On Mon, Apr 07, 2003 at 09:33:53PM +0300, netbsd@ns.purk.ee wrote:
> pciide1 at pci0 dev 19 function 0: Triones/Highpoint HPT374 IDE Controller
> pciide1: bus-master DMA support present
> pciide1: primary channel wired to native-PCI mode
> pciide1: using irq 11 for native-PCI interrupt
> -------------------------------------------------------> Here 10-15/sec
> pciide1: secondary channel wired to native-PCI mode
> pciide2 at pci0 dev 19 function 1: Triones/Highpoint HPT374 IDE Controller
> pciide2: bus-master DMA support present
> pciide2: primary channel wired to native-PCI mode
> pciide2: using irq 11 for native-PCI interrupt
> -------------------------------------------------------> Here 10-15/sec

Definitely stinks of waiting for the slave reset!
The code would have to be hacked to detemine exactly where/why.

> David..i run couple of boxen with x86/Solaris and i don't have any delays but
> Solaris is *DEAD* slow in x/86 ;-)

That won't have 4 IDE controllers!  also solaris x86 used to come from
a separate group (San Diago?) from the main sun kernel people now in
East Palo Alto prison.

	David

-- 
David Laight: david@l8s.co.uk