Hi, I have a system setup with Promise PDC40718 (SATA 300 TX4) serial-ata card (4-ports) and also with 4 Seagate hard-disks. The problem is that I seem to get every once in a while the following errors to syslog, especially when the system is loaded (log from 2.6.20-git15 snapshot): Feb 21 12:37:58 alderan kernel: ata4.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0 Feb 21 12:37:58 alderan kernel: ata4.00: cmd 25/00:f8:cf:9f:f4/00:00:22:00:00/e0 tag 0 cdb 0x0 data 126976 in Feb 21 12:37:58 alderan kernel: res 50/00:00:c6:a0:f4/00:00:00:00:00/e0 Emask 0x1 (device error) Feb 21 12:37:58 alderan kernel: ata4.00: configured for UDMA/133 Feb 21 12:37:58 alderan kernel: ata4: EH complete Feb 21 12:37:58 alderan kernel: SCSI device sdc: 976773168 512-byte hdwr sectors (500108 MB) Feb 21 12:37:58 alderan kernel: sdc: Write Protect is off Feb 21 12:37:58 alderan kernel: SCSI device sdc: write cache: enabled, read cache: enabled, doesn't support DPO or FUA At the moment all the errors are coming from the two new 500GB ST3500630AS model Seagate sata-drives and none from the older 200GB ST3200822AS-models. The motherboard is Asus A7v880. I have already done some of the basic hardware checks, ie. I have changed power source from Antec to Nexus, changed sata-cables and even tried the 400GB model of Seagate hard-disk without success. It doesn't look like the there would be a problem with a single problematic cable or hard-disk --> I hooked up one of the 500GB Seagates to via_6420 sata-controller (patched for bar5 bug) and there has been no errors from that drive since after. NCQ is disabled for all the drives (cat /sys/block/sd*/device/queue_depth is 1 for all the drives). Does the Promise Sata 300 TX4 work currently without problems for others ? Regards, Tomi Orava --
Attachment:
dmesg
Description: Binary data
Attachment:
lspci.alderan
Description: Binary data
