2.4.9: PDC20267 not working

2.4.9: PDC20267 not working

[Joe Fago]

Hi,

System hangs on boot:


Uniform Multi-Platform E_IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with
idebus=xx
PDC20267: IDE controller on PCI bus 00 dev 40
PCI: Assigned IRQ 10 for device 00:08.0
PDC20267: chipset revision 2
PDC20267: not 100% native mode: will probe irqs later
PDC20267: (U)DMA Burst Bit ENABLED Primary PCI Mode Secondary PCI Mode
  ide0: BM-DMA at 0xe800-0xe807, BIOS settings: hda: pio, hdb: pio
  ide1: BM-DMA at 0xe808-0xe80f, BIOS settings: hdc: pio, hdd: DMA
hda: Maxtor 2B020H1, ATA DISK drive


This is the only device attached to the controller. Any suggestions?

Thanks,
Joe

Re: 2.4.9: PDC20267 not working

[Jacob Luna Lundberg]

On Sun, 9 Sep 2001, Joe Fago wrote:

> System hangs on boot:

> PDC20267: IDE controller on PCI bus 00 dev 40

> This is the only device attached to the controller. Any suggestions?

I have seen this before.  I have a system that will do it every time right
now, in fact.  You can try setting interrupts to edge-triggered in your
BIOS if it has such an option; this usually ``fixes'' the problem for me.
Of course, it will mean you can't share PCI interrupts, if I understand it
correctly.  However, I'm not sure what's going on and nobody has commented
on it thus far that I know of.  :(

-Jacob

Re: 2.4.9: PDC20267 not working

[David Grant]

----- Original Message -----
From: "Jacob Luna Lundberg" <kernel@gnifty.net>
To: "Joe Fago" <cfago@tconl.com>
Cc: <linux-kernel@vger.kernel.org>
Sent: Sunday, September 09, 2001 6:46 PM
Subject: Re: 2.4.9: PDC20267 not working


>
> On Sun, 9 Sep 2001, Joe Fago wrote:
>
> > System hangs on boot:
>
> > PDC20267: IDE controller on PCI bus 00 dev 40
>
> > This is the only device attached to the controller. Any suggestions?
>
> I have seen this before.  I have a system that will do it every time right
> now, in fact.  You can try setting interrupts to edge-triggered in your
> BIOS if it has such an option; this usually ``fixes'' the problem for me.
> Of course, it will mean you can't share PCI interrupts, if I understand it
> correctly.  However, I'm not sure what's going on and nobody has commented
> on it thus far that I know of.  :(

Mine has problems as well.  I know there are workarounds, but I'm not
experienced enough to do these.  Joe: what error messages do you get?  I get
the "hde: timeout waiting for dma" errors.  After that I have to cold start
my PC to get it working.  BTW, I have a PDC20265, but I think these are
probably almost identical (or at least the behaviour we are getting is
probably due to the same problem in the code).

I also suspect that maybe what kind of hard drive you have makes a
difference as well, because some people can get their Promise IDE to work
fine, while others can't.  I have a Quantum Fireball Plus AS.

In case you didn't know, Andre Hedrick met with some dude from Promise
called Craig Lyons on Wednesday.  I haven't heard of any news from the
meeting, but supposedly Craig was going to give Andre some API for the
Ultra/Fastrack line of chips.  Hopefully it went down on Wednesday.

David Grant

Re: 2.4.9: PDC20267 not working

[Joe Fago]

> >
> > On Sun, 9 Sep 2001, Joe Fago wrote:
> >
> > > System hangs on boot:
> >
> > > PDC20267: IDE controller on PCI bus 00 dev 40
> >
> > > This is the only device attached to the controller. Any suggestions?
> >
> > You can try setting interrupts to edge-triggered in your
> > BIOS if it has such an option; 
>
> Joe: what error messages do you get?  I get
> the "hde: timeout waiting for dma" errors.  After that I have to cold start

I had no more information, the system just hung after the `hda:' line. Setting
the
bios to interrupts triggered by `level' seems to get things going in my case.

I guess these are workarounds rather than solutions. I was just happy to
get the system booted. It seems that we can reproduce this particular problem,
and I'd be glad to participate in any testing that might assist Andre's efforts.

Thanks,
Joe

Re: 2.4.9: PDC20267 not working

[Joe Fago]

Jacob Luna Lundberg wrote:
> 
> On Sun, 9 Sep 2001, Joe Fago wrote:
> 
> > System hangs on boot:
> 
> > PDC20267: IDE controller on PCI bus 00 dev 40
> 
> > This is the only device attached to the controller. Any suggestions?
> 
> I have seen this before.  I have a system that will do it every time right
> now, in fact.  You can try setting interrupts to edge-triggered in your
> BIOS if it has such an option; this usually ``fixes'' the problem for me.
> Of course, it will mean you can't share PCI interrupts, if I understand it
> correctly.  However, I'm not sure what's going on and nobody has commented
> on it thus far that I know of.  :(
> 
> -Jacob
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Thanks, that worked -- well the opposite worked. I had to set mine to 
triggered by `level' rather than `edge'. But I never would have suspected
it if you hadn't brought it up.

Thanks Again,
Joe

Re: 2.4.9: PDC20267 not working

[Rok Papež]

Hi!

On Monday 10 September 2001 03:26, Joe Fago wrote:

> PDC20267: (U)DMA Burst Bit ENABLED Primary PCI Mode Secondary PCI Mode
>   ide0: BM-DMA at 0xe800-0xe807, BIOS settings: hda: pio, hdb: pio
>   ide1: BM-DMA at 0xe808-0xe80f, BIOS settings: hdc: pio, hdd: DMA
> hda: Maxtor 2B020H1, ATA DISK drive
>
>
> This is the only device attached to the controller. Any suggestions?

ASUS A7V with on-board PDC20265, kernel 2.4.9 with ACPI support - it hangs on 
detection of the last drive (hde). I've disabled APM and ACPI altogether in 
the kernel and now it works.
I didn't play with edge/level setting of interrupts.

-- 
best regards,
Rok Papež.