Re: [RFC] Making compound pages mandatory

[David Howells]

> Yes, it's just the single pointer chase.  Probably that's the common case
> now, because everyone will be enabling hugepages on lots of architectures.

No, it's not just the single pointer chase. See my email to Linus.

I would, however, be willing to endorse the use of PG_compound, especially
under !MMU conditions for two reasons:

 (1) Safety. The current mechanism of managing intermediate pages is
     potentially fragile under uClinux.

 (2) put_page() and get_page() aren't actually called all that often compared
     to other things, especially under uClinux.

David
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/