Promise SATA 300 TX4

[Arno J. Klaassen]

Hello,


Jeremie Le Hen <jeremie at le-hen.org> writes:

> Hi Ulf,
> 
> On Mon, Aug 13, 2007 at 11:18:02AM +0200, Ulf Lilleengen wrote:
> > I have problems with this card in -CURRENT too. Details can be found
> > here:
> > http://lists.freebsd.org/pipermail/freebsd-current/2007-July/074960.html
> > 
> > I never bothered trying with earlier versions of -CURRENT, but it
> > works good in 6.2.
> 
> I'm planning to switch my own server from 6.2 to -CURRENT soon.
> Are you still experiencing such problems?

might be related : I see similar problems (tons of 'taskqueue timeout'
'READ_DMA48 retrying' and finaly 'adX : FAILURE device detached' )
on 6-STABLE-amd64' (verified on two different mother-boards).

And I remarked a curious message in de dmesg :

atapci0: <Promise PDC40718 SATA300 controller> port 0xe400-0xe47f,0xe500-0xe5ff mem 0xf60c2000-0xf60c2fff,0xf60a0000-0xf60bffff irq 16 at device 8.0 on pci0
pci0: child atapci0 requested type 4 for rid 0x20, but the BAR says it is an memio

...

atapci1: <Promise PDC40718 SATA300 controller> port 0xe600-0xe67f,0xe700-0xe7ff mem 0xf60c5000-0xf60c5fff,0xf6080000-0xf609ffff irq 18 at device 10.0 on pci0
pci0: child atapci1 requested type 4 for rid 0x20, but the BAR says it is an memio


The ' .. but the BAR says .. ' messages seem to come from pci_alloc_map()
and the code has the following comment :

         * Note: atapci in legacy mode are special and handled elsewhere
         * in the code.  If you have a atapci device in legacy mode and
         * it fails here, that other code is broken.
         */

I can provide more info or test ideas/patches.

Thanx in advance, regards,

Arno