[PATCH] fs/super.c fixes - second series (1/11)

From: Alexander Viro (viro@math.psu.edu)
Date: Sun Aug 12 2001 - 20:25:58 EST

        Since the second part of fs/super.c fixes (finally killing the
get_super() races) got public testing in -ac (since 2.4.7-ac6 - i.e. it had
been a week with 0 complaints so far), how about I start feeding it
right now? That one removes the cruft from FS_SINGLE handling, gives
the final variant of refcounting on superblocks (not dependent upon
mount_sem) and has final (as far as I'm concerned) exclusion on
->s_umount, which closes {u,}mount/get_super races for good.

Basically, after that point we will have sane mount handling. There are
things that can be cleaned up after that, but that's it - trivial cleanups.
Neurosurgery ends here.

Part 1/11

We grab exclusive lock on ->s_umount() in get_sb_...() and release it
once vfsmount is formed (by do_add_mount()). No deadlocks, since all
activity in protected area is already not allowed to lead to anything
that would grab ->s_umount. We hold ->s_lock over the whole non-trivial part
and anything of that kind would deadlock on it.

We are almost done with the get_super() races by now.

diff -urN S9-pre1/fs/super.c S9-pre1-s_umount/fs/super.c
--- S9-pre1/fs/super.c Sat Aug 11 14:59:24 2001
+++ S9-pre1-s_umount/fs/super.c Sun Aug 12 20:45:49 2001
@@ -820,6 +820,7 @@
         spin_lock(&sb_lock);
         list_add (&s->s_list, super_blocks.prev);
         spin_unlock(&sb_lock);
+ down_write(&s->s_umount);
         lock_super(s);
         if (!type->read_super(s, data, silent))
                 goto out_fail;
@@ -839,7 +840,7 @@
         spin_lock(&sb_lock);
         list_del(&s->s_list);
         spin_unlock(&sb_lock);
- __put_super(s);
+ put_super(s);
         return NULL;
 }
 
@@ -919,7 +920,9 @@
                                 spin_unlock(&sb_lock);
                         }
                         atomic_inc(&sb->s_active);
+ /* Next chunk will drop it */
                         up_read(&sb->s_umount);
+ down_write(&sb->s_umount);
                         path_release(&nd);
                         return sb;
                 }
@@ -986,6 +989,7 @@
                 BUG();
         atomic_inc(&sb->s_active);
         do_remount_sb(sb, flags, data);
+ down_write(&sb->s_umount);
         return sb;
 }
 
@@ -1104,6 +1108,7 @@
         mnt->mnt_mountpoint = mnt->mnt_root;
         mnt->mnt_parent = mnt;
         type->kern_mnt = mnt;
+ up_write(&sb->s_umount);
         return mnt;
 }
 
@@ -1379,6 +1384,7 @@
         mnt->mnt_root = dget(sb->s_root);
         mnt->mnt_mountpoint = mnt->mnt_root;
         mnt->mnt_parent = mnt;
+ up_write(&sb->s_umount);
 
         /* Something was mounted here while we slept */
         while(d_mountpoint(nd->dentry) && follow_down(&nd->mnt, &nd->dentry))
@@ -1639,6 +1645,7 @@
                 fs_type = sb->s_type;
                 atomic_inc(&sb->s_active);
                 up_read(&sb->s_umount);
+ down_write(&sb->s_umount);
                 goto mount_it;
         }
 
@@ -1659,6 +1666,8 @@
         panic("VFS: Unable to mount root fs on %s", kdevname(ROOT_DEV));
 
 mount_it:
+ /* FIXME */
+ up_write(&sb->s_umount);
         printk ("VFS: Mounted root (%s filesystem)%s.\n",
                 fs_type->name,
                 (sb->s_flags & MS_RDONLY) ? " readonly" : "");

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

This archive was generated by hypermail 2b29 : Wed Aug 15 2001 - 21:00:43 EST