[PATCH] Fixup Orlov block allocator for ext2

I finally had time to look at the Orlov patches, and found a memory
leak; sbi->s_debts wasn't getting freed when the filesystem was
getting unmounted, or in the error path.

This patch also makes the following cleanups/changes:

1) Use sbi->s_debts instead of sbi->debts --- all other fields in
	struct ext2_sb_info are prefixed by "s_", so this makes things
	consistent.

2) Add support for a new inode flag, EXT2_TOPDIR_FL, which tells tells
	the Orlov allocator to treat that directory as the top of
	directory hierarchies, so that new subdirectories created in
	that directory should be spread apart.  System administrators
	should set this flag on directories like /usr/src, /usr/home, etc.

3) Add a mount-time flag, -o oldalloc, which forces the use of the old
	inode (pre-Orlov) allocator.  This makes it easier to do
	comparison benchmarks, and in case people want to use the old
	algorithm.

diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c
index 6b615fb184c60d692afe554ee38b4250ca7ca44b..470f3252b2cbc4ef5a404687b9745548d7c9e84e 100644 (file)
--- a/fs/ext2/ialloc.c
+++ b/fs/ext2/ialloc.c
@@ -209,9 +209,7 @@ static void ext2_preread_inode(struct inode *inode)
  * For other inodes, search forward from the parent directory\'s block
  * group to find a free inode.
  */
-#if 0
-
-static int find_group_dir(struct super_block *sb, int parent_group)
+static int find_group_dir(struct super_block *sb, struct inode *parent)
 {
        struct ext2_super_block * es = EXT2_SB(sb)->s_es;
        int ngroups = EXT2_SB(sb)->s_groups_count;
@@ -243,7 +241,6 @@ static int find_group_dir(struct super_block *sb, int parent_group)
        mark_buffer_dirty(best_bh);
        return best_group;
 }
-#endif
 
 /* 
  * Orlov's allocator for directories. 
@@ -289,7 +286,8 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent)
        struct ext2_group_desc *desc;
        struct buffer_head *bh;
 
-       if (parent == sb->s_root->d_inode) {
+       if ((parent == sb->s_root->d_inode) ||
+           (parent->i_flags & EXT2_TOPDIR_FL)) {
                struct ext2_group_desc *best_desc = NULL;
                struct buffer_head *best_bh = NULL;
                int best_ndir = inodes_per_group;
@@ -342,7 +340,7 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent)
                desc = ext2_get_group_desc (sb, group, &bh);
                if (!desc || !desc->bg_free_inodes_count)
                        continue;
-               if (sbi->debts[group] >= max_debt)
+               if (sbi->s_debts[group] >= max_debt)
                        continue;
                if (le16_to_cpu(desc->bg_used_dirs_count) >= max_dirs)
                        continue;
@@ -447,9 +445,12 @@ struct inode * ext2_new_inode(struct inode * dir, int mode)
        lock_super (sb);
        es = EXT2_SB(sb)->s_es;
 repeat:
-       if (S_ISDIR(mode))
-               group = find_group_orlov(sb, dir);
-       else 
+       if (S_ISDIR(mode)) {
+               if (test_opt (sb, OLDALLOC))
+                       group = find_group_dir(sb, dir);
+               else
+                       group = find_group_orlov(sb, dir);
+       } else 
                group = find_group_other(sb, dir);
 
        err = -ENOSPC;
@@ -488,11 +489,11 @@ repeat:
                cpu_to_le32(le32_to_cpu(es->s_free_inodes_count) - 1);
 
        if (S_ISDIR(mode)) {
-               if (EXT2_SB(sb)->debts[group] < 255)
-                       EXT2_SB(sb)->debts[group]++;
+               if (EXT2_SB(sb)->s_debts[group] < 255)
+                       EXT2_SB(sb)->s_debts[group]++;
        } else {
-               if (EXT2_SB(sb)->debts[group])
-                       EXT2_SB(sb)->debts[group]--;
+               if (EXT2_SB(sb)->s_debts[group])
+                       EXT2_SB(sb)->s_debts[group]--;
        }
 
        mark_buffer_dirty(EXT2_SB(sb)->s_sbh);

diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index 1fd20127164b59ddf427910305a64cf135f5c24e..e65cc4d0b3f536d04713425c054c091a86d89fd4 100644 (file)
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -140,6 +140,7 @@ static void ext2_put_super (struct super_block * sb)
                if (sbi->s_group_desc[i])
                        brelse (sbi->s_group_desc[i]);
        kfree(sbi->s_group_desc);
+       kfree(sbi->s_debts);
        brelse (sbi->s_sbh);
        sb->s_fs_info = NULL;
        kfree(sbi);
@@ -385,6 +386,10 @@ static int parse_options (char * options,
                                return 0;
                        sbi->s_resuid = v;
                }
+               else if (!strcmp (this_char, "oldalloc"))
+                       set_opt (sbi->s_mount_opt, OLDALLOC);
+               else if (!strcmp (this_char, "orlov"))
+                       clear_opt (sbi->s_mount_opt, OLDALLOC);
                /* Silently ignore the quota options */
                else if (!strcmp (this_char, "grpquota")
                         || !strcmp (this_char, "noquota")
@@ -756,13 +761,13 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
                printk ("EXT2-fs: not enough memory\n");
                goto failed_mount;
        }
-       sbi->debts = kmalloc(sbi->s_groups_count * sizeof(*sbi->debts),
-                       GFP_KERNEL);
-       if (!sbi->debts) {
+       sbi->s_debts = kmalloc(sbi->s_groups_count * sizeof(*sbi->s_debts),
+                              GFP_KERNEL);
+       if (!sbi->s_debts) {
                printk ("EXT2-fs: not enough memory\n");
                goto failed_mount_group_desc;
        }
-       memset(sbi->debts, 0, sbi->s_groups_count * sizeof(*sbi->debts));
+       memset(sbi->s_debts, 0, sbi->s_groups_count * sizeof(*sbi->s_debts));
        for (i = 0; i < db_count; i++) {
                block = descriptor_loc(sb, logic_sb_block, i);
                sbi->s_group_desc[i] = sb_bread(sb, block);
@@ -771,7 +776,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
                                brelse (sbi->s_group_desc[j]);
                        kfree(sbi->s_group_desc);
                        printk ("EXT2-fs: unable to read group descriptors\n");
-                       goto failed_mount;
+                       goto failed_mount_group_desc;
                }
        }
        if (!ext2_check_descriptors (sb)) {
@@ -808,6 +813,8 @@ failed_mount2:
                brelse(sbi->s_group_desc[i]);
 failed_mount_group_desc:
        kfree(sbi->s_group_desc);
+       if (sbi->s_debts)
+               kfree(sbi->s_debts);
 failed_mount:
        brelse(bh);
 failed_sbi:

diff --git a/include/linux/ext2_fs.h b/include/linux/ext2_fs.h
index 7e34cc7518000779b7a71d7d725488cc98a0645f..0d008cfb99a855ee488b44b637d132acd001a93f 100644 (file)
--- a/include/linux/ext2_fs.h
+++ b/include/linux/ext2_fs.h
@@ -191,10 +191,11 @@ struct ext2_group_desc
 #define EXT2_JOURNAL_DATA_FL           0x00004000 /* Reserved for ext3 */
 #define EXT2_NOTAIL_FL                 0x00008000 /* file tail should not be merged */
 #define EXT2_DIRSYNC_FL                        0x00010000 /* dirsync behaviour (directories only) */
+#define EXT2_TOPDIR_FL                 0x00020000 /* Top of directory hierarchies*/
 #define EXT2_RESERVED_FL               0x80000000 /* reserved for ext2 lib */
 
-#define EXT2_FL_USER_VISIBLE           0x00011FFF /* User visible flags */
-#define EXT2_FL_USER_MODIFIABLE                0x000100FF /* User modifiable flags */
+#define EXT2_FL_USER_VISIBLE           0x0003DFFF /* User visible flags */
+#define EXT2_FL_USER_MODIFIABLE                0x000380FF /* User modifiable flags */
 
 /*
  * ioctl commands
@@ -300,6 +301,7 @@ struct ext2_inode {
  * Mount flags
  */
 #define EXT2_MOUNT_CHECK               0x0001  /* Do mount-time checks */
+#define EXT2_MOUNT_OLDALLOC            0x0002  /* Don't use the new Orlov allocator */
 #define EXT2_MOUNT_GRPID               0x0004  /* Create files with directory's group */
 #define EXT2_MOUNT_DEBUG               0x0008  /* Some debugging messages */
 #define EXT2_MOUNT_ERRORS_CONT         0x0010  /* Continue on errors */

diff --git a/include/linux/ext2_fs_sb.h b/include/linux/ext2_fs_sb.h
index 434924a64fe474cc28630b303889d5cb59658e51..3c07d4ecf898daa0022cfb96b82479e37407412f 100644 (file)
--- a/include/linux/ext2_fs_sb.h
+++ b/include/linux/ext2_fs_sb.h
@@ -44,7 +44,7 @@ struct ext2_sb_info {
        int s_first_ino;
        u32 s_next_generation;
        unsigned long s_dir_count;
-       u8 *debts;
+       u8 *s_debts;
 };
 
 #endif /* _LINUX_EXT2_FS_SB */