ext2 reservations

Val's cross-port of the ext3 reservations code into ext2.

[mbligh@mbligh.org: Small type error for printk
[akpm@linux-foundation.org: fix types, sync with ext3]
[mbligh@mbligh.org: Bring ext2 reservations code in line with latest ext3]
[akpm@linux-foundation.org: kill noisy printk]
[akpm@linux-foundation.org: remember to dirty the gdp's block]
[akpm@linux-foundation.org: cross-port the missed 5dea5176]
[akpm@linux-foundation.org: cross-port e6022603]
[akpm@linux-foundation.org: Port the omitted 08fb306f]
[akpm@linux-foundation.org: Backport the missed 20acaa18]
[akpm@linux-foundation.org: fixes]
[cmm@us.ibm.com: fix reservation extension]
[bunk@stusta.de: make ext2_get_blocks() static]
[hugh@veritas.com: fix hang]
[hugh@veritas.com: ext2_new_blocks should reset the reservation window size]
[hugh@veritas.com: ext2 balloc: fix off-by-one against rsv_end]
[hugh@veritas.com: grp_goal 0 is a genuine goal (unlike -1), so ext2_try_to_allocate_with_rsv should treat it as such]
[hugh@veritas.com: rbtree usage cleanup]
[pbadari@us.ibm.com: Fix for ext2 reservation]
[bunk@kernel.org: remove fs/ext2/balloc.c:reserve_blocks()]
[hugh@veritas.com: ext2 balloc: use io_error label]
Cc: "Martin J. Bligh" <mbligh@mbligh.org>
Cc: Valerie Henson <val_henson@linux.intel.com>
Cc: Mingming Cao <cmm@us.ibm.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

fs/ext2/ext2.h

@@ -33,22 +33,9 @@ struct ext2_inode_info {
 	 */
 	__u32	i_block_group;
 
-	/*
-	 * i_next_alloc_block is the logical (file-relative) number of the
-	 * most-recently-allocated block in this file.  Yes, it is misnamed.
-	 * We use this for detecting linearly ascending allocation requests.
-	 */
-	__u32	i_next_alloc_block;
+	/* block reservation info */
+	struct ext2_block_alloc_info *i_block_alloc_info;
 
-	/*
-	 * i_next_alloc_goal is the *physical* companion to i_next_alloc_block.
-	 * it the the physical block number of the block which was most-recently
-	 * allocated to this file.  This give us the goal (target) for the next
-	 * allocation when we detect linearly ascending requests.
-	 */
-	__u32	i_next_alloc_goal;
-	__u32	i_prealloc_block;
-	__u32	i_prealloc_count;
 	__u32	i_dir_start_lookup;
 #ifdef CONFIG_EXT2_FS_XATTR
 	/*
@@ -65,7 +52,16 @@ struct ext2_inode_info {
 	struct posix_acl	*i_default_acl;
 #endif
 	rwlock_t i_meta_lock;
+
+	/*
+	 * truncate_mutex is for serialising ext2_truncate() against
+	 * ext2_getblock().  It also protects the internals of the inode's
+	 * reservation data structures: ext2_reserve_window and
+	 * ext2_reserve_window_node.
+	 */
+	struct mutex truncate_mutex;
 	struct inode	vfs_inode;
+	struct list_head i_orphan;	/* unlinked but open inodes */
 };
 
 /*
@@ -91,8 +87,9 @@ static inline struct ext2_inode_info *EXT2_I(struct inode *inode)
 /* balloc.c */
 extern int ext2_bg_has_super(struct super_block *sb, int group);
 extern unsigned long ext2_bg_num_gdb(struct super_block *sb, int group);
-extern int ext2_new_block (struct inode *, unsigned long,
-			   __u32 *, __u32 *, int *);
+extern ext2_fsblk_t ext2_new_block(struct inode *, unsigned long, int *);
+extern ext2_fsblk_t ext2_new_blocks(struct inode *, unsigned long,
+				unsigned long *, int *);
 extern void ext2_free_blocks (struct inode *, unsigned long,
 			      unsigned long);
 extern unsigned long ext2_count_free_blocks (struct super_block *);
@@ -101,6 +98,10 @@ extern void ext2_check_blocks_bitmap (struct super_block *);
 extern struct ext2_group_desc * ext2_get_group_desc(struct super_block * sb,
 						    unsigned int block_group,
 						    struct buffer_head ** bh);
+extern void ext2_discard_reservation (struct inode *);
+extern int ext2_should_retry_alloc(struct super_block *sb, int *retries);
+extern void ext2_init_block_alloc_info(struct inode *);
+extern void ext2_rsv_window_add(struct super_block *sb, struct ext2_reserve_window_node *rsv);
 
 /* dir.c */
 extern int ext2_add_link (struct dentry *, struct inode *);
@@ -128,7 +129,6 @@ extern int ext2_write_inode (struct inode *, int);
 extern void ext2_put_inode (struct inode *);
 extern void ext2_delete_inode (struct inode *);
 extern int ext2_sync_inode (struct inode *);
-extern void ext2_discard_prealloc (struct inode *);
 extern int ext2_get_block(struct inode *, sector_t, struct buffer_head *, int);
 extern void ext2_truncate (struct inode *);
 extern int ext2_setattr (struct dentry *, struct iattr *);

fs/ext2/file.c

@@ -30,8 +30,11 @@
  */
 static int ext2_release_file (struct inode * inode, struct file * filp)
 {
-	if (filp->f_mode & FMODE_WRITE)
-		ext2_discard_prealloc (inode);
+	if (filp->f_mode & FMODE_WRITE) {
+		mutex_lock(&EXT2_I(inode)->truncate_mutex);
+		ext2_discard_reservation(inode);
+		mutex_unlock(&EXT2_I(inode)->truncate_mutex);
+	}
 	return 0;
 }
 

fs/ext2/ialloc.c

@@ -581,11 +581,8 @@ got:
 	ei->i_file_acl = 0;
 	ei->i_dir_acl = 0;
 	ei->i_dtime = 0;
+	ei->i_block_alloc_info = NULL;
 	ei->i_block_group = group;
-	ei->i_next_alloc_block = 0;
-	ei->i_next_alloc_goal = 0;
-	ei->i_prealloc_block = 0;
-	ei->i_prealloc_count = 0;
 	ei->i_dir_start_lookup = 0;
 	ei->i_state = EXT2_STATE_NEW;
 	ext2_set_inode_flags(inode);

fs/ext2/ioctl.c

@@ -22,6 +22,7 @@ int ext2_ioctl (struct inode * inode, struct file * filp, unsigned int cmd,
 {
 	struct ext2_inode_info *ei = EXT2_I(inode);
 	unsigned int flags;
+	unsigned short rsv_window_size;
 
 	ext2_debug ("cmd = %u, arg = %lu\n", cmd, arg);
 
@@ -83,6 +84,50 @@ int ext2_ioctl (struct inode * inode, struct file * filp, unsigned int cmd,
 		inode->i_ctime = CURRENT_TIME_SEC;
 		mark_inode_dirty(inode);
 		return 0;
+	case EXT2_IOC_GETRSVSZ:
+		if (test_opt(inode->i_sb, RESERVATION)
+			&& S_ISREG(inode->i_mode)
+			&& ei->i_block_alloc_info) {
+			rsv_window_size = ei->i_block_alloc_info->rsv_window_node.rsv_goal_size;
+			return put_user(rsv_window_size, (int __user *)arg);
+		}
+		return -ENOTTY;
+	case EXT2_IOC_SETRSVSZ: {
+
+		if (!test_opt(inode->i_sb, RESERVATION) ||!S_ISREG(inode->i_mode))
+			return -ENOTTY;
+
+		if (IS_RDONLY(inode))
+			return -EROFS;
+
+		if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER))
+			return -EACCES;
+
+		if (get_user(rsv_window_size, (int __user *)arg))
+			return -EFAULT;
+
+		if (rsv_window_size > EXT2_MAX_RESERVE_BLOCKS)
+			rsv_window_size = EXT2_MAX_RESERVE_BLOCKS;
+
+		/*
+		 * need to allocate reservation structure for this inode
+		 * before set the window size
+		 */
+		/*
+		 * XXX What lock should protect the rsv_goal_size?
+		 * Accessed in ext2_get_block only.  ext3 uses i_truncate.
+		 */
+		mutex_lock(&ei->truncate_mutex);
+		if (!ei->i_block_alloc_info)
+			ext2_init_block_alloc_info(inode);
+
+		if (ei->i_block_alloc_info){
+			struct ext2_reserve_window_node *rsv = &ei->i_block_alloc_info->rsv_window_node;
+			rsv->rsv_goal_size = rsv_window_size;
+		}
+		mutex_unlock(&ei->truncate_mutex);
+		return 0;
+	}
 	default:
 		return -ENOTTY;
 	}

fs/ext2/super.c

@@ -149,6 +149,7 @@ static struct inode *ext2_alloc_inode(struct super_block *sb)
 	ei->i_acl = EXT2_ACL_NOT_CACHED;
 	ei->i_default_acl = EXT2_ACL_NOT_CACHED;
 #endif
+	ei->i_block_alloc_info = NULL;
 	ei->vfs_inode.i_version = 1;
 	return &ei->vfs_inode;
 }
@@ -166,6 +167,7 @@ static void init_once(struct kmem_cache * cachep, void *foo)
 #ifdef CONFIG_EXT2_FS_XATTR
 	init_rwsem(&ei->xattr_sem);
 #endif
+	mutex_init(&ei->truncate_mutex);
 	inode_init_once(&ei->vfs_inode);
 }
 
@@ -188,6 +190,7 @@ static void destroy_inodecache(void)
 
 static void ext2_clear_inode(struct inode *inode)
 {
+	struct ext2_block_alloc_info *rsv = EXT2_I(inode)->i_block_alloc_info;
 #ifdef CONFIG_EXT2_FS_POSIX_ACL
 	struct ext2_inode_info *ei = EXT2_I(inode);
 
@@ -200,6 +203,10 @@ static void ext2_clear_inode(struct inode *inode)
 		ei->i_default_acl = EXT2_ACL_NOT_CACHED;
 	}
 #endif
+	ext2_discard_reservation(inode);
+	EXT2_I(inode)->i_block_alloc_info = NULL;
+	if (unlikely(rsv))
+		kfree(rsv);
 }
 
 static int ext2_show_options(struct seq_file *seq, struct vfsmount *vfs)
@@ -291,7 +298,6 @@ static const struct super_operations ext2_sops = {
 	.destroy_inode	= ext2_destroy_inode,
 	.read_inode	= ext2_read_inode,
 	.write_inode	= ext2_write_inode,
-	.put_inode	= ext2_put_inode,
 	.delete_inode	= ext2_delete_inode,
 	.put_super	= ext2_put_super,
 	.write_super	= ext2_write_super,
@@ -379,7 +385,7 @@ enum {
 	Opt_err_ro, Opt_nouid32, Opt_nocheck, Opt_debug,
 	Opt_oldalloc, Opt_orlov, Opt_nobh, Opt_user_xattr, Opt_nouser_xattr,
 	Opt_acl, Opt_noacl, Opt_xip, Opt_ignore, Opt_err, Opt_quota,
-	Opt_usrquota, Opt_grpquota
+	Opt_usrquota, Opt_grpquota, Opt_reservation, Opt_noreservation
 };
 
 static match_table_t tokens = {
@@ -411,6 +417,8 @@ static match_table_t tokens = {
 	{Opt_ignore, "noquota"},
 	{Opt_quota, "quota"},
 	{Opt_usrquota, "usrquota"},
+	{Opt_reservation, "reservation"},
+	{Opt_noreservation, "noreservation"},
 	{Opt_err, NULL}
 };
 
@@ -543,6 +551,14 @@ static int parse_options (char * options,
 			break;
 #endif
 
+		case Opt_reservation:
+			set_opt(sbi->s_mount_opt, RESERVATION);
+			printk("reservations ON\n");
+			break;
+		case Opt_noreservation:
+			clear_opt(sbi->s_mount_opt, RESERVATION);
+			printk("reservations OFF\n");
+			break;
 		case Opt_ignore:
 			break;
 		default:
@@ -784,6 +800,8 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 	sbi->s_resuid = le16_to_cpu(es->s_def_resuid);
 	sbi->s_resgid = le16_to_cpu(es->s_def_resgid);
 	
+	set_opt(sbi->s_mount_opt, RESERVATION);
+
 	if (!parse_options ((char *) data, sbi))
 		goto failed_mount;
 
@@ -965,6 +983,21 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 	get_random_bytes(&sbi->s_next_generation, sizeof(u32));
 	spin_lock_init(&sbi->s_next_gen_lock);
 
+	/* per fileystem reservation list head & lock */
+	spin_lock_init(&sbi->s_rsv_window_lock);
+	sbi->s_rsv_window_root = RB_ROOT;
+	/*
+	 * Add a single, static dummy reservation to the start of the
+	 * reservation window list --- it gives us a placeholder for
+	 * append-at-start-of-list which makes the allocation logic
+	 * _much_ simpler.
+	 */
+	sbi->s_rsv_window_head.rsv_start = EXT2_RESERVE_WINDOW_NOT_ALLOCATED;
+	sbi->s_rsv_window_head.rsv_end = EXT2_RESERVE_WINDOW_NOT_ALLOCATED;
+	sbi->s_rsv_window_head.rsv_alloc_hit = 0;
+	sbi->s_rsv_window_head.rsv_goal_size = 0;
+	ext2_rsv_window_add(sb, &sbi->s_rsv_window_head);
+
 	err = percpu_counter_init(&sbi->s_freeblocks_counter,
 				ext2_count_free_blocks(sb));
 	if (!err) {
@@ -1260,7 +1293,7 @@ static ssize_t ext2_quota_read(struct super_block *sb, int type, char *data,
 
 		tmp_bh.b_state = 0;
 		err = ext2_get_block(inode, blk, &tmp_bh, 0);
-		if (err)
+		if (err < 0)
 			return err;
 		if (!buffer_mapped(&tmp_bh))	/* A hole? */
 			memset(data, 0, tocopy);
@@ -1299,7 +1332,7 @@ static ssize_t ext2_quota_write(struct super_block *sb, int type,
 
 		tmp_bh.b_state = 0;
 		err = ext2_get_block(inode, blk, &tmp_bh, 1);
-		if (err)
+		if (err < 0)
 			goto out;
 		if (offset || tocopy != EXT2_BLOCK_SIZE(sb))
 			bh = sb_bread(sb, tmp_bh.b_blocknr);

fs/ext2/xattr.c

@@ -664,8 +664,7 @@ ext2_xattr_set2(struct inode *inode, struct buffer_head *old_bh,
 						           s_first_data_block) +
 				   EXT2_I(inode)->i_block_group *
 				   EXT2_BLOCKS_PER_GROUP(sb);
-			int block = ext2_new_block(inode, goal,
-						   NULL, NULL, &error);
+			int block = ext2_new_block(inode, goal, &error);
 			if (error)
 				goto cleanup;
 			ea_idebug(inode, "creating block %d", block);

include/linux/ext2_fs.h

@@ -29,11 +29,12 @@
 #undef EXT2FS_DEBUG
 
 /*
- * Define EXT2_PREALLOCATE to preallocate data blocks for expanding files
+ * Define EXT2_RESERVATION to reserve data blocks for expanding files
  */
-#define EXT2_PREALLOCATE
-#define EXT2_DEFAULT_PREALLOC_BLOCKS	8
-
+#define EXT2_DEFAULT_RESERVE_BLOCKS     8
+/*max window size: 1024(direct blocks) + 3([t,d]indirect blocks) */
+#define EXT2_MAX_RESERVE_BLOCKS         1027
+#define EXT2_RESERVE_WINDOW_NOT_ALLOCATED 0
 /*
  * The second extended file system version
  */
@@ -200,6 +201,8 @@ struct ext2_group_desc
 #define	EXT2_IOC_SETFLAGS		FS_IOC_SETFLAGS
 #define	EXT2_IOC_GETVERSION		FS_IOC_GETVERSION
 #define	EXT2_IOC_SETVERSION		FS_IOC_SETVERSION
+#define	EXT2_IOC_GETRSVSZ		_IOR('f', 5, long)
+#define	EXT2_IOC_SETRSVSZ		_IOW('f', 6, long)
 
 /*
  * ioctl commands in 32 bit emulation
@@ -319,6 +322,7 @@ struct ext2_inode {
 #define EXT2_MOUNT_XIP			0x010000  /* Execute in place */
 #define EXT2_MOUNT_USRQUOTA		0x020000  /* user quota */
 #define EXT2_MOUNT_GRPQUOTA		0x040000  /* group quota */
+#define EXT2_MOUNT_RESERVATION		0x080000  /* Preallocation */
 
 
 #define clear_opt(o, opt)		o &= ~EXT2_MOUNT_##opt
@@ -558,4 +562,11 @@ enum {
 #define EXT2_DIR_REC_LEN(name_len)	(((name_len) + 8 + EXT2_DIR_ROUND) & \
 					 ~EXT2_DIR_ROUND)
 
+static inline ext2_fsblk_t
+ext2_group_first_block_no(struct super_block *sb, unsigned long group_no)
+{
+	return group_no * (ext2_fsblk_t)EXT2_BLOCKS_PER_GROUP(sb) +
+		le32_to_cpu(EXT2_SB(sb)->s_es->s_first_data_block);
+}
+
 #endif	/* _LINUX_EXT2_FS_H */

include/linux/ext2_fs_sb.h

@@ -18,6 +18,52 @@
 
 #include <linux/blockgroup_lock.h>
 #include <linux/percpu_counter.h>
+#include <linux/rbtree.h>
+
+/* XXX Here for now... not interested in restructing headers JUST now */
+
+/* data type for block offset of block group */
+typedef int ext2_grpblk_t;
+
+/* data type for filesystem-wide blocks number */
+typedef unsigned long ext2_fsblk_t;
+
+#define E2FSBLK "%lu"
+
+struct ext2_reserve_window {
+	ext2_fsblk_t		_rsv_start;	/* First byte reserved */
+	ext2_fsblk_t		_rsv_end;	/* Last byte reserved or 0 */
+};
+
+struct ext2_reserve_window_node {
+	struct rb_node	 	rsv_node;
+	__u32			rsv_goal_size;
+	__u32			rsv_alloc_hit;
+	struct ext2_reserve_window	rsv_window;
+};
+
+struct ext2_block_alloc_info {
+	/* information about reservation window */
+	struct ext2_reserve_window_node	rsv_window_node;
+	/*
+	 * was i_next_alloc_block in ext2_inode_info
+	 * is the logical (file-relative) number of the
+	 * most-recently-allocated block in this file.
+	 * We use this for detecting linearly ascending allocation requests.
+	 */
+	__u32			last_alloc_logical_block;
+	/*
+	 * Was i_next_alloc_goal in ext2_inode_info
+	 * is the *physical* companion to i_next_alloc_block.
+	 * it the the physical block number of the block which was most-recentl
+	 * allocated to this file.  This give us the goal (target) for the next
+	 * allocation when we detect linearly ascending requests.
+	 */
+	ext2_fsblk_t		last_alloc_physical_block;
+};
+
+#define rsv_start rsv_window._rsv_start
+#define rsv_end rsv_window._rsv_end
 
 /*
  * second extended-fs super-block data in memory
@@ -56,6 +102,10 @@ struct ext2_sb_info {
 	struct percpu_counter s_freeinodes_counter;
 	struct percpu_counter s_dirs_counter;
 	struct blockgroup_lock s_blockgroup_lock;
+	/* root of the per fs reservation window tree */
+	spinlock_t s_rsv_window_lock;
+	struct rb_root s_rsv_window_root;
+	struct ext2_reserve_window_node s_rsv_window_head;
 };
 
 #endif	/* _LINUX_EXT2_FS_SB */