Re: spurious NCQ completions from FUJITSU MHW2120BH

[Robert Hancock]

Cal Peake wrote:
> On Fri, 3 Aug 2007, Meelis Roos wrote:
>
> > Todays git gave me some EH errors during fsck (30 times mounted and time 
> > to check), then recovered and works fine. Seems to be spurious NCQ 
> > completion. The disk is
> >
> > Device Model:     FUJITSU MHW2120BH
> > Firmware Version: 00000012
> > User Capacity:    120ï034ï123ï776 bytes
>
> As a data point: I've got the same drive w/ same revision in my laptop and 
> haven't had any NCQ errors pop up. Currently running 2.6.23-rc2.
>
>  ata1.00: ATA-8: FUJITSU MHW2120BH, 00000012, max UDMA/100
>  ata1.00: 234441648 sectors, multi 16: LBA48 NCQ (depth 0/32)
>  ata1.00: configured for UDMA/100
>
> Might be dependent on the controller and/or the driver. Got a nVidia MCP51 
> using sata_nv here.

That would be because you're not using NCQ. Patches to add NCQ support 
for MCP51 are only in -mm tree, not in mainline.

Also, AHCI is the only controller type that is easily able to detect this..

--
Robert Hancock      Saskatoon, SK, Canada
To email, remove "nospam" from hancockr@xxxxxxxxxxxxx
Home Page: http://www.roberthancock.com/

-
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/