[Bug 98171] [Regression] Marvell SE91xx SATA 3 controllers not recognized correctly

[bugzilla-daemon@bugzilla.kernel.org]

https://bugzilla.kernel.org/show_bug.cgi?id=98171

jrickman@myamigos.us changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jrickman@myamigos.us

--- Comment #6 from jrickman@myamigos.us ---
I see almost the exact same error across multiple (qty 4, all the same HW
configs) Supermicro X10SBA motherboards that I have.

Jan 30 06:52:52 tank2 kernel: [    1.689428] ata12.00: exception Emask 0x0 SAct
0x0 SErr 0x0 action 0x6
Jan 30 06:52:52 tank2 kernel: [    1.689430] ata12.00: irq_stat 0x40000001
Jan 30 06:52:52 tank2 kernel: [    1.689440] ata12.00: cmd
a0/01:00:00:00:01/00:00:00:00:00/a0 tag 2 dma 16640 in
Jan 30 06:52:52 tank2 kernel: [    1.689440]          opcode=0x12 12 01 00 00
ff 00res 00/00:00:00:00:00/00:00:00:00:00/00 Emask 0x3 (HSM violation)
Jan 30 06:52:52 tank2 kernel: [    1.689446] ata12: hard resetting link

These motherboards use the Marvell 9230 chip to add 4 SATA ports to the 2 SATA
ports provided by the BayTrail SoC.

I can sometimes recreate this error when placing the filesystem under heavy
load. In this case it is a multi-terabyte ZFS "tank" and "zpool" is doing a
"scrub".

Gentoo Linux running gentoo-sources-4.1.15-r1

I do not remember when I first started to see this error appear.

-- 
You are receiving this mail because:
You are watching the assignee of the bug.