On Wed, 2007-01-10 at 19:23 +0000, Tvrtko A. Ursulin wrote:
> Anyway, the main stuff... could it be that the interrupt problems it mentions
> early is to blame for all unpredictability and later mmc failures?
[...]
>start_kernel(): bug: interrupts were enabled early
This sounds like a haret problem, maybe it left interrupts enabled and
shouldn't have done? That would also explain the unpredictability...
Does it do this with all the versions of haret you have?
> Kernel command line: root=/dev/mmcblk0p2 rw mem=64M cachepolicy=writeback
As a random guess, you could also try removing the cachepolicy=writeback
as I think that might cause problems on the pxa.
Regards,
Richard
Received on Wed Jan 10 2007 - 14:58:45 EST
This archive was generated by hypermail 2.2.0 : Wed Jan 10 2007 - 14:58:59 EST