[PATCH] FAT: the inode hash from per module to per sb (4/4)

From: OGAWA Hirofumi
Date: Sun Sep 05 2004 - 06:38:14 EST

Next message: Tonnerre: "Re: silent semantic changes with reiser4"
Previous message: OGAWA Hirofumi: "[PATCH] FAT: cache lock from per sb to per inode (3/4)"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Signed-off-by: OGAWA Hirofumi <hirofumi@xxxxxxxxxxxxxxxxxx>
---

fs/fat/fatfs_syms.c | 2 -
fs/fat/inode.c | 57 +++++++++++++++++++++++---------------------
include/linux/msdos_fs.h | 1
include/linux/msdos_fs_sb.h | 7 +++++
4 files changed, 39 insertions(+), 28 deletions(-)

diff -puN fs/fat/fatfs_syms.c~fat_per-sb-inodehash fs/fat/fatfs_syms.c
--- linux-2.6.9-rc1/fs/fat/fatfs_syms.c~fat_per-sb-inodehash 2004-09-05 18:52:34.000000000 +0900
+++ linux-2.6.9-rc1-hirofumi/fs/fat/fatfs_syms.c 2004-09-05 18:52:34.000000000 +0900
@@ -40,7 +40,7 @@ void __exit fat_destroy_inodecache(void)
static int __init init_fat_fs(void)
{
int ret;
- fat_hash_init();
+
ret = fat_cache_init();
if (ret < 0)
return ret;
diff -puN fs/fat/inode.c~fat_per-sb-inodehash fs/fat/inode.c
--- linux-2.6.9-rc1/fs/fat/inode.c~fat_per-sb-inodehash 2004-09-05 18:52:34.000000000 +0900
+++ linux-2.6.9-rc1-hirofumi/fs/fat/inode.c 2004-09-05 18:54:16.000000000 +0900
@@ -55,17 +55,14 @@ static char fat_default_iocharset[] = CO
* and consider negative result as cache miss.
*/

-#define FAT_HASH_BITS 8
-#define FAT_HASH_SIZE (1UL << FAT_HASH_BITS)
-#define FAT_HASH_MASK (FAT_HASH_SIZE-1)
-static struct hlist_head fat_inode_hashtable[FAT_HASH_SIZE];
-static spinlock_t fat_inode_lock = SPIN_LOCK_UNLOCKED;
-
-void fat_hash_init(void)
+static void fat_hash_init(struct super_block *sb)
{
+ struct msdos_sb_info *sbi = MSDOS_SB(sb);
int i;
+
+ spin_lock_init(&sbi->inode_hash_lock);
for (i = 0; i < FAT_HASH_SIZE; i++)
- INIT_HLIST_HEAD(&fat_inode_hashtable[i]);
+ INIT_HLIST_HEAD(&sbi->inode_hashtable[i]);
}

static inline unsigned long fat_hash(struct super_block *sb, loff_t i_pos)
@@ -77,39 +74,43 @@ static inline unsigned long fat_hash(str

void fat_attach(struct inode *inode, loff_t i_pos)
{
- spin_lock(&fat_inode_lock);
+ struct super_block *sb = inode->i_sb;
+ struct msdos_sb_info *sbi = MSDOS_SB(sb);
+
+ spin_lock(&sbi->inode_hash_lock);
MSDOS_I(inode)->i_pos = i_pos;
hlist_add_head(&MSDOS_I(inode)->i_fat_hash,
- fat_inode_hashtable + fat_hash(inode->i_sb, i_pos));
- spin_unlock(&fat_inode_lock);
+ sbi->inode_hashtable + fat_hash(sb, i_pos));
+ spin_unlock(&sbi->inode_hash_lock);
}

void fat_detach(struct inode *inode)
{
- spin_lock(&fat_inode_lock);
+ struct msdos_sb_info *sbi = MSDOS_SB(inode->i_sb);
+ spin_lock(&sbi->inode_hash_lock);
MSDOS_I(inode)->i_pos = 0;
hlist_del_init(&MSDOS_I(inode)->i_fat_hash);
- spin_unlock(&fat_inode_lock);
+ spin_unlock(&sbi->inode_hash_lock);
}

struct inode *fat_iget(struct super_block *sb, loff_t i_pos)
{
- struct hlist_head *head = fat_inode_hashtable + fat_hash(sb, i_pos);
+ struct msdos_sb_info *sbi = MSDOS_SB(sb);
+ struct hlist_head *head = sbi->inode_hashtable + fat_hash(sb, i_pos);
struct hlist_node *_p;
struct msdos_inode_info *i;
struct inode *inode = NULL;

- spin_lock(&fat_inode_lock);
+ spin_lock(&sbi->inode_hash_lock);
hlist_for_each_entry(i, _p, head, i_fat_hash) {
- if (i->vfs_inode.i_sb != sb)
- continue;
+ BUG_ON(i->vfs_inode.i_sb != sb);
if (i->i_pos != i_pos)
continue;
inode = igrab(&i->vfs_inode);
if (inode)
break;
}
- spin_unlock(&fat_inode_lock);
+ spin_unlock(&sbi->inode_hash_lock);
return inode;
}

@@ -152,13 +153,15 @@ void fat_delete_inode(struct inode *inod

void fat_clear_inode(struct inode *inode)
{
+ struct msdos_sb_info *sbi = MSDOS_SB(inode->i_sb);
+
if (is_bad_inode(inode))
return;
lock_kernel();
- spin_lock(&fat_inode_lock);
+ spin_lock(&sbi->inode_hash_lock);
fat_cache_inval_inode(inode);
hlist_del_init(&MSDOS_I(inode)->i_fat_hash);
- spin_unlock(&fat_inode_lock);
+ spin_unlock(&sbi->inode_hash_lock);
unlock_kernel();
}

@@ -819,6 +822,7 @@ int fat_fill_super(struct super_block *s
goto out_fail;

/* set up enough so that it can read an inode */
+ fat_hash_init(sb);
init_MUTEX(&sbi->fat_lock);

error = -EIO;
@@ -1228,6 +1232,7 @@ static int fat_fill_inode(struct inode *
void fat_write_inode(struct inode *inode, int wait)
{
struct super_block *sb = inode->i_sb;
+ struct msdos_sb_info *sbi = MSDOS_SB(sb);
struct buffer_head *bh;
struct msdos_dir_entry *raw_entry;
loff_t i_pos;
@@ -1238,22 +1243,22 @@ retry:
return;
}
lock_kernel();
- if (!(bh = sb_bread(sb, i_pos >> MSDOS_SB(sb)->dir_per_block_bits))) {
+ if (!(bh = sb_bread(sb, i_pos >> sbi->dir_per_block_bits))) {
printk(KERN_ERR "FAT: unable to read inode block "
"for updating (i_pos %lld)\n", i_pos);
unlock_kernel();
return /* -EIO */;
}
- spin_lock(&fat_inode_lock);
+ spin_lock(&sbi->inode_hash_lock);
if (i_pos != MSDOS_I(inode)->i_pos) {
- spin_unlock(&fat_inode_lock);
+ spin_unlock(&sbi->inode_hash_lock);
brelse(bh);
unlock_kernel();
goto retry;
}

raw_entry = &((struct msdos_dir_entry *) (bh->b_data))
- [i_pos & (MSDOS_SB(sb)->dir_per_block - 1)];
+ [i_pos & (sbi->dir_per_block - 1)];
if (S_ISDIR(inode->i_mode)) {
raw_entry->attr = ATTR_DIR;
raw_entry->size = 0;
@@ -1267,11 +1272,11 @@ retry:
raw_entry->start = CT_LE_W(MSDOS_I(inode)->i_logstart);
raw_entry->starthi = CT_LE_W(MSDOS_I(inode)->i_logstart >> 16);
fat_date_unix2dos(inode->i_mtime.tv_sec, &raw_entry->time, &raw_entry->date);
- if (MSDOS_SB(sb)->options.isvfat) {
+ if (sbi->options.isvfat) {
fat_date_unix2dos(inode->i_ctime.tv_sec,&raw_entry->ctime,&raw_entry->cdate);
raw_entry->ctime_ms = MSDOS_I(inode)->i_ctime_ms; /* use i_ctime.tv_nsec? */
}
- spin_unlock(&fat_inode_lock);
+ spin_unlock(&sbi->inode_hash_lock);
mark_buffer_dirty(bh);
brelse(bh);
unlock_kernel();
diff -puN include/linux/msdos_fs.h~fat_per-sb-inodehash include/linux/msdos_fs.h
--- linux-2.6.9-rc1/include/linux/msdos_fs.h~fat_per-sb-inodehash 2004-09-05 18:52:34.000000000 +0900
+++ linux-2.6.9-rc1-hirofumi/include/linux/msdos_fs.h 2004-09-05 18:52:34.000000000 +0900
@@ -262,7 +262,6 @@ extern int fat_get_block(struct inode *i
extern void fat_truncate(struct inode *inode);

/* fat/inode.c */
-extern void fat_hash_init(void);
extern void fat_attach(struct inode *inode, loff_t i_pos);
extern void fat_detach(struct inode *inode);
extern struct inode *fat_iget(struct super_block *sb, loff_t i_pos);
diff -puN include/linux/msdos_fs_sb.h~fat_per-sb-inodehash include/linux/msdos_fs_sb.h
--- linux-2.6.9-rc1/include/linux/msdos_fs_sb.h~fat_per-sb-inodehash 2004-09-05 18:52:34.000000000 +0900
+++ linux-2.6.9-rc1-hirofumi/include/linux/msdos_fs_sb.h 2004-09-05 18:52:34.000000000 +0900
@@ -26,6 +26,10 @@ struct fat_mount_options {
nocase:1; /* Does this need case conversion? 0=need case conversion*/
};

+#define FAT_HASH_BITS 8
+#define FAT_HASH_SIZE (1UL << FAT_HASH_BITS)
+#define FAT_HASH_MASK (FAT_HASH_SIZE-1)
+
struct msdos_sb_info {
unsigned short sec_per_clus; /* sectors/cluster */
unsigned short cluster_bits; /* log2(cluster_size) */
@@ -48,6 +52,9 @@ struct msdos_sb_info {
void *dir_ops; /* Opaque; default directory operations */
int dir_per_block; /* dir entries per block */
int dir_per_block_bits; /* log2(dir_per_block) */
+
+ spinlock_t inode_hash_lock;
+ struct hlist_head inode_hashtable[FAT_HASH_SIZE];
};

#endif
_

--
OGAWA Hirofumi <hirofumi@xxxxxxxxxxxxxxxxxx>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Next message: Tonnerre: "Re: silent semantic changes with reiser4"
Previous message: OGAWA Hirofumi: "[PATCH] FAT: cache lock from per sb to per inode (3/4)"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]