Re: + expose-range-checking-functions-from-arch-specific.patch added to -mm tree

[Rusty Russell <rusty@rustcorp.com.au>]

On Wed, 2007-04-11 at 18:03 +0100, David Howells wrote:
> Rusty Russell <rusty@rustcorp.com.au> wrote:
> 
> > static inline bool range_over_limit(unsigned long start,
> > 				    unsigned long len,
> > 				    unsigned long limit)
> 
> I'm still not sure the name is entirely clear, but it's better.  I'd still
> stick the word "check" in there personally, perhaps check_range_limit(), but
> that's just my preference.

"if (check_range_limit())" seems like the reverse of "if
(range_over_limit())" tho.

> > PS.  Previously this identical function was called __range_ok() (and
> > returned 0 if it was not ok...)
> 
> Ummm...  Didn't __range_ok() implicitly involve get_addr_limit() rather than
> taking an explicit range?  Certainly i386 thinks so:

Yep, my bad.  I was thinking of the sense of the return value.

Thanks,
Rusty.