Re: A10 bad image

Started by DangeMask, January 03, 2018, 04:36:37 PM

Previous topic - Next topic

DangeMask

Not sure if I should start a new thread, but my issue seems to be the same. I have already sent an email to support, but maybe somebody can suggest something here.

We already use cca 250 devices based on A10 LIME with customized OS (based on official image). Last suply of new devices was sent to us from suplier with info: "There is a new HW revision E and our testing OS is not working, but we send it anyway".
Yesterday I got hand to the fisrt of this devices with "new board" and it didn't boot. I checked the HW revision, but all our working boards are also Rev.E. The only difference is RAM manufacturer (old - Samsung, new - Kingston).

Today I tried to boot with UART connected. our customized OS fails to boot with this output going in loop:
QuoteU-Boot SPL 2014.04-10675-g44b53fd (May 23 2014 - 13:02:55)
Board: A10-OLinuXino-Lime
DRAM: 512 MiB
CPU: 1008000000Hz, AXI/AHB/APB: 3/2/2
spl: not an uImage at 1600


U-Boot 2014.04-10675-g44b53fd (May 23 2014 - 13:02:55) Allwinner Technology

CPU:   Allwinner A10 (SUN4I)
Board: A10-OLinuXino-Lime
I2C:   ready
DRAM:  512 MiB
MMC:   SUNXI SD/MMC: 0
In:    serial
Out:   serial
Err:   serial
Net:   emac
Hit any key to stop autoboot:  0
reading uEnv.txt
116 bytes read in 17 ms (5.9 KiB/s)
Loaded environment from uEnv.txt
reading boot.scr
347 bytes read in 16 ms (20.5 KiB/s)
Jumping to boot.scr
## Executing script at 44000000
reading script.bin
Error reading cluster
** Unable to read file script.bin **
reading uImage
data abort
pc : [<5ff87518>]          lr : [<5ff9c968>]
sp : 5fb63f10  ip : 0000020f     fp : 5fb66094
r10: 00000000  r9 : 5fb66f0c     r8 : 80003340
r7 : 00000000  r6 : 5fb660b4     r5 : 5ffbc738  r4 : 5fb660a4
r3 : 48000000  r2 : 00000040     r1 : 4841c800  r0 : 48000000
Flags: Nzcv  IRQs off  FIQs off  Mode SVC_32
Resetting CPU ...

resetting ...

I have donwloaded brand new official image from Olimex wiki and tried again. The error message is different, but the result is the same.

QuoteU-Boot SPL 2014.04-10675-g44b53fd (May 23 2014 - 13:02:55)
Board: A10-OLinuXino-Lime
DRAM: 512 MiB
CPU: 1008000000Hz, AXI/AHB/APB: 3/2/2
spl: not an uImage at 1600


U-Boot 2014.04-10675-g44b53fd (May 23 2014 - 13:02:55) Allwinner Technology

CPU:   Allwinner A10 (SUN4I)
Board: A10-OLinuXino-Lime
I2C:   ready
DRAM:  512 MiB
MMC:   SUNXI SD/MMC: 0
In:    serial
Out:   serial
Err:   serial
Net:   emac
Hit any key to stop autoboot:  0
reading uEnv.txt
104 bytes read in 17 ms (5.9 KiB/s)
Loaded environment from uEnv.txt
reading boot.scr
337 bytes read in 16 ms (20.5 KiB/s)
Jumping to boot.scr
## Executing script at 44000000
reading script.bin
54504 bytes read in 38 ms (1.4 MiB/s)
reading uImage
Error reading cluster
** Unable to read file uImage **
## Booting kernel from Legacy Image at 48000000 ...
   Image Name:   Linux-3.4.90+
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4447440 Bytes = 4.2 MiB
   Load Address: 40008000
   Entry Point:  40008000
   Verifying Checksum ... OK
   Loading Kernel Image ... data abort
pc : [<5ffa8410>]          lr : [<5ff8d474>]
sp : 5fb66b38  ip : 00000000     fp : 48000040
r10: 5ffb9a80  r9 : 5fb66f0c     r8 : 5fb68ac4
r7 : 00000000  r6 : 00000000     r5 : 0000070f  r4 : 40008000
r3 : 00000000  r2 : 0043dcd0     r1 : 48000040  r0 : 40008000
Flags: Nzcv  IRQs off  FIQs off  Mode SVC_32
Resetting CPU ...

Both cards work perfectly in "older" board, so I assume there is no problem with the SD card.

I have a box of 30 new devices so I will test them tomorrow to be sure the issue is only with the "new" ones with Kingston RAM.

PS:
QuoteError reading cluster
** Unable to read file uImage **
This error message is shown even in my working combination but with no negative efect. The "main" error message is "data abort" in bold.

LubOlimex

It might be another problem, unrelated to the RAM memory. A hardware fault or something else. We haven't modified the official image for the A10 boards since 2014. It also seems to work fine with the boards with Kingston memories without issues here. Comparing your boot logs with the boot logs I get here when using the image from the wiki, I get same versions for u-boot, so it seem you used the same image as a base for your image (the u-boot build time is the same). My test with a brand new A10 board with Kingston memory and the image from the wiki (torrent) went fine and you can find my log here: https://justpaste.it/1f9fe

How do you power the board? Can you test powering a board in question stand-alone (not via the IPC) using the board's power jack and a 5V power supply adapter capable of providing at least 2A of current (5x2 = at least 10W)?

I've seen exact same issue mentioned before we changed memories here: https://www.olimex.com/forum/index.php?topic=5388.0
Technical support and documentation manager at Olimex

DangeMask

The IPC is powered by 24V 1A power adapter and has its own 9-30V to 5V converter. I only tried to power the olinuxino directly with mobile charger and miniUSB cable with the same results, but the charger might have been weak.

I'll try a few more "new" IPCs today to test if there is any difference with "old" and "new" boards or if the issue stays the same for the whole batch. Then I'll try to switch one of the "not working" boards to "working" IPC and we will see.

LubOlimex

Testing more than one board would also help a lot. Let me know when you test more boards and with more powerful adapter.
Technical support and documentation manager at Olimex