[2.6.22.y] {01**/17} - do_anonymous_page-race - series for stable kernel

[Oliver Pinter (Pintér Olivér)]

NOT IN MAINLINE

Linus it's go or drop it?



-- 
Thanks,
Oliver
> --- Begin Message ---
> Additional Comment #103 From Andrea Arcangeli 2004-10-15 19:41
> the last patch I attached is the safest I believe.
>
> I'm not sure if a lock_unlock or lock_unlock is always guaranteed to happen
> after the I/O, and this makes sure no race can happen anymore.
>
> ---
>  fs/bio.c |    9 +++++++++
>  1 file changed, 9 insertions(+)
>
> --- a/fs/bio.c	2007-07-08 19:32:17.000000000 -0400
> +++ b/fs/bio.c	2007-08-27 14:01:21.000000000 -0400
> @@ -1028,6 +1028,15 @@ void bio_endio(struct bio *bio, unsigned
>  	bio->bi_size -= bytes_done;
>  	bio->bi_sector += (bytes_done >> 9);
>  
> +	if (bio_data_dir(bio) == READ)
> +		/*
> +		 * If the current cpu has written to the page by hand
> +		 * without dma, we must enforce ordering to be sure
> +		 * this written data will be visible before we expose
> +		 * the page contents to other cpus (for example with
> +		 * a set_pte).
> +		 */
> +		smp_wmb();
>  	if (bio->bi_end_io)
>  		bio->bi_end_io(bio, bytes_done, error);
>  }
>
> --- End Message ---