I do get

ex0: uplistptr was 0

in messages a few times a week, which is of concern, but the box (2.0.2/i386
p4 1024 user 64k nmbclusters) runs ok, albeit at a rather high load -
consistent 5-10.

Uptime has not passed 30 days.

Stef

On Fri, 20 May 2005, Jukka Salmi wrote:

> Hello,
>
> today I found a NetBSD/i386 2.0_STABLE system dropped into ddb:
>
> db> dmesg
> [...]
> ex0: uplistptr was 0
> ex0: uplistptr was 0
> viaide0:0:1: lost interrupt
>         type: ata tc_bcount: 0 tc_skip: 0
> db> trace
> ltsleep(c038adb8,10,c02c6de8,1,0) at netbsd:ltsleep+0x6b
> wdcwait(c0c2b8fc,48,40,3e8,9) at netbsd:wdcwait+0xa5
> __wdccommand_start(c0c2b8fc,c0d76000,0,c0d77904,c0d77904) at netbsd:__wdccommand_start+0x73
> wdcstart(c0c2b8fc,c0d76034,c038ae50,c0d1dce0,800) at netbsd:wdcstart+0xb1
> wdc_ata_bio_intr(c0c2b8fc,c0d76034,1,c0d1a000,c038ae7c) at netbsd:wdc_ata_bio_intr+0x18a
> wdcintr(c0c2b8fc,0,ad510010,30030,580e0010) at netbsd:wdcintr+0xda
> Xintr_legacy14() at netbsd:Xintr_legacy14+0xa8
> --- interrupt ---
> cpu_switch(c033ddc0,0,cb7f6000,c0249965,1) at netbsd:cpu_switch+0x9f
> ltsleep(c033dc20,4,c02cc1b3,0,0) at netbsd:ltsleep+0x323
> uvm_scheduler(c033dc00,0,c034219c,c02c9051,0) at netbsd:uvm_scheduler+0x74
> main(0,0,0,0,0) at netbsd:main+0x67f
> db>
>
> I executed some other ddb commands (`ps' using all available modifiers, and
> `show registers'), but I'm not sure which of them make sense to be included
> in a PR.
>
> Does anybody recognise this problem as an already reported one?
>
> A crash dump is available, in case anybody cares.
>
>
> TIA, Jukka
>
> --
> bashian roulette:
> $ ((RANDOM%6)) || rm -rf ~
>