Herbert,

On Tue, May 15, 2018 at 3:29 PM, Gilad Ben-Yossef [off-list ref] wrote:

We were using the content of the signature register as a sanity
check for the hardware functioning but it turns out not all
implementers use the same values so the check is giving false
negative on certain SoCs and so we drop it.

Please drop this patch. I have found a better fix and will send a v2 soon.

Thanks,
Gilad


-- 
Gilad Ben-Yossef
Chief Coffee Drinker

"If you take a class in large-scale robotics, can you end up in a
situation where the homework eats your dog?"
 -- Jean-Baptiste Queru