[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: RES: State of XOrg in testing/unstable



On Mon, 2 Oct 2006, Grant Grundler wrote:

On Sun, Oct 01, 2006 at 02:15:42AM -0400, Ivan Jager wrote:
...
Then it will start cycling between:
FLT 500B SYS BD bus timeout
FLT CB74 SYS BD bad os HPMC cksm
FLT CBFC SYS BD OS HPMC br err
FLT CBF0 SYS BD HPMC initiated

To paraphrase, the chassis codes mean:
o Someone accessed an address that didn't respond.
o CPU timed out the transaction.
o HPMC == High Priority machince Check. Try "ser pim" at
 the firmware prompt to see which address it tried to reference.

Ok, I'll try that sometime soon.

Pressing the (soft) power button causes a clean shutdown after which it
powers down, so it's not completely dead.

An HPMC means the box is dead from an OS point of view.
We can't access the IO subsystem after an HPMC and the
OS gets control but only attempts dump a tombstone to the
console (which might fail too).

Can you clarify what you mean by "clean shutdown"?

Oops, I was looking at the wrong part of /var/log/messages. :( No clean shutdown. It just takes about half a minute before turning off.

...
I'm currently using the fbdev driver with the G200. I can't vouch for
it's stability, but it seems to be working.

wow! coool!

Yes. Um, for people on debian-alpha: I guess I should have explicitly said I was crossposting to avoid confusion. The fbdev driver with G200 is working on hppa now. (I only tried fbdev with the radeon on alpha, and that didn't work.)



Reply to: