On Tue, Nov 20, 2001 at 07:04:27AM -0800, Jason R Thorpe wrote:
>  > At first I expected that, but the result shows that mmap(2) doesn't
>  > act like that: the third mmap call returned 0x50001000, not 0x50000000.
>  > If the manpage was right, mmap should zap the previous mapping at
>  > 0x50000000.
> 
> No, because you did not use MAP_FIXED.  The first argument to mmap is
> only used as a hint if MAP_FIXED is not supplied...

That's what I want, always using a hint as starting address if it's 
valid and usable. The current implementation ignores any hint smaller
than a certain value even if it's valid and usable.

Jun-Young

-- 
Bang Jun-Young <bjy@mogua.org>