Dear Rasmus,
In message 56b724ef-8051-8a46-5df8-fc5b00bd8408@prevas.dk you wrote:
Yes, but I'm not talking about or using the SOC watchdog. Also, this is not really about the particular watchdog device at all. It is about what generic code can expect from the __udelay primitive, which is not currently that on ppc.
Maybe what you expect from __udelay() is not what other people expect(ed) from it? I don't think there is any formal description of that __udelay() is supposed to do in any existing U-Boot documentation...
Please just face facts: for nearly two decades this code has been doing what it does, and I guess there were good reasons at that time (like memory footprint and resources to fix a specific problem). I agree that from a design point of view this is not nice, but that's not the point here.
Best regards,
Wolfgang Denk