> > This would work, but only if it is mlock'd (could be done by the
> > kernel, just set the appropriate vm_flags during the mmap) else the
> > swapper can toss it to disk at any time.
>
> You need to bring the pages in. Also mmap() doesnt technically guarantee
> pages will not move physically. It means they wont wander onto disk.
If the kernel has a pointer to the pages the page count
can never get below 1 and the page will stay firmly in
it's place.
I agree, however, that a kernel-space alloc() with an
mmap() is a much cleaner solution...
Rik -- Open Source: you deserve to be in control of your data.
+-------------------------------------------------------------------+
| Linux Memory Management site: http://humbolt.geo.uu.nl/Linux-MM/ |
| Nederlandse Linux documentatie: http://www.nl.linux.org/ |
+-------------------------------------------------------------------+
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/