Re: locking question: do_mmap(), do_munmap()

[Dominic Sweetman]

Ralf,

> (On the R3000 which allows swapping of i-cache and d-cache for
> diagnostic purposes it actually is possible to have dirty lines in
> the d-cache when it's used as the i-cache...

Not really; the R3000 caches were always write-through, so the 
d-cache can never be dirty and in fact has exactly the same possible
states as the i-cache.

> But that's not a consideration for Linux.)

Well, OK then.

Dominic Sweetman
[EMAIL PROTECTED]