Messages in this thread | | | From | "Robbert Kouprie" <> | Subject | Re: ASUS TUSL2-C and Promise Ultra100 TX2 | Date | Sat, 26 Oct 2002 03:59:27 +0200 |
| |
Cajoline writes:
> I recently setup a box with the following components: > Intel Celeron 1300 MHz > ASUS TUSL2-C motherboard > 2 x Promise Ultra100 TX2 controllers
I have a CUSL2-C board, P3-800 Coppermine and a Ultra133TX2 controller.
> Any 2.4 kernel I have tried on this machine displays this strange > a > behavior: any drives attached to the PDC controllers only work at udma > mode 2 (UDMA33).
I have the same problem. This is a known problem in the vanilla kernels (still in 2.4.20-pre11). You can force the right UDMA setting by giving a "ideX=ata66" kernel boot parameter, where the "X" is your interface number. This is fixed in recent -ac kernels (I tested with 2.4.20-pre10-ac2).
> What's even funnier is that if I try to copy files from a > filesystem on > a > > drive attached to a PDC20268 and a drive attached to the motherboard > controller (PIIX4 chipset), the system eventually locks up > (after about > 3 > GB). > What I mean by this is that there are no errors whatsoever, from the > kernel ide driver, from the filesystem, nothing at all. It just stops > responding to anything: login at the console, shell commands, network > daemons, everything stops working. You can't even reboot it - a hard > reset > is required.
This is nasty, I experience this too. This is different from the problem you describe earlier. I already checked different recent kernels, BIOS versions, NICs, memory, processors, and still it hangs. I suspect it's a unknown bug in the driver or a hardware bug in the controller. The problem is that it hangs completely dead, giving no information to start debugging. :(
> So I have come to the conclusion there must be some rather bizarre > > incompatibility between the PDCs and this motherboard. > Let me note that the PDC controllers do work just fine with > other older > motherboards.
Like you, I also have other boxes with Promise Ultra66/100/133 controllers, with _different_ motherboards, which indeed don't have such problem, so the combination of motherboard <-> controller looks important here.
> And another thing, during boot-up, the PDCs do show the > drives attached to it, detected at the right udma mode.
Ditto.
> I was wondering if anyone has come across this specific problem. I > browsed > > thoroughly through the list archives, but I didn't find any mention of > the > specific motherboard, or even the PIIX4 chipset and these controllers. > I know there is probably no way I can get this hardware to work > together, > yet I'm curious to know if this has occurred to someone else as well.
Well, it has. And I'm still hoping to solve this one. I an open to any suggestions, patches or tests.
Regards, - Robbert Kouprie
- 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/
| |