ATA driver races with interrupts
Daniel Eriksson
daniel_k_eriksson at telia.com
Mon Aug 2 10:33:43 PDT 2004
Ville-Pertti Keinonen wrote:
> Repeating the problem should be trivial with an unpatched
> -current with
> PREEMPTION enabled and hardware similar to mine - ASUS K8V
> Deluxe, two
> SATA disks on the VIA 6420:
>
> ...
> atapci1: <VIA 6420 SATA150 controller> port
> 0xd000-0xd0ff,0xd400-0xd40f,0xd800-0xd803,0xe000-0xe007,0xe400
> -0xe403,0xe800-0xe807
> irq 20 at device 15.0 on pci0
> ata5: at 0xe800 on atapci1
> ata6: at 0xe000 on atapci1
> ...
> ad10: 152627MB <SAMSUNG SP1614C> [310101/16/63] at ata5-master SATA150
> ad12: 152627MB <SAMSUNG SP1614C> [310101/16/63] at ata6-master SATA150
This looks exactly like the problems I have had with similar hardware (ASUS
A7V600-X, a KT-600 based mobo with an on-board 6420 controller). With
preemption turned off it sort of worked, but with preemption turned on the
discs would lock up pretty quick just as you described it.
I have just applied the patch and will try it out later tonight once I get
everything recompiled. I'll let you know how it turns out.
/Daniel Eriksson
More information about the freebsd-current
mailing list