Re: Another fatfs patch

[Marcus Brinkmann]

Hi,

most of the patch should go in soon (hi Jeff ;), but please keep
paying attention to the coding standard.  If you split a line, make
sure operators go at the beginning of the next line and not at the end
of the current line.

> I also removed some dead code in pager.c that only produced some
> warnings.

We still might want the stats, so I fixed the warning by casting 0 to void.
 
> I hope all other stuff is obvious (well, it is to me ;)). If it is
> not, please ask me! :)

Yeah, the following doesn't make sense to me.  You claim it is about
fixing the case where cluster size is larger than vm_page_size, but I
don't see that being addressed here, nor do I know why it's an issue.
Please explain that :)

> @@ -106,7 +104,8 @@ find_cluster (struct node *node, vm_offs
>        rwlock_reader_lock (*lock);
>      }
>  
> -  if (offset + bytes_per_cluster > node->allocsize)
> +  if ((offset + vm_page_size > node->allocsize) &&
> +      (offset + bytes_per_cluster > node->allocsize))
>      return EIO;
>  
>    err = fat_getcluster (node, offset >> log2_bytes_per_cluster, 0, cluster);

Thanks,
Marcus