2017-01-26 12:27:58 +00:00
|
|
|
/*
|
|
|
|
* This program is free software; you can redistribute it and/or
|
|
|
|
* modify it under the terms of the GNU General Public
|
|
|
|
* License v2 as published by the Free Software Foundation.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|
|
|
* General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public
|
|
|
|
* License along with this program; if not, write to the
|
|
|
|
* Free Software Foundation, Inc., 59 Temple Place - Suite 330,
|
|
|
|
* Boston, MA 021110-1307, USA.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include "kerncompat.h"
|
|
|
|
#include <sys/utsname.h>
|
2021-09-07 14:20:17 +00:00
|
|
|
#include <sys/ioctl.h>
|
2017-01-26 12:27:58 +00:00
|
|
|
#include <linux/version.h>
|
2020-10-13 01:06:02 +00:00
|
|
|
#include <unistd.h>
|
2022-09-15 11:59:39 +00:00
|
|
|
#include <errno.h>
|
|
|
|
#include <stdbool.h>
|
|
|
|
#include <stdlib.h>
|
|
|
|
#include <string.h>
|
|
|
|
#include "kernel-lib/sizes.h"
|
2023-04-19 21:17:12 +00:00
|
|
|
#include "kernel-shared/uapi/btrfs.h"
|
2020-08-18 13:56:04 +00:00
|
|
|
#include "kernel-shared/ctree.h"
|
2022-09-14 15:06:52 +00:00
|
|
|
#include "common/fsfeatures.h"
|
2022-09-15 21:15:17 +00:00
|
|
|
#include "common/string-utils.h"
|
2023-08-21 19:13:24 +00:00
|
|
|
#include "common/sysfs-utils.h"
|
2019-06-19 23:46:21 +00:00
|
|
|
#include "common/utils.h"
|
2022-09-15 11:59:39 +00:00
|
|
|
#include "common/messages.h"
|
2017-01-26 12:27:58 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Insert a root item for temporary tree root
|
|
|
|
*
|
|
|
|
* Only used in make_btrfs_v2().
|
|
|
|
*/
|
2021-09-29 13:05:26 +00:00
|
|
|
#define VERSION_TO_STRING3(name, a,b,c) \
|
|
|
|
.name ## _str = #a "." #b "." #c, \
|
|
|
|
.name ## _ver = KERNEL_VERSION(a,b,c)
|
|
|
|
#define VERSION_TO_STRING2(name, a,b) \
|
|
|
|
.name ## _str = #a "." #b, \
|
|
|
|
.name ## _ver = KERNEL_VERSION(a,b,0)
|
|
|
|
#define VERSION_NULL(name) \
|
|
|
|
.name ## _str = NULL, \
|
|
|
|
.name ## _ver = 0
|
2017-01-26 12:27:58 +00:00
|
|
|
|
2018-05-08 06:31:52 +00:00
|
|
|
enum feature_source {
|
|
|
|
FS_FEATURES,
|
|
|
|
RUNTIME_FEATURES,
|
|
|
|
};
|
|
|
|
|
2017-01-26 12:27:58 +00:00
|
|
|
/*
|
|
|
|
* Feature stability status and versions: compat <= safe <= default
|
|
|
|
*/
|
2018-05-08 06:31:52 +00:00
|
|
|
struct btrfs_feature {
|
2017-01-26 12:27:58 +00:00
|
|
|
const char *name;
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* At least one of the bit must be set in the following *_flag member.
|
|
|
|
*
|
|
|
|
* For features like list-all and quota which don't have any
|
|
|
|
* incompat/compat_ro bit set, it go to runtime_flag.
|
|
|
|
*/
|
|
|
|
u64 incompat_flag;
|
|
|
|
u64 compat_ro_flag;
|
|
|
|
u64 runtime_flag;
|
|
|
|
|
2017-01-26 12:27:58 +00:00
|
|
|
const char *sysfs_name;
|
|
|
|
/*
|
|
|
|
* Compatibility with kernel of given version. Filesystem can be
|
|
|
|
* mounted.
|
|
|
|
*/
|
|
|
|
const char *compat_str;
|
|
|
|
u32 compat_ver;
|
|
|
|
/*
|
|
|
|
* Considered safe for use, but is not on by default, even if the
|
|
|
|
* kernel supports the feature.
|
|
|
|
*/
|
|
|
|
const char *safe_str;
|
|
|
|
u32 safe_ver;
|
|
|
|
/*
|
|
|
|
* Considered safe for use and will be turned on by default if
|
|
|
|
* supported by the running kernel.
|
|
|
|
*/
|
|
|
|
const char *default_str;
|
|
|
|
u32 default_ver;
|
|
|
|
const char *desc;
|
2018-05-08 06:31:52 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
static const struct btrfs_feature mkfs_features[] = {
|
2021-09-29 13:05:26 +00:00
|
|
|
{
|
|
|
|
.name = "mixed-bg",
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.incompat_flag = BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS,
|
2021-09-29 13:05:26 +00:00
|
|
|
.sysfs_name = "mixed_groups",
|
|
|
|
VERSION_TO_STRING3(compat, 2,6,37),
|
|
|
|
VERSION_TO_STRING3(safe, 2,6,37),
|
|
|
|
VERSION_NULL(default),
|
|
|
|
.desc = "mixed data and metadata block groups"
|
2022-09-10 19:26:35 +00:00
|
|
|
},
|
|
|
|
{
|
|
|
|
.name = "quota",
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.runtime_flag = BTRFS_FEATURE_RUNTIME_QUOTA,
|
2022-09-10 19:26:35 +00:00
|
|
|
.sysfs_name = NULL,
|
|
|
|
VERSION_TO_STRING2(compat, 3,4),
|
|
|
|
VERSION_NULL(safe),
|
|
|
|
VERSION_NULL(default),
|
|
|
|
.desc = "quota support (qgroups)"
|
|
|
|
},
|
2023-09-27 17:46:46 +00:00
|
|
|
{
|
|
|
|
.name = "squota",
|
|
|
|
.incompat_flag = BTRFS_FEATURE_INCOMPAT_SIMPLE_QUOTA,
|
|
|
|
.sysfs_name = "squota",
|
|
|
|
VERSION_TO_STRING2(compat, 6,7),
|
|
|
|
VERSION_NULL(safe),
|
|
|
|
VERSION_NULL(default),
|
|
|
|
.desc = "squota support (simple accounting qgroups)"
|
|
|
|
},
|
2022-09-10 19:26:35 +00:00
|
|
|
{
|
2021-09-29 13:05:26 +00:00
|
|
|
.name = "extref",
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.incompat_flag = BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF,
|
2021-09-29 13:05:26 +00:00
|
|
|
.sysfs_name = "extended_iref",
|
|
|
|
VERSION_TO_STRING2(compat, 3,7),
|
|
|
|
VERSION_TO_STRING2(safe, 3,12),
|
|
|
|
VERSION_TO_STRING2(default, 3,12),
|
|
|
|
.desc = "increased hardlink limit per file to 65536"
|
|
|
|
}, {
|
|
|
|
.name = "raid56",
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.incompat_flag = BTRFS_FEATURE_INCOMPAT_RAID56,
|
2021-09-29 13:05:26 +00:00
|
|
|
.sysfs_name = "raid56",
|
|
|
|
VERSION_TO_STRING2(compat, 3,9),
|
|
|
|
VERSION_NULL(safe),
|
|
|
|
VERSION_NULL(default),
|
|
|
|
.desc = "raid56 extended format"
|
|
|
|
}, {
|
|
|
|
.name = "skinny-metadata",
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.incompat_flag = BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA,
|
2021-09-29 13:05:26 +00:00
|
|
|
.sysfs_name = "skinny_metadata",
|
|
|
|
VERSION_TO_STRING2(compat, 3,10),
|
|
|
|
VERSION_TO_STRING2(safe, 3,18),
|
|
|
|
VERSION_TO_STRING2(default, 3,18),
|
|
|
|
.desc = "reduced-size metadata extent refs"
|
|
|
|
}, {
|
|
|
|
.name = "no-holes",
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.incompat_flag = BTRFS_FEATURE_INCOMPAT_NO_HOLES,
|
2021-09-29 13:05:26 +00:00
|
|
|
.sysfs_name = "no_holes",
|
|
|
|
VERSION_TO_STRING2(compat, 3,14),
|
|
|
|
VERSION_TO_STRING2(safe, 4,0),
|
2021-09-29 13:12:21 +00:00
|
|
|
VERSION_TO_STRING2(default, 5,15),
|
2021-09-29 13:05:26 +00:00
|
|
|
.desc = "no explicit hole extents for files"
|
2022-09-10 19:26:35 +00:00
|
|
|
},
|
|
|
|
{
|
|
|
|
.name = "free-space-tree",
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.compat_ro_flag = BTRFS_FEATURE_COMPAT_RO_FREE_SPACE_TREE |
|
|
|
|
BTRFS_FEATURE_COMPAT_RO_FREE_SPACE_TREE_VALID,
|
2022-09-10 19:26:35 +00:00
|
|
|
.sysfs_name = "free_space_tree",
|
|
|
|
VERSION_TO_STRING2(compat, 4,5),
|
|
|
|
VERSION_TO_STRING2(safe, 4,9),
|
|
|
|
VERSION_TO_STRING2(default, 5,15),
|
|
|
|
.desc = "free space tree (space_cache=v2)"
|
|
|
|
},
|
|
|
|
{
|
2021-09-29 13:05:26 +00:00
|
|
|
.name = "raid1c34",
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.incompat_flag = BTRFS_FEATURE_INCOMPAT_RAID1C34,
|
2021-09-29 13:05:26 +00:00
|
|
|
.sysfs_name = "raid1c34",
|
|
|
|
VERSION_TO_STRING2(compat, 5,5),
|
|
|
|
VERSION_NULL(safe),
|
|
|
|
VERSION_NULL(default),
|
|
|
|
.desc = "RAID1 with 3 or 4 copies"
|
|
|
|
},
|
2021-04-26 06:27:20 +00:00
|
|
|
#ifdef BTRFS_ZONED
|
2021-09-29 13:05:26 +00:00
|
|
|
{
|
|
|
|
.name = "zoned",
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.incompat_flag = BTRFS_FEATURE_INCOMPAT_ZONED,
|
2021-09-29 13:05:26 +00:00
|
|
|
.sysfs_name = "zoned",
|
|
|
|
VERSION_TO_STRING2(compat, 5,12),
|
|
|
|
VERSION_NULL(safe),
|
|
|
|
VERSION_NULL(default),
|
|
|
|
.desc = "support zoned devices"
|
|
|
|
},
|
2021-11-05 20:40:27 +00:00
|
|
|
#endif
|
2022-09-10 19:26:35 +00:00
|
|
|
{
|
|
|
|
.name = "block-group-tree",
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.compat_ro_flag = BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE,
|
|
|
|
.sysfs_name = "block_group_tree",
|
2022-10-17 13:46:37 +00:00
|
|
|
VERSION_TO_STRING2(compat, 6,1),
|
2022-09-10 19:26:35 +00:00
|
|
|
VERSION_NULL(safe),
|
|
|
|
VERSION_NULL(default),
|
|
|
|
.desc = "block group tree to reduce mount time"
|
|
|
|
},
|
2023-04-11 02:31:06 +00:00
|
|
|
#if EXPERIMENTAL
|
2021-11-05 20:40:27 +00:00
|
|
|
{
|
|
|
|
.name = "extent-tree-v2",
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.incompat_flag = BTRFS_FEATURE_INCOMPAT_EXTENT_TREE_V2,
|
2021-11-05 20:40:27 +00:00
|
|
|
.sysfs_name = "extent_tree_v2",
|
|
|
|
VERSION_TO_STRING2(compat, 5,15),
|
|
|
|
VERSION_NULL(safe),
|
|
|
|
VERSION_NULL(default),
|
|
|
|
.desc = "new extent tree format"
|
2023-09-14 16:05:35 +00:00
|
|
|
} , {
|
|
|
|
.name = "raid-stripe-tree",
|
|
|
|
.incompat_flag = BTRFS_FEATURE_INCOMPAT_RAID_STRIPE_TREE,
|
|
|
|
.sysfs_name = NULL,
|
|
|
|
VERSION_TO_STRING2(compat, 6,7),
|
|
|
|
VERSION_NULL(safe),
|
|
|
|
VERSION_NULL(default),
|
|
|
|
.desc = "raid stripe tree"
|
2021-11-05 20:40:27 +00:00
|
|
|
},
|
2021-04-26 06:27:20 +00:00
|
|
|
#endif
|
2017-01-26 12:27:58 +00:00
|
|
|
/* Keep this one last */
|
2021-09-29 13:05:26 +00:00
|
|
|
{
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.name = "list-all",
|
|
|
|
.runtime_flag = BTRFS_FEATURE_RUNTIME_LIST_ALL,
|
|
|
|
.sysfs_name = NULL,
|
2021-09-29 13:05:26 +00:00
|
|
|
VERSION_NULL(compat),
|
|
|
|
VERSION_NULL(safe),
|
|
|
|
VERSION_NULL(default),
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.desc = NULL
|
2021-09-29 13:05:26 +00:00
|
|
|
}
|
2017-01-26 12:27:58 +00:00
|
|
|
};
|
|
|
|
|
2018-05-08 06:31:52 +00:00
|
|
|
static const struct btrfs_feature runtime_features[] = {
|
2021-09-29 13:05:26 +00:00
|
|
|
{
|
|
|
|
.name = "quota",
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.runtime_flag = BTRFS_FEATURE_RUNTIME_QUOTA,
|
2021-09-29 13:05:26 +00:00
|
|
|
.sysfs_name = NULL,
|
|
|
|
VERSION_TO_STRING2(compat, 3,4),
|
|
|
|
VERSION_NULL(safe),
|
|
|
|
VERSION_NULL(default),
|
|
|
|
.desc = "quota support (qgroups)"
|
|
|
|
}, {
|
|
|
|
.name = "free-space-tree",
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.compat_ro_flag = BTRFS_FEATURE_COMPAT_RO_FREE_SPACE_TREE |
|
|
|
|
BTRFS_FEATURE_COMPAT_RO_FREE_SPACE_TREE_VALID,
|
|
|
|
.sysfs_name = "free_space_tree",
|
2021-09-29 13:05:26 +00:00
|
|
|
VERSION_TO_STRING2(compat, 4,5),
|
|
|
|
VERSION_TO_STRING2(safe, 4,9),
|
2021-09-29 13:16:47 +00:00
|
|
|
VERSION_TO_STRING2(default, 5,15),
|
2021-09-29 13:05:26 +00:00
|
|
|
.desc = "free space tree (space_cache=v2)"
|
2022-09-12 03:12:01 +00:00
|
|
|
},
|
2018-05-08 06:31:52 +00:00
|
|
|
/* Keep this one last */
|
2021-09-29 13:05:26 +00:00
|
|
|
{
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.name = "list-all",
|
|
|
|
.runtime_flag = BTRFS_FEATURE_RUNTIME_LIST_ALL,
|
|
|
|
.sysfs_name = NULL,
|
2021-09-29 13:05:26 +00:00
|
|
|
VERSION_NULL(compat),
|
|
|
|
VERSION_NULL(safe),
|
|
|
|
VERSION_NULL(default),
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
.desc = NULL
|
2021-09-29 13:05:26 +00:00
|
|
|
}
|
2018-05-08 06:31:52 +00:00
|
|
|
};
|
|
|
|
|
btrfs-progs: mkfs: fix a stack over-flow when features string are too long
[BUG]
Even with chunk_objectid bug fixed, mkfs.btrfs can still caused stack
overflow when enabling extent-tree-v2 feature (need experimental
features enabled):
# ./mkfs.btrfs -f -O extent-tree-v2 ~/test.img
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
NOTE: several default settings have changed in version 5.15, please make sure
this does not affect your deployments:
- DUP for metadata (-m dup)
- enabled no-holes (-O no-holes)
- enabled free-space-tree (-R free-space-tree)
Label: (null)
UUID: 205c61e7-f58e-4e8f-9dc2-38724f5c554b
Node size: 16384
Sector size: 4096
Filesystem size: 512.00MiB
Block group profiles:
Data: single 8.00MiB
Metadata: DUP 32.00MiB
System: DUP 8.00MiB
SSD detected: no
Zoned device: no
=================================================================
[... Skip full ASAN output ...]
==65655==ABORTING
[CAUSE]
For experimental build, we have unified feature output, but the old
buffer size is only 64 bytes, which is too small to cover the new full
feature string:
extref, skinny-metadata, no-holes, free-space-tree, block-group-tree, extent-tree-v2
Above feature string is already 84 bytes, over the 64 on-stack memory
size.
This can also be proved by the ASAN output:
==65655==ERROR: AddressSanitizer: stack-buffer-overflow on address 0x7ffc4e03b1d0 at pc 0x7ff0fc05fafe bp 0x7ffc4e03ac60 sp 0x7ffc4e03a408
WRITE of size 17 at 0x7ffc4e03b1d0 thread T0
#0 0x7ff0fc05fafd in __interceptor_strcat /usr/src/debug/gcc/libsanitizer/asan/asan_interceptors.cpp:377
#1 0x55cdb7b06ca5 in parse_features_to_string common/fsfeatures.c:316
#2 0x55cdb7b06ce1 in btrfs_parse_fs_features_to_string common/fsfeatures.c:324
#3 0x55cdb7a37226 in main mkfs/main.c:1783
#4 0x7ff0fbe3c28f (/usr/lib/libc.so.6+0x2328f)
#5 0x7ff0fbe3c349 in __libc_start_main (/usr/lib/libc.so.6+0x23349)
#6 0x55cdb7a2cb34 in _start ../sysdeps/x86_64/start.S:115
[FIX]
Introduce a new macro, BTRFS_FEATURE_STRING_BUF_SIZE, along with a new
sanity check helper, btrfs_assert_feature_buf_size().
The problem is I can not find a build time method to verify
BTRFS_FEATURE_STRING_BUF_SIZE is large enough to contain all feature
names, thus have to go the runtime function to do the BUG_ON() to verify
the macro size.
Now the minimal buffer size for experimental build is 138 bytes, just
bump it to 160 for future expansion.
And if further features go beyond that number, mkfs.btrfs/btrfs-convert
will immediately crash at that BUG_ON(), so we can definitely detect it.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Tested-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-07 12:03:01 +00:00
|
|
|
/*
|
|
|
|
* This is a sanity check to make sure BTRFS_FEATURE_STRING_BUF_SIZE is large
|
|
|
|
* enough to contain all strings.
|
|
|
|
*
|
|
|
|
* All callers using btrfs_parse_*_features_to_string() should call this first.
|
|
|
|
*/
|
|
|
|
void btrfs_assert_feature_buf_size(void)
|
|
|
|
{
|
|
|
|
int total_size = 0;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This is a little over-calculated, as we include ", list-all".
|
|
|
|
* But 10 extra bytes should not be a big deal.
|
|
|
|
*/
|
|
|
|
for (i = 0; i < ARRAY_SIZE(mkfs_features); i++)
|
|
|
|
/* The extra 2 bytes are for the ", " prefix. */
|
|
|
|
total_size += strlen(mkfs_features[i].name) + 2;
|
|
|
|
BUG_ON(BTRFS_FEATURE_STRING_BUF_SIZE < total_size);
|
|
|
|
|
|
|
|
total_size = 0;
|
|
|
|
for (i = 0; i < ARRAY_SIZE(runtime_features); i++)
|
|
|
|
total_size += strlen(runtime_features[i].name) + 2;
|
|
|
|
BUG_ON(BTRFS_FEATURE_STRING_BUF_SIZE < total_size);
|
|
|
|
}
|
|
|
|
|
2018-05-08 06:31:52 +00:00
|
|
|
static size_t get_feature_array_size(enum feature_source source)
|
|
|
|
{
|
|
|
|
if (source == FS_FEATURES)
|
|
|
|
return ARRAY_SIZE(mkfs_features);
|
|
|
|
if (source == RUNTIME_FEATURES)
|
|
|
|
return ARRAY_SIZE(runtime_features);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static const struct btrfs_feature *get_feature(int i, enum feature_source source)
|
|
|
|
{
|
|
|
|
if (source == FS_FEATURES)
|
|
|
|
return &mkfs_features[i];
|
|
|
|
if (source == RUNTIME_FEATURES)
|
|
|
|
return &runtime_features[i];
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
static int parse_one_fs_feature(const char *name,
|
|
|
|
struct btrfs_mkfs_features *features,
|
2018-05-08 06:31:52 +00:00
|
|
|
enum feature_source source)
|
2017-01-26 12:27:58 +00:00
|
|
|
{
|
2018-05-08 06:31:52 +00:00
|
|
|
const int array_size = get_feature_array_size(source);
|
2017-01-26 12:27:58 +00:00
|
|
|
int i;
|
|
|
|
int found = 0;
|
|
|
|
|
2018-05-08 06:31:52 +00:00
|
|
|
for (i = 0; i < array_size; i++) {
|
|
|
|
const struct btrfs_feature *feat = get_feature(i, source);
|
|
|
|
|
|
|
|
if (name[0] == '^' && !strcmp(feat->name, name + 1)) {
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
features->compat_ro_flags &= ~feat->compat_ro_flag;
|
|
|
|
features->incompat_flags &= ~feat->incompat_flag;
|
|
|
|
features->runtime_flags &= ~feat->runtime_flag;
|
2017-01-26 12:27:58 +00:00
|
|
|
found = 1;
|
2018-05-08 06:31:52 +00:00
|
|
|
} else if (!strcmp(feat->name, name)) {
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
features->compat_ro_flags |= feat->compat_ro_flag;
|
|
|
|
features->incompat_flags |= feat->incompat_flag;
|
|
|
|
features->runtime_flags |= feat->runtime_flag;
|
2017-01-26 12:27:58 +00:00
|
|
|
found = 1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return !found;
|
|
|
|
}
|
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
static void parse_features_to_string(char *buf,
|
|
|
|
const struct btrfs_mkfs_features *features,
|
2018-05-08 06:31:52 +00:00
|
|
|
enum feature_source source)
|
2017-01-26 12:27:58 +00:00
|
|
|
{
|
2018-05-08 06:31:52 +00:00
|
|
|
const int array_size = get_feature_array_size(source);
|
2017-01-26 12:27:58 +00:00
|
|
|
int i;
|
|
|
|
|
|
|
|
buf[0] = 0;
|
|
|
|
|
2018-05-08 06:31:52 +00:00
|
|
|
for (i = 0; i < array_size; i++) {
|
|
|
|
const struct btrfs_feature *feat = get_feature(i, source);
|
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
if (features->compat_ro_flags & feat->compat_ro_flag ||
|
|
|
|
features->incompat_flags & feat->incompat_flag ||
|
|
|
|
features->runtime_flags & feat->runtime_flag) {
|
2017-01-26 12:27:58 +00:00
|
|
|
if (*buf)
|
|
|
|
strcat(buf, ", ");
|
2018-05-08 06:31:52 +00:00
|
|
|
strcat(buf, feat->name);
|
2017-01-26 12:27:58 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
void btrfs_parse_fs_features_to_string(char *buf,
|
|
|
|
const struct btrfs_mkfs_features *features)
|
2018-05-08 06:31:52 +00:00
|
|
|
{
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
parse_features_to_string(buf, features, FS_FEATURES);
|
2018-05-08 06:31:52 +00:00
|
|
|
}
|
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
void btrfs_parse_runtime_features_to_string(char *buf,
|
|
|
|
const struct btrfs_mkfs_features *features)
|
2018-05-08 06:31:52 +00:00
|
|
|
{
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
parse_features_to_string(buf, features, RUNTIME_FEATURES);
|
2018-05-08 06:31:52 +00:00
|
|
|
}
|
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
static void process_features(struct btrfs_mkfs_features *features,
|
|
|
|
enum feature_source source)
|
2017-01-26 12:27:58 +00:00
|
|
|
{
|
2018-05-08 06:31:52 +00:00
|
|
|
const int array_size = get_feature_array_size(source);
|
2017-01-26 12:27:58 +00:00
|
|
|
int i;
|
|
|
|
|
2018-05-08 06:31:52 +00:00
|
|
|
for (i = 0; i < array_size; i++) {
|
|
|
|
const struct btrfs_feature *feat = get_feature(i, source);
|
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
if ((features->compat_ro_flags & feat->compat_ro_flag ||
|
|
|
|
features->incompat_flags & feat->incompat_flag ||
|
|
|
|
features->runtime_flags & feat->runtime_flag) &&
|
|
|
|
feat->name && feat->desc) {
|
2017-01-26 12:27:58 +00:00
|
|
|
printf("Turning ON incompat feature '%s': %s\n",
|
2018-05-08 06:31:52 +00:00
|
|
|
feat->name, feat->desc);
|
2017-01-26 12:27:58 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
void btrfs_process_fs_features(struct btrfs_mkfs_features *features)
|
2018-05-08 06:31:52 +00:00
|
|
|
{
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
process_features(features, FS_FEATURES);
|
2018-05-08 06:31:52 +00:00
|
|
|
}
|
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
void btrfs_process_runtime_features(struct btrfs_mkfs_features *features)
|
2017-01-26 12:27:58 +00:00
|
|
|
{
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
process_features(features, RUNTIME_FEATURES);
|
2018-05-08 06:31:52 +00:00
|
|
|
}
|
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
static void list_all_features(const struct btrfs_mkfs_features *allowed,
|
|
|
|
enum feature_source source)
|
2018-05-08 06:31:52 +00:00
|
|
|
{
|
|
|
|
const int array_size = get_feature_array_size(source);
|
2017-01-26 12:27:58 +00:00
|
|
|
int i;
|
2018-05-08 06:31:52 +00:00
|
|
|
char *prefix;
|
2017-01-26 12:27:58 +00:00
|
|
|
|
2018-05-08 06:31:52 +00:00
|
|
|
if (source == FS_FEATURES)
|
|
|
|
prefix = "Filesystem";
|
|
|
|
else if (source == RUNTIME_FEATURES)
|
|
|
|
prefix = "Runtime";
|
|
|
|
else
|
|
|
|
prefix = "UNKNOWN";
|
|
|
|
|
|
|
|
fprintf(stderr, "%s features available:\n", prefix);
|
|
|
|
for (i = 0; i < array_size - 1; i++) {
|
|
|
|
const struct btrfs_feature *feat = get_feature(i, source);
|
2022-09-10 19:18:01 +00:00
|
|
|
const char *sep = "";
|
2017-01-26 12:27:58 +00:00
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
/* The feature is not in the allowed one, skip it. */
|
|
|
|
if (allowed &&
|
|
|
|
!(feat->compat_ro_flag & allowed->compat_ro_flags ||
|
|
|
|
feat->incompat_flag & allowed->incompat_flags ||
|
|
|
|
feat->runtime_flag & allowed->runtime_flags))
|
2017-01-26 12:27:58 +00:00
|
|
|
continue;
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
|
2022-09-10 19:18:01 +00:00
|
|
|
fprintf(stderr, "%-20s- %s (", feat->name, feat->desc);
|
|
|
|
if (feat->compat_ver) {
|
|
|
|
fprintf(stderr, "compat=%s", feat->compat_str);
|
|
|
|
sep = ", ";
|
|
|
|
}
|
|
|
|
if (feat->safe_ver) {
|
|
|
|
fprintf(stderr, "%ssafe=%s", sep, feat->safe_str);
|
|
|
|
sep = ", ";
|
|
|
|
}
|
2017-01-26 12:27:58 +00:00
|
|
|
if (feat->default_ver)
|
2022-09-10 19:18:01 +00:00
|
|
|
fprintf(stderr, "%sdefault=%s", sep, feat->default_str);
|
2017-01-26 12:27:58 +00:00
|
|
|
fprintf(stderr, ")\n");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
/* @allowed can be null, then all features will be listed. */
|
|
|
|
void btrfs_list_all_fs_features(const struct btrfs_mkfs_features *allowed)
|
2018-05-08 06:31:52 +00:00
|
|
|
{
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
list_all_features(allowed, FS_FEATURES);
|
2018-05-08 06:31:52 +00:00
|
|
|
}
|
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
/* @allowed can be null, then all runtime features will be listed. */
|
|
|
|
void btrfs_list_all_runtime_features(const struct btrfs_mkfs_features *allowed)
|
2018-05-08 06:31:52 +00:00
|
|
|
{
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
list_all_features(allowed, RUNTIME_FEATURES);
|
2018-05-08 06:31:52 +00:00
|
|
|
}
|
|
|
|
|
2017-01-26 12:27:58 +00:00
|
|
|
/*
|
|
|
|
* Return NULL if all features were parsed fine, otherwise return the name of
|
|
|
|
* the first unparsed.
|
|
|
|
*/
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
static char *parse_features(char *namelist,
|
|
|
|
struct btrfs_mkfs_features *features,
|
2018-05-08 06:31:52 +00:00
|
|
|
enum feature_source source)
|
2017-01-26 12:27:58 +00:00
|
|
|
{
|
|
|
|
char *this_char;
|
|
|
|
char *save_ptr = NULL; /* Satisfy static checkers */
|
|
|
|
|
|
|
|
for (this_char = strtok_r(namelist, ",", &save_ptr);
|
|
|
|
this_char != NULL;
|
|
|
|
this_char = strtok_r(NULL, ",", &save_ptr)) {
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
if (parse_one_fs_feature(this_char, features, source))
|
2017-01-26 12:27:58 +00:00
|
|
|
return this_char;
|
|
|
|
}
|
|
|
|
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
char *btrfs_parse_fs_features(char *namelist,
|
|
|
|
struct btrfs_mkfs_features *features)
|
2018-05-08 06:31:52 +00:00
|
|
|
{
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
return parse_features(namelist, features, FS_FEATURES);
|
2018-05-08 06:31:52 +00:00
|
|
|
}
|
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
char *btrfs_parse_runtime_features(char *namelist,
|
|
|
|
struct btrfs_mkfs_features *features)
|
2018-05-08 06:31:52 +00:00
|
|
|
{
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
return parse_features(namelist, features, RUNTIME_FEATURES);
|
2018-05-08 06:31:52 +00:00
|
|
|
}
|
|
|
|
|
2017-01-26 12:27:58 +00:00
|
|
|
void print_kernel_version(FILE *stream, u32 version)
|
|
|
|
{
|
|
|
|
u32 v[3];
|
|
|
|
|
|
|
|
v[0] = version & 0xFF;
|
|
|
|
v[1] = (version >> 8) & 0xFF;
|
|
|
|
v[2] = version >> 16;
|
|
|
|
fprintf(stream, "%u.%u", v[2], v[1]);
|
|
|
|
if (v[0])
|
|
|
|
fprintf(stream, ".%u", v[0]);
|
|
|
|
}
|
|
|
|
|
|
|
|
u32 get_running_kernel_version(void)
|
|
|
|
{
|
|
|
|
struct utsname utsbuf;
|
|
|
|
char *tmp;
|
|
|
|
char *saveptr = NULL;
|
|
|
|
u32 version;
|
|
|
|
|
|
|
|
uname(&utsbuf);
|
|
|
|
if (strcmp(utsbuf.sysname, "Linux") != 0) {
|
|
|
|
error("unsupported system: %s", utsbuf.sysname);
|
|
|
|
exit(1);
|
|
|
|
}
|
|
|
|
/* 1.2.3-4-name */
|
|
|
|
tmp = strchr(utsbuf.release, '-');
|
|
|
|
if (tmp)
|
|
|
|
*tmp = 0;
|
|
|
|
|
|
|
|
tmp = strtok_r(utsbuf.release, ".", &saveptr);
|
|
|
|
if (!string_is_numerical(tmp))
|
|
|
|
return (u32)-1;
|
|
|
|
version = atoi(tmp) << 16;
|
|
|
|
tmp = strtok_r(NULL, ".", &saveptr);
|
|
|
|
if (!string_is_numerical(tmp))
|
|
|
|
return (u32)-1;
|
|
|
|
version |= atoi(tmp) << 8;
|
|
|
|
tmp = strtok_r(NULL, ".", &saveptr);
|
2019-02-25 18:16:43 +00:00
|
|
|
/* Relaxed format accepts eg. 1.2.3+ */
|
|
|
|
if (tmp && string_is_numerical(tmp))
|
2017-01-26 12:27:58 +00:00
|
|
|
version |= atoi(tmp);
|
|
|
|
|
|
|
|
return version;
|
|
|
|
}
|
2021-04-15 05:30:11 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* The buffer size is strlen of "4096 8192 16384 32768 65536", which is 28,
|
|
|
|
* then round up to 32.
|
|
|
|
*/
|
|
|
|
#define SUPPORTED_SECTORSIZE_BUF_SIZE 32
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Check if current kernel supports the given size
|
|
|
|
*/
|
|
|
|
static bool check_supported_sectorsize(u32 sectorsize)
|
|
|
|
{
|
|
|
|
char supported_buf[SUPPORTED_SECTORSIZE_BUF_SIZE] = { 0 };
|
|
|
|
char sectorsize_buf[SUPPORTED_SECTORSIZE_BUF_SIZE] = { 0 };
|
|
|
|
char *this_char;
|
|
|
|
char *save_ptr = NULL;
|
|
|
|
int fd;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
fd = sysfs_open_file("features/supported_sectorsizes");
|
|
|
|
if (fd < 0)
|
|
|
|
return false;
|
|
|
|
ret = sysfs_read_file(fd, supported_buf, SUPPORTED_SECTORSIZE_BUF_SIZE);
|
|
|
|
close(fd);
|
|
|
|
if (ret < 0)
|
|
|
|
return false;
|
|
|
|
snprintf(sectorsize_buf, SUPPORTED_SECTORSIZE_BUF_SIZE, "%u", sectorsize);
|
|
|
|
|
|
|
|
for (this_char = strtok_r(supported_buf, " ", &save_ptr);
|
|
|
|
this_char != NULL;
|
|
|
|
this_char = strtok_r(NULL, " ", &save_ptr)) {
|
|
|
|
/*
|
|
|
|
* Also check the terminal '\0' to handle cases like
|
|
|
|
* "4096" and "40960".
|
|
|
|
*/
|
|
|
|
if (!strncmp(this_char, sectorsize_buf, strlen(sectorsize_buf) + 1))
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2020-10-13 01:06:02 +00:00
|
|
|
int btrfs_check_sectorsize(u32 sectorsize)
|
|
|
|
{
|
2021-04-15 05:30:11 +00:00
|
|
|
bool sectorsize_checked = false;
|
2020-10-13 01:06:02 +00:00
|
|
|
u32 page_size = (u32)sysconf(_SC_PAGESIZE);
|
|
|
|
|
|
|
|
if (!is_power_of_2(sectorsize)) {
|
|
|
|
error("invalid sectorsize %u, must be power of 2", sectorsize);
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
if (sectorsize < SZ_4K || sectorsize > SZ_64K) {
|
|
|
|
error("invalid sectorsize %u, expected range is [4K, 64K]",
|
|
|
|
sectorsize);
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
2021-04-15 05:30:11 +00:00
|
|
|
if (page_size == sectorsize)
|
|
|
|
sectorsize_checked = true;
|
|
|
|
else
|
|
|
|
sectorsize_checked = check_supported_sectorsize(sectorsize);
|
|
|
|
|
|
|
|
if (!sectorsize_checked)
|
2020-10-13 01:06:02 +00:00
|
|
|
warning(
|
|
|
|
"the filesystem may not be mountable, sectorsize %u doesn't match page size %u",
|
|
|
|
sectorsize, page_size);
|
|
|
|
return 0;
|
|
|
|
}
|
2017-01-26 12:27:58 +00:00
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
int btrfs_check_nodesize(u32 nodesize, u32 sectorsize,
|
|
|
|
struct btrfs_mkfs_features *features)
|
2018-12-05 06:40:15 +00:00
|
|
|
{
|
|
|
|
if (nodesize < sectorsize) {
|
|
|
|
error("illegal nodesize %u (smaller than %u)",
|
|
|
|
nodesize, sectorsize);
|
|
|
|
return -1;
|
|
|
|
} else if (nodesize > BTRFS_MAX_METADATA_BLOCKSIZE) {
|
|
|
|
error("illegal nodesize %u (larger than %u)",
|
|
|
|
nodesize, BTRFS_MAX_METADATA_BLOCKSIZE);
|
|
|
|
return -1;
|
|
|
|
} else if (nodesize & (sectorsize - 1)) {
|
|
|
|
error("illegal nodesize %u (not aligned to %u)",
|
|
|
|
nodesize, sectorsize);
|
|
|
|
return -1;
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
} else if (features->incompat_flags &
|
|
|
|
BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS &&
|
2018-12-05 06:40:15 +00:00
|
|
|
nodesize != sectorsize) {
|
|
|
|
error(
|
|
|
|
"illegal nodesize %u (not equal to %u for mixed block group)",
|
|
|
|
nodesize, sectorsize);
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
2021-09-07 14:20:17 +00:00
|
|
|
|
btrfs-progs: fsfeatures: properly merge -O and -R options
[BUG]
Commit "btrfs-progs: prepare merging compat feature lists" tries to
merged "-O" and "-R" options, as they don't correctly represents
btrfs features.
But that commit caused the following bug during mkfs for experimental
build:
$ mkfs.btrfs -f -O block-group-tree /dev/nvme0n1
btrfs-progs v5.19.1
See http://btrfs.wiki.kernel.org for more information.
ERROR: superblock magic doesn't match
ERROR: illegal nodesize 16384 (not equal to 4096 for mixed block group)
[CAUSE]
Currently btrfs_parse_fs_features() will return a u64, and reuse the
same u64 for both incompat and compat RO flags for experimental branch.
This can easily leads to conflicts, as
BTRFS_FEATURE_INCOMPAT_MIXED_BLOCK_GROUP and
BTRFS_FEATURE_COMPAT_RO_BLOCK_GROUP_TREE both share the same bit
(1 << 2).
Thus for above case, mkfs.btrfs believe it has set MIXED_BLOCK_GROUP
feature, but what we really want is BLOCK_GROUP_TREE.
[FIX]
Instead of incorrectly re-using the same bits in btrfs_feature, split
the old flags into 3 flags:
- incompat_flag
- compat_ro_flag
- runtime_flag
The first two flags are easy to understand, the corresponding flag of
each feature.
The last runtime_flag is to compensate features which doesn't have any
on-disk flag set, like QUOTA and LIST_ALL.
And since we're no longer using a single u64 as features, we have to
introduce a new structure, btrfs_mkfs_features, to contain above 3
flags.
This also mean, things like default mkfs features must be converted to
use the new structure, thus those old macros are all converted to
const static structures:
- BTRFS_MKFS_DEFAULT_FEATURES + BTRFS_MKFS_DEFAULT_RUNTIME_FEATURES
-> btrfs_mkfs_default_features
- BTRFS_CONVERT_ALLOWED_FEATURES -> btrfs_convert_allowed_features
And since we're using a structure, it's not longer as easy to implement
a disallowed mask.
Thus functions with @mask_disallowed are all changed to using
an @allowed structure pointer (which can be NULL).
Finally if we have experimental features enabled, all features can be
specified by -O options, and we can output a unified feature list,
instead of the old split ones.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-10-05 01:48:07 +00:00
|
|
|
int btrfs_check_features(const struct btrfs_mkfs_features *features,
|
|
|
|
const struct btrfs_mkfs_features *allowed)
|
|
|
|
{
|
|
|
|
if (features->compat_ro_flags & ~allowed->compat_ro_flags ||
|
|
|
|
features->incompat_flags & ~allowed->incompat_flags ||
|
|
|
|
features->runtime_flags & ~allowed->runtime_flags)
|
|
|
|
return -EINVAL;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2021-09-07 14:20:17 +00:00
|
|
|
/*
|
|
|
|
* Check if the BTRFS_IOC_TREE_SEARCH_V2 ioctl is supported on a given
|
|
|
|
* filesystem, opened at fd
|
|
|
|
*/
|
|
|
|
int btrfs_tree_search2_ioctl_supported(int fd)
|
|
|
|
{
|
|
|
|
struct btrfs_ioctl_search_args_v2 *args2;
|
|
|
|
struct btrfs_ioctl_search_key *sk;
|
|
|
|
int args2_size = 1024;
|
|
|
|
char args2_buf[args2_size];
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
args2 = (struct btrfs_ioctl_search_args_v2 *)args2_buf;
|
|
|
|
sk = &(args2->key);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Search for the extent tree item in the root tree.
|
|
|
|
*/
|
|
|
|
sk->tree_id = BTRFS_ROOT_TREE_OBJECTID;
|
|
|
|
sk->min_objectid = BTRFS_EXTENT_TREE_OBJECTID;
|
|
|
|
sk->max_objectid = BTRFS_EXTENT_TREE_OBJECTID;
|
|
|
|
sk->min_type = BTRFS_ROOT_ITEM_KEY;
|
|
|
|
sk->max_type = BTRFS_ROOT_ITEM_KEY;
|
|
|
|
sk->min_offset = 0;
|
|
|
|
sk->max_offset = (u64)-1;
|
|
|
|
sk->min_transid = 0;
|
|
|
|
sk->max_transid = (u64)-1;
|
|
|
|
sk->nr_items = 1;
|
|
|
|
args2->buf_size = args2_size - sizeof(struct btrfs_ioctl_search_args_v2);
|
|
|
|
ret = ioctl(fd, BTRFS_IOC_TREE_SEARCH_V2, args2);
|
|
|
|
if (ret == -EOPNOTSUPP)
|
|
|
|
return 0;
|
|
|
|
else if (ret == 0)
|
|
|
|
return 1;
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|