Merge branch 'bkl-removal' into next

7 files changed, 83 insertions, 77 deletions

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index b1940660502f..28a238dab23a 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -139,10 +139,8 @@ nfs_opendir(struct inode *inode, struct file *filp)
 
         nfs_inc_stats(inode, NFSIOS_VFSOPEN);
 
-        lock_kernel();
         /* Call generic open code in order to cache credentials */
         res = nfs_open(inode, filp);
-        unlock_kernel();
         return res;
 }
 
@@ -536,8 +534,6 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
                         (long long)filp->f_pos);
         nfs_inc_stats(inode, NFSIOS_VFSGETDENTS);
 
-        lock_kernel();
-
         /*
          * filp->f_pos points to the dirent entry number.
          * *desc->dir_cookie has the cookie for the next entry. We have
@@ -595,7 +591,6 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
         }
 out:
         nfs_unblock_sillyrename(dentry);
-        unlock_kernel();
         if (res > 0)
                 res = 0;
         dfprintk(FILE, "NFS: readdir(%s/%s) returns %ld\n",
@@ -779,7 +774,6 @@ static int nfs_lookup_revalidate(struct dentry * dentry, struct nameidata *nd)
         struct nfs_fattr fattr;
 
         parent = dget_parent(dentry);
-        lock_kernel();
         dir = parent->d_inode;
         nfs_inc_stats(dir, NFSIOS_DENTRYREVALIDATE);
         inode = dentry->d_inode;
@@ -817,7 +811,6 @@ static int nfs_lookup_revalidate(struct dentry * dentry, struct nameidata *nd)
 
         nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
  out_valid:
-        unlock_kernel();
         dput(parent);
         dfprintk(LOOKUPCACHE, "NFS: %s(%s/%s) is valid\n",
                         __func__, dentry->d_parent->d_name.name,
@@ -836,7 +829,6 @@ out_zap_parent:
                 shrink_dcache_parent(dentry);
         }
         d_drop(dentry);
-        unlock_kernel();
         dput(parent);
         dfprintk(LOOKUPCACHE, "NFS: %s(%s/%s) is invalid\n",
                         __func__, dentry->d_parent->d_name.name,
@@ -870,6 +862,14 @@ static int nfs_dentry_delete(struct dentry *dentry)
 
 }
 
+static void nfs_drop_nlink(struct inode *inode)
+{
+        spin_lock(&inode->i_lock);
+        if (inode->i_nlink > 0)
+                drop_nlink(inode);
+        spin_unlock(&inode->i_lock);
+}
+
 /*
  * Called when the dentry loses inode.
  * We use it to clean up silly-renamed files.
@@ -881,10 +881,8 @@ static void nfs_dentry_iput(struct dentry *dentry, struct inode *inode)
                 NFS_I(inode)->cache_validity |= NFS_INO_INVALID_DATA;
 
         if (dentry->d_flags & DCACHE_NFSFS_RENAMED) {
-                lock_kernel();
                 drop_nlink(inode);
                 nfs_complete_unlink(dentry, inode);
-                unlock_kernel();
         }
         iput(inode);
 }
@@ -915,8 +913,6 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru
         res = ERR_PTR(-ENOMEM);
         dentry->d_op = NFS_PROTO(dir)->dentry_ops;
 
-        lock_kernel();
-
         /*
          * If we're doing an exclusive create, optimize away the lookup
          * but don't hash the dentry.
@@ -924,7 +920,7 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru
         if (nfs_is_exclusive_create(dir, nd)) {
                 d_instantiate(dentry, NULL);
                 res = NULL;
-                goto out_unlock;
+                goto out;
         }
 
         parent = dentry->d_parent;
@@ -952,8 +948,6 @@ no_entry:
         nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
 out_unblock_sillyrename:
         nfs_unblock_sillyrename(parent);
-out_unlock:
-        unlock_kernel();
 out:
         return res;
 }
@@ -1011,9 +1005,7 @@ static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry
         }
 
         /* Open the file on the server */
-        lock_kernel();
         res = nfs4_atomic_open(dir, dentry, nd);
-        unlock_kernel();
         if (IS_ERR(res)) {
                 error = PTR_ERR(res);
                 switch (error) {
@@ -1075,9 +1067,7 @@ static int nfs_open_revalidate(struct dentry *dentry, struct nameidata *nd)
          * operations that change the directory. We therefore save the
          * change attribute *before* we do the RPC call.
          */
-        lock_kernel();
         ret = nfs4_open_revalidate(dir, dentry, openflags, nd);
-        unlock_kernel();
 out:
         dput(parent);
         if (!ret)
@@ -1230,14 +1220,11 @@ static int nfs_create(struct inode *dir, struct dentry *dentry, int mode,
         if ((nd->flags & LOOKUP_CREATE) != 0)
                 open_flags = nd->intent.open.flags;
 
-        lock_kernel();
         error = NFS_PROTO(dir)->create(dir, dentry, &attr, open_flags, nd);
         if (error != 0)
                 goto out_err;
-        unlock_kernel();
         return 0;
 out_err:
-        unlock_kernel();
         d_drop(dentry);
         return error;
 }
@@ -1260,14 +1247,11 @@ nfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev)
         attr.ia_mode = mode;
         attr.ia_valid = ATTR_MODE;
 
-        lock_kernel();
         status = NFS_PROTO(dir)->mknod(dir, dentry, &attr, rdev);
         if (status != 0)
                 goto out_err;
-        unlock_kernel();
         return 0;
 out_err:
-        unlock_kernel();
         d_drop(dentry);
         return status;
 }
@@ -1286,15 +1270,12 @@ static int nfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
         attr.ia_valid = ATTR_MODE;
         attr.ia_mode = mode | S_IFDIR;
 
-        lock_kernel();
         error = NFS_PROTO(dir)->mkdir(dir, dentry, &attr);
         if (error != 0)
                 goto out_err;
-        unlock_kernel();
         return 0;
 out_err:
         d_drop(dentry);
-        unlock_kernel();
         return error;
 }
 
@@ -1311,14 +1292,12 @@ static int nfs_rmdir(struct inode *dir, struct dentry *dentry)
         dfprintk(VFS, "NFS: rmdir(%s/%ld), %s\n",
                         dir->i_sb->s_id, dir->i_ino, dentry->d_name.name);
 
-        lock_kernel();
         error = NFS_PROTO(dir)->rmdir(dir, &dentry->d_name);
         /* Ensure the VFS deletes this inode */
         if (error == 0 && dentry->d_inode != NULL)
                 clear_nlink(dentry->d_inode);
         else if (error == -ENOENT)
                 nfs_dentry_handle_enoent(dentry);
-        unlock_kernel();
 
         return error;
 }
@@ -1420,7 +1399,7 @@ static int nfs_safe_remove(struct dentry *dentry)
                 error = NFS_PROTO(dir)->remove(dir, &dentry->d_name);
                 /* The VFS may want to delete this inode */
                 if (error == 0)
-                        drop_nlink(inode);
+                        nfs_drop_nlink(inode);
                 nfs_mark_for_revalidate(inode);
         } else
                 error = NFS_PROTO(dir)->remove(dir, &dentry->d_name);
@@ -1443,7 +1422,6 @@ static int nfs_unlink(struct inode *dir, struct dentry *dentry)
         dfprintk(VFS, "NFS: unlink(%s/%ld, %s)\n", dir->i_sb->s_id,
                 dir->i_ino, dentry->d_name.name);
 
-        lock_kernel();
         spin_lock(&dcache_lock);
         spin_lock(&dentry->d_lock);
         if (atomic_read(&dentry->d_count) > 1) {
@@ -1452,7 +1430,6 @@ static int nfs_unlink(struct inode *dir, struct dentry *dentry)
                 /* Start asynchronous writeout of the inode */
                 write_inode_now(dentry->d_inode, 0);
                 error = nfs_sillyrename(dir, dentry);
-                unlock_kernel();
                 return error;
         }
         if (!d_unhashed(dentry)) {
@@ -1466,7 +1443,6 @@ static int nfs_unlink(struct inode *dir, struct dentry *dentry)
                 nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
         } else if (need_rehash)
                 d_rehash(dentry);
-        unlock_kernel();
         return error;
 }
 
@@ -1503,13 +1479,9 @@ static int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *sym
         attr.ia_mode = S_IFLNK | S_IRWXUGO;
         attr.ia_valid = ATTR_MODE;
 
-        lock_kernel();
-
         page = alloc_page(GFP_HIGHUSER);
-        if (!page) {
-                unlock_kernel();
+        if (!page)
                 return -ENOMEM;
-        }
 
         kaddr = kmap_atomic(page, KM_USER0);
         memcpy(kaddr, symname, pathlen);
@@ -1524,7 +1496,6 @@ static int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *sym
                         dentry->d_name.name, symname, error);
                 d_drop(dentry);
                 __free_page(page);
-                unlock_kernel();
                 return error;
         }
 
@@ -1542,7 +1513,6 @@ static int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *sym
         } else
                 __free_page(page);
 
-        unlock_kernel();
         return 0;
 }
 
@@ -1556,14 +1526,12 @@ nfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry)
                 old_dentry->d_parent->d_name.name, old_dentry->d_name.name,
                 dentry->d_parent->d_name.name, dentry->d_name.name);
 
-        lock_kernel();
         d_drop(dentry);
         error = NFS_PROTO(dir)->link(inode, dir, &dentry->d_name);
         if (error == 0) {
                 atomic_inc(&inode->i_count);
                 d_add(dentry, inode);
         }
-        unlock_kernel();
         return error;
 }
 
@@ -1603,7 +1571,6 @@ static int nfs_rename(struct inode *old_dir, struct dentry *old_dentry,
          * To prevent any new references to the target during the rename,
          * we unhash the dentry and free the inode in advance.
          */
-        lock_kernel();
         if (!d_unhashed(new_dentry)) {
                 d_drop(new_dentry);
                 rehash = new_dentry;
@@ -1647,7 +1614,7 @@ static int nfs_rename(struct inode *old_dir, struct dentry *old_dentry,
                         /* dentry still busy? */
                         goto out;
         } else
-                drop_nlink(new_inode);
+                nfs_drop_nlink(new_inode);
 
 go_ahead:
         /*
@@ -1681,7 +1648,6 @@ out:
         /* new dentry created? */
         if (dentry)
                 dput(dentry);
-        unlock_kernel();
         return error;
 }
 
@@ -1974,8 +1940,6 @@ int nfs_permission(struct inode *inode, int mask, struct nameidata *nd)
         }
 
 force_lookup:
-        lock_kernel();
-
         if (!NFS_PROTO(inode)->access)
                 goto out_notsup;
 
@@ -1985,7 +1949,6 @@ force_lookup:
                 put_rpccred(cred);
         } else
                 res = PTR_ERR(cred);
-        unlock_kernel();
 out:
         dfprintk(VFS, "NFS: permission(%s/%ld), mask=0x%x, res=%d\n",
                 inode->i_sb->s_id, inode->i_ino, mask, res);
@@ -1994,7 +1957,6 @@ out_notsup:
         res = nfs_revalidate_inode(NFS_SERVER(inode), inode);
         if (res == 0)
                 res = generic_permission(inode, mask, NULL);
-        unlock_kernel();
         goto out;
 }
 
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 43164fe86069..78460657f5cb 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -128,9 +128,7 @@ nfs_file_open(struct inode *inode, struct file *filp)
                 return res;
 
         nfs_inc_stats(inode, NFSIOS_VFSOPEN);
-        lock_kernel();
         res = nfs_open(inode, filp);
-        unlock_kernel();
         return res;
 }
 
@@ -398,9 +396,7 @@ static int nfs_write_end(struct file *file, struct address_space *mapping,
                         zero_user_segment(page, pglen, PAGE_CACHE_SIZE);
         }
 
-        lock_kernel();
         status = nfs_updatepage(file, page, offset, copied);
-        unlock_kernel();
 
         unlock_page(page);
         page_cache_release(page);
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 2c23d067e2a6..df23f987da6b 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -370,7 +370,6 @@ nfs_setattr(struct dentry *dentry, struct iattr *attr)
         if ((attr->ia_valid & ~ATTR_FILE) == 0)
                 return 0;
 
-        lock_kernel();
         /* Write all dirty data */
         if (S_ISREG(inode->i_mode)) {
                 filemap_write_and_wait(inode->i_mapping);
@@ -384,11 +383,66 @@ nfs_setattr(struct dentry *dentry, struct iattr *attr)
         error = NFS_PROTO(inode)->setattr(dentry, &fattr, attr);
         if (error == 0)
                 nfs_refresh_inode(inode, &fattr);
-        unlock_kernel();
         return error;
 }
 
 /**
+ * nfs_vmtruncate - unmap mappings "freed" by truncate() syscall
+ * @inode: inode of the file used
+ * @offset: file offset to start truncating
+ *
+ * This is a copy of the common vmtruncate, but with the locking
+ * corrected to take into account the fact that NFS requires
+ * inode->i_size to be updated under the inode->i_lock.
+ */
+static int nfs_vmtruncate(struct inode * inode, loff_t offset)
+{
+        if (i_size_read(inode) < offset) {
+                unsigned long limit;
+
+                limit = current->signal->rlim[RLIMIT_FSIZE].rlim_cur;
+                if (limit != RLIM_INFINITY && offset > limit)
+                        goto out_sig;
+                if (offset > inode->i_sb->s_maxbytes)
+                        goto out_big;
+                spin_lock(&inode->i_lock);
+                i_size_write(inode, offset);
+                spin_unlock(&inode->i_lock);
+        } else {
+                struct address_space *mapping = inode->i_mapping;
+
+                /*
+                 * truncation of in-use swapfiles is disallowed - it would
+                 * cause subsequent swapout to scribble on the now-freed
+                 * blocks.
+                 */
+                if (IS_SWAPFILE(inode))
+                        return -ETXTBSY;
+                spin_lock(&inode->i_lock);
+                i_size_write(inode, offset);
+                spin_unlock(&inode->i_lock);
+
+                /*
+                 * unmap_mapping_range is called twice, first simply for
+                 * efficiency so that truncate_inode_pages does fewer
+                 * single-page unmaps.  However after this first call, and
+                 * before truncate_inode_pages finishes, it is possible for
+                 * private pages to be COWed, which remain after
+                 * truncate_inode_pages finishes, hence the second
+                 * unmap_mapping_range call must be made for correctness.
+                 */
+                unmap_mapping_range(mapping, offset + PAGE_SIZE - 1, 0, 1);
+                truncate_inode_pages(mapping, offset);
+                unmap_mapping_range(mapping, offset + PAGE_SIZE - 1, 0, 1);
+        }
+        return 0;
+out_sig:
+        send_sig(SIGXFSZ, current, 0);
+out_big:
+        return -EFBIG;
+}
+
+/**
  * nfs_setattr_update_inode - Update inode metadata after a setattr call.
  * @inode: pointer to struct inode
  * @attr: pointer to struct iattr
@@ -414,8 +468,7 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr)
         }
         if ((attr->ia_valid & ATTR_SIZE) != 0) {
                 nfs_inc_stats(inode, NFSIOS_SETATTRTRUNC);
-                inode->i_size = attr->ia_size;
-                vmtruncate(inode, attr->ia_size);
+                nfs_vmtruncate(inode, attr->ia_size);
         }
 }
 
@@ -645,7 +698,6 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
                 inode->i_sb->s_id, (long long)NFS_FILEID(inode));
 
         nfs_inc_stats(inode, NFSIOS_INODEREVALIDATE);
-        lock_kernel();
         if (is_bad_inode(inode))
                 goto out_nowait;
         if (NFS_STALE(inode))
@@ -694,7 +746,6 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
         nfs_wake_up_inode(inode);
 
  out_nowait:
-        unlock_kernel();
         return status;
 }
 
@@ -829,9 +880,9 @@ static void nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr *fattr)
                         if (S_ISDIR(inode->i_mode))
                                 nfsi->cache_validity |= NFS_INO_INVALID_DATA;
                 }
-                if (inode->i_size == nfs_size_to_loff_t(fattr->pre_size) &&
+                if (i_size_read(inode) == nfs_size_to_loff_t(fattr->pre_size) &&
                     nfsi->npages == 0)
-                        inode->i_size = nfs_size_to_loff_t(fattr->size);
+                        i_size_write(inode, nfs_size_to_loff_t(fattr->size));
         }
 }
 
@@ -972,7 +1023,7 @@ int nfs_post_op_update_inode_force_wcc(struct inode *inode, struct nfs_fattr *fa
                         (fattr->valid & NFS_ATTR_WCC) == 0) {
                 memcpy(&fattr->pre_ctime, &inode->i_ctime, sizeof(fattr->pre_ctime));
                 memcpy(&fattr->pre_mtime, &inode->i_mtime, sizeof(fattr->pre_mtime));
-                fattr->pre_size = inode->i_size;
+                fattr->pre_size = i_size_read(inode);
                 fattr->valid |= NFS_ATTR_WCC;
         }
         return nfs_post_op_update_inode(inode, fattr);
@@ -1057,7 +1108,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
                 /* Do we perhaps have any outstanding writes, or has
                  * the file grown beyond our last write? */
                 if (nfsi->npages == 0 || new_isize > cur_isize) {
-                        inode->i_size = new_isize;
+                        i_size_write(inode, new_isize);
                         invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA;
                 }
                 dprintk("NFS: isize change on server for file %s/%ld\n",
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 4451287a81d1..c910413eaeca 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -451,9 +451,7 @@ static struct nfs4_state *nfs4_try_open_cached(struct nfs4_opendata *opendata)
                 /* Save the delegation */
                 memcpy(stateid.data, delegation->stateid.data, sizeof(stateid.data));
                 rcu_read_unlock();
-                lock_kernel();
                 ret = nfs_may_open(state->inode, state->owner->so_cred, open_mode);
-                unlock_kernel();
                 if (ret != 0)
                         goto out;
                 ret = -EAGAIN;
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 856a8934f610..401ef8b28f97 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -940,7 +940,6 @@ static int reclaimer(void *ptr)
         allow_signal(SIGKILL);
 
         /* Ensure exclusive access to NFSv4 state */
-        lock_kernel();
         down_write(&clp->cl_sem);
         /* Are there any NFS mounts out there? */
         if (list_empty(&clp->cl_superblocks))
@@ -1000,7 +999,6 @@ restart_loop:
         nfs_delegation_reap_unclaimed(clp);
 out:
         up_write(&clp->cl_sem);
-        unlock_kernel();
         if (status == -NFS4ERR_CB_PATH_DOWN)
                 nfs_handle_cb_pathdown(clp);
         nfs4_clear_recover_bit(clp);
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 47cf83e917be..1b94e3650f5c 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -374,8 +374,6 @@ static int nfs_statfs(struct dentry *dentry, struct kstatfs *buf)
         };
         int error;
 
-        lock_kernel();
-
         error = server->nfs_client->rpc_ops->statfs(server, fh, &res);
         if (error < 0)
                 goto out_err;
@@ -407,12 +405,10 @@ static int nfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 
         buf->f_namelen = server->namelen;
 
-        unlock_kernel();
         return 0;
 
  out_err:
         dprintk("%s: statfs error = %d\n", __func__, -error);
-        unlock_kernel();
         return error;
 }
 
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index feca8c648766..3229e217c773 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -133,16 +133,21 @@ static struct nfs_page *nfs_page_find_request(struct page *page)
 static void nfs_grow_file(struct page *page, unsigned int offset, unsigned int count)
 {
         struct inode *inode = page->mapping->host;
-        loff_t end, i_size = i_size_read(inode);
-        pgoff_t end_index = (i_size - 1) >> PAGE_CACHE_SHIFT;
+        loff_t end, i_size;
+        pgoff_t end_index;
 
+        spin_lock(&inode->i_lock);
+        i_size = i_size_read(inode);
+        end_index = (i_size - 1) >> PAGE_CACHE_SHIFT;
         if (i_size > 0 && page->index < end_index)
-                return;
+                goto out;
         end = ((loff_t)page->index << PAGE_CACHE_SHIFT) + ((loff_t)offset+count);
         if (i_size >= end)
-                return;
-        nfs_inc_stats(inode, NFSIOS_EXTENDWRITE);
+                goto out;
         i_size_write(inode, end);
+        nfs_inc_stats(inode, NFSIOS_EXTENDWRITE);
+out:
+        spin_unlock(&inode->i_lock);
 }
 
 /* A writeback failed: mark the page as bad, and invalidate the page cache */