Only one SATA interface works on C6A816X EVM board

Hi,

Are you using the Linux Driver or a standalone test?

No, there is no switch. As you know, both ports have dedicated port control registers and when accessing SATA1, you should be accessing from its own register space. The HBA registers however are common to both.

Best regards, Zegeye

Hi,

Which means standalone test? I test SATA in linux os(and kernel version is the tag TI816XPSP_04.00.00.07), so the test should use linux driver.

SATA0 can detect hard disk well and mount disk, so I'm curious why SATA1 failed?

Best Regards

Hi Gary,

We have validated both SATA ports on the Linux PSP release 04.00.07. Is it possible to try latest Linux PSP release 04.00.00.10 ?

Regards

Mansoor

Hi,

I tested SATA with SD demo image and arago image(tag 2.6.34_TI816XPSP_04.00.00.07),

No interrupt is generated when insert SATA disk in SATA1/J11, however SATA0/J12 works ok with the same disk.

As you have said, both ports have been validated, I suspect there has some hardware problem on my board.

BTW: the board can't be boot with image(tag v2.6.37_TI816XPSP_04.00.00.10, is it same as v2.6.37_TI816XPSP_04.00.00.10?).

Best Regards,

Hi Gary,

Yes, tag v2.6.37_TI816XPSP_04.00.00.10 is for 04.00.00.10 release and it should work on Rev C EVM. Also, did you upgrade both u-boot and Linux to 04.00.00.10? This Linux release requires a u-boot upgrade.

Regards

Mansoor

Hi,

Hmm, it could be uboot problem,

TI8168_EVM#version

U-Boot 2010.06 (Nov 18 2010 - 22:36:59)

I have rebuilt u-boot-2010.06-psp04.00.00.09, use this uboot image and kernel image(tag 2.6.37-rc3_TI816XPSP_04.00.00.09),

but this combo also failed, console stop with "Starting kernel ... Uncompressing Linux... done, booting the kernel."

So do I don't update both uboot and linux kernel anymore.

Hi Gary,

The bootargs have changed after 04.00.00.08 release. Please check the Linux PSP Userguide for the list of changes you need to take care while moving to newer release version.

Also, 1-bit H/W ecc is used post 04.00.00.08 release. Hence even the filesystem have to be re-flashed. Enable 'earlyprintk' in bootargs for early debug prints.

Regards

Mansoor

Hi,

I have changed ttyS2 to ttyO2, and tried to reset machine ID to af0, but still failed boot board.

Any other boot args need to be modifed?

PS: the uboot is located in nand flash, kernel is downloaded from network.

Best Regards,

Hi Gary,

Could you add "earlyprintk" to bootargs and send us the bootlog. Also, please do not set machid environment variable in u-boot.

Regards

Mansoor

Hi,

Thanks for your help, log as follows.

Filename 'uImage'.
Load address: 0x81000000
Loading: #################################################################
         #################################################################
         #################################################################
         #################################################################
         #################################################################
         ########################################################
done
Bytes transferred = 1947236 (1db664 hex)
## Booting kernel from Legacy Image at 81000000 ...
   Image Name:
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    1947172 Bytes = 1.9 MiB
   Load Address: 80008000
   Entry Point:  80008000
   Verifying Checksum ... OK
   Loading Kernel Image ... OK
OK

Starting kernel ...

Uncompressing Linux... done, booting the kernel.
Linux version 2.6.37-rc3-24124-gde15396 (root@root-desktop) (gcc version 4.4.1 (Sourcery G++ 4.4a-323) ) #29 Wed Mar 23 15:56:06 CST 2011
CPU: ARMv7 Processor [413fc082] revision 2 (ARMv7), cr=10c53c7f
CPU: VIPT nonaliasing data cache, VIPT aliasing instruction cache
Machine: ti8168evm
bootconsole [earlycon0] enabled
reserved size = 0 at 0
Memory policy: ECC disabled, Data cache writeback
Truncating RAM at 80000000-bfffffff to -afffffff (vmalloc region overlap).
Ignoring RAM at c0000000-ffffffff (vmalloc region overlap).
Kernel panic - not syncing: ERROR: Failed to allocate 0x1000 bytes below 0x0.

Backtrace:
[<c003e274>] (dump_backtrace+0x0/0x110) from [<c02c7398>] (dump_stack+0x18/0x1c)
 r6:00000001 r5:00000000 r4:c03e5d10 r3:c03c653c
[<c02c7380>] (dump_stack+0x0/0x1c) from [<c02c73f8>] (panic+0x5c/0x178)
[<c02c739c>] (panic+0x0/0x178) from [<c00153dc>] (memblock_alloc_base+0x38/0x44)
 r3:00000000 r2:00001000 r1:00000000 r0:c0369916
 r7:00000002
[<c00153a4>] (memblock_alloc_base+0x0/0x44) from [<c0015430>] (memblock_alloc+0x14/0x18)
 r5:00000018 r4:c0409a78
[<c001541c>] (memblock_alloc+0x0/0x18) from [<c000d148>] (T.324+0x18/0x30)
[<c000d130>] (T.324+0x0/0x30) from [<c000da58>] (paging_init+0x578/0x6e0)
 r4:c0409a78 r3:80000000
[<c000d4e0>] (paging_init+0x0/0x6e0) from [<c000bc7c>] (setup_arch+0x390/0x618)
[<c000b8ec>] (setup_arch+0x0/0x618) from [<c0008a64>] (start_kernel+0x64/0x2a8)
[<c0008a00>] (start_kernel+0x0/0x2a8) from [<80008034>] (0x80008034)
 r5:c03e589c r4:10c53c7d

BTW: I see some error infos "Error: Bad compare! failed" when start uboot, but I don't think it is the root cause

for kernel hang.

Best Regards,

Hi Mansoor,

Great! you are genius, it is ok after pass "mem=128M".

Thank a lot!

Best Regards,

Hi Gary,

Could you confirm if the 2nd SATA port worked?

Regards

Mansoor

Hi Mansoor,

Unfortunately, the SATA port still doesn't work, no interrupt is generated when insert SATA disk.

Best Regards,