Patch turns s_lock+s_wait combination into semaphore.
wait_on_super() is dead. lock_super() is down(&sb->s_lock),
unlock_super() is up(...).
One place is still ugly - get_super(), but that'll have to
wait until we add reference counters on superblocks. For the time being
get_super() behaviour stays unchanged (i.e. all races with umount()
are still there).
Please, apply it - it doesn't break anything in tree,
unlikely to break anything 3rd-party and any potential breakage there
will be caught by compiler.
Al
diff -urN S5-pre2/fs/ext2/super.c S5-pre2-s_lock/fs/ext2/super.c
--- S5-pre2/fs/ext2/super.c Sat Apr 28 02:12:56 2001
+++ S5-pre2-s_lock/fs/ext2/super.c Tue May 15 18:12:35 2001
@@ -76,9 +76,6 @@
va_start (args, fmt);
vsprintf (error_buf, fmt, args);
va_end (args);
- /* this is to prevent panic from syncing this filesystem */
- if (sb->s_lock)
- sb->s_lock=0;
sb->s_flags |= MS_RDONLY;
panic ("EXT2-fs panic (device %s): %s: %s\n",
bdevname(sb->s_dev), function, error_buf);
diff -urN S5-pre2/fs/reiserfs/prints.c S5-pre2-s_lock/fs/reiserfs/prints.c
--- S5-pre2/fs/reiserfs/prints.c Sat Apr 28 02:12:56 2001
+++ S5-pre2-s_lock/fs/reiserfs/prints.c Tue May 15 18:12:35 2001
@@ -349,8 +349,6 @@
#endif
/* this is to prevent panic from syncing this filesystem */
- if (sb && sb->s_lock)
- sb->s_lock=0;
if (sb)
sb->s_flags |= MS_RDONLY;
diff -urN S5-pre2/fs/super.c S5-pre2-s_lock/fs/super.c
--- S5-pre2/fs/super.c Tue May 15 03:51:04 2001
+++ S5-pre2-s_lock/fs/super.c Tue May 15 18:12:35 2001
@@ -580,30 +580,7 @@
#undef MANGLE
#undef FREEROOM
}
-
-/**
- * __wait_on_super - wait on a superblock
- * @sb: superblock to wait on
- *
- * Waits for a superblock to become unlocked and then returns. It does
- * not take the lock. This is an internal function. See wait_on_super().
- */
-void __wait_on_super(struct super_block * sb)
-{
- DECLARE_WAITQUEUE(wait, current);
-
- add_wait_queue(&sb->s_wait, &wait);
-repeat:
- set_current_state(TASK_UNINTERRUPTIBLE);
- if (sb->s_lock) {
- schedule();
- goto repeat;
- }
- remove_wait_queue(&sb->s_wait, &wait);
- current->state = TASK_RUNNING;
-}
-
/*
* Note: check the dirty flag before waiting, so we don't
* hold up the sync while mounting a device. (The newly
@@ -648,7 +625,9 @@
s = sb_entry(super_blocks.next);
while (s != sb_entry(&super_blocks))
if (s->s_dev == dev) {
- wait_on_super(s);
+ /* Yes, it sucks. As soon as we get refcounting... */
+ lock_super(s);
+ unlock_super(s);
if (s->s_dev == dev)
return s;
goto restart;
@@ -700,9 +679,7 @@
s = sb_entry(s->s_list.next)) {
if (s->s_dev)
continue;
- if (!s->s_lock)
- return s;
- printk("VFS: empty superblock %p locked!\n", s);
+ return s;
}
/* Need a new one... */
if (nr_super_blocks >= max_super_blocks)
@@ -714,10 +691,14 @@
INIT_LIST_HEAD(&s->s_dirty);
INIT_LIST_HEAD(&s->s_locked_inodes);
list_add (&s->s_list, super_blocks.prev);
- init_waitqueue_head(&s->s_wait);
INIT_LIST_HEAD(&s->s_files);
INIT_LIST_HEAD(&s->s_mounts);
init_rwsem(&s->s_umount);
+ sema_init(&s->s_lock, 1);
+ sema_init(&s->s_vfs_rename_sem,1);
+ sema_init(&s->s_nfsd_free_path_sem,1);
+ sema_init(&s->s_dquot.dqio_sem, 1);
+ sema_init(&s->s_dquot.dqoff_sem, 1);
}
return s;
}
@@ -734,11 +715,7 @@
s->s_bdev = bdev;
s->s_flags = flags;
s->s_dirt = 0;
- sema_init(&s->s_vfs_rename_sem,1);
- sema_init(&s->s_nfsd_free_path_sem,1);
s->s_type = type;
- sema_init(&s->s_dquot.dqio_sem, 1);
- sema_init(&s->s_dquot.dqoff_sem, 1);
s->s_dquot.flags = 0;
s->s_maxbytes = MAX_NON_LFS;
lock_super(s);
diff -urN S5-pre2/fs/ufs/super.c S5-pre2-s_lock/fs/ufs/super.c
--- S5-pre2/fs/ufs/super.c Sat Apr 28 02:12:56 2001
+++ S5-pre2-s_lock/fs/ufs/super.c Tue May 15 18:12:35 2001
@@ -230,9 +230,6 @@
va_start (args, fmt);
vsprintf (error_buf, fmt, args);
va_end (args);
- /* this is to prevent panic from syncing this filesystem */
- if (sb->s_lock)
- sb->s_lock = 0;
sb->s_flags |= MS_RDONLY;
printk (KERN_CRIT "UFS-fs panic (device %s): %s: %s\n",
kdevname(sb->s_dev), function, error_buf);
diff -urN S5-pre2/include/linux/fs.h S5-pre2-s_lock/include/linux/fs.h
--- S5-pre2/include/linux/fs.h Tue May 15 03:51:04 2001
+++ S5-pre2-s_lock/include/linux/fs.h Tue May 15 18:12:35 2001
@@ -652,7 +652,6 @@
kdev_t s_dev;
unsigned long s_blocksize;
unsigned char s_blocksize_bits;
- unsigned char s_lock;
unsigned char s_dirt;
unsigned long long s_maxbytes; /* Max file size */
struct file_system_type *s_type;
@@ -662,7 +661,7 @@
unsigned long s_magic;
struct dentry *s_root;
struct rw_semaphore s_umount;
- wait_queue_head_t s_wait;
+ struct semaphore s_lock;
struct list_head s_dirty; /* dirty inodes */
struct list_head s_locked_inodes;/* inodes being synced */
diff -urN S5-pre2/include/linux/locks.h S5-pre2-s_lock/include/linux/locks.h
--- S5-pre2/include/linux/locks.h Thu May 3 15:24:14 2001
+++ S5-pre2-s_lock/include/linux/locks.h Tue May 15 18:12:35 2001
@@ -39,30 +39,15 @@
* a super-block (although even this isn't done right now.
* nfs may need it).
*/
-extern void __wait_on_super(struct super_block *);
-
-extern inline void wait_on_super(struct super_block * sb)
-{
- if (sb->s_lock)
- __wait_on_super(sb);
-}
extern inline void lock_super(struct super_block * sb)
{
- if (sb->s_lock)
- __wait_on_super(sb);
- sb->s_lock = 1;
+ down(&sb->s_lock);
}
extern inline void unlock_super(struct super_block * sb)
{
- sb->s_lock = 0;
- /*
- * No need of any barrier, we're protected by
- * the big kernel lock here... unfortunately :)
- */
- if (waitqueue_active(&sb->s_wait))
- wake_up(&sb->s_wait);
+ up(&sb->s_lock);
}
#endif /* _LINUX_LOCKS_H */
diff -urN S5-pre2/kernel/ksyms.c S5-pre2-s_lock/kernel/ksyms.c
--- S5-pre2/kernel/ksyms.c Tue May 15 03:51:04 2001
+++ S5-pre2-s_lock/kernel/ksyms.c Tue May 15 18:12:35 2001
@@ -478,7 +478,6 @@
/* Added to make file system as module */
EXPORT_SYMBOL(sys_tz);
-EXPORT_SYMBOL(__wait_on_super);
EXPORT_SYMBOL(file_fsync);
EXPORT_SYMBOL(fsync_inode_buffers);
EXPORT_SYMBOL(clear_inode);
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
