On my amd64 system with the kernel 2.6.17-r8 I have the following errors in dmesg: ata7: PIO error ata7: status=0x50 { DriveReady SeekComplete } ata7: PIO error ata7: status=0x50 { DriveReady SeekComplete } ata8: PIO error ata8: status=0x50 { DriveReady SeekComplete } ata8: PIO error ata8: status=0x50 { DriveReady SeekComplete } Supposing that my drives aren't broken, and verified that 2.6.17-r7 doesn't say anything strange, I suppose something is going broken in serial ata drivers. On 2.6.18 the messages becomes more verbose: ata7.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen ata7.00: tag 0 cmd 0xb0 Emask 0x2 stat 0x50 err 0x0 (HSM violation) ata7: soft resetting port ata7: SATA link up 1.5 Gbps (SStatus 113 SControl 300) ata7.00: configured for UDMA/133 ata7: EH complete ata7.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen ata7.00: tag 0 cmd 0xb0 Emask 0x2 stat 0x50 err 0x0 (HSM violation) skge eth0: Link is down. ata7: soft resetting port ata7: SATA link up 1.5 Gbps (SStatus 113 SControl 300) ata7.00: configured for UDMA/133 ata7: EH complete ata7.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen ata7.00: tag 0 cmd 0xb0 Emask 0x2 stat 0x50 err 0x0 (HSM violation) ata7: soft resetting port [lot of other exception] ata7: SATA link up 1.5 Gbps (SStatus 113 SControl 300) ata7.00: configured for UDMA/133 ata7: EH complete ata7.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen ata7.00: tag 0 cmd 0xb0 Emask 0x2 stat 0x50 err 0x0 (HSM violation) ata7: soft resetting port ata7: SATA link up 1.5 Gbps (SStatus 113 SControl 300) ata7.00: configured for UDMA/133 ata7: EH complete SCSI device sda: drive cache: write back SCSI device sda: 398297088 512-byte hdwr sectors (203928 MB) sda: Write Protect is off sda: Mode Sense: 00 3a 00 00 SCSI device sda: drive cache: write back Then the same with ata8 ata8.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen ata8.00: tag 0 cmd 0xb0 Emask 0x2 stat 0x50 err 0x0 (HSM violation) ata8: soft resetting port [lot of exception like that] ata8: SATA link up 1.5 Gbps (SStatus 113 SControl 300) ata8.00: configured for UDMA/133 ata8: EH complete SCSI device sdb: 398297088 512-byte hdwr sectors (203928 MB) sdb: Write Protect is off sdb: Mode Sense: 00 3a 00 00 SCSI device sdb: drive cache: write back SCSI device sdb: 398297088 512-byte hdwr sectors (203928 MB) sdb: Write Protect is off sdb: Mode Sense: 00 3a 00 00 SCSI device sdb: drive cache: write back The system seems to work anyway also with these mesage, but the presence of exceptions are symptoms of something that doesn't work as supposed to be. Opening a bug letting kernel people know of the problem so they can choose if mask the related ebuild or not.
2.6.18 introduces new libata error handling. It is more verbose than before. If you're convinced that you disks are fine then you should file a bug at http://bugzilla.kernel.org for this. However the word 'exception' is nothing to worry about.
