[dm-devel] Re: [PATCH] Fix panic in 2.6 with bounced bio and dm

[Andrew Morton]

Mark Haverkamp <markh at osdl.org> wrote:
>
> 
> Last September a fix was checked in for a memory leak problem in
> bounce_end_io causing the entire bio to be checked.  This ended up
> causing some dm cloned bios that had bounce buffers to free NULL pages
> because their bi_idx can be non-zero.   This patch skips NULL pages in
> the bio's bio_vec.  I'm not sure if this is the most optimal fix but I
> think that it is safe since bvec_alloc memsets the bio_vec to zero.
> 

Thanks, we should get fixed for 2.6.11.

It seems very weird for dm to be shoving NULL page*'s into the middle of a
bio's bvec array, so your fix might end up being a workaround pending a
closer look at what's going on in there.

> 
> ===== mm/highmem.c 1.55 vs edited =====
> --- 1.55/mm/highmem.c	2005-01-07 21:44:13 -08:00
> +++ edited/mm/highmem.c	2005-02-25 07:54:21 -08:00
> @@ -319,7 +319,7 @@
>  	 */
>  	__bio_for_each_segment(bvec, bio, i, 0) {
>  		org_vec = bio_orig->bi_io_vec + i;
> -		if (bvec->bv_page == org_vec->bv_page)
> +		if (!bvec->bv_page || bvec->bv_page == org_vec->bv_page)
>  			continue;
>  
>  		mempool_free(bvec->bv_page, pool);	
> 
> -- 
> Mark Haverkamp <markh at osdl.org>
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo at vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/