Re: [PATCH] Add basic BIGALLOC support for cluster-based allocation

[Ted Ts'o <tytso@mit.edu>]

On Sat, Feb 26, 2011 at 09:20:55AM +0200, Amir Goldstein wrote:
> Just to be clear, my alternative suggestion to on-disk format change was:
> 
>         __u32   s_log_block_size;       /* Block size */
> -       __s32   s_log_frag_size;        /* Fragment size */
> +       __s32   s_log_cluster_size;     /* Allocation cluster size */
> -       __u32   s_blocks_per_group;     /* # Blocks per group */
> +       __u32   s_clusters_per_group;   /* # Clusters per group */
> -       __u32   s_frags_per_group;      /* # Fragments per group */
> +       __u32   s_blocks_per_group;   /* # Blocks per group */
>         __u32   s_inodes_per_group;     /* # Inodes per group */
> 
> This way, old kernels see a sane value in what used to be
> s_blocks_per_group (32K) and new kernels (even without the BIGALLOC
> feature) make sure to write correct values in both new
> s_blocks_per_group and s_clusters_per_group (old s_blocks_per_group)

Unfortunately, that's likely to confuse old kernels even more, since
s_blocks_per_group is used to calculate the number of block groups,
i.e. s_blocks_count / s_blocks_per_group.  In BIGALLOC mode, the
number of block groups go down, and so if we change s_blocks_per_group
to s_clusters_per_group, it's likely to cause much mischief, for
example in ext4_check_descriptors().

					- Ted