Hi Nick,

Today's linux-next merge of the vfs-scale tree got a conflict in
fs/9p/vfs_inode.c between commit 3d21652a1d23591e7f0bbbbedae29ce78c2c1113
("fs/9p: Move dotl inode operations into a seperate file") from the v9fs
tree and various commits from the vfs-scale tree.

I fixed it up by using the v9fs changes to that file and then applying
the following merge fixup patch (which I can carry as necessary).

Someone will need to fix this up before one of these trees is merged by
Linus, or to send this merge fix to Linus.

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 4 Jan 2011 12:33:54 +1100
Subject: [PATCH] v9fs: merge fix for changes in the vfs-scale tree

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 fs/9p/vfs_inode_dotl.c |   22 +++++++++++-----------
 1 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c
index 38d5880..9dd534b 100644
--- a/fs/9p/vfs_inode_dotl.c
+++ b/fs/9p/vfs_inode_dotl.c
@@ -78,11 +78,11 @@ static struct dentry *v9fs_dentry_from_dir_inode(struct inode *inode)
 {
 	struct dentry *dentry;
 
-	spin_lock(&dcache_lock);
+	spin_lock(&inode->i_lock);
 	/* Directory should have only one entry. */
 	BUG_ON(S_ISDIR(inode->i_mode) && !list_is_singular(&inode->i_dentry));
 	dentry = list_entry(inode->i_dentry.next, struct dentry, d_alias);
-	spin_unlock(&dcache_lock);
+	spin_unlock(&inode->i_lock);
 	return dentry;
 }
 
@@ -215,7 +215,7 @@ v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int omode,
 				err);
 			goto error;
 		}
-		dentry->d_op = &v9fs_cached_dentry_operations;
+		d_set_d_op(dentry, &v9fs_cached_dentry_operations);
 		d_instantiate(dentry, inode);
 		err = v9fs_fid_add(dentry, fid);
 		if (err < 0)
@@ -233,7 +233,7 @@ v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int omode,
 			err = PTR_ERR(inode);
 			goto error;
 		}
-		dentry->d_op = &v9fs_dentry_operations;
+		d_set_d_op(dentry, &v9fs_dentry_operations);
 ...
[ message continues ]

If Linus does pull vfs-scale, and if it is before the v9fs tree, I can
help with the conversion. Otherwise I will be happy to fix up fallout
in my tree.

Thanks,
Nick

--

[ message continues ]

Are we doing away with dcache_lock?

I am not sure if the i_lock can serve the same purpose..but looks like with the
current code
there may not need any lock around this code. Aneesh/Eric do you guys have any

Assuming that this is a macro to the same operation.. rest of the changes look
fine to me.

Thanks,


--

[ message continues ]