2.6 PReP issues, still freezing

[Remco Post]

On Monday, Dec 1, 2003, at 20:16 Europe/Amsterdam, Tom Rini wrote:

> On Mon, Dec 01, 2003 at 12:11:19PM +0100, Remco Post wrote:
>>
>> On Sun, 30 Nov 2003 14:21:10 +0100
>> Christian Kujau <evil at g-house.de> wrote:
>>
>>>
>>> hi,
>>>
>>> after more days of compiling (at least now i know my ppc has good RAM
>>> :-) ) and testing serveral kernels, i think now it's not a
>>> network-related but a ppc-specific problem.
>>>
>>> Andrew Morton gave me the hint to just copy drivers/net/tulip/ from
>>> the
>>> working 2.5.30 kernel to a recent one. i've done so; i've copied all
>>> tulip stuff to a fresh 2.6.0-test11 kernel, after booting i was able
>>> to
>>> load the module, but a following "ifconfig eth1 192.168.2.1" freezed
>>> the
>>> machine -- as every kernel after 2.5.30 i was able to compile.
>>>
>>> a fix from Jeff Garzik went to linuxppc-dev (i saw it on lkml too),
>>> subject was "2.6 PReP fix? (Was: [jgarzik at pobox.com: [PATCH/CFT]
>>> de2104x
>>> fixes])". i found out, that i can load 3 different modules for my
>>> card:
>>>
>>> 00:0e.0 Ethernet controller: Digital Equipment Corporation DECchip
>>> 21140
>>> [FasterNet] (rev 22)
>>>          Flags: bus master, medium devsel, latency 0, IRQ 11
>>>          I/O ports at 1800 [size=128]
>>>          Memory at c300a000 (32-bit, non-prefetchable) [size=128]
>>>          Expansion ROM at c1040000 [disabled] [size=256K]
>>>
>>>
>>> /lib/modules/2.6.0-test11/kernel/drivers/net/tulip:
>>> de2104x.ko  de4x5.ko  tulip.ko
>>>
>>> /lib/modules/2.5.30/kernel/drivers/net/tulip:
>>> de2104x.o  de4x5.o  tulip.o
>>>
>>> and every module gave me a freeze with 2.6.0-test11 (oh, i forgot to
>>> test, if de2104x.o and de4x5.o are working correctly with 2.5.30
>>> since
>>> i've used always "tulip")
>>>
>>> and so, this makes me think it's a ppc-specific problem. no, i don't
>>> say
>>> it's easier now to fix, perhaps it's even harder, because a lot has
>>> changed in the ppc port too. if we could blame the device-driver, the
>>> codesize to play with would be a lot smaller...
>>>
>>> but since really no other PReP people complain, i think i'll stay
>>> with
>>> 2.4. a while. the machine is not that fast and i need a faster one
>>> anyway...
>>>
>>
>> I stopped complaining a while back (2.5.60 something) since I got
>> absolutely
>> no response. I'm one of the lucky (?) few who have a motorola
>> powerstack
>> (II) mobo.
>
> I suppose part of the problem is that my PowerStack (Series E) doesn't
> have any problems. :)
>

I just booted my powermac 8500 to build a new kernel (test11). after
that's finished, I'll see if I can reproduce the problem...

> What I've just done is put up all of the patches I have pending to try
> and get into 2.6.0, once Linus either becomes more open to !ia32
> patches
> or Andrew Morton takes over.  They're over at:
> http://stop.crashing.org:16080/~trini/
>
> The combination of 011 and 012 should allow for kgdb to work on any
> PReP
> box, and this should help people debug problems, if they have the time.
>
> --
> Tom Rini
> http://gate.crashing.org/~trini/
>
--
Met vriendelijke groeten,

Remco Post

SARA - Reken- en Netwerkdiensten                      http://www.sara.nl
High Performance Computing  Tel. +31 20 592 8008    Fax. +31 20 668 3167

"I really didn't foresee the Internet. But then, neither did the
computer
industry. Not that that tells us very much of course - the computer
industry
didn't even foresee that the century was going to end." -- Douglas Adams


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/