SATA378 / SATA150 RAID controller supported by 5.4?

[Doug White]

On Tue, 17 May 2005, Rob wrote:

>
> Hi,
>
> The 'pciconf -lv' tells me:
>
>  atapci0 at pci2:4:0: class=0x010400 card=0x80f51043
>                   chip=0x3373105a rev=0x02 hdr=0x00
>     vendor   = 'Promise Technology Inc'
>     device   = 'PDC20378 FastTrak 378/SATA 378 RAID
>                 Controller'
>     class    = mass storage
>     subclass = RAID
>
> But I get this in my dmesg output:
>
>  atapci0: <Promise PDC20378 SATA150 controller>
>     port 0xd880-0xd8ff,0xdfa0-0xdfaf,0xdf00-0xdf3f
>     mem 0xfeac0000-0xfeadffff,0xfeafe000-0xfeafefff
>     irq 10 at device 4.0 on pci2
>  atapci0: failed: rid 0x20 is memory, requested 4

That's the PCI busmaster register, although it seems to be the wrong
resource type. It should be I/O space, which maps to PCI config space, and
not memory-mapped.  I can't seem to find what function emits that message.

A full dmesg would be useful.

>  rl0: <RealTek 8139 10/100BaseTX>
>      port 0xd000-0xd0ff
>      mem 0xfeaff400-0xfeaff4ff
>      irq 10 at device 9.0 on pci2
>
> At present I use a single harddisk on the regular
> IDE connector (atapci1 UDMA100-controller).
>
> Is the RAID controllor on atapci0 supported by 5.4?
> What is the 'failed' message about in dmesg?

It seems to be a bit error...

> Also note the seemingly interrupt conflict of both,
> rl0 and atapci0, claiming irq 10 ?!?!

Thats normal for PIC mode. PCI cards can share interrupts, but ISA can't
share with anything else, including other ISA cards.

-- 
Doug White                    |  FreeBSD: The Power to Serve
dwhite at gumbysoft.com          |  www.FreeBSD.org