> > ioremap_nocache(). You can also use ioremap_nocache() on allocated pages
> > (it's a page granularity thing) via doing something like:
> >
> > page = __get_free_pages(whatever, order);
> > ptr = ioremap_nocache(virt_to_phys(page), 1 <<
> > (order+PAGE_SHIFT));
>
> But the problem is that this mapping is never released if you call
> iounmap() and if you remove that check there, vfree tries to free not only
> the mapping but also any non reserved page. I think we really need a real
> memremap() interface...
sigh. if you read the rest of my mail which you quote, i gave the exact
changes which are needed to get around this, and i suggested a new
memmap_nocache() interface.
-- mingo
-
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/