Re: Dmesg garbage after kernel recompile
Monique Y. Mudama wrote:
Are you using identical hardware for all of the machines exhibiting
these symptoms?
Not really, only two of these were identical, all the others were
totally different.
From PII-400 to P4-~2GHz, single processor or SMP, with or without
SCSI, server or workstation, the palette was quite large.
I'm wondering if one of the options you're using
doesn't properly support some piece of hardware.
It's possible, but i've got no prove.
The most evil thing I can think of is FreeSWAN, that patches the kernel
really hard. So hard, that it even can't patch the debianized kernel
versions (at least not the few that we tried), that's why we are using
stock kernel sources. But seen from the point of view that the compiled
kernels are working perfectly (except the bad kernel printouts) - even
FreeSWAN is working well (we have a private network from more then half
a dozen machines got together with FreeSWAN), I wouldn't dare to say
that this is the problem. And after a lot of experiments (which were not
really so-to-say "controlled" experiments) still no result.
From a point on we simply gave up, and didn't care for.
But we reached the point when we faced a new, unknown machine, that had
some boot-up problems (and a very forced dead-line:). And without proper
kernel output, who knows what's the problem. That's was why I decided to
write here.
In the meantime we solved the problem with the server (creativity and
mind-reading can do real magic:).
Anyway if I'll got some spare hours I will try some brain-killing kernel
compiling marathon. :-D
BTW any other ideas (that will save me from killing my brain) are welcome.:)
Thanks.
Bye,
Tylla
Reply to: