[patch 2/3] dentry->d_count fixes

From: Maneesh Soni (maneesh@in.ibm.com)
Date: Thu Jun 19 2003 - 08:10:46 EST

- nfs_unlink() can race with lockless d_lookup() as d_lookup() can
  successfully lookup a dentry for which nfs_unlink() can assume that no one
  else is using and can go ahead and do nfs_safe_remove() on it. By using
  per dentry lock, it is solved as we d_lookup() will fail the lookup for
  unhashed dentries.

 fs/nfs/dir.c | 3 +++
 1 files changed, 3 insertions(+)

diff -puN fs/nfs/dir.c~nfs_unlink-d_count-fix fs/nfs/dir.c
--- linux-2.5.72-mm2/fs/nfs/dir.c~nfs_unlink-d_count-fix 2003-06-19 17:38:51.000000000 +0530
+++ linux-2.5.72-mm2-maneesh/fs/nfs/dir.c 2003-06-19 17:38:56.000000000 +0530
@@ -1015,7 +1015,9 @@ static int nfs_unlink(struct inode *dir,
 
         lock_kernel();
         spin_lock(&dcache_lock);
+ spin_lock(&dentry->d_lock);
         if (atomic_read(&dentry->d_count) > 1) {
+ spin_unlock(&dentry->d_lock);
                 spin_unlock(&dcache_lock);
                 error = nfs_sillyrename(dir, dentry);
                 unlock_kernel();
@@ -1025,6 +1027,7 @@ static int nfs_unlink(struct inode *dir,
                 __d_drop(dentry);
                 need_rehash = 1;
         }
+ spin_unlock(&dentry->d_lock);
         spin_unlock(&dcache_lock);
         error = nfs_safe_remove(dentry);
         if (!error)

_ 

-- 
Maneesh Soni
IBM Linux Technology Center, 
IBM India Software Lab, Bangalore.
Phone: +91-80-5044999 email: maneesh@in.ibm.com
http://lse.sourceforge.net/
-
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 : Mon Jun 23 2003 - 22:00:29 EST