[2.6.34-rc1][libata] regression?

From: Etienne Basset
Date: Tue Mar 09 2010 - 16:56:46 EST


hello,


with .34-rc1 i see some "stalls"; systems is frozen for say 30seconds, then comes backs to life
The only clue i have is strange ATA logs that I didn't have before, see below.
My disk may be dying, it's 6 years old, but i shows the messages only in .34-rc1, never in .33 or .32

If the stalls are unrelated to ata, how could I debug that with perf? i got lots of 'WARNING: failed to keep up with mmap data' when using perf top....
'

thx
Etienne

lscpi
00:1f.2 IDE interface: Intel Corporation 82801FB/FW (ICH6/ICH6W) SATA Controller (rev 03)

cat /var/log/messages
Mar 9 21:13:08 etienne-desktop kernel: ata1: clearing spurious IRQ
Mar 9 21:13:38 etienne-desktop kernel: ata1: lost interrupt (Status 0x50)
Mar 9 21:13:38 etienne-desktop kernel: ata1: soft resetting link
Mar 9 21:13:39 etienne-desktop kernel: ata1.00: configured for UDMA/133
Mar 9 21:13:39 etienne-desktop kernel: ata1.00: device reported invalid CHS sector 0
Mar 9 21:13:39 etienne-desktop kernel: ata1: EH complete
Mar 9 21:19:01 etienne-desktop kernel: ata1: clearing spurious IRQ
Mar 9 21:19:31 etienne-desktop kernel: ata1: lost interrupt (Status 0x50)
Mar 9 21:19:31 etienne-desktop kernel: ata1: soft resetting link
Mar 9 21:19:32 etienne-desktop kernel: ata1.00: configured for UDMA/133
Mar 9 21:19:32 etienne-desktop kernel: ata1.00: device reported invalid CHS sector 0
Mar 9 21:19:32 etienne-desktop kernel: ata1: EH complete

dmesg
libata version 3.00 loaded.
ata_piix 0000:00:1f.2: version 2.13
ata_piix 0000:00:1f.2: PCI INT C -> GSI 20 (level, low) -> IRQ 20
ata_piix 0000:00:1f.2: MAP [ P0 P2 P1 P3 ]
ata_piix 0000:00:1f.2: setting latency timer to 64
scsi0 : ata_piix
scsi1 : ata_piix
ata1: SATA max UDMA/133 cmd 0xfe00 ctl 0xfe10 bmdma 0xfea0 irq 20
ata2: SATA max UDMA/133 cmd 0xfe20 ctl 0xfe30 bmdma 0xfea8 irq 20
ata1.00: ATA-6: ST3160023AS, 8.05, max UDMA/133
ata1.00: 312581808 sectors, multi 8: LBA48
ata1.00: configured for UDMA/133


--
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/