+This patch implements the large EA support in ext4. If the size of
+an EA value is larger than the blocksize, then the EA value would
+not be saved in the external EA block, instead it would be saved
+in an external EA inode. So, the patch also helps support a larger
+number of EAs.
+
Index: linux-stage/fs/ext4/ext4.h
===================================================================
--- linux-stage.orig/fs/ext4/ext4.h
+++ linux-stage/fs/ext4/ext4.h
-@@ -1412,6 +1412,7 @@ static inline void ext4_clear_state_flag
+@@ -1408,6 +1408,7 @@ static inline void ext4_clear_state_flag
EXT4_FEATURE_INCOMPAT_EXTENTS| \
EXT4_FEATURE_INCOMPAT_64BIT| \
EXT4_FEATURE_INCOMPAT_FLEX_BG| \
EXT4_FEATURE_INCOMPAT_MMP| \
EXT4_FEATURE_INCOMPAT_DIRDATA)
-@@ -1775,6 +1776,12 @@ struct mmpd_data {
- #endif
+@@ -1764,6 +1765,12 @@ struct mmpd_data {
+ #define EXT4_MMP_MAX_CHECK_INTERVAL 300UL
/*
+ * Maximum size of xattr attributes for FEATURE_INCOMPAT_EA_INODE 1Mb
* Function prototypes
*/
-@@ -1786,6 +1793,10 @@ struct mmpd_data {
+@@ -1775,6 +1782,10 @@ struct mmpd_data {
# define ATTRIB_NORET __attribute__((noreturn))
# define NORET_AND noreturn,
/* bitmap.c */
extern unsigned int ext4_count_free(char *bitmap, unsigned numchars);
-@@ -1905,6 +1916,7 @@ extern void ext4_set_inode_flags(struct
+@@ -1893,6 +1904,7 @@ extern void ext4_set_inode_flags(struct
extern void ext4_get_inode_flags(struct ext4_inode_info *);
extern int ext4_alloc_da_blocks(struct inode *inode);
extern void ext4_set_aops(struct inode *inode);
trace_ext4_evict_inode(inode);
-@@ -197,6 +199,9 @@ void ext4_evict_inode(struct inode *inod
- goto no_delete;
- }
-
-+ if (!IS_NOQUOTA(inode))
-+ extra_credits += 2 * EXT4_QUOTA_DEL_BLOCKS(inode->i_sb);
-+
- if (!is_bad_inode(inode))
- dquot_initialize(inode);
-
-@@ -207,7 +212,10 @@ void ext4_evict_inode(struct inode *inod
+@@ -207,7 +209,10 @@ void ext4_evict_inode(struct inode *inod
if (is_bad_inode(inode))
goto no_delete;
if (IS_ERR(handle)) {
ext4_std_error(inode->i_sb, PTR_ERR(handle));
/*
-@@ -218,9 +226,30 @@ void ext4_evict_inode(struct inode *inod
+@@ -218,9 +223,33 @@ void ext4_evict_inode(struct inode *inod
ext4_orphan_del(NULL, inode);
goto no_delete;
}
+ goto stop_handle;
+ }
+
++ if (!IS_NOQUOTA(inode))
++ extra_credits += 2 * EXT4_QUOTA_DEL_BLOCKS(inode->i_sb);
++
+ if (!ext4_handle_has_enough_credits(handle,
+ blocks_for_truncate(inode) + extra_credits)) {
+ err = ext4_journal_extend(handle,
return;
no_delete:
ext4_clear_inode(inode); /* We must guarantee clearing of inode... */
-@@ -5551,7 +5584,7 @@ static int ext4_index_trans_blocks(struc
+@@ -5552,7 +5585,7 @@ static int ext4_index_trans_blocks(struc
*
* Also account for superblock, inode, quota and xattr blocks
*/
size_t value_size = le32_to_cpu(entry->e_value_size);
- if (entry->e_value_block != 0 || value_size > size ||
-- le16_to_cpu(entry->e_value_offs) + value_size > size)
-+ if ((entry->e_value_inum == 0) &&
-+ (le16_to_cpu(entry->e_value_offs) + value_size > size))
++ if (!entry->e_value_inum &&
+ le16_to_cpu(entry->e_value_offs) + value_size > size)
+ return -EIO;
-+ if (entry->e_value_inum != 0 &&
++ if (entry->e_value_inum &&
+ (le32_to_cpu(entry->e_value_inum) < EXT4_FIRST_INO(inode->i_sb) ||
+ le32_to_cpu(entry->e_value_inum) >
+ le32_to_cpu(EXT4_SB(inode->i_sb)->s_es->s_inodes_count)))
{
struct ext4_xattr_entry *entry;
size_t name_len;
-@@ -200,11 +207,103 @@ ext4_xattr_find_entry(struct ext4_xattr_
+@@ -200,11 +207,104 @@ ext4_xattr_find_entry(struct ext4_xattr_
break;
}
*pentry = entry;
+ return err;
+}
+
-+struct inode *ext4_xattr_inode_iget(struct inode *parent, int ea_ino, int *err)
++struct inode *ext4_xattr_inode_iget(struct inode *parent, unsigned long ea_ino, int *err)
+{
+ struct inode *ea_inode = NULL;
+
+ ea_inode = ext4_iget(parent->i_sb, ea_ino);
+ if (IS_ERR(ea_inode) || is_bad_inode(ea_inode)) {
-+ ext4_error(parent->i_sb, "error while reading EA inode %d",
-+ ea_ino);
-+ *err = -EIO;
++ int rc = IS_ERR(ea_inode) ? PTR_ERR(ea_inode) : 0;
++ ext4_error(parent->i_sb, "error while reading EA inode %lu "
++ "/ %d %d", ea_ino, rc, is_bad_inode(ea_inode));
++ *err = rc != 0 ? rc : -EIO;
+ return NULL;
+ }
+
-+ if (ea_inode->i_xattr_inode_parent != parent->i_ino ||
++ if (EXT4_XATTR_INODE_GET_PARENT(ea_inode) != parent->i_ino ||
+ ea_inode->i_generation != parent->i_generation) {
-+ ext4_error(parent->i_sb, "Backpointer from EA inode %d "
++ ext4_error(parent->i_sb, "Backpointer from EA inode %lu "
+ "to parent invalid.", ea_ino);
+ *err = -EINVAL;
+ goto error;
+ }
+
+ if (!(EXT4_I(ea_inode)->i_flags & EXT4_EA_INODE_FL)) {
-+ ext4_error(parent->i_sb, "EA inode %d does not have "
++ ext4_error(parent->i_sb, "EA inode %lu does not have "
+ "EXT4_EA_INODE_FL flag set.\n", ea_ino);
+ *err = -EINVAL;
+ goto error;
+ * Read the value from the EA inode.
+ */
+static int
-+ext4_xattr_inode_get(struct inode *inode, int ea_ino, void *buffer,
++ext4_xattr_inode_get(struct inode *inode, unsigned long ea_ino, void *buffer,
+ size_t *size)
+{
+ struct inode *ea_inode = NULL;
goto cleanup;
- memcpy(buffer, bh->b_data + le16_to_cpu(entry->e_value_offs),
- size);
-+ if (entry->e_value_inum != 0) {
++ if (entry->e_value_inum) {
+ error = ext4_xattr_inode_get(inode,
+ le32_to_cpu(entry->e_value_inum),
+ buffer, &size);
goto cleanup;
- memcpy(buffer, (void *)IFIRST(header) +
- le16_to_cpu(entry->e_value_offs), size);
-+ if (entry->e_value_inum != 0) {
++ if (entry->e_value_inum) {
+ error = ext4_xattr_inode_get(inode,
+ le32_to_cpu(entry->e_value_inum),
+ buffer, &size);
for (; !IS_LAST_ENTRY(last); last = EXT4_XATTR_NEXT(last)) {
*total += EXT4_XATTR_LEN(last->e_name_len);
- if (!last->e_value_block && last->e_value_size) {
-+ if (last->e_value_inum == 0 && last->e_value_size > 0) {
++ if (!last->e_value_inum && last->e_value_size) {
size_t offs = le16_to_cpu(last->e_value_offs);
if (offs < *min_offs)
*min_offs = offs;
+ * A back-pointer from EA inode to parent inode will be useful
+ * for e2fsck.
+ */
-+ ea_inode->i_xattr_inode_parent = inode->i_ino;
++ EXT4_XATTR_INODE_SET_PARENT(ea_inode, inode->i_ino);
+ unlock_new_inode(ea_inode);
+ }
+
+ * Unlink the inode storing the value of the EA.
+ */
+int
-+ext4_xattr_inode_unlink(struct inode *inode, int ea_ino)
++ext4_xattr_inode_unlink(struct inode *inode, unsigned long ea_ino)
+{
+ struct inode *ea_inode = NULL;
+ int err;
+ * Add value of the EA in an inode.
+ */
+static int
-+ext4_xattr_inode_set(handle_t *handle, struct inode *inode, int *ea_ino,
++ext4_xattr_inode_set(handle_t *handle, struct inode *inode, unsigned long *ea_ino,
+ const void *value, size_t value_len)
+{
+ struct inode *ea_inode = NULL;
last = s->first;
for (; !IS_LAST_ENTRY(last); last = EXT4_XATTR_NEXT(last)) {
- if (!last->e_value_block && last->e_value_size) {
-+ if (last->e_value_inum == 0 && last->e_value_size > 0) {
++ if (!last->e_value_inum && last->e_value_size) {
size_t offs = le16_to_cpu(last->e_value_offs);
if (offs < min_offs)
min_offs = offs;
free = min_offs - ((void *)last - s->base) - sizeof(__u32);
if (!s->not_found) {
- if (!s->here->e_value_block && s->here->e_value_size) {
-+ if (!in_inode && s->here->e_value_inum == 0 &&
-+ s->here->e_value_size > 0) {
++ if (!in_inode &&
++ !s->here->e_value_inum && s->here->e_value_size) {
size_t size = le32_to_cpu(s->here->e_value_size);
free += EXT4_XATTR_SIZE(size);
}
memcpy(s->here->e_name, i->name, name_len);
} else {
- if (!s->here->e_value_block && s->here->e_value_size) {
-+ if (s->here->e_value_offs > 0 && s->here->e_value_inum == 0 &&
-+ s->here->e_value_size > 0) {
++ if (!s->here->e_value_inum && s->here->e_value_size &&
++ s->here->e_value_offs > 0) {
void *first_val = s->base + min_offs;
size_t offs = le16_to_cpu(s->here->e_value_offs);
void *val = s->base + offs;
-@@ -607,13 +888,17 @@ ext4_xattr_set_entry(struct ext4_xattr_i
+@@ -607,13 +888,18 @@ ext4_xattr_set_entry(struct ext4_xattr_i
last = s->first;
while (!IS_LAST_ENTRY(last)) {
size_t o = le16_to_cpu(last->e_value_offs);
- if (!last->e_value_block &&
-- last->e_value_size && o < offs)
-+ if (last->e_value_size > 0 && o < offs)
++ if (!last->e_value_inum &&
+ last->e_value_size && o < offs)
last->e_value_offs =
cpu_to_le16(o + size);
last = EXT4_XATTR_NEXT(last);
}
}
-+ if (s->here->e_value_inum != 0) {
++ if (s->here->e_value_inum) {
+ ext4_xattr_inode_unlink(inode,
+ le32_to_cpu(s->here->e_value_inum));
+ s->here->e_value_inum = 0;
s->here->e_value_size = cpu_to_le32(i->value_len);
- if (i->value_len) {
+ if (in_inode) {
-+ int ea_ino = le32_to_cpu(s->here->e_value_inum);
++ unsigned long ea_ino = le32_to_cpu(s->here->e_value_inum);
+ ext4_xattr_inode_set(handle, inode, &ea_ino, i->value,
+ i->value_len);
+ s->here->e_value_inum = cpu_to_le32(ea_ino);
if (error)
goto cleanup;
if (!is.s.not_found) {
-@@ -1089,10 +1388,25 @@ ext4_xattr_set(struct inode *inode, int
+@@ -1088,10 +1387,25 @@ ext4_xattr_set(struct inode *inode, int
const void *value, size_t value_len, int flags)
{
handle_t *handle;
if (IS_ERR(handle)) {
error = PTR_ERR(handle);
} else {
-@@ -1102,7 +1416,7 @@ retry:
+@@ -1101,7 +1415,7 @@ retry:
value, value_len, flags);
error2 = ext4_journal_stop(handle);
if (error == -ENOSPC &&
goto retry;
if (error == 0)
error = error2;
-@@ -1124,7 +1438,7 @@ static void ext4_xattr_shift_entries(str
+@@ -1123,7 +1437,7 @@ static void ext4_xattr_shift_entries(str
/* Adjust the value offsets of the entries */
for (; !IS_LAST_ENTRY(last); last = EXT4_XATTR_NEXT(last)) {
- if (!last->e_value_block && last->e_value_size) {
-+ if (last->e_value_inum == 0 && last->e_value_size > 0) {
++ if (!last->e_value_inum && last->e_value_size) {
new_offs = le16_to_cpu(last->e_value_offs) +
value_offs_shift;
BUG_ON(new_offs + le32_to_cpu(last->e_value_size)
-@@ -1359,20 +1673,92 @@ cleanup:
+@@ -1358,22 +1672,135 @@ cleanup:
return error;
}
--
+#define EIA_INCR 16 /* must be 2^n */
+#define EIA_MASK (EIA_INCR - 1)
+/* Add the large xattr @ino into @lea_ino_array for later deletion.
+ return -ENOMEM;
+ memcpy(new_array, *lea_ino_array,
+ offsetof(struct ext4_xattr_ino_array,
-+ xia_inodes[count]));
++ xia_inodes[count]));
+ kfree(*lea_ino_array);
+ *lea_ino_array = new_array;
+ }
+ (*lea_ino_array)->xia_inodes[(*lea_ino_array)->xia_count++] = ino;
+ return 0;
+}
++
++/**
++ * Add xattr inode to orphan list
++ */
++static int
++ext4_xattr_inode_orphan_add(handle_t *handle, struct inode *inode,
++ int credits, struct ext4_xattr_ino_array *lea_ino_array)
++{
++ struct inode *ea_inode = NULL;
++ int idx = 0, error = 0;
++
++ if (lea_ino_array == NULL)
++ return 0;
+
++ for (; idx < lea_ino_array->xia_count; ++idx) {
++ if (!ext4_handle_has_enough_credits(handle, credits)) {
++ error = ext4_journal_extend(handle, credits);
++ if (error > 0)
++ error = ext4_journal_restart(handle, credits);
++
++ if (error != 0) {
++ ext4_warning(inode->i_sb,
++ "couldn't extend journal "
++ "(err %d)", error);
++ return error;
++ }
++ }
++ ea_inode = ext4_xattr_inode_iget(inode,
++ lea_ino_array->xia_inodes[idx], &error);
++ if (error)
++ continue;
++ ext4_orphan_add(handle, ea_inode);
++ /* the inode's i_count will be released by caller */
++ }
++
++ return 0;
++}
/*
* ext4_xattr_delete_inode()
+ struct ext4_inode *raw_inode;
+ struct ext4_iloc iloc;
+ struct ext4_xattr_entry *entry;
-+ int error = 0;
-+
++ int credits = 3, error = 0;
+
+- if (!EXT4_I(inode)->i_file_acl)
+ if (!ext4_test_inode_state(inode, EXT4_STATE_XATTR))
+ goto delete_external_ea;
+
+ error = ext4_get_inode_loc(inode, &iloc);
+ if (error)
-+ goto cleanup;
+ goto cleanup;
+ raw_inode = ext4_raw_inode(&iloc);
+ header = IHDR(inode, raw_inode);
-+ entry = IFIRST(header);
-+ for (; !IS_LAST_ENTRY(entry); entry = EXT4_XATTR_NEXT(entry)) {
-+ if (entry->e_value_inum == 0)
++ for (entry = IFIRST(header); !IS_LAST_ENTRY(entry);
++ entry = EXT4_XATTR_NEXT(entry)) {
++ if (!entry->e_value_inum)
+ continue;
+ if (ext4_expand_ino_array(lea_ino_array,
+ entry->e_value_inum) != 0) {
+ entry->e_value_inum = 0;
+ }
+ brelse(iloc.bh);
-
++
+delete_external_ea:
- if (!EXT4_I(inode)->i_file_acl)
- goto cleanup;
++ if (!EXT4_I(inode)->i_file_acl) {
++ /* add xattr inode to orphan list */
++ ext4_xattr_inode_orphan_add(handle, inode, credits,
++ *lea_ino_array);
++ goto cleanup;
++ }
bh = sb_bread(inode->i_sb, EXT4_I(inode)->i_file_acl);
-@@ -1387,11 +1770,74 @@ ext4_xattr_delete_inode(handle_t *handle
+ if (!bh) {
+ EXT4_ERROR_INODE(inode, "block %llu read error",
+@@ -1386,11 +1813,69 @@ ext4_xattr_delete_inode(handle_t *handle
EXT4_I(inode)->i_file_acl);
goto cleanup;
}
+
-+ entry = BFIRST(bh);
-+ for (; !IS_LAST_ENTRY(entry); entry = EXT4_XATTR_NEXT(entry)) {
-+ if (entry->e_value_inum == 0)
++ for (entry = BFIRST(bh); !IS_LAST_ENTRY(entry);
++ entry = EXT4_XATTR_NEXT(entry)) {
++ if (!entry->e_value_inum)
+ continue;
+ if (ext4_expand_ino_array(lea_ino_array,
+ entry->e_value_inum) != 0)
+ }
+
+ /* adding xattr inode to orphan list */
-+ if (*lea_ino_array != NULL) {
-+ struct inode *ea_inode = NULL;
-+ int idx = (*lea_ino_array)->xia_count;
-+
-+ for (idx = 0; idx < (*lea_ino_array)->xia_count; ++idx) {
-+ if (!ext4_handle_has_enough_credits(handle, 3)) {
-+ error = ext4_journal_extend(handle, 3);
-+ if (error > 0)
-+ error = ext4_journal_restart(handle, 3);
-+ if (error != 0) {
-+ ext4_warning(inode->i_sb,
-+ "couldn't extend journal "
-+ "(err %d)", error);
-+ goto cleanup;
-+ }
-+ }
-+ ea_inode = ext4_xattr_inode_iget(inode,
-+ (*lea_ino_array)->xia_inodes[idx], &error);
-+ if (error)
-+ continue;
-+ ext4_orphan_add(handle, ea_inode);
-+ /* the inode's i_count will be released by caller */
++ error = ext4_xattr_inode_orphan_add(handle, inode, credits,
++ *lea_ino_array);
++ if (error != 0)
++ goto cleanup;
++
++ if (!IS_NOQUOTA(inode))
++ credits += 2 * EXT4_QUOTA_DEL_BLOCKS(inode->i_sb);
++
++ if (!ext4_handle_has_enough_credits(handle, credits)) {
++ error = ext4_journal_extend(handle, credits);
++ if (error > 0)
++ error = ext4_journal_restart(handle, credits);
++ if (error != 0) {
++ ext4_warning(inode->i_sb,
++ "couldn't extend journal (err %d)", error);
++ goto cleanup;
+ }
+ }
+
+ lea_ino_array->xia_inodes[idx], &err);
+ if (err)
+ continue;
-+ ea_inode->i_nlink = 0;
-+ iput(ea_inode);
+ /* for inode's i_count get from ext4_xattr_delete_inode */
++ if (!list_empty(&EXT4_I(ea_inode)->i_orphan))
++ iput(ea_inode);
++ ea_inode->i_nlink = 0;
+ iput(ea_inode);
+ }
+ kfree(lea_ino_array);
}
/*
-@@ -1461,10 +1907,9 @@ ext4_xattr_cmp(struct ext4_xattr_header
+@@ -1460,10 +1945,9 @@ ext4_xattr_cmp(struct ext4_xattr_header
entry1->e_name_index != entry2->e_name_index ||
entry1->e_name_len != entry2->e_name_len ||
entry1->e_value_size != entry2->e_value_size ||
if (memcmp((char *)header1 + le16_to_cpu(entry1->e_value_offs),
(char *)header2 + le16_to_cpu(entry2->e_value_offs),
le32_to_cpu(entry1->e_value_size)))
-@@ -1548,7 +1993,7 @@ static inline void ext4_xattr_hash_entry
+@@ -1547,7 +2031,7 @@ static inline void ext4_xattr_hash_entry
*name++;
}
- if (entry->e_value_block == 0 && entry->e_value_size != 0) {
-+ if (entry->e_value_inum == 0 && entry->e_value_size != 0) {
++ if (!entry->e_value_inum && entry->e_value_size) {
__le32 *value = (__le32 *)((char *)header +
le16_to_cpu(entry->e_value_offs));
for (n = (le32_to_cpu(entry->e_value_size) +
__le32 e_value_size; /* size of attribute value */
__le32 e_hash; /* hash value of name and value */
char e_name[0]; /* attribute name */
-@@ -63,6 +63,15 @@ struct ext4_xattr_entry {
+@@ -63,6 +63,26 @@ struct ext4_xattr_entry {
EXT4_I(inode)->i_extra_isize))
#define IFIRST(hdr) ((struct ext4_xattr_entry *)((hdr)+1))
-+#define i_xattr_inode_parent i_mtime.tv_sec
++/*
++ * Link EA inode back to parent one using i_mtime field.
++ * Extra integer type conversion added to ignore higher
++ * bits in i_mtime.tv_sec which might be set by ext4_get()
++ */
++#define EXT4_XATTR_INODE_SET_PARENT(inode, inum) \
++do { \
++ (inode)->i_mtime.tv_sec = inum; \
++} while(0)
++
++#define EXT4_XATTR_INODE_GET_PARENT(inode) \
++((__u32)(inode)->i_mtime.tv_sec)
+
+/*
+ * The minimum size of EA value when you start storing it in an external inode
extern int ext4_xattr_set_handle(handle_t *, struct inode *, int, const char *, const void *, size_t, int);
-extern void ext4_xattr_delete_inode(handle_t *, struct inode *);
-+extern struct inode *ext4_xattr_inode_iget(struct inode *parent, int ea_ino,
++extern struct inode *ext4_xattr_inode_iget(struct inode *parent, unsigned long ea_ino,
+ int *err);
-+extern int ext4_xattr_inode_unlink(struct inode *inode, int ea_ino);
++extern int ext4_xattr_inode_unlink(struct inode *inode, unsigned long ea_ino);
+extern int ext4_xattr_delete_inode(handle_t *handle, struct inode *inode,
+ struct ext4_xattr_ino_array **array);
+extern void ext4_xattr_inode_array_free(struct inode *inode,
===================================================================
--- linux-stage.orig/fs/ext4/ialloc.c
+++ linux-stage/fs/ext4/ialloc.c
-@@ -222,7 +222,6 @@ void ext4_free_inode(handle_t *handle, s
+@@ -221,7 +221,6 @@ void ext4_free_inode(handle_t *handle, s
* as writing the quota to disk may need the lock as well.
*/
dquot_initialize(inode);
git://git.whamcloud.com / fs / lustre-release.git / blobdiff