@@ -621,6 +621,17 @@ struct ext2_super_block {
#define EXT3_JNL_BACKUP_BLOCKS 1
/*
+ * 'big journal' needs to accomodate extra snapshot COW credits
+ * default size accomodates maximum possible COW credits
+ * minimum required size accomodates the avarage COW credits
+ */
+#define EXT3_DEF_JOURNAL_BLOCKS 32768
+#define NEXT3_AVG_COW_CREDITS 16
+#define NEXT3_MAX_COW_CREDITS 24
+#define NEXT3_MIN_JOURNAL_BLOCKS (EXT3_DEF_JOURNAL_BLOCKS*NEXT3_AVG_COW_CREDITS)
+#define NEXT3_DEF_JOURNAL_BLOCKS (EXT3_DEF_JOURNAL_BLOCKS*NEXT3_MAX_COW_CREDITS)
+
+/*
* Feature set definitions
*/
@@ -1222,6 +1222,8 @@ extern errcode_t ext2fs_add_journal_device(ext2_filsys fs,
extern errcode_t ext2fs_add_journal_inode(ext2_filsys fs, blk_t size,
int flags);
extern int ext2fs_default_journal_size(__u64 blocks);
+extern int ext2fs_big_journal_size(__u64 blocks);
+extern int ext2fs_check_journal_size(ext2_filsys fs);
/* openfs.c */
extern errcode_t ext2fs_open(const char *name, int flags, int superblock,
@@ -400,6 +400,54 @@ int ext2fs_default_journal_size(__u64 blocks)
return 32768;
}
+/*
+ * Big journal is up to 24 times bigger than the default journal
+ * to accomodate snapshot COW credits in transactions.
+ * journal size is restricted to 1/32 of the filesystem size
+ */
+int ext2fs_big_journal_size(__u64 blocks)
+{
+ int mega_blocks = blocks >> 20;
+ if (!mega_blocks)
+ return -1;
+
+ if (mega_blocks < NEXT3_MAX_COW_CREDITS)
+ /* 32K/1M = 1/32 of filesystem size */
+ return 32768*mega_blocks;
+
+ /* 24 times bigger than the default journal */
+ return 32768*NEXT3_MAX_COW_CREDITS;
+}
+
+/*
+ * Find the number of blocks in the journal inode
+ * and adjust the file system 'big_journal' feature accordingy
+ */
+int ext2fs_check_journal_size(ext2_filsys fs)
+{
+ struct ext2_inode j_inode;
+ int j_blocks;
+
+ if (!(fs->super->s_feature_compat &
+ EXT3_FEATURE_COMPAT_HAS_JOURNAL) ||
+ !fs->super->s_journal_inum)
+ return 0;
+
+ if (ext2fs_read_inode(fs, fs->super->s_journal_inum, &j_inode))
+ return -1;
+
+ /* read journal inode size */
+ j_blocks = j_inode.i_size >> EXT2_BLOCK_SIZE_BITS(fs->super);
+
+ /* fix the 'big_journal' feature */
+ if (j_blocks >= NEXT3_MIN_JOURNAL_BLOCKS)
+ fs->super->s_feature_compat |= NEXT3_FEATURE_COMPAT_BIG_JOURNAL;
+ else
+ fs->super->s_feature_compat &= ~NEXT3_FEATURE_COMPAT_BIG_JOURNAL;
+
+ return j_blocks;
+}
+
/*
* This function adds a journal device to a filesystem
*/
@@ -553,6 +601,7 @@ errcode_t ext2fs_add_journal_inode(ext2_filsys fs, blk_t size, int flags)
sizeof(fs->super->s_journal_uuid));
fs->super->s_feature_compat |= EXT3_FEATURE_COMPAT_HAS_JOURNAL;
+ ext2fs_check_journal_size(fs);
ext2fs_mark_super_dirty(fs);
return 0;
errout:
@@ -781,6 +781,7 @@ static void parse_extended_opts(struct ext2_super_block *param,
static __u32 ok_features[3] = {
/* Compat */
EXT3_FEATURE_COMPAT_HAS_JOURNAL |
+ NEXT3_FEATURE_COMPAT_BIG_JOURNAL |
EXT2_FEATURE_COMPAT_RESIZE_INODE |
EXT2_FEATURE_COMPAT_DIR_INDEX |
EXT2_FEATURE_COMPAT_EXT_ATTR,
@@ -119,6 +119,7 @@ static void usage(void)
static __u32 ok_features[3] = {
/* Compat */
EXT3_FEATURE_COMPAT_HAS_JOURNAL |
+ NEXT3_FEATURE_COMPAT_BIG_JOURNAL |
EXT2_FEATURE_COMPAT_DIR_INDEX,
/* Incompat */
EXT2_FEATURE_INCOMPAT_FILETYPE |
@@ -136,6 +137,7 @@ static __u32 ok_features[3] = {
static __u32 clear_ok_features[3] = {
/* Compat */
EXT3_FEATURE_COMPAT_HAS_JOURNAL |
+ NEXT3_FEATURE_COMPAT_BIG_JOURNAL |
EXT2_FEATURE_COMPAT_RESIZE_INODE |
EXT2_FEATURE_COMPAT_DIR_INDEX,
/* Incompat */
@@ -403,6 +405,23 @@ static void update_feature_set(ext2_filsys fs, char *features)
sb->s_feature_compat &= ~EXT3_FEATURE_COMPAT_HAS_JOURNAL;
}
+ if (FEATURE_CHANGED(E2P_FEATURE_COMPAT, NEXT3_FEATURE_COMPAT_BIG_JOURNAL)) {
+ if (sb->s_feature_compat &
+ EXT3_FEATURE_COMPAT_HAS_JOURNAL) {
+ /* update 'big_journal' flag according to existing journal size */
+ ext2fs_check_journal_size(fs);
+ if (!FEATURE_CHANGED(E2P_FEATURE_COMPAT, NEXT3_FEATURE_COMPAT_BIG_JOURNAL))
+ fputs(_("the filesystem already has a journal.\n"
+ "Please remove it before changing "
+ "the big_journal flag.\n"), stderr);
+ }
+ else if (sb->s_feature_compat & NEXT3_FEATURE_COMPAT_BIG_JOURNAL) {
+ /* create 'big_journal' */
+ if (!journal_size)
+ journal_size = -1;
+ }
+ }
+
if (FEATURE_ON(E2P_FEATURE_COMPAT, EXT2_FEATURE_COMPAT_DIR_INDEX)) {
if (!sb->s_def_hash_version)
sb->s_def_hash_version = EXT2_HASH_HALF_MD4;
@@ -1792,8 +1811,10 @@ retry_open:
}
}
- if (l_flag)
+ if (l_flag) {
+ ext2fs_check_journal_size(fs);
list_super(sb);
+ }
if (stride_set) {
sb->s_raid_stride = stride;
ext2fs_mark_super_dirty(fs);
@@ -259,6 +259,20 @@ unsigned int figure_journal_size(int size, ext2_filsys fs)
return 0;
}
+ if (fs->super->s_feature_compat & NEXT3_FEATURE_COMPAT_BIG_JOURNAL) {
+ /* big journal requested */
+ j_blocks = ext2fs_big_journal_size(fs->super->s_blocks_count);
+ if (j_blocks < NEXT3_MIN_JOURNAL_BLOCKS) {
+ fputs(_("\nFilesystem too small for a big journal. "),
+ stderr);
+ if (j_blocks < 0) {
+ fputs(_("Aborting.\n"), stderr);
+ exit(1);
+ }
+ fputs(_("Creating a smaller journal.\n"), stderr);
+ }
+ }
+
if (size > 0) {
j_blocks = size * 1024 / (fs->blocksize / 1024);
if (j_blocks < 1024 || j_blocks > 10240000) {
Next3 transactions reserve up to 24 times more credits than Ext3 transactions for the same operation. On mke2fs and tune2fs, if the 'big_journal' feature is set along with the 'has_journal' feature, increase the default journal size by a factor of 24. Signed-off-by: Amir Goldstein <amir73il@users.sf.net> --- lib/ext2fs/ext2_fs.h | 11 ++++++++++ lib/ext2fs/ext2fs.h | 2 + lib/ext2fs/mkjournal.c | 49 ++++++++++++++++++++++++++++++++++++++++++++++++ misc/mke2fs.c | 1 + misc/tune2fs.c | 23 +++++++++++++++++++++- misc/util.c | 14 +++++++++++++ 6 files changed, 99 insertions(+), 1 deletions(-)