On 2017年12月05日 16:39, Nikolay Borisov wrote:
> Signed-off-by: Nikolay Borisov <nborisov@xxxxxxxx>
> ---
> disk-io.c | 17 +++++++++++++++++
> 1 file changed, 17 insertions(+)
>
> diff --git a/disk-io.c b/disk-io.c
> index 3d8785d5bb37..40077d4919c6 100644
> --- a/disk-io.c
> +++ b/disk-io.c
> @@ -1419,6 +1419,23 @@ static int check_super(struct btrfs_super_block *sb, unsigned sbflags)
> return -EIO;
> }
>
> +/*
> + * btrfs_read_dev_super - read a valid superblock from a block device
> + * @fd: file descrioptor of the device
> + * @sb: buffer where the superblock is going to be read in
> + * @sb_bytenr: offset of the particular superblock copie we want
> + * @sbflags: flags controlling how the superblock is read.
> + *
> + * This function is used by various btrfs comands to obtain a valid superblock.
> + *
> + * It's mode of operation is controlled by the @sb_bytenr and @sbdflags
> + * parameters. If SBREAD_RECOVER flag is set and @sb_bytenr is
> + * BTRFS_SUPER_INFO_OFFSET then the function reads all 3 superblock copies and
> + * returns the newest one. If SBREAD_RECOVER is not set then only a single
> + * copy is read, which one is decided by @sb_bytenr. If @sb_bytenr !=
> + * BTRFS_SUPER_INFO_OFFSET then the sbflags is effectively ignored and only a
> + * single copy is read.
Although the logic is not as straightforward, it's still acceptable and
the comment does makes it clearer.
Reviewed-by: Qu Wenruo <wqu@xxxxxxxx>
Thanks,
Qu
> + */
> int btrfs_read_dev_super(int fd, struct btrfs_super_block *sb, u64 sb_bytenr,
> unsigned sbflags)
> {
>
Attachment:
signature.asc
Description: OpenPGP digital signature
