Re: Oops in __free_pages_ok (pre7-1) (Long) (backtrace)

[Linus Torvalds <torvalds@transmeta.com>]

On Wed, 3 May 2000, Rajagopal Ananthanarayanan wrote:
> 
> One other problem with having the page locked in
> try_to_swapout() is in the call to 
> prepare_highmem_swapout() when the incoming
> page is in highmem. 

Look at how I handled this in pre7-4.

Just unlocking the old page and returning with the new page locked is
quite acceptable. The "prepare_highmem_swapout()" thing breaks the
association with the pages anyway, and as such there is no race (and this
is allowable only exactly because of the anonymous and non-shared nature
of a private COW-mapping - which is the only thing we accept in that
code-path anyway).

Doing it that way means that there are no special cases in vmscan.c.

		Linus

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux.eu.org/Linux-MM/