Most recent kernel where this bug did not occur: 2.6.21.5
Distribution: LFS
Hardware Environment: Thinkpad R60
Software Environment: 
Problem Description: In the system log file I noticed the following sequences 
-----
Jul  9 11:38:07 localhost kernel: ata1.00: exception Emask 0x2 SAct 0x7ffffff8 SErr 0x0 action 0x2 frozen
Jul  9 11:38:07 localhost kernel: ata1.00: (spurious completions during NCQ issue=0x0 SAct=0x7ffffff8 FIS=004040a1:00000004)
Jul  9 11:38:07 localhost kernel: ata1.00: cmd 61/08:18:6f:c3:58/00:00:00:00:00/40 tag 3 cdb 0x0 data 4096 out
Jul  9 11:38:07 localhost kernel:          res 40/00:f0:47:c4:58/00:00:00:00:00/40 Emask 0x2 (HSM violation)
.
.
Jul  9 11:38:07 localhost kernel: ata1.00: cmd 61/08:f0:47:c4:58/00:00:00:00:00/40 tag 30 cdb 0x0 data 4096 out
Jul  9 11:38:07 localhost kernel:          res 40/00:f0:47:c4:58/00:00:00:00:00/40 Emask 0x2 (HSM violation)
Jul  9 11:38:07 localhost kernel: ata1: soft resetting port
Jul  9 11:38:07 localhost kernel: ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
Jul  9 11:38:07 localhost kernel: ata1.00: configured for UDMA/100
Jul  9 11:38:07 localhost kernel: ata1: EH complete
Jul  9 11:38:07 localhost kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
Jul  9 11:38:07 localhost kernel: sda: Write Protect is off
Jul  9 11:38:07 localhost kernel: SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
-----

The harddisk that is in my laptop is :
ATA device, with non-removable media
        Model Number:       FUJITSU MHV2080BH                       
        Serial Number:      NW05T6827L0T
        Firmware Revision:  00840028

Which I included in the ncq blacklist in libata-core.c. This doesn't show the error anymore, and I don't notice any speed differences. What I do not know if this harddrive has the same problem as hitachi drives that I see a lot more about when I search for th the above error.


Steps to reproduce: irregular. (Downloading-, Copying files)