Second IDE port not working on PC164
Alan Young
ayoung at teleport.com
Fri Apr 9 15:37:41 UTC 2004
On Fri, 9 Apr 2004 10:57:05 -0400 Jay Estabrook <Jay.Estabrook at hp.com> wrote:
> On Thu, Apr 08, 2004 at 08:28:12PM -0700, Alan Young wrote:
> > On Thu, 8 Apr 2004 16:40:30 -0400 Jay Estabrook <Jay.Estabrook at hp.com> wrote:
> >
> > > PYXIS is the core logic (north bridge equiv) on LX and SX (and MIATA).
> > > Under certain conditions, DMA traffic could cause the hardware to hang.
> > > Early MIATA had the most problems, LX and SX many fewer, and final MIATA
> > > (aka MIATA-GL) none to speak of, due to amount of add-on hardware fixes.
> >
> > Are there any issues with a LX and cards with a PCI bridge chip on it?
>
> Actually, bridges usually help the situation, by breaking up the DMA
> data transfers into smaller/shorter bursts. That's one reason why the
> MIATA has a built-in bridge.
That is interesting. I've seen hard locks when I plug a USB hard drive
into the card. It can take a minute or two under load for it to happen.
I'm not sure if this is driver related though. The drive also has a
firewire connection and as I recall is much more stable.
I have a some lock ups with just a Apple USB keyboard.
> > The reason I ask is that I have a card that has USB, Firewire and
> > SATA on it. They are behind the bridge and I never got the card
> > to work quite right. USB 1.1 seems stable. USB2 does a good
> > job at locking the box. The SATA seems to get an I/O conflict -
> > "Error, MMIO ports already in use". /proc/iomem shows the only
> > other device in the range is the bridge. The PCI bridge chip is a
> > Hint Corp. HB1-SE33.
>
> Any chance of seeing an "lspci -v" with that combo card installed?
Please see the attachement to this email.
> Also, what kernels have you tried on it?
2.4.22 was the last 2.4 kernel I tried with it. I have also tried
2.6.0 and 2.6.1.
Thanks,
Alan
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: list_pci_verbose.txt
URL: <http://listman.redhat.com/archives/axp-list/attachments/20040409/f28b2e45/attachment.txt>
More information about the axp-list
mailing list