2.6.24 + ICH8M + high SATA load == death

[Jeffrey Baker]

On 2.6.24 with a SATA controller: Intel Corporation 82801HBM/HEM
(ICH8M/ICH8M-E) SATA AHCI Controller (rev 03) and a   Vendor: ATA
Model: SAMSUNG MCBQE32G Rev: PS10 flash disk, I get this error when
doing 32 parallel runs of pgbench:

ata1.00: exception Emask 0x10 SAct 0x0 SErr 0x50000 action 0xa frozen
ata1.00: irq_stat 0x00400001, PHY RDY changed
ata1: SError: { PHYRdyChg CommWake }
ata1.00: cmd c8/00:10:67:38:97/00:00:00:00:00/e1 tag 0 dma 8192 in
         res 50/00:00:76:38:97/00:00:00:00:00/e1 Emask 0x10 (ATA bus error)
ata1.00: status: { DRDY }

Afterwards the machine was in some kind of bad state where it would do
only about 1MB/s to the disk, and I had to power it off.

Basically I have no idea what any of that gibberish means.  Note that
this device is about 80 times faster than the spinning disk it
replaced, so it may be stressing parts of the software that are not
normally stressed.  Note also that it could just be crap hardware.  I
don't really know.  However, I do note that someone recently posted a
very similar error using Western Digital disks and the same SATA
controller.  I don't think the problem is cables, since this is a
laptop.  Any advice welcome.

-jwb
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/