[RFC,v1,3/4] mke2fs: add options extended option for changing first inode

All currently reserved special inodes are assigned to some features. Some of
them never been actually implemented so could be reused. But anyway there are
only few special inodes left. Index of first normal inode is already stored in
the super-block thus we can reserve more without changing disk layout.

This patch add extended option 'first_inode' for overriding default first inode
(mostly for debug purpose). Features which want special inodes above 10 could
change first_inode automatically.

Changing default first_inode is possible, but this breaks too much tests where
inode numbers are present in expected output or even hardcoded: for example in
"tst_inline_data".

Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
---
 misc/mke2fs.8.in |    8 ++++++++
 misc/mke2fs.c    |   18 ++++++++++++++++++
 2 files changed, 26 insertions(+)


--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

On Thu, Mar 12, 2015 at 07:20:15PM +0300, Konstantin Khlebnikov wrote:
> All currently reserved special inodes are assigned to some features. Some of
> them never been actually implemented so could be reused. But anyway there are
> only few special inodes left. Index of first normal inode is already stored in
> the super-block thus we can reserve more without changing disk layout.
> 
> This patch add extended option 'first_inode' for overriding default first inode
> (mostly for debug purpose). Features which want special inodes above 10 could
> change first_inode automatically.
> 
> Changing default first_inode is possible, but this breaks too much tests where
> inode numbers are present in expected output or even hardcoded: for example in
> "tst_inline_data".

Hmmm, that's going to make things interesting whenever we *do* need to
increase the default # of reserved inodes.

> Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
> ---
>  misc/mke2fs.8.in |    8 ++++++++
>  misc/mke2fs.c    |   18 ++++++++++++++++++
>  2 files changed, 26 insertions(+)
> 
> diff --git a/misc/mke2fs.8.in b/misc/mke2fs.8.in
> index aeb5caf6e869..13f0c7a2c9d2 100644
> --- a/misc/mke2fs.8.in
> +++ b/misc/mke2fs.8.in
> @@ -369,6 +369,14 @@ as default.
>  .TP
>  .BI nodiscard
>  Do not attempt to discard blocks at mkfs time.
> +.TP
> +.BI first_inode= inode-number
> +This defines first normal inode, lower inodes are reserved for internal use.
> +For filesystem revision 0 this valus is fixed to 11, that is default and
> +minimal value for revision 1. This could be changed at a later time with the

I doubt there are very many people formatting a rev0 FS these days. :)

"valus" -> "value" above.

> +.B \-I
> +command-line option to
> +.BR resize2fs (8).
>  @QUOTA_MAN_COMMENT@.TP
>  @QUOTA_MAN_COMMENT@.BI quotatype
>  @QUOTA_MAN_COMMENT@Specify which quota type ('usr' or 'grp') is to be
> diff --git a/misc/mke2fs.c b/misc/mke2fs.c
> index ec450adf9e12..73d148206c69 100644
> --- a/misc/mke2fs.c
> +++ b/misc/mke2fs.c
> @@ -1005,6 +1005,23 @@ static void parse_extended_opts(struct ext2_super_block *param,
>  			discard = 1;
>  		} else if (!strcmp(token, "nodiscard")) {
>  			discard = 0;
> +		} else if (!strcmp(token, "first_inode")) {
> +			if (!arg) {
> +				r_usage++;
> +				badopt = token;
> +				continue;
> +			}
> +			param->s_first_ino = strtoul(arg, &p, 0);
> +			if (*p ||
> +			    param->s_first_ino < EXT2_GOOD_OLD_FIRST_INO ||
> +			    (param->s_rev_level == EXT2_GOOD_OLD_REV &&
> +			     param->s_first_ino != EXT2_GOOD_OLD_FIRST_INO)) {
> +				fprintf(stderr,
> +					_("Invalid first_ino parameter: %s\n"),

"Invalid first_inode parameter"; please keep the parameter name consistent in
the program output.

> +					arg);
> +				r_usage++;
> +				continue;
> +			}
>  		} else if (!strcmp(token, "quotatype")) {
>  			if (!arg) {
>  				r_usage++;
> @@ -1047,6 +1064,7 @@ static void parse_extended_opts(struct ext2_super_block *param,
>  			"\ttest_fs\n"
>  			"\tdiscard\n"
>  			"\tnodiscard\n"
> +			"\tfirst_inode=<inode number>\n"
>  			"\tquotatype=<usr OR grp>\n\n"),
>  			badopt ? badopt : "");
>  		free(buf);
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html