ext3-nanosecond-2.6-rhel4.patch

非常经典的一个分布式系统

+			if (sbi->s_want_extra_isize <
+			    le16_to_cpu(es->s_want_extra_isize))
+				sbi->s_want_extra_isize =
+					le16_to_cpu(es->s_want_extra_isize);
+			if (sbi->s_want_extra_isize <
+			    le16_to_cpu(es->s_min_extra_isize))
+				sbi->s_want_extra_isize =
+					le16_to_cpu(es->s_min_extra_isize);
+		}
+	}
+	/* Check if enough inode space is available */
+	if (EXT3_GOOD_OLD_INODE_SIZE + sbi->s_want_extra_isize >
+							sbi->s_inode_size) {
+		sbi->s_want_extra_isize = sizeof(struct ext3_inode) -
+						       EXT3_GOOD_OLD_INODE_SIZE;
+		printk(KERN_INFO "EXT3-fs: required extra inode space not"
+			"available.\n");
+	}
+
 	/*
 	 * akpm: core read_super() calls in here with the superblock locked.
 	 * That deadlocks, because orphan cleanup needs to lock the superblock
Index: linux-stage/fs/ext3/xattr.c
===================================================================
--- linux-stage.orig/fs/ext3/xattr.c
+++ linux-stage/fs/ext3/xattr.c
@@ -1305,7 +1305,7 @@ getblk_failed:
 
 	/* Update the inode. */
 	EXT3_I(inode)->i_file_acl = new_bh ? new_bh->b_blocknr : 0;
-	inode->i_ctime = CURRENT_TIME_SEC;
+	inode->i_ctime = ext3_current_time(inode);
 	ext3_mark_inode_dirty(handle, inode);
 	if (IS_SYNC(inode))
 		handle->h_sync = 1;
Index: linux-stage/include/linux/ext3_fs.h
===================================================================
--- linux-stage.orig/include/linux/ext3_fs.h
+++ linux-stage/include/linux/ext3_fs.h
@@ -280,7 +280,7 @@ struct ext3_inode {
 	__le16	i_uid;		/* Low 16 bits of Owner Uid */
 	__le32	i_size;		/* Size in bytes */
 	__le32	i_atime;	/* Access time */
-	__le32	i_ctime;	/* Creation time */
+	__le32	i_ctime;	/* Inode Change time */
 	__le32	i_mtime;	/* Modification time */
 	__le32	i_dtime;	/* Deletion Time */
 	__le16	i_gid;		/* Low 16 bits of Group Id */
@@ -329,10 +329,73 @@ struct ext3_inode {
 	} osd2;				/* OS dependent 2 */
 	__u16	i_extra_isize;
 	__u16	i_pad1;
+	__le32  i_ctime_extra;	/* extra Change time      (nsec << 2 | epoch) */
+	__le32  i_mtime_extra;	/* extra Modification time(nsec << 2 | epoch) */
+	__le32  i_atime_extra;	/* extra Access time      (nsec << 2 | epoch) */
+	__le32  i_crtime;	/* File Creation time */
+	__le32  i_crtime_extra;	/* extra File Creation time (nsec << 2 | epoch) */
 };
 
 #define i_size_high	i_dir_acl
 
+#define EXT3_EPOCH_BITS 2
+#define EXT3_EPOCH_MASK ((1 << EXT3_EPOCH_BITS) - 1)
+#define EXT3_NSEC_MASK  (~0UL << EXT3_EPOCH_BITS)
+
+#define EXT3_FITS_IN_INODE(ext3_inode, einode, field)   \
+	((offsetof(typeof(*ext3_inode), field) +	\
+	 sizeof((ext3_inode)->field))		 	\
+	 <= (EXT3_GOOD_OLD_INODE_SIZE +	       		\
+	     (einode)->i_extra_isize))			\
+
+static inline __le32 ext3_encode_extra_time(struct timespec *time)
+{
+	return cpu_to_le32((sizeof(time->tv_sec) > 4 ?
+			   time->tv_sec >> 32 : 0) |
+			   ((time->tv_nsec << 2) & EXT3_NSEC_MASK));
+}
+
+static inline void ext3_decode_extra_time(struct timespec *time, __le32 extra) {
+       if (sizeof(time->tv_sec) > 4)
+	       time->tv_sec |= (__u64)(le32_to_cpu(extra) & EXT3_EPOCH_MASK)
+				       << 32;
+       time->tv_nsec = (le32_to_cpu(extra) & EXT3_NSEC_MASK) >> 2;
+}
+
+#define EXT3_INODE_SET_XTIME(xtime, inode, raw_inode)			 \
+do {									 \
+       (raw_inode)->xtime = cpu_to_le32((inode)->xtime.tv_sec);		 \
+       if (EXT3_FITS_IN_INODE(raw_inode, EXT3_I(inode), xtime ## _extra))\
+	       (raw_inode)->xtime ## _extra =				 \
+			       ext3_encode_extra_time(&(inode)->xtime);  \
+} while (0)
+
+#define EXT3_EINODE_SET_XTIME(xtime, einode, raw_inode)			 \
+do {									 \
+       if (EXT3_FITS_IN_INODE(raw_inode, einode, xtime))		 \
+	       (raw_inode)->xtime = cpu_to_le32((einode)->xtime.tv_sec); \
+       if (EXT3_FITS_IN_INODE(raw_inode, einode, xtime ## _extra))	 \
+	       (raw_inode)->xtime ## _extra =				 \
+			       ext3_encode_extra_time(&(einode)->xtime); \
+} while (0)
+
+#define EXT3_INODE_GET_XTIME(xtime, inode, raw_inode)			 \
+do {									 \
+       (inode)->xtime.tv_sec = le32_to_cpu((raw_inode)->xtime);		 \
+       if (EXT3_FITS_IN_INODE(raw_inode, EXT3_I(inode), xtime ## _extra))\
+	       ext3_decode_extra_time(&(inode)->xtime,			 \
+				      raw_inode->xtime ## _extra);	 \
+} while (0)
+
+#define EXT3_EINODE_GET_XTIME(xtime, einode, raw_inode)		       	 \
+do {									 \
+       if (EXT3_FITS_IN_INODE(raw_inode, einode, xtime))		 \
+	       (einode)->xtime.tv_sec = le32_to_cpu((raw_inode)->xtime); \
+       if (EXT3_FITS_IN_INODE(raw_inode, einode, xtime ## _extra))	 \
+	       ext3_decode_extra_time(&(einode)->xtime,			 \
+				      raw_inode->xtime ## _extra);	 \
+} while (0)
+
 #if defined(__KERNEL__) || defined(__linux__)
 #define i_reserved1	osd1.linux1.l_i_reserved1
 #define i_frag		osd2.linux2.l_i_frag
@@ -500,11 +563,19 @@ struct ext3_super_block {
 	__le32	s_last_orphan;		/* start of list of inodes to delete */
 	__le32	s_hash_seed[4];		/* HTREE hash seed */
 	__u8	s_def_hash_version;	/* Default hash version to use */
-	__u8	s_reserved_char_pad;
-	__u16	s_reserved_word_pad;
+	__u8	s_jnl_backup_type;	/* Default type of journal backup */
+	__le16	s_desc_size;		/* Group desc. size: INCOMPAT_64BIT */
 	__le32	s_default_mount_opts;
-	__le32	s_first_meta_bg; 	/* First metablock block group */
-	__u32	s_reserved[190];	/* Padding to the end of the block */
+	__le32	s_first_meta_bg;	/* First metablock block group */
+	__le32	s_mkfs_time;		/* When the filesystem was created */
+	__le32	s_jnl_blocks[17];	/* Backup of the journal inode */
+	__le32	s_blocks_count_hi;	/* Blocks count high 32 bits */
+	__le32	s_r_blocks_count_hi;	/* Reserved blocks count high 32 bits*/
+	__le32	s_free_blocks_hi;	/* Free blocks count high 32 bits */
+	__le16	s_min_extra_isize;	/* All inodes have at least # bytes */
+	__le16	s_want_extra_isize;	/* New inodes should reserve # bytes */
+	__le32	s_flags;		/* Miscellaneous flags */
+	__u32	s_reserved[167];	/* Padding to the end of the block */
 };
 
 #ifdef __KERNEL__
@@ -580,6 +651,7 @@ static inline struct ext3_inode_info *EX
 #define EXT3_FEATURE_RO_COMPAT_BTREE_DIR	0x0004
 #define EXT4_FEATURE_RO_COMPAT_GDT_CSUM		0x0010
 #define EXT4_FEATURE_RO_COMPAT_DIR_NLINK	0x0020
+#define EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE	0x0040
 
 #define EXT3_FEATURE_INCOMPAT_COMPRESSION	0x0001
 #define EXT3_FEATURE_INCOMPAT_FILETYPE		0x0002
@@ -597,6 +669,7 @@ static inline struct ext3_inode_info *EX
 					 EXT3_FEATURE_RO_COMPAT_LARGE_FILE| \
 					 EXT4_FEATURE_RO_COMPAT_GDT_CSUM| \
 					 EXT4_FEATURE_RO_COMPAT_DIR_NLINK| \
+					 EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE| \
 					 EXT3_FEATURE_RO_COMPAT_BTREE_DIR)
 
 /*
@@ -724,6 +797,12 @@ static inline struct ext3_inode *ext3_ra
 	return (struct ext3_inode *) (iloc->bh->b_data + iloc->offset);
 }
 
+static inline struct timespec ext3_current_time(struct inode *inode)
+{
+	return (inode->i_sb->s_time_gran < NSEC_PER_SEC) ?
+		current_fs_time(inode->i_sb) : CURRENT_TIME_SEC;
+}
+
 /*
  * This structure is stuffed into the struct file's private_data field
  * for directories.  It is where we put information so that we can do
Index: linux-stage/include/linux/ext3_fs_i.h
===================================================================
--- linux-stage.orig/include/linux/ext3_fs_i.h
+++ linux-stage/include/linux/ext3_fs_i.h
@@ -130,6 +130,7 @@ struct ext3_inode_info {
 
 	/* on-disk additional length */
 	__u16 i_extra_isize;
+	struct timespec i_crtime;
 
 	/*
 	 * truncate_sem is for serialising ext3_truncate() against
Index: linux-stage/include/linux/ext3_fs_sb.h
===================================================================
--- linux-stage.orig/include/linux/ext3_fs_sb.h
+++ linux-stage/include/linux/ext3_fs_sb.h
@@ -71,6 +71,8 @@ struct ext3_sb_info {
 	/* Last group used to allocate inode */
 	int s_last_alloc_group;
 
+	unsigned int s_want_extra_isize; /* New inodes should reserve # bytes */
+
 	/* root of the per fs reservation window tree */
 	spinlock_t s_rsv_window_lock;
 	struct rb_root s_rsv_window_root;