SATA (Hitachi 500GB) would not be detected by kernel

[SimonasL_]

Hello,

With latest ltib uboot and kernel (L2.6.35_11.09.01_ER_source with L2.6.35_MX53_201112_Patches.tar.gz applied do not detect SATA HDD (other kernels from Linaro succeed to boot from this HDD...), all CONFIG_SATA_* options are enabled (here's the .config: http://pastebin.com/aS2Mcahr). The output speaks for itself:

U-Boot 2009.08 (Mar 06 2012 - 10:14:53)
...
MX53-LOCO U-Boot > sata init
AHCI 0001.0100 32 slots 1 ports 3 Gbps 0x1 impl SATA mode
flags: ncq stag pm led clo only pmp pio slum part
SATA Device Info:
S/N: 081006BK0F00VEGBM9AB
Product model number: Hitachi HTS545050KTA300
Firmware version: BKFOC60M
Capacity: 976773168 sectors
...
Linux version 2.6.35.3-1129-g691c08a (sledges@quercus) (gcc version 4.4.4 (4.4.4_09.06.2010) ) #5 PREEMPT Wed Mar 21 14:53:34 GMT 2012
...
ata1: link is slow to respond, please be patient (ready=0)
ata1: softreset failed (device not ready)


Kind regards, hope we could fix this, many thanks in advance;

Simonas

[Danial]

I have the same problem.

[SimonasL_]

Thanks for reply Mike,

Bear in mind that 2.6.38 ltib kernel inits SATA fine. Just tried Samsung SP2504C 250GB same behaviour.. So it is not disk related. Probably worth backporting fsl_sata to 2.6.35 , as that kernel looks most supported and patched (touchscreen works etc)

Kind regards,

Simonas

[MikeThompson]

I've had a very similar the with a 2.5" Western Digital WD5000BEVT 500GB SATA drive.  Other systems access the drive fine, but this kernel has problems.  I traced it through the kernel and wrote some notes on what I found.  The drive produces PORT_IRQ_CONNECT for a status which is interpreted by the existing AHCI code (libahci.c) as an error and the port is frozen.

I believe the issue is related to using drives that support hot swap capability confusing the AHCI code.  Other drives not intended for hot swap capability don't seem to have this problem.  Either the AHCI code or the drive is not following the serial ata hci spec v1.1 -- I suspect the AHCI code. Useful information on how the AHCI code should be handling drives with hot swap capabilities here:

http://www.intel.com/content/www/us/en/io/serial-ata/serial-ata-ahc...

I never did find a fix for this, but if someone does have a fix, please pass it along.

Also, at least for my drive, u-boot had similar issues accessing it.  I suspect for the same reason.

Mike