u_long static_start versus resource_size_t

[Steven A. Falco]

Larry Finger wrote:
> Steven A. Falco wrote:
>> So, has anyone else run into this?  Is my hack at all correct, or
>> should I attempt to work up a patch changing static_start to a
>> phys_addr_t or resource_size_t?
> 
> I waited to see if anyone more knowledgeable would answer, but as no
> one seems to have bit, my recommendation is that you change
> static_start from a u_long to a phys_addr_t, and not do a hack for
> your driver. For most architectures, this should be a no-op.
> 
> In testing this, I suggest that you configure the ten, or so, drivers
> in drivers/pcmcia that actually use this quantity, and run a sparse
> ("make C=2") run on that directory to make certain that the value
> isn't being written into a 32-bit word somewhere.
> 
> Larry

Thanks!  It looks promising.  I didn't see anything bad in the
sparse after changing the type to phys_addr_t.  But I want to
do more testing before I post a patch.

	Steve