Merge ../linux-2.6

14 files changed, 76 insertions, 56 deletions

diff --git a/fs/buffer.c b/fs/buffer.c
index 62cfd17dc5fe..a9b399402007 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -3060,6 +3060,7 @@ int buffer_migrate_page(struct page *newpage, struct page *page)
 {
         struct address_space *mapping = page->mapping;
         struct buffer_head *bh, *head;
+        int rc;
 
         if (!mapping)
                 return -EAGAIN;
@@ -3069,8 +3070,9 @@ int buffer_migrate_page(struct page *newpage, struct page *page)
 
         head = page_buffers(page);
 
-        if (migrate_page_remove_references(newpage, page, 3))
-                return -EAGAIN;
+        rc = migrate_page_remove_references(newpage, page, 3);
+        if (rc)
+                return rc;
 
         bh = head;
         do {
diff --git a/fs/direct-io.c b/fs/direct-io.c
index 848044af7e16..27f3e787faca 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -1155,15 +1155,16 @@ direct_io_worker(int rw, struct kiocb *iocb, struct inode *inode,
  * For writes, i_mutex is not held on entry; it is never taken.
  *
  * DIO_LOCKING (simple locking for regular files)
- * For writes we are called under i_mutex and return with i_mutex held, even though
- * it is internally dropped.
+ * For writes we are called under i_mutex and return with i_mutex held, even
+ * though it is internally dropped.
  * For reads, i_mutex is not held on entry, but it is taken and dropped before
  * returning.
  *
  * DIO_OWN_LOCKING (filesystem provides synchronisation and handling of
  *      uninitialised data, allowing parallel direct readers and writers)
  * For writes we are called without i_mutex, return without it, never touch it.
- * For reads, i_mutex is held on entry and will be released before returning.
+ * For reads we are called under i_mutex and return with i_mutex held, even
+ * though it may be internally dropped.
  *
  * Additional i_alloc_sem locking requirements described inline below.
  */
@@ -1182,7 +1183,8 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
         ssize_t retval = -EINVAL;
         loff_t end = offset;
         struct dio *dio;
-        int reader_with_isem = (rw == READ && dio_lock_type == DIO_OWN_LOCKING);
+        int release_i_mutex = 0;
+        int acquire_i_mutex = 0;
 
         if (rw & WRITE)
                 current->flags |= PF_SYNCWRITE;
@@ -1225,7 +1227,6 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
          *      writers need to grab i_alloc_sem only (i_mutex is already held)
          * For regular files using DIO_OWN_LOCKING,
          *      neither readers nor writers take any locks here
-         *      (i_mutex is already held and release for writers here)
          */
         dio->lock_type = dio_lock_type;
         if (dio_lock_type != DIO_NO_LOCKING) {
@@ -1236,7 +1237,7 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
                         mapping = iocb->ki_filp->f_mapping;
                         if (dio_lock_type != DIO_OWN_LOCKING) {
                                 mutex_lock(&inode->i_mutex);
-                                reader_with_isem = 1;
+                                release_i_mutex = 1;
                         }
 
                         retval = filemap_write_and_wait_range(mapping, offset,
@@ -1248,7 +1249,7 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
 
                         if (dio_lock_type == DIO_OWN_LOCKING) {
                                 mutex_unlock(&inode->i_mutex);
-                                reader_with_isem = 0;
+                                acquire_i_mutex = 1;
                         }
                 }
 
@@ -1269,11 +1270,13 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
                                 nr_segs, blkbits, get_blocks, end_io, dio);
 
         if (rw == READ && dio_lock_type == DIO_LOCKING)
-                reader_with_isem = 0;
+                release_i_mutex = 0;
 
 out:
-        if (reader_with_isem)
+        if (release_i_mutex)
                 mutex_unlock(&inode->i_mutex);
+        else if (acquire_i_mutex)
+                mutex_lock(&inode->i_mutex);
         if (rw & WRITE)
                 current->flags &= ~PF_SYNCWRITE;
         return retval;
diff --git a/fs/ext2/dir.c b/fs/ext2/dir.c
index 7442bdd1267a..b3dbd716cd3a 100644
--- a/fs/ext2/dir.c
+++ b/fs/ext2/dir.c
@@ -256,11 +256,10 @@ ext2_readdir (struct file * filp, void * dirent, filldir_t filldir)
         unsigned long npages = dir_pages(inode);
         unsigned chunk_mask = ~(ext2_chunk_size(inode)-1);
         unsigned char *types = NULL;
-        int need_revalidate = (filp->f_version != inode->i_version);
-        int ret;
+        int need_revalidate = filp->f_version != inode->i_version;
 
         if (pos > inode->i_size - EXT2_DIR_REC_LEN(1))
-                goto success;
+                return 0;
 
         if (EXT2_HAS_INCOMPAT_FEATURE(sb, EXT2_FEATURE_INCOMPAT_FILETYPE))
                 types = ext2_filetype_table;
@@ -275,12 +274,15 @@ ext2_readdir (struct file * filp, void * dirent, filldir_t filldir)
                                    "bad page in #%lu",
                                    inode->i_ino);
                         filp->f_pos += PAGE_CACHE_SIZE - offset;
-                        ret = -EIO;
-                        goto done;
+                        return -EIO;
                 }
                 kaddr = page_address(page);
-                if (need_revalidate) {
-                        offset = ext2_validate_entry(kaddr, offset, chunk_mask);
+                if (unlikely(need_revalidate)) {
+                        if (offset) {
+                                offset = ext2_validate_entry(kaddr, offset, chunk_mask);
+                                filp->f_pos = (n<<PAGE_CACHE_SHIFT) + offset;
+                        }
+                        filp->f_version = inode->i_version;
                         need_revalidate = 0;
                 }
                 de = (ext2_dirent *)(kaddr+offset);
@@ -289,9 +291,8 @@ ext2_readdir (struct file * filp, void * dirent, filldir_t filldir)
                         if (de->rec_len == 0) {
                                 ext2_error(sb, __FUNCTION__,
                                         "zero-length directory entry");
-                                ret = -EIO;
                                 ext2_put_page(page);
-                                goto done;
+                                return -EIO;
                         }
                         if (de->inode) {
                                 int over;
@@ -306,19 +307,14 @@ ext2_readdir (struct file * filp, void * dirent, filldir_t filldir)
                                                 le32_to_cpu(de->inode), d_type);
                                 if (over) {
                                         ext2_put_page(page);
-                                        goto success;
+                                        return 0;
                                 }
                         }
                         filp->f_pos += le16_to_cpu(de->rec_len);
                 }
                 ext2_put_page(page);
         }
-
-success:
-        ret = 0;
-done:
-        filp->f_version = inode->i_version;
-        return ret;
+        return 0;
 }
 
 /*
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 3fc4238e9703..0384e539b88f 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -1624,15 +1624,14 @@ static int ext3_block_truncate_page(handle_t *handle, struct page *page,
          * For "nobh" option,  we can only work if we don't need to
          * read-in the page - otherwise we create buffers to do the IO.
          */
-        if (!page_has_buffers(page) && test_opt(inode->i_sb, NOBH)) {
-                if (PageUptodate(page)) {
-                        kaddr = kmap_atomic(page, KM_USER0);
-                        memset(kaddr + offset, 0, length);
-                        flush_dcache_page(page);
-                        kunmap_atomic(kaddr, KM_USER0);
-                        set_page_dirty(page);
-                        goto unlock;
-                }
+        if (!page_has_buffers(page) && test_opt(inode->i_sb, NOBH) &&
+             ext3_should_writeback_data(inode) && PageUptodate(page)) {
+                kaddr = kmap_atomic(page, KM_USER0);
+                memset(kaddr + offset, 0, length);
+                flush_dcache_page(page);
+                kunmap_atomic(kaddr, KM_USER0);
+                set_page_dirty(page);
+                goto unlock;
         }
 
         if (!page_has_buffers(page))
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 8bd8ac077704..b8f5cd1e540d 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -2141,7 +2141,8 @@ retry:
                  * We have a transaction open.  All is sweetness.  It also sets
                  * i_size in generic_commit_write().
                  */
-                err = page_symlink(inode, symname, l);
+                err = __page_symlink(inode, symname, l,
+                                mapping_gfp_mask(inode->i_mapping) & ~__GFP_FS);
                 if (err) {
                         ext3_dec_count(handle, inode);
                         ext3_mark_inode_dirty(handle, inode);
diff --git a/fs/jffs2/nodelist.c b/fs/jffs2/nodelist.c
index b635e167a3fa..d4d0c41490cd 100644
--- a/fs/jffs2/nodelist.c
+++ b/fs/jffs2/nodelist.c
@@ -406,7 +406,8 @@ static int check_node_data(struct jffs2_sb_info *c, struct jffs2_tmp_dnode_info
         int err = 0, pointed = 0;
         struct jffs2_eraseblock *jeb;
         unsigned char *buffer;
-        uint32_t crc, ofs, retlen, len;
+        uint32_t crc, ofs, len;
+        size_t retlen;
 
         BUG_ON(tn->csize == 0);
 
diff --git a/fs/jffs2/readinode.c b/fs/jffs2/readinode.c
index 5f0652df5d47..f1695642d0f7 100644
--- a/fs/jffs2/readinode.c
+++ b/fs/jffs2/readinode.c
@@ -112,7 +112,7 @@ static struct jffs2_raw_node_ref *jffs2_first_valid_node(struct jffs2_raw_node_r
  *          negative error code on failure.
  */
 static inline int read_direntry(struct jffs2_sb_info *c, struct jffs2_raw_node_ref *ref,
-                                struct jffs2_raw_dirent *rd, uint32_t read, struct jffs2_full_dirent **fdp,
+                                struct jffs2_raw_dirent *rd, size_t read, struct jffs2_full_dirent **fdp,
                                 uint32_t *latest_mctime, uint32_t *mctime_ver)
 {
         struct jffs2_full_dirent *fd;
diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c
index 2967b7393415..79b5404db100 100644
--- a/fs/jfs/jfs_dmap.c
+++ b/fs/jfs/jfs_dmap.c
@@ -532,10 +532,10 @@ dbUpdatePMap(struct inode *ipbmap,
 
                 lastlblkno = lblkno;
 
+                LOGSYNC_LOCK(log, flags);
                 if (mp->lsn != 0) {
                         /* inherit older/smaller lsn */
                         logdiff(diffp, mp->lsn, log);
-                        LOGSYNC_LOCK(log, flags);
                         if (difft < diffp) {
                                 mp->lsn = lsn;
 
@@ -548,20 +548,17 @@ dbUpdatePMap(struct inode *ipbmap,
                         logdiff(diffp, mp->clsn, log);
                         if (difft > diffp)
                                 mp->clsn = tblk->clsn;
-                        LOGSYNC_UNLOCK(log, flags);
                 } else {
                         mp->log = log;
                         mp->lsn = lsn;
 
                         /* insert bp after tblock in logsync list */
-                        LOGSYNC_LOCK(log, flags);
-
                         log->count++;
                         list_add(&mp->synclist, &tblk->synclist);
 
                         mp->clsn = tblk->clsn;
-                        LOGSYNC_UNLOCK(log, flags);
                 }
+                LOGSYNC_UNLOCK(log, flags);
         }
 
         /* write the last buffer. */
diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c
index 31b4aa13dd4b..4efa0d0eec39 100644
--- a/fs/jfs/jfs_imap.c
+++ b/fs/jfs/jfs_imap.c
@@ -2844,11 +2844,11 @@ diUpdatePMap(struct inode *ipimap,
          */
         lsn = tblk->lsn;
         log = JFS_SBI(tblk->sb)->log;
+        LOGSYNC_LOCK(log, flags);
         if (mp->lsn != 0) {
                 /* inherit older/smaller lsn */
                 logdiff(difft, lsn, log);
                 logdiff(diffp, mp->lsn, log);
-                LOGSYNC_LOCK(log, flags);
                 if (difft < diffp) {
                         mp->lsn = lsn;
                         /* move mp after tblock in logsync list */
@@ -2860,17 +2860,15 @@ diUpdatePMap(struct inode *ipimap,
                 logdiff(diffp, mp->clsn, log);
                 if (difft > diffp)
                         mp->clsn = tblk->clsn;
-                LOGSYNC_UNLOCK(log, flags);
         } else {
                 mp->log = log;
                 mp->lsn = lsn;
                 /* insert mp after tblock in logsync list */
-                LOGSYNC_LOCK(log, flags);
                 log->count++;
                 list_add(&mp->synclist, &tblk->synclist);
                 mp->clsn = tblk->clsn;
-                LOGSYNC_UNLOCK(log, flags);
         }
+        LOGSYNC_UNLOCK(log, flags);
         write_metapage(mp);
         return (0);
 }
diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c
index 220058d8616d..970b6a6aa337 100644
--- a/fs/lockd/clntproc.c
+++ b/fs/lockd/clntproc.c
@@ -662,12 +662,18 @@ nlmclnt_unlock(struct nlm_rqst *req, struct file_lock *fl)
          * reclaimed while we're stuck in the unlock call. */
         fl->fl_u.nfs_fl.flags &= ~NFS_LCK_GRANTED;
 
+        /*
+         * Note: the server is supposed to either grant us the unlock
+         * request, or to deny it with NLM_LCK_DENIED_GRACE_PERIOD. In either
+         * case, we want to unlock.
+         */
+        do_vfs_lock(fl);
+
         if (req->a_flags & RPC_TASK_ASYNC) {
                 status = nlmclnt_async_call(req, NLMPROC_UNLOCK,
                                         &nlmclnt_unlock_ops);
                 /* Hrmf... Do the unlock early since locks_remove_posix()
                  * really expects us to free the lock synchronously */
-                do_vfs_lock(fl);
                 if (status < 0) {
                         nlmclnt_release_lockargs(req);
                         kfree(req);
@@ -680,7 +686,6 @@ nlmclnt_unlock(struct nlm_rqst *req, struct file_lock *fl)
         if (status < 0)
                 return status;
 
-        do_vfs_lock(fl);
         if (resp->status == NLM_LCK_GRANTED)
                 return 0;
 
diff --git a/fs/namei.c b/fs/namei.c
index 557dcf395ca1..8dc2b038d5d9 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -2613,13 +2613,15 @@ void page_put_link(struct dentry *dentry, struct nameidata *nd, void *cookie)
         }
 }
 
-int page_symlink(struct inode *inode, const char *symname, int len)
+int __page_symlink(struct inode *inode, const char *symname, int len,
+                gfp_t gfp_mask)
 {
         struct address_space *mapping = inode->i_mapping;
-        struct page *page = grab_cache_page(mapping, 0);
+        struct page *page;
         int err = -ENOMEM;
         char *kaddr;
 
+        page = find_or_create_page(mapping, 0, gfp_mask);
         if (!page)
                 goto fail;
         err = mapping->a_ops->prepare_write(NULL, page, 0, len-1);
@@ -2654,6 +2656,12 @@ fail:
         return err;
 }
 
+int page_symlink(struct inode *inode, const char *symname, int len)
+{
+        return __page_symlink(inode, symname, len,
+                        mapping_gfp_mask(inode->i_mapping));
+}
+
 struct inode_operations page_symlink_inode_operations = {
         .readlink       = generic_readlink,
         .follow_link    = page_follow_link_light,
@@ -2672,6 +2680,7 @@ EXPORT_SYMBOL(lookup_one_len);
 EXPORT_SYMBOL(page_follow_link_light);
 EXPORT_SYMBOL(page_put_link);
 EXPORT_SYMBOL(page_readlink);
+EXPORT_SYMBOL(__page_symlink);
 EXPORT_SYMBOL(page_symlink);
 EXPORT_SYMBOL(page_symlink_inode_operations);
 EXPORT_SYMBOL(path_lookup);
diff --git a/fs/namespace.c b/fs/namespace.c
index 058a44865beb..39c81a8d6316 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -1338,7 +1338,7 @@ struct namespace *dup_namespace(struct task_struct *tsk, struct fs_struct *fs)
 
         new_ns = kmalloc(sizeof(struct namespace), GFP_KERNEL);
         if (!new_ns)
-                goto out;
+                return NULL;
 
         atomic_set(&new_ns->count, 1);
         INIT_LIST_HEAD(&new_ns->list);
@@ -1352,7 +1352,7 @@ struct namespace *dup_namespace(struct task_struct *tsk, struct fs_struct *fs)
         if (!new_ns->root) {
                 up_write(&namespace_sem);
                 kfree(new_ns);
-                goto out;
+                return NULL;
         }
         spin_lock(&vfsmount_lock);
         list_add_tail(&new_ns->list, &new_ns->root->mnt_list);
@@ -1393,7 +1393,6 @@ struct namespace *dup_namespace(struct task_struct *tsk, struct fs_struct *fs)
         if (altrootmnt)
                 mntput(altrootmnt);
 
-out:
         return new_ns;
 }
 
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 04ab2fc360e7..4e9b3a1b36c5 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -57,6 +57,7 @@
 #define NFSDBG_FACILITY         NFSDBG_VFS
 #define MAX_DIRECTIO_SIZE       (4096UL << PAGE_SHIFT)
 
+static void nfs_free_user_pages(struct page **pages, int npages, int do_dirty);
 static kmem_cache_t *nfs_direct_cachep;
 
 /*
@@ -107,6 +108,15 @@ nfs_get_user_pages(int rw, unsigned long user_addr, size_t size,
                                         page_count, (rw == READ), 0,
                                         *pages, NULL);
                 up_read(&current->mm->mmap_sem);
+                /*
+                 * If we got fewer pages than expected from get_user_pages(),
+                 * the user buffer runs off the end of a mapping; return EFAULT.
+                 */
+                if (result >= 0 && result < page_count) {
+                        nfs_free_user_pages(*pages, result, 0);
+                        *pages = NULL;
+                        result = -EFAULT;
+                }
         }
         return result;
 }
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 984ca3454d04..f8c0066e02e1 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1430,7 +1430,7 @@ static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
         if (status == 0)
                 status = nfs4_do_fsinfo(server, fhandle, info);
 out:
-        return status;
+        return nfs4_map_errors(status);
 }
 
 static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr)