[PATCH RFC v2 4/6] ext4: support project id and project quota
From: Konstantin Khlebnikov
Date: Tue Mar 10 2015 - 13:25:25 EST
This patch implements project quota feature.
If EXT4_FEATURE_RO_COMPAT_PROJECT is set then inode field i_faddr contains
project id (otherwise all project id are set to invalid: -1).
I_faddr is obsolete for a long time, it seems linux never used it.
New files always inherit project id from directory where they are created.
Link/rename works only if projects of inode and target directory are matched,
or if project id of directory is mapped to zero. Otherwise -EXDEV is returned.
User-space tool 'mv' handles EXDEV as cross-filesystem move: it makes a copy of
files and unlinks source, this files will be transferred into target project.
(next patch adds optimization for that)
Signed-off-by: Konstantin Khlebnikov <khlebnikov@xxxxxxxxxxxxxx>
---
fs/ext4/ext4.h | 15 ++++++++++----
fs/ext4/ialloc.c | 3 +++
fs/ext4/inode.c | 15 ++++++++++++++
fs/ext4/namei.c | 25 +++++++++++++++++++++++
fs/ext4/super.c | 59 ++++++++++++++++++++++++++++++++++++++++++++++++++----
5 files changed, 109 insertions(+), 8 deletions(-)
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index f63c3d5805c4..8ba523ff5f84 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -212,6 +212,8 @@ struct ext4_io_submit {
/* First non-reserved inode for old ext4 filesystems */
#define EXT4_GOOD_OLD_FIRST_INO 11
+#define EXT4_PRJ_QUOTA_INO 11 /* Project quota inode */
+
/*
* Maximal count of links to a file
*/
@@ -653,7 +655,7 @@ struct ext4_inode {
__le32 i_generation; /* File version (for NFS) */
__le32 i_file_acl_lo; /* File ACL */
__le32 i_size_high;
- __le32 i_obso_faddr; /* Obsoleted fragment address */
+ __le32 i_project; /* Formerly i_faddr, fragment address */
union {
struct {
__le16 l_i_blocks_high; /* were l_i_reserved1 */
@@ -939,6 +941,7 @@ struct ext4_inode_info {
/* Precomputed uuid+inum+igen checksum for seeding inode checksums */
__u32 i_csum_seed;
+ kprojid_t i_project;
};
/*
@@ -1169,7 +1172,8 @@ struct ext4_super_block {
__le32 s_overhead_clusters; /* overhead blocks/clusters in fs */
__le32 s_backup_bgs[2]; /* groups with sparse_super2 SBs */
__u8 s_encrypt_algos[4]; /* Encryption algorithms in use */
- __le32 s_reserved[105]; /* Padding to the end of the block */
+ __le32 s_prj_quota_inum; /* inode for tracking project quota */
+ __le32 s_reserved[104]; /* Padding to the end of the block */
__le32 s_checksum; /* crc32c(superblock) */
};
@@ -1184,7 +1188,7 @@ struct ext4_super_block {
#define EXT4_MF_FS_ABORTED 0x0002 /* Fatal error detected */
/* Number of quota types we support */
-#define EXT4_MAXQUOTAS 2
+#define EXT4_MAXQUOTAS 3
/*
* fourth extended-fs super-block data in memory
@@ -1373,6 +1377,7 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino)
return ino == EXT4_ROOT_INO ||
ino == EXT4_USR_QUOTA_INO ||
ino == EXT4_GRP_QUOTA_INO ||
+ ino == EXT4_PRJ_QUOTA_INO ||
ino == EXT4_BOOT_LOADER_INO ||
ino == EXT4_JOURNAL_INO ||
ino == EXT4_RESIZE_INO ||
@@ -1536,6 +1541,7 @@ static inline void ext4_clear_state_flags(struct ext4_inode_info *ei)
*/
#define EXT4_FEATURE_RO_COMPAT_METADATA_CSUM 0x0400
#define EXT4_FEATURE_RO_COMPAT_READONLY 0x1000
+#define EXT4_FEATURE_RO_COMPAT_PROJECT 0x2000
#define EXT4_FEATURE_INCOMPAT_COMPRESSION 0x0001
#define EXT4_FEATURE_INCOMPAT_FILETYPE 0x0002
@@ -1586,7 +1592,8 @@ static inline void ext4_clear_state_flags(struct ext4_inode_info *ei)
EXT4_FEATURE_RO_COMPAT_HUGE_FILE |\
EXT4_FEATURE_RO_COMPAT_BIGALLOC |\
EXT4_FEATURE_RO_COMPAT_METADATA_CSUM|\
- EXT4_FEATURE_RO_COMPAT_QUOTA)
+ EXT4_FEATURE_RO_COMPAT_QUOTA |\
+ EXT4_FEATURE_RO_COMPAT_PROJECT)
/*
* Default values for user and/or group using reserved blocks
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index ac644c31ca67..5d812caf4bef 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -756,6 +756,9 @@ struct inode *__ext4_new_inode(handle_t *handle, struct inode *dir,
inode->i_gid = dir->i_gid;
} else
inode_init_owner(inode, dir, mode);
+
+ ei->i_project = EXT4_I(dir)->i_project;
+
dquot_initialize(inode);
if (!goal)
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 5cb9a212b86f..fc55590437af 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -3968,6 +3968,14 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
}
i_uid_write(inode, i_uid);
i_gid_write(inode, i_gid);
+
+ if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_PROJECT)) {
+ projid_t project = le32_to_cpu(raw_inode->i_project);
+
+ ei->i_project = make_kprojid(&init_user_ns, project);
+ } else
+ ei->i_project = INVALID_PROJID;
+
set_nlink(inode, le16_to_cpu(raw_inode->i_links_count));
ext4_clear_state_flags(ei); /* Only relevant on 32-bit archs */
@@ -4293,6 +4301,13 @@ static int ext4_do_update_inode(handle_t *handle,
raw_inode->i_uid_high = 0;
raw_inode->i_gid_high = 0;
}
+
+ if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_PROJECT)) {
+ projid_t i_project = from_kprojid(&init_user_ns, ei->i_project);
+
+ raw_inode->i_project = cpu_to_le32(i_project);
+ }
+
raw_inode->i_links_count = cpu_to_le16(inode->i_nlink);
EXT4_INODE_SET_XTIME(i_ctime, inode, raw_inode);
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index 8110dd20ad3f..094f7096a41c 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -2934,6 +2934,17 @@ err_drop_inode:
return err;
}
+static bool ext4_check_project(struct inode *dir, struct inode *inode)
+{
+ /*
+ * Allow link/rename only into directory with matched project or
+ * if direcorty belongs to default 'zero' project. Without feature
+ * RO_COMPAT_PROJECT all i_project are 'invalid' thus always equal.
+ */
+ return projid_eq(EXT4_I(dir)->i_project, EXT4_I(inode)->i_project) ||
+ from_kprojid(current_user_ns(), EXT4_I(dir)->i_project) == 0;
+}
+
static int ext4_link(struct dentry *old_dentry,
struct inode *dir, struct dentry *dentry)
{
@@ -2944,6 +2955,9 @@ static int ext4_link(struct dentry *old_dentry,
if (inode->i_nlink >= EXT4_LINK_MAX)
return -EMLINK;
+ if (!ext4_check_project(dir, inode))
+ return -EXDEV;
+
dquot_initialize(dir);
retry:
@@ -3260,6 +3274,11 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry,
if (new.inode && !test_opt(new.dir->i_sb, NO_AUTO_DA_ALLOC))
ext4_alloc_da_blocks(old.inode);
+ if (!ext4_check_project(new.dir, old.inode)) {
+ retval = -EXDEV;
+ goto end_rename;
+ }
+
credits = (2 * EXT4_DATA_TRANS_BLOCKS(old.dir->i_sb) +
EXT4_INDEX_EXTRA_TRANS_BLOCKS + 2);
if (!(flags & RENAME_WHITEOUT)) {
@@ -3436,6 +3455,12 @@ static int ext4_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
if (!new.bh || le32_to_cpu(new.de->inode) != new.inode->i_ino)
goto end_rename;
+ if (!ext4_check_project(new.dir, old.inode) ||
+ !ext4_check_project(old.dir, new.inode)) {
+ retval = -EXDEV;
+ goto end_rename;
+ }
+
handle = ext4_journal_start(old.dir, EXT4_HT_DIR,
(2 * EXT4_DATA_TRANS_BLOCKS(old.dir->i_sb) +
2 * EXT4_INDEX_EXTRA_TRANS_BLOCKS + 2));
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index e061e66c8280..c62ed5b554ae 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1035,9 +1035,54 @@ static int bdev_try_to_free_page(struct super_block *sb, struct page *page,
return try_to_free_buffers(page);
}
+static kprojid_t ext4_get_project(struct inode *inode)
+{
+ /* Without feature RO_COMPAT_PROJECT all i_project are set to invalid */
+ return EXT4_I(inode)->i_project;
+}
+
+static int ext4_set_project(struct inode *inode, kprojid_t project)
+{
+ struct super_block *sb = inode->i_sb;
+ struct ext4_iloc iloc;
+ handle_t *handle;
+ int ret;
+
+ if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_PROJECT))
+ return -EOPNOTSUPP;
+
+ if (projid_eq(EXT4_I(inode)->i_project, project))
+ return 0;
+
+ dquot_initialize(inode);
+
+ handle = ext4_journal_start(inode, EXT4_HT_INODE, 1 +
+ EXT4_QUOTA_INIT_BLOCKS(sb) +
+ EXT4_QUOTA_DEL_BLOCKS(sb));
+ if (IS_ERR(handle))
+ return PTR_ERR(handle);
+
+ ret = dquot_transfer_project(inode, project);
+ if (ret)
+ goto out;
+
+ ret = ext4_reserve_inode_write(handle, inode, &iloc);
+ if (!ret) {
+ inode->i_ctime = ext4_current_time(inode);
+ EXT4_I(inode)->i_project = project;
+ ret = ext4_mark_iloc_dirty(handle, inode, &iloc);
+ }
+ if (IS_SYNC(inode))
+ ext4_handle_sync(handle);
+out:
+ ext4_journal_stop(handle);
+
+ return ret;
+}
+
#ifdef CONFIG_QUOTA
-#define QTYPE2NAME(t) ((t) == USRQUOTA ? "user" : "group")
-#define QTYPE2MOPT(on, t) ((t) == USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA))
+static char *quotatypes[] = INITQFNAMES;
+#define QTYPE2NAME(t) (quotatypes[t])
static int ext4_write_dquot(struct dquot *dquot);
static int ext4_acquire_dquot(struct dquot *dquot);
@@ -1103,6 +1148,8 @@ static const struct super_operations ext4_sops = {
.get_dquots = ext4_get_dquots,
#endif
.bdev_try_to_free_page = bdev_try_to_free_page,
+ .get_project = ext4_get_project,
+ .set_project = ext4_set_project,
};
static const struct export_operations ext4_export_ops = {
@@ -3965,6 +4012,8 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
else
sb->s_qcop = &ext4_qctl_operations;
sb->s_quota_types = QTYPE_MASK_USR | QTYPE_MASK_GRP;
+ if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_PROJECT))
+ sb->s_quota_types |= QTYPE_MASK_PRJ;
#endif
memcpy(sb->s_uuid, es->s_uuid, sizeof(es->s_uuid));
@@ -5269,7 +5318,8 @@ static int ext4_quota_enable(struct super_block *sb, int type, int format_id,
struct inode *qf_inode;
unsigned long qf_inums[EXT4_MAXQUOTAS] = {
le32_to_cpu(EXT4_SB(sb)->s_es->s_usr_quota_inum),
- le32_to_cpu(EXT4_SB(sb)->s_es->s_grp_quota_inum)
+ le32_to_cpu(EXT4_SB(sb)->s_es->s_grp_quota_inum),
+ le32_to_cpu(EXT4_SB(sb)->s_es->s_prj_quota_inum),
};
BUG_ON(!EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA));
@@ -5297,7 +5347,8 @@ static int ext4_enable_quotas(struct super_block *sb)
int type, err = 0;
unsigned long qf_inums[EXT4_MAXQUOTAS] = {
le32_to_cpu(EXT4_SB(sb)->s_es->s_usr_quota_inum),
- le32_to_cpu(EXT4_SB(sb)->s_es->s_grp_quota_inum)
+ le32_to_cpu(EXT4_SB(sb)->s_es->s_grp_quota_inum),
+ le32_to_cpu(EXT4_SB(sb)->s_es->s_prj_quota_inum),
};
sb_dqopt(sb)->flags |= DQUOT_QUOTA_SYS_FILE;
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/