HoneyComb SATA performance issues

NXP LX2160
1

I’ve run into the same issue as this thread on two LX2160A HoneyComb boards, but only specifically on SATA ports SATA0 and SATA2 (U32) — the other two function just fine and reliably. With certain SATA drives, SATA0 is practically unusable; and SATA2 is unreliable due to the amount of errors, immediately being throttled to SATA v2 or even SATA v1 by the kernel.

Is there any known cause for this behavior? We will be designing a custom carrier card modeled after the ClearFog, and would like to be able to work around this problem if possible.

An example of the behavior:

root@localhost:~# mount /dev/sda /mnt/
[  276.179457] ata1.00: Read log 0x30 page 0x00 failed, Emask 0x5
[  276.185356] ata1.00: failed to set xfermode (err_mask=0x40)
[  276.668797] I/O error, dev sda, sector 512 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 2
[  276.719433] ata1.00: exception Emask 0x12 SAct 0x100 SErr 0x380500 action 0x6 frozen
[  276.727222] ata1.00: irq_stat 0x08000000, interface fatal error
[  276.733175] ata1: SError: { UnrecovData Proto 10B8B Dispar BadCRC }
[  276.739483] ata1.00: failed command: READ FPDMA QUEUED
[  276.744636] ata1.00: cmd 60/08:40:00:08:00/00:00:00:00:00/40 tag 8 ncq dma 4096 in
[  276.744636]          res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x12 (ATA bus error)
[  276.760239] ata1.00: status: { DRDY }
[  277.240784] I/O error, dev sda, sector 2048 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 2
[  277.303430] ata1.00: exception Emask 0x12 SAct 0x8000000 SErr 0x380500 action 0x6 frozen
[  277.311548] ata1.00: irq_stat 0x08000000, interface fatal error
[  277.317481] ata1: SError: { UnrecovData Proto 10B8B Dispar BadCRC }
[  277.323764] ata1.00: failed command: READ FPDMA QUEUED
[  277.328913] ata1.00: cmd 60/08:d8:08:00:00/00:00:00:00:00/40 tag 27 ncq dma 4096 in
[  277.328913]          res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x12 (ATA bus error)
[  277.344581] ata1.00: status: { DRDY }
[  277.824747] I/O error, dev sda, sector 8 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 2
[  277.875441] ata1.00: exception Emask 0x12 SAct 0x4000 SErr 0x380500 action 0x6 frozen
[  277.883295] ata1.00: irq_stat 0x08000004, interface fatal error
[  277.889227] ata1: SError: { UnrecovData Proto 10B8B Dispar BadCRC }
[  277.895515] ata1.00: failed command: READ FPDMA QUEUED
[  277.900663] ata1.00: cmd 60/08:70:a8:92:3e/00:00:7e:03:00/40 tag 14 ncq dma 4096 in
[  277.900663]          res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x12 (ATA bus error)
[  277.916330] ata1.00: status: { DRDY }
[  278.396765] I/O error, dev sda, sector 15002931880 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 2
mount: /mnt: wrong fs type, bad option, bad superblock on /dev/sda, missing codepage or helper program, or other error.
       dmesg(1) may have more information after failed mount system call.