the reason for your problem.
 
@mgb
Your input has been appreciated and I hope you don't take offence at this but I think we may be missing each others objective here.
I can flash the box with any image I choose. It will accept the new image.
It will start up and I can see that the box has a new image in the LCD...but it keeps looping on start up.
That's why I'm looking more closely at the boot process and what it calls. This is where my Linux expert will come in.
The question about the bootloader stems from the fact that my sportster image on my working box let's me take a copy of my BR bootloader. I would like to somehow put this on the looping box to see what happens. Any idea how this can be done ?
 
After booting up a piece of software the job is done for the bootloader.
If you get a loop from the sportster image the bootloader has nothing to do with that.
If you flash without null modem you get no flash log. You cannot see if anything went wrong.
If you flash with null modem cable you get a flash log with all commands and with all answers to that commands, but it looks like you are not able to flash with null modem cable.
 
OK. Here we go.
I can't flash using the null modem because it loops.
I can flash (I think) using just the cross over but box loops when started in normal mode. ie. No pc attached.
I am assuming the cross over only method works for two reasons.
1. The LCD gives me the displayed status's I would expect from a 'working' box during the flash process.
2. When starting up, just after the BMON LADE LCD I get a different display stating the name of the new image I've just flashed just before it reboots which indicates that the image was flashed succesfully. Granted. At this point I don't have a log to prove it was successful but if it hadn't flashed properly, would I not get the original image displayed on the LCD everytime ?
It is this logic that point me to the boot process as being corrupt.
The question is at what point does the boot loader hand over control to the image ?
 
The bootloader is booting up the u-boot from the image.
If you see message from the u-boot, the job of the bootloader is done and the u-boot overtake the control. If the image or parts of the image are corrupted for instance nothing go further.
You can try the flashing method from dietmar-h.
http://www.dietmar-h.net/alt_flash_eng.html
Let your null modem cable connected and you should get also a flash log.
Select the correct com port and 9600.
 
Interesting detail about the bootloader and u-boot.
I know the images I'm flashing are working ok on other boxes.
If the u-boot is part of the working flashed images, then the looping could be indicating a hardware fault !?

Thanks for the link. I was looking at this web page today.
I'll give it a go and keep you posted.
Cheers
pelyo
 
After flashing type
cmp.l 120000 10020000 1F8000
behind dbox2-ppcboot>
That command compare the image in the flash chips with the datas transfered via network.
 
dbox2-ppcboot> cmp 120000 10020000 1F8000
word at 0x00120000 (0xdbcda9f3) != word at 0x10020000 (0x73717368)
Total of 0 words were the same

What does this mean ?
Full log attached.
 
The command is cmp.l and then the numbers.
l stands for long.
 
Sorry about that. Miss read the command.

dbox2-ppcboot> cmp.l 120000 10020000 1F8000
word at 0x00120000 (0xdbcda9f3) != word at 0x10020000 (0x73717368)
Total of 0 words were the same

Same output though.
 
There was nothing written to the flash chips.
Faulty flash chips or maybe programming voltage too low.
 
So we finally come to the end of the road ! The conclusion is that the box is knackered.
I must admit that at times it's been frustrating but also interesting.
I think this exercise has expanded my knowledge and hopefully I'll be able to contribute more to this forum.
Many thanks to all that participated and BIG BIG thanks to mgb.
Cheers for the help.
pelyo
 
You can still use that box with a yadd image booting up via network.
 
So it's not all over !

That sounds a bit impractical for just basic use (watching TV).
What sort of setup are we looking at ?

Got to sign off for now. Be back tomorrow.
 
A yadi image is a normal legal image created from the official dbox linux developers. The official linux developers don't want any relationship to illegal issues and such a image don't include emus. If you want to use a yadi image you have to builtin an emu by your own.
Yadd images are something different and are designed to be stored on the pc and booted up via network. The box get all datas via a nfs share from the pc hard disk. You have to builtin emus also by your own.
@pelyo
http://www.dietmar-h.net/yadd_eng.html
The name for the bootfile changed in the meantime from ppcboot to u-boot.
 
You must log in or register to reply here.

Similar threads

chookey
Humidifier
Replies
11
Views
543
D 8 RCS
D 8 RCS
Ferret
    • Like
Qviart Dual 4k Uhd Linux Enigma2+Android 9.0 Combo Review & Setup Guide
Replies
14
Views
3K
Ferret
Ferret
alimac
    • Like
CleanSweep 2.3.1
Replies
0
Views
211
alimac
alimac
alimac
    • Like
Disable TPM, Secure Boot and RAM requirements for Windows 11 with the latest Rufus beta
Replies
0
Views
635
alimac
alimac
G
Video Query
Replies
0
Views
194
goalseeker
G
Back
Top