Complete bollocks. As long as the pte chains are consistent while
refill_inactive_zone holds pte_chain_lock (they darn well should be),
concurrent page_referenced() and page_convert_anon() is fine.
It could be that page_referenced() returns an inappropriate answer, but it's
so rare we don't care.
Which is good. We really don't want to lock pages in refill_inactive_zone()
to keep the extremely rare page_convert_anon() away. refill_inactive_zone()
is more a bath-temperature path than a hotpath, but still...
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to firstname.lastname@example.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/