12 files changed, 89 insertions, 29 deletions

diff --git a/fs/dcache.c b/fs/dcache.c
index 37f72ee5bf7c..6e4ea6d87774 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -2213,14 +2213,15 @@ static void dentry_unlock_parents_for_move(struct dentry *dentry,
  * The hash value has to match the hash queue that the dentry is on..
  */
 /*
- * d_move - move a dentry
+ * __d_move - move a dentry
  * @dentry: entry to move
  * @target: new dentry
  *
  * Update the dcache to reflect the move of a file name. Negative
- * dcache entries should not be moved in this way.
+ * dcache entries should not be moved in this way.  Caller hold
+ * rename_lock.
  */
-void d_move(struct dentry * dentry, struct dentry * target)
+static void __d_move(struct dentry * dentry, struct dentry * target)
 {
         if (!dentry->d_inode)
                 printk(KERN_WARNING "VFS: moving negative dcache entry\n");
@@ -2228,8 +2229,6 @@ void d_move(struct dentry * dentry, struct dentry * target)
         BUG_ON(d_ancestor(dentry, target));
         BUG_ON(d_ancestor(target, dentry));
 
-        write_seqlock(&rename_lock);
-
         dentry_lock_for_move(dentry, target);
 
         write_seqcount_begin(&dentry->d_seq);
@@ -2275,6 +2274,20 @@ void d_move(struct dentry * dentry, struct dentry * target)
         spin_unlock(&target->d_lock);
         fsnotify_d_move(dentry);
         spin_unlock(&dentry->d_lock);
+}
+
+/*
+ * d_move - move a dentry
+ * @dentry: entry to move
+ * @target: new dentry
+ *
+ * Update the dcache to reflect the move of a file name. Negative
+ * dcache entries should not be moved in this way.
+ */
+void d_move(struct dentry *dentry, struct dentry *target)
+{
+        write_seqlock(&rename_lock);
+        __d_move(dentry, target);
         write_sequnlock(&rename_lock);
 }
 EXPORT_SYMBOL(d_move);
@@ -2302,7 +2315,7 @@ struct dentry *d_ancestor(struct dentry *p1, struct dentry *p2)
  * This helper attempts to cope with remotely renamed directories
  *
  * It assumes that the caller is already holding
- * dentry->d_parent->d_inode->i_mutex and the inode->i_lock
+ * dentry->d_parent->d_inode->i_mutex, inode->i_lock and rename_lock
  *
  * Note: If ever the locking in lock_rename() changes, then please
  * remember to update this too...
@@ -2317,11 +2330,6 @@ static struct dentry *__d_unalias(struct inode *inode,
         if (alias->d_parent == dentry->d_parent)
                 goto out_unalias;
 
-        /* Check for loops */
-        ret = ERR_PTR(-ELOOP);
-        if (d_ancestor(alias, dentry))
-                goto out_err;
-
         /* See lock_rename() */
         ret = ERR_PTR(-EBUSY);
         if (!mutex_trylock(&dentry->d_sb->s_vfs_rename_mutex))
@@ -2331,7 +2339,7 @@ static struct dentry *__d_unalias(struct inode *inode,
                 goto out_err;
         m2 = &alias->d_parent->d_inode->i_mutex;
 out_unalias:
-        d_move(alias, dentry);
+        __d_move(alias, dentry);
         ret = alias;
 out_err:
         spin_unlock(&inode->i_lock);
@@ -2416,15 +2424,24 @@ struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode)
                 alias = __d_find_alias(inode, 0);
                 if (alias) {
                         actual = alias;
-                        /* Is this an anonymous mountpoint that we could splice
-                         * into our tree? */
-                        if (IS_ROOT(alias)) {
+                        write_seqlock(&rename_lock);
+
+                        if (d_ancestor(alias, dentry)) {
+                                /* Check for loops */
+                                actual = ERR_PTR(-ELOOP);
+                        } else if (IS_ROOT(alias)) {
+                                /* Is this an anonymous mountpoint that we
+                                 * could splice into our tree? */
                                 __d_materialise_dentry(dentry, alias);
+                                write_sequnlock(&rename_lock);
                                 __d_drop(alias);
                                 goto found;
+                        } else {
+                                /* Nope, but we must(!) avoid directory
+                                 * aliasing */
+                                actual = __d_unalias(inode, dentry, alias);
                         }
-                        /* Nope, but we must(!) avoid directory aliasing */
-                        actual = __d_unalias(inode, dentry, alias);
+                        write_sequnlock(&rename_lock);
                         if (IS_ERR(actual))
                                 dput(alias);
                         goto out_nolock;
diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c
index 802ac5eeba28..f9fbbe96c222 100644
--- a/fs/gfs2/aops.c
+++ b/fs/gfs2/aops.c
@@ -1069,6 +1069,7 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask)
                 return 0;
 
         gfs2_log_lock(sdp);
+        spin_lock(&sdp->sd_ail_lock);
         head = bh = page_buffers(page);
         do {
                 if (atomic_read(&bh->b_count))
@@ -1080,6 +1081,7 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask)
                         goto not_possible;
                 bh = bh->b_this_page;
         } while(bh != head);
+        spin_unlock(&sdp->sd_ail_lock);
         gfs2_log_unlock(sdp);
 
         head = bh = page_buffers(page);
@@ -1112,6 +1114,7 @@ not_possible: /* Should never happen */
         WARN_ON(buffer_dirty(bh));
         WARN_ON(buffer_pinned(bh));
 cannot_release:
+        spin_unlock(&sdp->sd_ail_lock);
         gfs2_log_unlock(sdp);
         return 0;
 }
diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
index 8ef70f464731..2cca29316bd6 100644
--- a/fs/gfs2/glops.c
+++ b/fs/gfs2/glops.c
@@ -47,10 +47,10 @@ static void __gfs2_ail_flush(struct gfs2_glock *gl)
                                 bd_ail_gl_list);
                 bh = bd->bd_bh;
                 gfs2_remove_from_ail(bd);
-                spin_unlock(&sdp->sd_ail_lock);
-
                 bd->bd_bh = NULL;
                 bh->b_private = NULL;
+                spin_unlock(&sdp->sd_ail_lock);
+
                 bd->bd_blkno = bh->b_blocknr;
                 gfs2_log_lock(sdp);
                 gfs2_assert_withdraw(sdp, !buffer_busy(bh));
@@ -221,8 +221,10 @@ static void inode_go_inval(struct gfs2_glock *gl, int flags)
                 }
         }
 
-        if (ip == GFS2_I(gl->gl_sbd->sd_rindex))
+        if (ip == GFS2_I(gl->gl_sbd->sd_rindex)) {
+                gfs2_log_flush(gl->gl_sbd, NULL);
                 gl->gl_sbd->sd_rindex_uptodate = 0;
+        }
         if (ip && S_ISREG(ip->i_inode.i_mode))
                 truncate_inode_pages(ip->i_inode.i_mapping, 0);
 }
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 0a064e91ac70..81206e70cbf6 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -17,6 +17,7 @@
 #include <linux/buffer_head.h>
 #include <linux/rcupdate.h>
 #include <linux/rculist_bl.h>
+#include <linux/completion.h>
 
 #define DIO_WAIT        0x00000010
 #define DIO_METADATA    0x00000020
@@ -546,6 +547,7 @@ struct gfs2_sbd {
         struct gfs2_glock *sd_trans_gl;
         wait_queue_head_t sd_glock_wait;
         atomic_t sd_glock_disposal;
+        struct completion sd_locking_init;
 
         /* Inode Stuff */
 
diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c
index 903115f2bb34..85c62923ee29 100644
--- a/fs/gfs2/log.c
+++ b/fs/gfs2/log.c
@@ -903,6 +903,7 @@ void gfs2_meta_syncfs(struct gfs2_sbd *sdp)
                 if (gfs2_ail1_empty(sdp))
                         break;
         }
+        gfs2_log_flush(sdp, NULL);
 }
 
 static inline int gfs2_jrnl_flush_reqd(struct gfs2_sbd *sdp)
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 8ac9ae189b53..2a77071fb7b6 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -72,6 +72,7 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb)
 
         init_waitqueue_head(&sdp->sd_glock_wait);
         atomic_set(&sdp->sd_glock_disposal, 0);
+        init_completion(&sdp->sd_locking_init);
         spin_lock_init(&sdp->sd_statfs_spin);
 
         spin_lock_init(&sdp->sd_rindex_spin);
@@ -1017,11 +1018,13 @@ hostdata_error:
                 fsname++;
         if (lm->lm_mount == NULL) {
                 fs_info(sdp, "Now mounting FS...\n");
+                complete(&sdp->sd_locking_init);
                 return 0;
         }
         ret = lm->lm_mount(sdp, fsname);
         if (ret == 0)
                 fs_info(sdp, "Joined cluster. Now mounting FS...\n");
+        complete(&sdp->sd_locking_init);
         return ret;
 }
 
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index ed540e7018be..fb0edf735483 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -757,13 +757,17 @@ static int gfs2_write_inode(struct inode *inode, struct writeback_control *wbc)
         struct timespec atime;
         struct gfs2_dinode *di;
         int ret = -EAGAIN;
+        int unlock_required = 0;
 
         /* Skip timestamp update, if this is from a memalloc */
         if (current->flags & PF_MEMALLOC)
                 goto do_flush;
-        ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
-        if (ret)
-                goto do_flush;
+        if (!gfs2_glock_is_locked_by_me(ip->i_gl)) {
+                ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
+                if (ret)
+                        goto do_flush;
+                unlock_required = 1;
+        }
         ret = gfs2_trans_begin(sdp, RES_DINODE, 0);
         if (ret)
                 goto do_unlock;
@@ -780,7 +784,8 @@ static int gfs2_write_inode(struct inode *inode, struct writeback_control *wbc)
         }
         gfs2_trans_end(sdp);
 do_unlock:
-        gfs2_glock_dq_uninit(&gh);
+        if (unlock_required)
+                gfs2_glock_dq_uninit(&gh);
 do_flush:
         if (wbc->sync_mode == WB_SYNC_ALL)
                 gfs2_log_flush(GFS2_SB(inode), ip->i_gl);
@@ -1427,7 +1432,20 @@ out:
         return error;
 }
 
-/*
+/**
+ * gfs2_evict_inode - Remove an inode from cache
+ * @inode: The inode to evict
+ *
+ * There are three cases to consider:
+ * 1. i_nlink == 0, we are final opener (and must deallocate)
+ * 2. i_nlink == 0, we are not the final opener (and cannot deallocate)
+ * 3. i_nlink > 0
+ *
+ * If the fs is read only, then we have to treat all cases as per #3
+ * since we are unable to do any deallocation. The inode will be
+ * deallocated by the next read/write node to attempt an allocation
+ * in the same resource group
+ *
  * We have to (at the moment) hold the inodes main lock to cover
  * the gap between unlocking the shared lock on the iopen lock and
  * taking the exclusive lock. I'd rather do a shared -> exclusive
@@ -1470,6 +1488,8 @@ static void gfs2_evict_inode(struct inode *inode)
         if (error)
                 goto out_truncate;
 
+        /* Case 1 starts here */
+
         if (S_ISDIR(inode->i_mode) &&
             (ip->i_diskflags & GFS2_DIF_EXHASH)) {
                 error = gfs2_dir_exhash_dealloc(ip);
@@ -1493,13 +1513,16 @@ static void gfs2_evict_inode(struct inode *inode)
         goto out_unlock;
 
 out_truncate:
+        /* Case 2 starts here */
         error = gfs2_trans_begin(sdp, 0, sdp->sd_jdesc->jd_blocks);
         if (error)
                 goto out_unlock;
-        gfs2_final_release_pages(ip);
+        /* Needs to be done before glock release & also in a transaction */
+        truncate_inode_pages(&inode->i_data, 0);
         gfs2_trans_end(sdp);
 
 out_unlock:
+        /* Error path for case 1 */
         if (test_bit(HIF_HOLDER, &ip->i_iopen_gh.gh_iflags))
                 gfs2_glock_dq(&ip->i_iopen_gh);
         gfs2_holder_uninit(&ip->i_iopen_gh);
@@ -1507,6 +1530,7 @@ out_unlock:
         if (error && error != GLR_TRYFAILED && error != -EROFS)
                 fs_warn(sdp, "gfs2_evict_inode: %d\n", error);
 out:
+        /* Case 3 starts here */
         truncate_inode_pages(&inode->i_data, 0);
         end_writeback(inode);
 
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index e20eab37bc80..443cabcfcd23 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -338,6 +338,9 @@ static ssize_t lkfirst_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
         rv = sscanf(buf, "%u", &first);
         if (rv != 1 || first > 1)
                 return -EINVAL;
+        rv = wait_for_completion_killable(&sdp->sd_locking_init);
+        if (rv)
+                return rv;
         spin_lock(&sdp->sd_jindex_spin);
         rv = -EBUSY;
         if (test_bit(SDF_NOJOURNALID, &sdp->sd_flags) == 0)
@@ -414,7 +417,9 @@ static ssize_t jid_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
         rv = sscanf(buf, "%d", &jid);
         if (rv != 1)
                 return -EINVAL;
-
+        rv = wait_for_completion_killable(&sdp->sd_locking_init);
+        if (rv)
+                return rv;
         spin_lock(&sdp->sd_jindex_spin);
         rv = -EINVAL;
         if (sdp->sd_lockstruct.ls_ops->lm_mount == NULL)
diff --git a/fs/namei.c b/fs/namei.c
index 0223c41fb114..5c867dd1c0b3 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -433,6 +433,8 @@ static int unlazy_walk(struct nameidata *nd, struct dentry *dentry)
                         goto err_parent;
                 BUG_ON(nd->inode != parent->d_inode);
         } else {
+                if (dentry->d_parent != parent)
+                        goto err_parent;
                 spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED);
                 if (!__d_rcu_to_refcount(dentry, nd->seq))
                         goto err_child;
diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c
index 0bafcc91c27f..f9d03abcd04c 100644
--- a/fs/nfs/nfs4filelayout.c
+++ b/fs/nfs/nfs4filelayout.c
@@ -398,7 +398,6 @@ filelayout_write_pagelist(struct nfs_write_data *data, int sync)
          * this offset and save the original offset.
          */
         data->args.offset = filelayout_get_dserver_offset(lseg, offset);
-        data->mds_offset = offset;
 
         /* Perform an asynchronous write */
         status = nfs_initiate_write(data, ds->ds_clp->cl_rpcclient,
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 6870bc61ceec..e6e8f3b9a1de 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -91,7 +91,7 @@ static int nfs4_stat_to_errno(int);
 #define encode_getfh_maxsz      (op_encode_hdr_maxsz)
 #define decode_getfh_maxsz      (op_decode_hdr_maxsz + 1 + \
                                 ((3+NFS4_FHSIZE) >> 2))
-#define nfs4_fattr_bitmap_maxsz 3
+#define nfs4_fattr_bitmap_maxsz 4
 #define encode_getattr_maxsz    (op_encode_hdr_maxsz + nfs4_fattr_bitmap_maxsz)
 #define nfs4_name_maxsz         (1 + ((3 + NFS4_MAXNAMLEN) >> 2))
 #define nfs4_path_maxsz         (1 + ((3 + NFS4_MAXPATHLEN) >> 2))
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index e268e3b23497..727168059684 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -864,6 +864,8 @@ static int nfs_write_rpcsetup(struct nfs_page *req,
 
         data->args.fh     = NFS_FH(inode);
         data->args.offset = req_offset(req) + offset;
+        /* pnfs_set_layoutcommit needs this */
+        data->mds_offset = data->args.offset;
         data->args.pgbase = req->wb_pgbase + offset;
         data->args.pages  = data->pagevec;
         data->args.count  = count;