Re: [PATCH v6 1/3] block: introduce compress filter driver

[Kevin Wolf <kwolf@redhat.com>]

Am 11.11.2019 um 17:04 hat Andrey Shinkevich geschrieben:
> Allow writing all the data compressed through the filter driver.
> The written data will be aligned by the cluster size.
> Based on the QEMU current implementation, that data can be written to
> unallocated clusters only. May be used for a backup job.
> 
> Suggested-by: Max Reitz <mreitz@redhat.com>
> Signed-off-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>

> +static BlockDriver bdrv_compress = {
> +    .format_name                        = "compress",
> +
> +    .bdrv_open                          = zip_open,
> +    .bdrv_child_perm                    = zip_child_perm,

Why do you call the functions zip_* when the driver is called compress?
I think zip would be a driver for zip archives, which we don't use here.

> +    .bdrv_getlength                     = zip_getlength,
> +    .bdrv_co_truncate                   = zip_co_truncate,
> +
> +    .bdrv_co_preadv                     = zip_co_preadv,
> +    .bdrv_co_preadv_part                = zip_co_preadv_part,
> +    .bdrv_co_pwritev                    = zip_co_pwritev,
> +    .bdrv_co_pwritev_part               = zip_co_pwritev_part,

If you implement .bdrv_co_preadv/pwritev_part, isn't the implementation
of .bdrv_co_preadv/pwritev (without _part) dead code?

> +    .bdrv_co_pwrite_zeroes              = zip_co_pwrite_zeroes,
> +    .bdrv_co_pdiscard                   = zip_co_pdiscard,
> +    .bdrv_refresh_limits                = zip_refresh_limits,
> +
> +    .bdrv_eject                         = zip_eject,
> +    .bdrv_lock_medium                   = zip_lock_medium,
> +
> +    .bdrv_co_block_status               = bdrv_co_block_status_from_backing,

Why not use bs->file? (Well, apart from the still not merged filter
series by Max...)

> +    .bdrv_recurse_is_first_non_filter   = zip_recurse_is_first_non_filter,
> +
> +    .has_variable_length                = true,
> +    .is_filter                          = true,
> +};

Kevin