Re: [PATCH] mm: use pagevec to rotate reclaimable page

[Hisashi Hifumi]

Thank you for your comment.

At 11:37 07/09/14, Andrew Morton wrote:

>The page_count() test here is a bit of a worry, too.  Why do we need it?
>The caller must have pinned the page in some fashion else we couldn't use
>it safely in this function at all.
>
>I assume that you discovered that once we've cleared PageWriteback(), the
>page can get reclaimed elsewhere?  If so, that could still happen
>immediately after the page_count() test.  It's all a bit of a worry.
>Deferring the ClearPageWriteback() will fix any race concerns, but I do
>think that we need to take a ref on the page for the pagevec ownership.
>

Actually, I considered taking a ref to pin pages. But this could prevent 
the page
reclaiming activity so I did not use it.

I reflect your comment and send you modified patch.

-
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/