Subject: wd0 latency
To: None <current-users@netbsd.org>
From: Arnaud Lacombe <lists-netbsd@sigfpe.info>
List: current-users
Date: 08/11/2006 16:28:23
Hi,

I've got a problem with my laptop at boot time with the detection of
wd0: there is a really huge latency before and after the probe of the
hard disk (about 10s after the probe, 2s or 3s before). He is the
relevant part of the dmesg:

NetBSD 3.99.21 (debug) #12: Sat Jul  8 22:25:27 CEST 2006
  al@nog:/obj/sys/arch/i386/compile/debug
[...]
timecounter: selected timecounter "TSC" frequency 697508100 Hz quality 800
timecounter: Timecounter "clockinterrupt" frequency 100 Hz quality 0
allocated pic pic0 type level pin 0 level 12 to cpu0 slot 0 idt entry 32
raidattach: Asked for 8 units
Kernelized RAIDframe activated
__wdcwait: timeout (time=11), status 80 error 80 (mask 0x8 bits 0x8)
__wdcwait: timeout (time=11), status 80 error 80 (mask 0x8 bits 0x8)
__wdcwait: timeout (time=11), status 1 error 4 (mask 0x8 bits 0x8)
[last message repeated 300 times]
__wdcwait: timeout (time=11), status d0 error d0 (mask 0x8 bits 0x8)
__wdcwait: timeout (time=11), status d0 error d0 (mask 0x8 bits 0x8)
__wdcwait: timeout (time=11), status 1 error 4 (mask 0x8 bits 0x8)
wd0 at atabus0 drive 0: <SAMSUNG MP0402H>
wd0: drive supports 16-sector PIO transfers, LBA48 addressing
wd0: 38204 MB, 77622 cyl, 16 head, 63 sec, 512 bytes/sect x 78242976 sectors
__wdcwait: timeout (time=11), status 1 error 4 (mask 0x8 bits 0x8)
wd0: 32-bit data port
wd0: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 5 (Ultra/100)
wd0(viaide0:0:0): using PIO mode 4, Ultra-DMA mode 4 (Ultra/66) (using DMA)
__wdcwait: timeout (time=11), status 1 error 4 (mask 0x8 bits 0x8)
[last message repeated 1790 times]

This problem occurs since March or April (I don't know exactly, sorry).
3.0 boots without such problem.

Any hints about it ?

Thanks in advance,

Arnaud

ps: I didn't test with 4.99.1 yet, but it is likely it will also happen