Re: [PATCH] btrfs: Fix BUG_ON condition in scrub_setup_recheck_block()

[Sanidhya Solanki <lkml.page@gmail.com> (by way of Sanidhya Solanki <lkml.page@gmail.com>)]

On Fri, 29 Apr 2016 18:33:59 -0700
Ashish Samant <ashish.samant@oracle.com> wrote:

> pagev array in scrub_block{} is of size SCRUB_MAX_PAGES_PER_BLOCK.
> page_index should be checked with the same to trigger BUG_ON().
> 
> Signed-off-by: Ashish Samant <ashish.samant@oracle.com>
> ---
>  fs/btrfs/scrub.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
> index 4678f03..376a6dd 100644
> --- a/fs/btrfs/scrub.c
> +++ b/fs/btrfs/scrub.c
> @@ -1350,7 +1350,7 @@ static int scrub_setup_recheck_block(struct scrub_block *original_sblock,
>  		recover->bbio = bbio;
>  		recover->map_length = mapped_length;
>  
> -		BUG_ON(page_index >= SCRUB_PAGES_PER_RD_BIO);
> +		BUG_ON(page_index >= SCRUB_MAX_PAGES_PER_BLOCK);
>  
>  		nmirrors = min(scrub_nr_raid_mirrors(bbio), BTRFS_MAX_MIRRORS);
>  

Hello

I am trying to implement a user-space IOCTL option to change the RAID Stripe Size.
I have run into an issue in the code included here, specifically in line 64 of scrub.c. 

#define SCRUB_MAX_PAGES_PER_BLOCK      16      /* 64k per node/leaf/sector */

This line should become a user-space variable if SCRUB_MAX_PAGES_PER_BLOCK is
related in any way to the definition of BTrFS Stripe Length (defined in
volume.h)  being equal to a product of Super_blk and RAID Stripe Length, as it was
before mid-December before the patch
"http://article.gmane.org/gmane.comp.file-systems.btrfs/51221/" modified it. If 
SCRUB_MAX_PAGES_PER_BLOCK is related to Super_blk, then it too will need to be
changed from a macro to something that is runtime configurable.

Could those more familiar with the code base let me know.

Thanks