On Fri, Jul 30, 2010 at 3:43 AM, KAMEZAWA Hiroyuki
<kamezawa.hiroyu@jp.fujitsu.com> wrote:
Absolutely. I didn't intend my broader anxieties to distract from
your immediate scan_swap_map() fix. If Ondrej finds that it does in
fact fix the corruption he's been seeing since 2.6.31, then I'd like
your fix to go straight to Linus for 2.6.35: once swap is 50% full,
your case is rather too likely to occur; whereas I'm worrying about
correctness in unlikely hypothetical circumstances. But please do
credit Ondrej, and make clear that you're fixing an observed
corrution; and feel free to add my
Acked-by: Hugh Dickins <hughd@google.com>
Hugh
--