SATA for kernel(s) 3.5+ has an issue with IRQ Balancing on AMD SB710 with motherboard ASRock m3a770d3.
The problem started while using Arch linux, after going back to Debian it persisted. I had tried to find problems with the bug but couldn't find any direct help online. I had a break when finding similar bugs on other sites stating that IRQ Balancing might be the fault. So I tried ACPI_IRQ_NOBALANCE in the kernel boot parameters and so far the issue has stopped. Linux version 3.6.0-rc2 (mboy@debian) (gcc version 4.7.1 (Debian 4.7.1-6) ) #1 SMP Sat Aug 18 00:46:06 CDT 2012 Aug 20 12:32:08 debian kernel: [42238.075368] ata4.00: exception Emask 0x0 SAct 0x7f SErr 0x0 action 0x6 frozen Aug 20 12:32:08 debian kernel: [42238.075373] ata4.00: failed command: WRITE FPDMA QUEUED Aug 20 12:32:08 debian kernel: [42238.075377] ata4.00: cmd 61/08:00:c8:0a:89/00:00:01:00:00/40 tag 0 ncq 4096 out Aug 20 12:32:08 debian kernel: [42238.075377] res 40/00:00:00:4f:c2/00:00:00:4f:c2/00 Emask 0x4 (timeout) Aug 20 12:32:08 debian kernel: [42238.075379] ata4.00: status: { DRDY } Aug 20 12:32:08 debian kernel: [42238.075381] ata4.00: failed command: WRITE FPDMA QUEUED Aug 20 12:32:08 debian kernel: [42238.075384] ata4.00: cmd 61/08:08:f8:cb:8a/00:00:03:00:00/40 tag 1 ncq 4096 out Aug 20 12:32:08 debian kernel: [42238.075384] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) Aug 20 12:32:08 debian kernel: [42238.075386] ata4.00: status: { DRDY } Aug 20 12:32:08 debian kernel: [42238.075388] ata4.00: failed command: WRITE FPDMA QUEUED Aug 20 12:32:08 debian kernel: [42238.075391] ata4.00: cmd 61/10:10:80:95:86/00:00:08:00:00/40 tag 2 ncq 8192 out Aug 20 12:32:08 debian kernel: [42238.075391] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) Aug 20 12:32:08 debian kernel: [42238.075393] ata4.00: status: { DRDY } Aug 20 12:32:08 debian kernel: [42238.075394] ata4.00: failed command: READ FPDMA QUEUED Aug 20 12:32:08 debian kernel: [42238.075397] ata4.00: cmd 60/60:18:60:71:62/00:00:00:00:00/40 tag 3 ncq 49152 in Aug 20 12:32:08 debian kernel: [42238.075397] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) Aug 20 12:32:08 debian kernel: [42238.075399] ata4.00: status: { DRDY } Aug 20 12:32:08 debian kernel: [42238.075401] ata4.00: failed command: READ FPDMA QUEUED Aug 20 12:32:08 debian kernel: [42238.075404] ata4.00: cmd 60/08:20:c8:71:62/00:00:00:00:00/40 tag 4 ncq 4096 in Aug 20 12:32:08 debian kernel: [42238.075404] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) Aug 20 12:32:08 debian kernel: [42238.075406] ata4.00: status: { DRDY } Aug 20 12:32:08 debian kernel: [42238.075407] ata4.00: failed command: READ FPDMA QUEUED Aug 20 12:32:08 debian kernel: [42238.075410] ata4.00: cmd 60/18:28:d8:71:62/00:00:00:00:00/40 tag 5 ncq 12288 in Aug 20 12:32:08 debian kernel: [42238.075410] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) Aug 20 12:32:08 debian kernel: [42238.075412] ata4.00: status: { DRDY } Aug 20 12:32:08 debian kernel: [42238.075413] ata4.00: failed command: READ FPDMA QUEUED Aug 20 12:32:08 debian kernel: [42238.075417] ata4.00: cmd 60/08:30:f8:71:62/00:00:00:00:00/40 tag 6 ncq 4096 in Aug 20 12:32:08 debian kernel: [42238.075417] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) Aug 20 12:32:08 debian kernel: [42238.075418] ata4.00: status: { DRDY } Aug 20 12:32:08 debian kernel: [42238.075422] ata4: hard resetting link After it would reset, it would try SATA I, fail, then mount the drive as RO. I'm running a Phenom II on Debian Wheezy with a bit of Sid, don't have the time at the moment to get a better grasp on what you need for a proper bug report but I hope this helps. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/