On Tue, Sep 28, 2021 at 08:18:07PM +0300, Konstantin Komarov wrote:
> If we continue to work in this case, then we can corrupt fs.
>
Remember to add fixes tag.
> Signed-off-by: Konstantin Komarov <almaz.alexandrovich@xxxxxxxxxxxxxxxxxxxx>
> ---
> fs/ntfs3/super.c | 13 ++++++++++++-
> 1 file changed, 12 insertions(+), 1 deletion(-)
>
> diff --git a/fs/ntfs3/super.c b/fs/ntfs3/super.c
> index aff90f70e7bf..890c5d9d6d60 100644
> --- a/fs/ntfs3/super.c
> +++ b/fs/ntfs3/super.c
> @@ -763,9 +763,20 @@ static int ntfs_init_from_boot(struct super_block *sb, u32 sector_size,
> sbi->mft.lbo = mlcn << sbi->cluster_bits;
> sbi->mft.lbo2 = mlcn2 << sbi->cluster_bits;
>
> - if (sbi->cluster_size < sbi->sector_size)
> + /* Compare boot's cluster and sector. */
> + if (sbi->cluster_size < boot_sector_size)
This should actually be in patch 3/3 as this has nothing to do with bug
fix and bug fix is most important thing to be seperated and as you have
do be front in the series. After that
Reviewed-by: Kari Argillander <kari.argillander@xxxxxxxxx>
> goto out;
>
> + /* Compare boot's cluster and media sector. */
> + if (sbi->cluster_size < sector_size) {
> + /* No way to use ntfs_get_block in this case. */
> + ntfs_err(
> + sb,
> + "Failed to mount 'cause NTFS's cluster size (%u) is less than media sector size (%u)",
> + sbi->cluster_size, sector_size);
> + goto out;
> + }
> +
> sbi->cluster_mask = sbi->cluster_size - 1;
> sbi->cluster_mask_inv = ~(u64)sbi->cluster_mask;
> sbi->record_size = record_size = boot->record_size < 0
> --
> 2.33.0
>
>
