At 12:04 PM 2/16/98 , Ignatios Souvatzis wrote:
>> Here is a different fix for the mentioned problem. It disallows procfs access
>> to regions mmap()ed from a device completely and thus avoids the situation
>> which led to the panic (or to data corruption without
>> MACHINE_NEW_NONCONTIG).
>> Would this be too restrictive?
>
>Hm... thinking about frame buffers, maybe. Thinking about the
>controlling registers for the frame buffer devices, no.

I don't think so.  If you didn't map the memory, there is little chance
of you knowing how to access it (think of Alpha's and sparse/dense memory).
If you need access, map it yourself.
-- 
Matt Thomas               Internet:   matt@3am-software.com
3am Software Foundry      WWW URL:    http://www.3am-software.com/bio/matt/
Nashua, NH                Disclaimer: I disavow all knowledge of this message