Re: NetBSD 5.0.2 can't handle 7455 CPU @ 1 GHz?

[Flavio Donadio <flavio%donadio.com.br@localhost>]

Donald,


If you can try it without the card, it may help us get an insight.

Also, you can rule out the graphics card by swapping the processor, which sure 
is more prone to accidental damage.

It looks like no one else had this particular problem, so I think you'll have 
to make those experiments yourself.

I wish you good luck!


Cheers,
Flavio

On 20/12/2010, at 14:37, Donald Lee wrote:

> I have not yet set up to build 5.0 kernels, and was hoping not to get
> (back) in that business.  I *could* put a different video card in the
> machine, but I don't have any non-ATI spares.  I have one from a G3
> that's much older that I could try, but that may not prove anything.
> 
> I could also put the original CPU back in the machine to see if it's
> the graphics or the 7455.  I'm reluctant to do that because the
> CPU swap process is touchy, and a little dust or clumsy move can
> make for broken HW.
> 
> I could run the machine headless for a while as an easy experiment.
> It would be easy enough to pull the graphics card, once I'm sure it will
> boot.........
> 
> Thanks,
> 
> -dgl-
> 
>> Donald,
>> 
>> 
>> I see that these machines you're trying have different video cards in 
>> different buses.
>> 
>> The one that locks loads the driver for the Rage128 frame buffer (r128fb). 
>> The other loads the driver for the generic frame buffer (genfb).
>> 
>> Try building a kernel with no ATI drivers, forcing it to load genfb. If you 
>> have a serial port that can be used as a console (I love my Stealth serial 
>> port!), I think it would suffice to just boot without the AGP graphics card 
>> installed. Am I right?
>> 
>> I think I had problems with ATI graphics cards in the past and using genfb 
>> solved them. In my case, the machine would run headless anyways...
>> 
>> 
>> Cheers,
>> Flavio
>> 
>> On 20/12/2010, at 02:07, Donald Lee wrote:
>> 
>>>>> Dec 19 08:30:10 charm /netbsd: cpu0: 1000.00 MHz, 256KB L2 cache no 
>>>>> parity parity enabled, 2MB no-parity L3 cache (PB2 SRAM) at 4:1 ratio
>>>> 
>>>> Are you sure the accelerator card's L3 can run at 250 MHz? Most of the 
>>>> accelerators I've seen run the L2 / L3 at around 200 MHz, and the only one 
>>>> I've seen which runs at 250 MHz is the L3 on the very last of the 1 GHz 
>>>> 9600-type upgrades from Sonnet...
>>>> 
>>>> What did NetBSD 2.x report for the L3?
>>>> 
>>>> John
>>> 
>>> Looks the same, I think.
>>> 
>>> excerpt from logs with NetBSD 2.1:
>>> 
>>> Dec  3 22:46:21 mercy /netbsd: NetBSD 2.1 (try9) #1: Tue Jan 30 23:29:09 
>>> CST 2007
>>> Dec  3 22:46:21 mercy /netbsd:  
>>> donlee@mercy:/usr/src.21/sys/arch/macppc/compile/try9
>>> Dec  3 22:46:21 mercy /netbsd: total memory = 1280 MB
>>> Dec  3 22:46:21 mercy /netbsd: avail memory = 1224 MB
>>> Dec  3 22:46:21 mercy /netbsd: mainbus0 (root)
>>> Dec  3 22:46:21 mercy /netbsd: cpu0 at mainbus0: 7455 (Revision 2.1), ID 0 
>>> (primary)
>>> Dec  3 22:46:21 mercy /netbsd: cpu0: HID0 
>>> 8450c0bc<EMCP,TBEN,NAP,DPM,ICE,DCE,SGE,BTIC,LRSTK,FOLD,BHT> 
>>> Dec  3 22:46:21 mercy /netbsd: cpu0: 1000.00 MHz, 256KB L2 cache, 2MB 
>>> no-parity L3 cache (PB2 SRAM) at 4:1 ratio 
>>> 
>>> 
>>> I note the ID of the kernel is "try9", which means I built it.
>>> I dont't remember what changes I put in it, but my testing
>>> today with installs from CD confirm that whatever it is, it
>>> is also present in the "stock" kernels on the CDs.
>>> 
>>> -dgl-
>