Skip to contentSkip to conversation

Uboot mtest hang on mx31

U-Boot alfred steele 2 participants 3 messages Apr 27, 2009

lore
Share & tools
lore
#0 lore raw

Hi All.

When i do a memtest without any args, it crashes after printing the first line. I know the output may vary depending on the processor and the memmap.
I just wanted to knoe is this tha appropiate behaviour for mtest.

CPU:   Freescale i.MX31 at 531 MHz
Board: MX31 3Stack
DRAM:  128 MB
NAND:  Bad block table found at page 65472, version 0x01

Bad block table found at page 65408, version 0x01
nand_read_bbt: Bad block at 0x01b00000
nand_read_bbt: Bad block at 0x02ae0000
nand_read_bbt: Bad block at 0x02c60000
nand_read_bbt: Bad block at 0x052a0000
nand_read_bbt: Bad block at 0x052c0000
128 MiB

In:    serial
Out:   serial
Err:   serial
Hit any key to stop autoboot:  0

=>
=> mtest
Pattern 00000000 Writing...
<The serial console just freezes here, meaning an evident crash>

After reboot, when i do and md for the base address 0x0, i get the following, which proabbly is the uboot image itself in RAM. So i am guessing, the memtest tries modifying the U-boot in memory and hence the crash. i am not sure though Please correct me if i am wrong.

Here's the message log.

CPU:   Freescale i.MX31 at 531 MHz
Board: MX31 3Stack
DRAM:  128 MB
NAND:  Bad block table found at page 65472, version 0x01

Bad block table found at page 65408, version 0x01
nand_read_bbt: Bad block at 0x01b00000
nand_read_bbt: Bad block at 0x02ae0000
nand_read_bbt: Bad block at 0x02c60000
nand_read_bbt: Bad block at 0x052a0000
nand_read_bbt: Bad block at 0x052c0000
128 MiB

In:    serial
Out:   serial
Err:   serial
Hit any key to stop autoboot:  0
=>md 0
00000000: e59ff00c e59ff018 e59ff018 e59ff018    ................
00000010: e59ff018 b8000000 e59ff014 e59ff014    ................
00000020: 00000090 1fffffd0 1fffffd4 1fffffd8    ................
00000030: 1fffffdc 1fffffe0 1fffffe4 ffffffff    ................
00000040: 79706f43 68676972 63282074 30322029    Copyright (c) 20
00000050: 4d203430 726f746f 20616c6f 2e636e49    04 Motorola Inc.
00000060: 6c6c4120 67697220 20737468 65736572     All rights rese
00000070: 64657672 0000002e ffffffff ffffffff    rved............
00000080: 000002b4 000004e0 00000524 ffffffff    ........$.......
00000090: e321f0d3 e59fd008 e59f0008 e12fff10    ..!.........../.
000000a0: ea000001 1fffffd0 000000e0 e321f0d3    ..............!.
000000b0: e59fd014 e59f0014 e3a0106c e5c01000    ........l.......
000000c0: e59f000c e12fff10 eafffff7 1fffffd0    ....../.........
000000d0: 1fffc800 000002b4 ffffffff ffffffff    ................
000000e0: e59f2064 e59f005c e59f1054 e3c22003    d ..\...T.... ..
000000f0: e0823000 e1500003 34912004 34802004    .0....P.. .4. .4

Thanks,
Alfred.

#1 lore raw
Quoted 8 lines

When i do a memtest without any args, it crashes after printing the first line.

=> mtest
Pattern 00000000 Writing...

After reboot, when i do and md for the base address 0x0, i get the following, which proabbly is the uboot image itself in RAM.

Not exactly. At address 0 you have the internal boot ROM. That code is not u-boot, as you can see from comparing with the u-boot.bin you compiled.

Actually, your ram start at 2GB physical (CSD0_BASE, 0x80000000) but the MX31 config file in current git says:

#define CONFIG_SYS_MEMTEST_START 0 /* memtest works on */
#define CONFIG_SYS_MEMTEST_END 0x10000

So, most likely memtest without arguments has never been used on the board.

[disclaimer: I have mx21 but not mx31, I just assume it's similar]

/alessandro

#2 lore raw

Hi Alessandro,

Thanks!

Quoted 2 lines

#define CONFIG_SYS_MEMTEST_START ? ? 0 ? ? ? ?/* memtest works on */ #define CONFIG_SYS_MEMTEST_END ? ? ? 0x10000

So, is the config file wrong?

Quoted 2 lines

So, most likely memtest without arguments has never been used on the board.

Is it the cause for the hang, then? I thought it could be something wrong with the way i built my u-boot (linker script etc.) which was causing the crash/hang.

Regards,
Alfred