weak performance after sata disk timeout

[Emmanuel Dreyfus <manu%netbsd.org@localhost>]

Hi

I run a RAID 1 on NetBSD-5.1.2. After a disk gets a device timout, it will 
have weak performances. sysctl wm reports it as being 100% busy while it
moves about 3KB/s of data, while the other disk is fine.

That occured on its own, but I can reproduce the problem by issuing 
        atactl /dev/atabus3 reset, 
This will cause a device timeout, and the disk will have weak performances
after that. I have not yet investigated wether RAID is required or not
 to exhibit the problem.

Here is what I get in dmesg before performances get bad:
  wd3a: channel reset writing fsbn 3273882368 of 3273882368-3273882399 
     (wd3 bn 3273882431; cn 3247899 tn 3 sn 50), retrying
  wd3: soft error (corrected)


And here is startup information on the disk:
  piixide0 at pci0 dev 31 function 2
  piixide0: Intel 82801I Serial ATA Controller (ICH9) (rev. 0x02)
  piixide0: bus-master DMA support present
  piixide0: primary channel configured to compatibility mode
  piixide0: primary channel interrupting at ioapic0 pin 14
  (...)
  piixide0: secondary channel configured to compatibility mode
  piixide0: secondary channel interrupting at ioapic0 pin 15
  atabus3 at piixide0 channel 1
  (...)
  wd3 at atabus3 drive 1: <WDC WD2003FYYS-02W0B1>
  wd3: drive supports 16-sector PIO transfers, LBA48 addressing
  wd3: 1863 GB, 3876021 cyl, 16 head, 63 sec, 512 bytes/sect x 3907029168 
      sectors
  wd3: 32-bit data port
  wd3: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 6 (Ultra/133)
  wd3(piixide0:1:1): using PIO mode 4, Ultra-DMA mode 6 (Ultra/133) (using DMA)

The only workaround I have is to reboot the machine. Is that a known problem?

-- 
Emmanuel Dreyfus
manu%netbsd.org@localhost