]> www.pilppa.org Git - linux-2.6-omap-h63xx.git/commitdiff
[XFS] Don't update i_size for directories and special files
authorChristoph Hellwig <hch@infradead.org>
Mon, 23 Jun 2008 03:25:25 +0000 (13:25 +1000)
committerNiv Sardi <xaiki@debian.org>
Mon, 28 Jul 2008 06:59:00 +0000 (16:59 +1000)
The core kernel uses vfs_getattr to look at the inode size and similar
attributes, so there is no need to keep i_size uptodate for directories or
special files. This means we can remove xfs_validate_fields because the
I/O path already keeps i_size uptodate for regular files.

SGI-PV: 981498

SGI-Modid: xfs-linux-melb:xfs-kern:31336a

Signed-off-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Barry Naujok <bnaujok@sgi.com>
Signed-off-by: Lachlan McIlroy <lachlan@sgi.com>
fs/xfs/linux-2.6/xfs_iops.c

index 1f89c19cd4c45198ee53fddf1c311a067eeeac57..7b42569968fe1aff876cb46d129ca438b5602749 100644 (file)
@@ -181,23 +181,6 @@ xfs_ichgtime_fast(
                mark_inode_dirty_sync(inode);
 }
 
-
-/*
- * Pull the link count and size up from the xfs inode to the linux inode
- */
-STATIC void
-xfs_validate_fields(
-       struct inode            *inode)
-{
-       struct xfs_inode        *ip = XFS_I(inode);
-       loff_t size;
-
-       /* we're under i_sem so i_size can't change under us */
-       size = XFS_ISIZE(ip);
-       if (i_size_read(inode) != size)
-               i_size_write(inode, size);
-}
-
 /*
  * Hook in SELinux.  This is not quite correct yet, what we really need
  * here (as we do for default ACLs) is a mechanism by which creation of
@@ -331,10 +314,7 @@ xfs_vn_mknod(
        }
 
 
-       if (S_ISDIR(mode))
-               xfs_validate_fields(inode);
        d_instantiate(dentry, inode);
-       xfs_validate_fields(dir);
        return -error;
 
  out_cleanup_inode:
@@ -450,7 +430,6 @@ xfs_vn_link(
        }
 
        xfs_iflags_set(XFS_I(dir), XFS_IMODIFIED);
-       xfs_validate_fields(inode);
        d_instantiate(dentry, inode);
        return 0;
 }
@@ -460,26 +439,23 @@ xfs_vn_unlink(
        struct inode    *dir,
        struct dentry   *dentry)
 {
-       struct inode    *inode;
        struct xfs_name name;
        int             error;
 
-       inode = dentry->d_inode;
        xfs_dentry_to_name(&name, dentry);
 
-       error = xfs_remove(XFS_I(dir), &name, XFS_I(inode));
-       if (likely(!error)) {
-               xfs_validate_fields(dir);       /* size needs update */
-               xfs_validate_fields(inode);
-               /*
-                * With unlink, the VFS makes the dentry "negative": no inode,
-                * but still hashed. This is incompatible with case-insensitive
-                * mode, so invalidate (unhash) the dentry in CI-mode.
-                */
-               if (xfs_sb_version_hasasciici(&XFS_M(dir->i_sb)->m_sb))
-                       d_invalidate(dentry);
-       }
-       return -error;
+       error = -xfs_remove(XFS_I(dir), &name, XFS_I(dentry->d_inode));
+       if (error)
+               return error;
+
+       /*
+        * With unlink, the VFS makes the dentry "negative": no inode,
+        * but still hashed. This is incompatible with case-insensitive
+        * mode, so invalidate (unhash) the dentry in CI-mode.
+        */
+       if (xfs_sb_version_hasasciici(&XFS_M(dir->i_sb)->m_sb))
+               d_invalidate(dentry);
+       return 0;
 }
 
 STATIC int
@@ -509,8 +485,6 @@ xfs_vn_symlink(
                goto out_cleanup_inode;
 
        d_instantiate(dentry, inode);
-       xfs_validate_fields(dir);
-       xfs_validate_fields(inode);
        return 0;
 
  out_cleanup_inode:
@@ -529,22 +503,13 @@ xfs_vn_rename(
        struct inode    *new_inode = ndentry->d_inode;
        struct xfs_name oname;
        struct xfs_name nname;
-       int             error;
 
        xfs_dentry_to_name(&oname, odentry);
        xfs_dentry_to_name(&nname, ndentry);
 
-       error = xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode),
+       return -xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode),
                           XFS_I(ndir), &nname, new_inode ?
                                                XFS_I(new_inode) : NULL);
-       if (likely(!error)) {
-               if (new_inode)
-                       xfs_validate_fields(new_inode);
-               xfs_validate_fields(odir);
-               if (ndir != odir)
-                       xfs_validate_fields(ndir);
-       }
-       return -error;
 }
 
 /*