ext4: Add multi block allocator for ext4

Signed-off-by: Alex Tomas <alex@clusterfs.com>
Signed-off-by: Andreas Dilger <adilger@clusterfs.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>

3 files changed, 131 insertions, 1 deletions

diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index d0b7ca99b91f..1852313fc7c7 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -20,6 +20,8 @@
 #include <linux/blkdev.h>
 #include <linux/magic.h>
 
+#include <linux/ext4_fs_i.h>
+
 /*
  * The second extended filesystem constants/structures
  */
@@ -51,6 +53,50 @@
 #define ext4_debug(f, a...)     do {} while (0)
 #endif
 
+#define EXT4_MULTIBLOCK_ALLOCATOR       1
+
+/* prefer goal again. length */
+#define EXT4_MB_HINT_MERGE              1
+/* blocks already reserved */
+#define EXT4_MB_HINT_RESERVED           2
+/* metadata is being allocated */
+#define EXT4_MB_HINT_METADATA           4
+/* first blocks in the file */
+#define EXT4_MB_HINT_FIRST              8
+/* search for the best chunk */
+#define EXT4_MB_HINT_BEST               16
+/* data is being allocated */
+#define EXT4_MB_HINT_DATA               32
+/* don't preallocate (for tails) */
+#define EXT4_MB_HINT_NOPREALLOC         64
+/* allocate for locality group */
+#define EXT4_MB_HINT_GROUP_ALLOC        128
+/* allocate goal blocks or none */
+#define EXT4_MB_HINT_GOAL_ONLY          256
+/* goal is meaningful */
+#define EXT4_MB_HINT_TRY_GOAL           512
+
+struct ext4_allocation_request {
+        /* target inode for block we're allocating */
+        struct inode *inode;
+        /* logical block in target inode */
+        ext4_lblk_t logical;
+        /* phys. target (a hint) */
+        ext4_fsblk_t goal;
+        /* the closest logical allocated block to the left */
+        ext4_lblk_t lleft;
+        /* phys. block for ^^^ */
+        ext4_fsblk_t pleft;
+        /* the closest logical allocated block to the right */
+        ext4_lblk_t lright;
+        /* phys. block for ^^^ */
+        ext4_fsblk_t pright;
+        /* how many blocks we want to allocate */
+        unsigned long len;
+        /* flags. see above EXT4_MB_HINT_* */
+        unsigned long flags;
+};
+
 /*
  * Special inodes numbers
  */
@@ -474,6 +520,7 @@ do {									       \
 #define EXT4_MOUNT_JOURNAL_CHECKSUM     0x800000 /* Journal checksums */
 #define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT 0x1000000 /* Journal Async Commit */
 #define EXT4_MOUNT_I_VERSION            0x2000000 /* i_version support */
+#define EXT4_MOUNT_MBALLOC              0x4000000 /* Buddy allocation support */
 /* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */
 #ifndef _LINUX_EXT2_FS_H
 #define clear_opt(o, opt)               o &= ~EXT4_MOUNT_##opt
@@ -912,7 +959,7 @@ extern ext4_fsblk_t ext4_new_blocks (handle_t *handle, struct inode *inode,
 extern ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode,
                         ext4_fsblk_t goal, unsigned long *count, int *errp);
 extern void ext4_free_blocks (handle_t *handle, struct inode *inode,
-                        ext4_fsblk_t block, unsigned long count);
+                        ext4_fsblk_t block, unsigned long count, int metadata);
 extern void ext4_free_blocks_sb (handle_t *handle, struct super_block *sb,
                                  ext4_fsblk_t block, unsigned long count,
                                 unsigned long *pdquot_freed_blocks);
@@ -950,6 +997,20 @@ extern unsigned long ext4_count_dirs (struct super_block *);
 extern void ext4_check_inodes_bitmap (struct super_block *);
 extern unsigned long ext4_count_free (struct buffer_head *, unsigned);
 
+/* mballoc.c */
+extern long ext4_mb_stats;
+extern long ext4_mb_max_to_scan;
+extern int ext4_mb_init(struct super_block *, int);
+extern int ext4_mb_release(struct super_block *);
+extern ext4_fsblk_t ext4_mb_new_blocks(handle_t *,
+                                struct ext4_allocation_request *, int *);
+extern int ext4_mb_reserve_blocks(struct super_block *, int);
+extern void ext4_mb_discard_inode_preallocations(struct inode *);
+extern int __init init_ext4_mballoc(void);
+extern void exit_ext4_mballoc(void);
+extern void ext4_mb_free_blocks(handle_t *, struct inode *,
+                unsigned long, unsigned long, int, unsigned long *);
+
 
 /* inode.c */
 int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode,
@@ -1080,6 +1141,19 @@ static inline void ext4_isize_set(struct ext4_inode *raw_inode, loff_t i_size)
         raw_inode->i_size_high = cpu_to_le32(i_size >> 32);
 }
 
+static inline
+struct ext4_group_info *ext4_get_group_info(struct super_block *sb,
+                                                        ext4_group_t group)
+{
+         struct ext4_group_info ***grp_info;
+         long indexv, indexh;
+         grp_info = EXT4_SB(sb)->s_group_info;
+         indexv = group >> (EXT4_DESC_PER_BLOCK_BITS(sb));
+         indexh = group & ((EXT4_DESC_PER_BLOCK(sb)) - 1);
+         return grp_info[indexv][indexh];
+}
+
+
 #define ext4_std_error(sb, errno)                               \
 do {                                                            \
         if ((errno))                                            \
diff --git a/include/linux/ext4_fs_i.h b/include/linux/ext4_fs_i.h
index 4377d249d378..d5508d3cf290 100644
--- a/include/linux/ext4_fs_i.h
+++ b/include/linux/ext4_fs_i.h
@@ -158,6 +158,10 @@ struct ext4_inode_info {
          * struct timespec i_{a,c,m}time in the generic inode.
          */
         struct timespec i_crtime;
+
+        /* mballoc */
+        struct list_head i_prealloc_list;
+        spinlock_t i_prealloc_lock;
 };
 
 #endif  /* _LINUX_EXT4_FS_I */
diff --git a/include/linux/ext4_fs_sb.h b/include/linux/ext4_fs_sb.h
index 38a47ec06df9..abaae2c8cccf 100644
--- a/include/linux/ext4_fs_sb.h
+++ b/include/linux/ext4_fs_sb.h
@@ -91,6 +91,58 @@ struct ext4_sb_info {
         unsigned long s_ext_blocks;
         unsigned long s_ext_extents;
 #endif
+
+        /* for buddy allocator */
+        struct ext4_group_info ***s_group_info;
+        struct inode *s_buddy_cache;
+        long s_blocks_reserved;
+        spinlock_t s_reserve_lock;
+        struct list_head s_active_transaction;
+        struct list_head s_closed_transaction;
+        struct list_head s_committed_transaction;
+        spinlock_t s_md_lock;
+        tid_t s_last_transaction;
+        unsigned short *s_mb_offsets, *s_mb_maxs;
+
+        /* tunables */
+        unsigned long s_stripe;
+        unsigned long s_mb_stream_request;
+        unsigned long s_mb_max_to_scan;
+        unsigned long s_mb_min_to_scan;
+        unsigned long s_mb_stats;
+        unsigned long s_mb_order2_reqs;
+        unsigned long s_mb_group_prealloc;
+        /* where last allocation was done - for stream allocation */
+        unsigned long s_mb_last_group;
+        unsigned long s_mb_last_start;
+
+        /* history to debug policy */
+        struct ext4_mb_history *s_mb_history;
+        int s_mb_history_cur;
+        int s_mb_history_max;
+        int s_mb_history_num;
+        struct proc_dir_entry *s_mb_proc;
+        spinlock_t s_mb_history_lock;
+        int s_mb_history_filter;
+
+        /* stats for buddy allocator */
+        spinlock_t s_mb_pa_lock;
+        atomic_t s_bal_reqs;    /* number of reqs with len > 1 */
+        atomic_t s_bal_success; /* we found long enough chunks */
+        atomic_t s_bal_allocated;       /* in blocks */
+        atomic_t s_bal_ex_scanned;      /* total extents scanned */
+        atomic_t s_bal_goals;   /* goal hits */
+        atomic_t s_bal_breaks;  /* too long searches */
+        atomic_t s_bal_2orders; /* 2^order hits */
+        spinlock_t s_bal_lock;
+        unsigned long s_mb_buddies_generated;
+        unsigned long long s_mb_generation_time;
+        atomic_t s_mb_lost_chunks;
+        atomic_t s_mb_preallocated;
+        atomic_t s_mb_discarded;
+
+        /* locality groups */
+        struct ext4_locality_group *s_locality_groups;
 };
 
 #endif  /* _LINUX_EXT4_FS_SB */