1 files changed, 453 insertions, 365 deletions

diff --git a/fs/afs/inode.c b/fs/afs/inode.c
index 281470fe1183..6d3a3dbe4928 100644
--- a/fs/afs/inode.c
+++ b/fs/afs/inode.c
@@ -27,7 +27,6 @@
 
 static const struct inode_operations afs_symlink_inode_operations = {
 	.get_link	= page_get_link,
-	.listxattr	= afs_listxattr,
 };
 
 static noinline void dump_vnode(struct afs_vnode *vnode, struct afs_vnode *parent_vnode)
@@ -55,28 +54,28 @@ static noinline void dump_vnode(struct afs_vnode *vnode, struct afs_vnode *paren
 }
 
 /*
- * Set the file size and block count.  Estimate the number of 512 bytes blocks
- * used, rounded up to nearest 1K for consistency with other AFS clients.
+ * Set parameters for the netfs library
  */
-static void afs_set_i_size(struct afs_vnode *vnode, u64 size)
+static void afs_set_netfs_context(struct afs_vnode *vnode)
 {
-	i_size_write(&vnode->vfs_inode, size);
-	vnode->vfs_inode.i_blocks = ((size + 1023) >> 10) << 1;
+	netfs_inode_init(&vnode->netfs, &afs_req_ops);
 }
 
 /*
  * Initialise an inode from the vnode status.
  */
-static int afs_inode_init_from_status(struct afs_vnode *vnode, struct key *key,
-				      struct afs_cb_interest *cbi,
-				      struct afs_vnode *parent_vnode,
-				      struct afs_status_cb *scb)
+static int afs_inode_init_from_status(struct afs_operation *op,
+				      struct afs_vnode_param *vp,
+				      struct afs_vnode *vnode)
 {
-	struct afs_cb_interest *old_cbi = NULL;
-	struct afs_file_status *status = &scb->status;
+	struct afs_file_status *status = &vp->scb.status;
 	struct inode *inode = AFS_VNODE_TO_I(vnode);
 	struct timespec64 t;
 
+	_enter("{%llx:%llu.%u} %s",
+	       vp->fid.vid, vp->fid.vnode, vp->fid.unique,
+	       op->type ? op->type->name : "???");
+
 	_debug("FS: ft=%d lk=%d sz=%llu ver=%Lu mod=%hu",
 	       status->type,
 	       status->nlink,
@@ -86,28 +85,33 @@ static int afs_inode_init_from_status(struct afs_vnode *vnode, struct key *key,
 
 	write_seqlock(&vnode->cb_lock);
 
+	vnode->cb_v_break = op->cb_v_break;
+	vnode->cb_s_break = op->cb_s_break;
 	vnode->status = *status;
 
 	t = status->mtime_client;
 	inode->i_ctime = t;
 	inode->i_mtime = t;
 	inode->i_atime = t;
+	inode->i_flags |= S_NOATIME;
 	inode->i_uid = make_kuid(&init_user_ns, status->owner);
 	inode->i_gid = make_kgid(&init_user_ns, status->group);
-	set_nlink(&vnode->vfs_inode, status->nlink);
+	set_nlink(&vnode->netfs.inode, status->nlink);
 
 	switch (status->type) {
 	case AFS_FTYPE_FILE:
-		inode->i_mode	= S_IFREG | status->mode;
+		inode->i_mode	= S_IFREG | (status->mode & S_IALLUGO);
 		inode->i_op	= &afs_file_inode_operations;
 		inode->i_fop	= &afs_file_operations;
-		inode->i_mapping->a_ops	= &afs_fs_aops;
+		inode->i_mapping->a_ops	= &afs_file_aops;
+		mapping_set_large_folios(inode->i_mapping);
 		break;
 	case AFS_FTYPE_DIR:
-		inode->i_mode	= S_IFDIR | status->mode;
+		inode->i_mode	= S_IFDIR |  (status->mode & S_IALLUGO);
 		inode->i_op	= &afs_dir_inode_operations;
 		inode->i_fop	= &afs_dir_file_operations;
 		inode->i_mapping->a_ops	= &afs_dir_aops;
+		mapping_set_large_folios(inode->i_mapping);
 		break;
 	case AFS_FTYPE_SYMLINK:
 		/* Symlinks with a mode of 0644 are actually mountpoints. */
@@ -119,57 +123,57 @@ static int afs_inode_init_from_status(struct afs_vnode *vnode, struct key *key,
 			inode->i_mode	= S_IFDIR | 0555;
 			inode->i_op	= &afs_mntpt_inode_operations;
 			inode->i_fop	= &afs_mntpt_file_operations;
-			inode->i_mapping->a_ops	= &afs_fs_aops;
+			inode->i_mapping->a_ops	= &afs_symlink_aops;
 		} else {
 			inode->i_mode	= S_IFLNK | status->mode;
 			inode->i_op	= &afs_symlink_inode_operations;
-			inode->i_mapping->a_ops	= &afs_fs_aops;
+			inode->i_mapping->a_ops	= &afs_symlink_aops;
 		}
 		inode_nohighmem(inode);
 		break;
 	default:
-		dump_vnode(vnode, parent_vnode);
+		dump_vnode(vnode, op->file[0].vnode != vnode ? op->file[0].vnode : NULL);
 		write_sequnlock(&vnode->cb_lock);
-		return afs_protocol_error(NULL, -EBADMSG, afs_eproto_file_type);
+		return afs_protocol_error(NULL, afs_eproto_file_type);
 	}
 
 	afs_set_i_size(vnode, status->size);
+	afs_set_netfs_context(vnode);
 
 	vnode->invalid_before	= status->data_version;
-	inode_set_iversion_raw(&vnode->vfs_inode, status->data_version);
+	inode_set_iversion_raw(&vnode->netfs.inode, status->data_version);
 
-	if (!scb->have_cb) {
+	if (!vp->scb.have_cb) {
 		/* it's a symlink we just created (the fileserver
 		 * didn't give us a callback) */
 		vnode->cb_expires_at = ktime_get_real_seconds();
 	} else {
-		vnode->cb_expires_at = scb->callback.expires_at;
-		old_cbi = rcu_dereference_protected(vnode->cb_interest,
-						    lockdep_is_held(&vnode->cb_lock.lock));
-		if (cbi != old_cbi)
-			rcu_assign_pointer(vnode->cb_interest, afs_get_cb_interest(cbi));
-		else
-			old_cbi = NULL;
+		vnode->cb_expires_at = vp->scb.callback.expires_at;
+		vnode->cb_server = op->server;
 		set_bit(AFS_VNODE_CB_PROMISED, &vnode->flags);
 	}
 
 	write_sequnlock(&vnode->cb_lock);
-	afs_put_cb_interest(afs_v2net(vnode), old_cbi);
 	return 0;
 }
 
 /*
  * Update the core inode struct from a returned status record.
  */
-static void afs_apply_status(struct afs_fs_cursor *fc,
-			     struct afs_vnode *vnode,
-			     struct afs_status_cb *scb,
-			     const afs_dataversion_t *expected_version)
+static void afs_apply_status(struct afs_operation *op,
+			     struct afs_vnode_param *vp)
 {
-	struct afs_file_status *status = &scb->status;
+	struct afs_file_status *status = &vp->scb.status;
+	struct afs_vnode *vnode = vp->vnode;
+	struct inode *inode = &vnode->netfs.inode;
 	struct timespec64 t;
 	umode_t mode;
 	bool data_changed = false;
+	bool change_size = vp->set_size;
+
+	_enter("{%llx:%llu.%u} %s",
+	       vp->fid.vid, vp->fid.vnode, vp->fid.unique,
+	       op->type ? op->type->name : "???");
 
 	BUG_ON(test_bit(AFS_VNODE_UNSET, &vnode->flags));
 
@@ -179,44 +183,44 @@ static void afs_apply_status(struct afs_fs_cursor *fc,
 			vnode->fid.vnode,
 			vnode->fid.unique,
 			status->type, vnode->status.type);
-		afs_protocol_error(NULL, -EBADMSG, afs_eproto_bad_status);
+		afs_protocol_error(NULL, afs_eproto_bad_status);
 		return;
 	}
 
 	if (status->nlink != vnode->status.nlink)
-		set_nlink(&vnode->vfs_inode, status->nlink);
+		set_nlink(inode, status->nlink);
 
 	if (status->owner != vnode->status.owner)
-		vnode->vfs_inode.i_uid = make_kuid(&init_user_ns, status->owner);
+		inode->i_uid = make_kuid(&init_user_ns, status->owner);
 
 	if (status->group != vnode->status.group)
-		vnode->vfs_inode.i_gid = make_kgid(&init_user_ns, status->group);
+		inode->i_gid = make_kgid(&init_user_ns, status->group);
 
 	if (status->mode != vnode->status.mode) {
-		mode = vnode->vfs_inode.i_mode;
+		mode = inode->i_mode;
 		mode &= ~S_IALLUGO;
-		mode |= status->mode;
-		WRITE_ONCE(vnode->vfs_inode.i_mode, mode);
+		mode |= status->mode & S_IALLUGO;
+		WRITE_ONCE(inode->i_mode, mode);
 	}
 
 	t = status->mtime_client;
-	vnode->vfs_inode.i_ctime = t;
-	vnode->vfs_inode.i_mtime = t;
-	vnode->vfs_inode.i_atime = t;
+	inode->i_mtime = t;
+	if (vp->update_ctime)
+		inode->i_ctime = op->ctime;
 
 	if (vnode->status.data_version != status->data_version)
 		data_changed = true;
 
 	vnode->status = *status;
 
-	if (expected_version &&
-	    *expected_version != status->data_version) {
+	if (vp->dv_before + vp->dv_delta != status->data_version) {
 		if (test_bit(AFS_VNODE_CB_PROMISED, &vnode->flags))
-			pr_warn("kAFS: vnode modified {%llx:%llu} %llx->%llx %s\n",
+			pr_warn("kAFS: vnode modified {%llx:%llu} %llx->%llx %s (op=%x)\n",
 				vnode->fid.vid, vnode->fid.vnode,
-				(unsigned long long)*expected_version,
+				(unsigned long long)vp->dv_before + vp->dv_delta,
 				(unsigned long long)status->data_version,
-				fc->type ? fc->type->name : "???");
+				op->type ? op->type->name : "???",
+				op->debug_id);
 
 		vnode->invalid_before = status->data_version;
 		if (vnode->status.type == AFS_FTYPE_DIR) {
@@ -225,6 +229,7 @@ static void afs_apply_status(struct afs_fs_cursor *fc,
 		} else {
 			set_bit(AFS_VNODE_ZAP_DATA, &vnode->flags);
 		}
+		change_size = true;
 	} else if (vnode->status.type == AFS_FTYPE_DIR) {
 		/* Expected directory change is handled elsewhere so
 		 * that we can locally edit the directory and save on a
@@ -232,33 +237,38 @@ static void afs_apply_status(struct afs_fs_cursor *fc,
 		 */
 		if (test_bit(AFS_VNODE_DIR_VALID, &vnode->flags))
 			data_changed = false;
+		change_size = true;
 	}
 
 	if (data_changed) {
-		inode_set_iversion_raw(&vnode->vfs_inode, status->data_version);
-		afs_set_i_size(vnode, status->size);
+		inode_set_iversion_raw(inode, status->data_version);
+
+		/* Only update the size if the data version jumped.  If the
+		 * file is being modified locally, then we might have our own
+		 * idea of what the size should be that's not the same as
+		 * what's on the server.
+		 */
+		vnode->netfs.remote_i_size = status->size;
+		if (change_size) {
+			afs_set_i_size(vnode, status->size);
+			inode->i_ctime = t;
+			inode->i_atime = t;
+		}
 	}
 }
 
 /*
  * Apply a callback to a vnode.
  */
-static void afs_apply_callback(struct afs_fs_cursor *fc,
-			       struct afs_vnode *vnode,
-			       struct afs_status_cb *scb,
-			       unsigned int cb_break)
+static void afs_apply_callback(struct afs_operation *op,
+			       struct afs_vnode_param *vp)
 {
-	struct afs_cb_interest *old;
-	struct afs_callback *cb = &scb->callback;
+	struct afs_callback *cb = &vp->scb.callback;
+	struct afs_vnode *vnode = vp->vnode;
 
-	if (!afs_cb_is_broken(cb_break, vnode, fc->cbi)) {
+	if (!afs_cb_is_broken(vp->cb_break_before, vnode)) {
 		vnode->cb_expires_at	= cb->expires_at;
-		old = rcu_dereference_protected(vnode->cb_interest,
-						lockdep_is_held(&vnode->cb_lock.lock));
-		if (old != fc->cbi) {
-			rcu_assign_pointer(vnode->cb_interest, afs_get_cb_interest(fc->cbi));
-			afs_put_cb_interest(afs_v2net(vnode), old);
-		}
+		vnode->cb_server	= op->server;
 		set_bit(AFS_VNODE_CB_PROMISED, &vnode->flags);
 	}
 }
@@ -267,106 +277,125 @@ static void afs_apply_callback(struct afs_fs_cursor *fc,
  * Apply the received status and callback to an inode all in the same critical
  * section to avoid races with afs_validate().
  */
-void afs_vnode_commit_status(struct afs_fs_cursor *fc,
-			     struct afs_vnode *vnode,
-			     unsigned int cb_break,
-			     const afs_dataversion_t *expected_version,
-			     struct afs_status_cb *scb)
+void afs_vnode_commit_status(struct afs_operation *op, struct afs_vnode_param *vp)
 {
-	if (fc->ac.error != 0)
-		return;
+	struct afs_vnode *vnode = vp->vnode;
+
+	_enter("");
 
 	write_seqlock(&vnode->cb_lock);
 
-	if (scb->have_error) {
-		if (scb->status.abort_code == VNOVNODE) {
+	if (vp->scb.have_error) {
+		/* A YFS server will return this from RemoveFile2 and AFS and
+		 * YFS will return this from InlineBulkStatus.
+		 */
+		if (vp->scb.status.abort_code == VNOVNODE) {
+			set_bit(AFS_VNODE_DELETED, &vnode->flags);
+			clear_nlink(&vnode->netfs.inode);
+			__afs_break_callback(vnode, afs_cb_break_for_deleted);
+			op->flags &= ~AFS_OPERATION_DIR_CONFLICT;
+		}
+	} else if (vp->scb.have_status) {
+		if (vp->speculative &&
+		    (test_bit(AFS_VNODE_MODIFYING, &vnode->flags) ||
+		     vp->dv_before != vnode->status.data_version))
+			/* Ignore the result of a speculative bulk status fetch
+			 * if it splits around a modification op, thereby
+			 * appearing to regress the data version.
+			 */
+			goto out;
+		afs_apply_status(op, vp);
+		if (vp->scb.have_cb)
+			afs_apply_callback(op, vp);
+	} else if (vp->op_unlinked && !(op->flags & AFS_OPERATION_DIR_CONFLICT)) {
+		drop_nlink(&vnode->netfs.inode);
+		if (vnode->netfs.inode.i_nlink == 0) {
 			set_bit(AFS_VNODE_DELETED, &vnode->flags);
-			clear_nlink(&vnode->vfs_inode);
 			__afs_break_callback(vnode, afs_cb_break_for_deleted);
 		}
-	} else {
-		if (scb->have_status)
-			afs_apply_status(fc, vnode, scb, expected_version);
-		if (scb->have_cb)
-			afs_apply_callback(fc, vnode, scb, cb_break);
 	}
 
+out:
 	write_sequnlock(&vnode->cb_lock);
 
-	if (fc->ac.error == 0 && scb->have_status)
-		afs_cache_permit(vnode, fc->key, cb_break, scb);
+	if (vp->scb.have_status)
+		afs_cache_permit(vnode, op->key, vp->cb_break_before, &vp->scb);
+}
+
+static void afs_fetch_status_success(struct afs_operation *op)
+{
+	struct afs_vnode_param *vp = &op->file[op->fetch_status.which];
+	struct afs_vnode *vnode = vp->vnode;
+	int ret;
+
+	if (vnode->netfs.inode.i_state & I_NEW) {
+		ret = afs_inode_init_from_status(op, vp, vnode);
+		op->error = ret;
+		if (ret == 0)
+			afs_cache_permit(vnode, op->key, vp->cb_break_before, &vp->scb);
+	} else {
+		afs_vnode_commit_status(op, vp);
+	}
 }
 
+const struct afs_operation_ops afs_fetch_status_operation = {
+	.issue_afs_rpc	= afs_fs_fetch_status,
+	.issue_yfs_rpc	= yfs_fs_fetch_status,
+	.success	= afs_fetch_status_success,
+	.aborted	= afs_check_for_remote_deletion,
+};
+
 /*
  * Fetch file status from the volume.
  */
 int afs_fetch_status(struct afs_vnode *vnode, struct key *key, bool is_new,
 		     afs_access_t *_caller_access)
 {
-	struct afs_status_cb *scb;
-	struct afs_fs_cursor fc;
-	int ret;
+	struct afs_operation *op;
 
 	_enter("%s,{%llx:%llu.%u,S=%lx}",
 	       vnode->volume->name,
 	       vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique,
 	       vnode->flags);
 
-	scb = kzalloc(sizeof(struct afs_status_cb), GFP_KERNEL);
-	if (!scb)
-		return -ENOMEM;
-
-	ret = -ERESTARTSYS;
-	if (afs_begin_vnode_operation(&fc, vnode, key, true)) {
-		afs_dataversion_t data_version = vnode->status.data_version;
+	op = afs_alloc_operation(key, vnode->volume);
+	if (IS_ERR(op))
+		return PTR_ERR(op);
 
-		while (afs_select_fileserver(&fc)) {
-			fc.cb_break = afs_calc_vnode_cb_break(vnode);
-			afs_fs_fetch_file_status(&fc, scb, NULL);
-		}
+	afs_op_set_vnode(op, 0, vnode);
 
-		if (fc.error) {
-			/* Do nothing. */
-		} else if (is_new) {
-			ret = afs_inode_init_from_status(vnode, key, fc.cbi,
-							 NULL, scb);
-			fc.error = ret;
-			if (ret == 0)
-				afs_cache_permit(vnode, key, fc.cb_break, scb);
-		} else {
-			afs_vnode_commit_status(&fc, vnode, fc.cb_break,
-						&data_version, scb);
-		}
-		afs_check_for_remote_deletion(&fc, vnode);
-		ret = afs_end_vnode_operation(&fc);
-	}
+	op->nr_files	= 1;
+	op->ops		= &afs_fetch_status_operation;
+	afs_begin_vnode_operation(op);
+	afs_wait_for_operation(op);
 
-	if (ret == 0 && _caller_access)
-		*_caller_access = scb->status.caller_access;
-	kfree(scb);
-	_leave(" = %d", ret);
-	return ret;
+	if (_caller_access)
+		*_caller_access = op->file[0].scb.status.caller_access;
+	return afs_put_operation(op);
 }
 
 /*
- * iget5() comparator
+ * ilookup() comparator
  */
-int afs_iget5_test(struct inode *inode, void *opaque)
+int afs_ilookup5_test_by_fid(struct inode *inode, void *opaque)
 {
-	struct afs_iget_data *iget_data = opaque;
 	struct afs_vnode *vnode = AFS_FS_I(inode);
+	struct afs_fid *fid = opaque;
 
-	return memcmp(&vnode->fid, &iget_data->fid, sizeof(iget_data->fid)) == 0;
+	return (fid->vnode == vnode->fid.vnode &&
+		fid->vnode_hi == vnode->fid.vnode_hi &&
+		fid->unique == vnode->fid.unique);
 }
 
 /*
- * iget5() comparator for inode created by autocell operations
- *
- * These pseudo inodes don't match anything.
+ * iget5() comparator
  */
-static int afs_iget5_pseudo_dir_test(struct inode *inode, void *opaque)
+static int afs_iget5_test(struct inode *inode, void *opaque)
 {
-	return 0;
+	struct afs_vnode_param *vp = opaque;
+	//struct afs_vnode *vnode = AFS_FS_I(inode);
+
+	return afs_ilookup5_test_by_fid(inode, &vp->fid);
 }
 
 /*
@@ -374,191 +403,169 @@ static int afs_iget5_pseudo_dir_test(struct inode *inode, void *opaque)
  */
 static int afs_iget5_set(struct inode *inode, void *opaque)
 {
-	struct afs_iget_data *iget_data = opaque;
+	struct afs_vnode_param *vp = opaque;
+	struct afs_super_info *as = AFS_FS_S(inode->i_sb);
 	struct afs_vnode *vnode = AFS_FS_I(inode);
 
-	vnode->fid		= iget_data->fid;
-	vnode->volume		= iget_data->volume;
-	vnode->cb_v_break	= iget_data->cb_v_break;
-	vnode->cb_s_break	= iget_data->cb_s_break;
+	vnode->volume		= as->volume;
+	vnode->fid		= vp->fid;
 
 	/* YFS supports 96-bit vnode IDs, but Linux only supports
 	 * 64-bit inode numbers.
 	 */
-	inode->i_ino		= iget_data->fid.vnode;
-	inode->i_generation	= iget_data->fid.unique;
+	inode->i_ino		= vnode->fid.vnode;
+	inode->i_generation	= vnode->fid.unique;
 	return 0;
 }
 
 /*
- * Create an inode for a dynamic root directory or an autocell dynamic
- * automount dir.
+ * Get a cache cookie for an inode.
  */
-struct inode *afs_iget_pseudo_dir(struct super_block *sb, bool root)
+static void afs_get_inode_cache(struct afs_vnode *vnode)
 {
-	struct afs_super_info *as;
-	struct afs_vnode *vnode;
-	struct inode *inode;
-	static atomic_t afs_autocell_ino;
+#ifdef CONFIG_AFS_FSCACHE
+	struct {
+		__be32 vnode_id;
+		__be32 unique;
+		__be32 vnode_id_ext[2];	/* Allow for a 96-bit key */
+	} __packed key;
+	struct afs_vnode_cache_aux aux;
 
-	struct afs_iget_data iget_data = {
-		.cb_v_break = 0,
-		.cb_s_break = 0,
-	};
+	if (vnode->status.type != AFS_FTYPE_FILE) {
+		vnode->netfs.cache = NULL;
+		return;
+	}
 
-	_enter("");
+	key.vnode_id		= htonl(vnode->fid.vnode);
+	key.unique		= htonl(vnode->fid.unique);
+	key.vnode_id_ext[0]	= htonl(vnode->fid.vnode >> 32);
+	key.vnode_id_ext[1]	= htonl(vnode->fid.vnode_hi);
+	afs_set_cache_aux(vnode, &aux);
+
+	afs_vnode_set_cache(vnode,
+			    fscache_acquire_cookie(
+				    vnode->volume->cache,
+				    vnode->status.type == AFS_FTYPE_FILE ?
+				    0 : FSCACHE_ADV_SINGLE_CHUNK,
+				    &key, sizeof(key),
+				    &aux, sizeof(aux),
+				    vnode->status.size));
+#endif
+}
 
-	as = sb->s_fs_info;
-	if (as->volume) {
-		iget_data.volume = as->volume;
-		iget_data.fid.vid = as->volume->vid;
-	}
-	if (root) {
-		iget_data.fid.vnode = 1;
-		iget_data.fid.unique = 1;
-	} else {
-		iget_data.fid.vnode = atomic_inc_return(&afs_autocell_ino);
-		iget_data.fid.unique = 0;
-	}
+/*
+ * inode retrieval
+ */
+struct inode *afs_iget(struct afs_operation *op, struct afs_vnode_param *vp)
+{
+	struct afs_vnode_param *dvp = &op->file[0];
+	struct super_block *sb = dvp->vnode->netfs.inode.i_sb;
+	struct afs_vnode *vnode;
+	struct inode *inode;
+	int ret;
+
+	_enter(",{%llx:%llu.%u},,", vp->fid.vid, vp->fid.vnode, vp->fid.unique);
 
-	inode = iget5_locked(sb, iget_data.fid.vnode,
-			     afs_iget5_pseudo_dir_test, afs_iget5_set,
-			     &iget_data);
+	inode = iget5_locked(sb, vp->fid.vnode, afs_iget5_test, afs_iget5_set, vp);
 	if (!inode) {
 		_leave(" = -ENOMEM");
 		return ERR_PTR(-ENOMEM);
 	}
 
-	_debug("GOT INODE %p { ino=%lu, vl=%llx, vn=%llx, u=%x }",
-	       inode, inode->i_ino, iget_data.fid.vid, iget_data.fid.vnode,
-	       iget_data.fid.unique);
-
 	vnode = AFS_FS_I(inode);
 
-	/* there shouldn't be an existing inode */
-	BUG_ON(!(inode->i_state & I_NEW));
+	_debug("GOT INODE %p { vl=%llx vn=%llx, u=%x }",
+	       inode, vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique);
 
-	inode->i_size		= 0;
-	inode->i_mode		= S_IFDIR | S_IRUGO | S_IXUGO;
-	if (root) {
-		inode->i_op	= &afs_dynroot_inode_operations;
-		inode->i_fop	= &simple_dir_operations;
-	} else {
-		inode->i_op	= &afs_autocell_inode_operations;
-	}
-	set_nlink(inode, 2);
-	inode->i_uid		= GLOBAL_ROOT_UID;
-	inode->i_gid		= GLOBAL_ROOT_GID;
-	inode->i_ctime = inode->i_atime = inode->i_mtime = current_time(inode);
-	inode->i_blocks		= 0;
-	inode_set_iversion_raw(inode, 0);
-	inode->i_generation	= 0;
-
-	set_bit(AFS_VNODE_PSEUDODIR, &vnode->flags);
-	if (!root) {
-		set_bit(AFS_VNODE_MOUNTPOINT, &vnode->flags);
-		inode->i_flags |= S_AUTOMOUNT;
+	/* deal with an existing inode */
+	if (!(inode->i_state & I_NEW)) {
+		_leave(" = %p", inode);
+		return inode;
 	}
 
-	inode->i_flags |= S_NOATIME;
+	ret = afs_inode_init_from_status(op, vp, vnode);
+	if (ret < 0)
+		goto bad_inode;
+
+	afs_get_inode_cache(vnode);
+
+	/* success */
+	clear_bit(AFS_VNODE_UNSET, &vnode->flags);
 	unlock_new_inode(inode);
 	_leave(" = %p", inode);
 	return inode;
+
+	/* failure */
+bad_inode:
+	iget_failed(inode);
+	_leave(" = %d [bad]", ret);
+	return ERR_PTR(ret);
 }
 
-/*
- * Get a cache cookie for an inode.
- */
-static void afs_get_inode_cache(struct afs_vnode *vnode)
+static int afs_iget5_set_root(struct inode *inode, void *opaque)
 {
-#ifdef CONFIG_AFS_FSCACHE
-	struct {
-		u32 vnode_id;
-		u32 unique;
-		u32 vnode_id_ext[2];	/* Allow for a 96-bit key */
-	} __packed key;
-	struct afs_vnode_cache_aux aux;
-
-	if (vnode->status.type == AFS_FTYPE_DIR) {
-		vnode->cache = NULL;
-		return;
-	}
+	struct afs_super_info *as = AFS_FS_S(inode->i_sb);
+	struct afs_vnode *vnode = AFS_FS_I(inode);
 
-	key.vnode_id		= vnode->fid.vnode;
-	key.unique		= vnode->fid.unique;
-	key.vnode_id_ext[0]	= vnode->fid.vnode >> 32;
-	key.vnode_id_ext[1]	= vnode->fid.vnode_hi;
-	aux.data_version	= vnode->status.data_version;
-
-	vnode->cache = fscache_acquire_cookie(vnode->volume->cache,
-					      &afs_vnode_cache_index_def,
-					      &key, sizeof(key),
-					      &aux, sizeof(aux),
-					      vnode, vnode->status.size, true);
-#endif
+	vnode->volume		= as->volume;
+	vnode->fid.vid		= as->volume->vid,
+	vnode->fid.vnode	= 1;
+	vnode->fid.unique	= 1;
+	inode->i_ino		= 1;
+	inode->i_generation	= 1;
+	return 0;
 }
 
 /*
- * inode retrieval
+ * Set up the root inode for a volume.  This is always vnode 1, unique 1 within
+ * the volume.
  */
-struct inode *afs_iget(struct super_block *sb, struct key *key,
-		       struct afs_iget_data *iget_data,
-		       struct afs_status_cb *scb,
-		       struct afs_cb_interest *cbi,
-		       struct afs_vnode *parent_vnode)
+struct inode *afs_root_iget(struct super_block *sb, struct key *key)
 {
-	struct afs_super_info *as;
+	struct afs_super_info *as = AFS_FS_S(sb);
+	struct afs_operation *op;
 	struct afs_vnode *vnode;
-	struct afs_fid *fid = &iget_data->fid;
 	struct inode *inode;
 	int ret;
 
-	_enter(",{%llx:%llu.%u},,", fid->vid, fid->vnode, fid->unique);
+	_enter(",{%llx},,", as->volume->vid);
 
-	as = sb->s_fs_info;
-	iget_data->volume = as->volume;
-
-	inode = iget5_locked(sb, fid->vnode, afs_iget5_test, afs_iget5_set,
-			     iget_data);
+	inode = iget5_locked(sb, 1, NULL, afs_iget5_set_root, NULL);
 	if (!inode) {
 		_leave(" = -ENOMEM");
 		return ERR_PTR(-ENOMEM);
 	}
 
-	_debug("GOT INODE %p { vl=%llx vn=%llx, u=%x }",
-	       inode, fid->vid, fid->vnode, fid->unique);
+	_debug("GOT ROOT INODE %p { vl=%llx }", inode, as->volume->vid);
+
+	BUG_ON(!(inode->i_state & I_NEW));
 
 	vnode = AFS_FS_I(inode);
+	vnode->cb_v_break = as->volume->cb_v_break,
+	afs_set_netfs_context(vnode);
 
-	/* deal with an existing inode */
-	if (!(inode->i_state & I_NEW)) {
-		_leave(" = %p", inode);
-		return inode;
+	op = afs_alloc_operation(key, as->volume);
+	if (IS_ERR(op)) {
+		ret = PTR_ERR(op);
+		goto error;
 	}
 
-	if (!scb) {
-		/* it's a remotely extant inode */
-		ret = afs_fetch_status(vnode, key, true, NULL);
-		if (ret < 0)
-			goto bad_inode;
-	} else {
-		ret = afs_inode_init_from_status(vnode, key, cbi, parent_vnode,
-						 scb);
-		if (ret < 0)
-			goto bad_inode;
-	}
+	afs_op_set_vnode(op, 0, vnode);
+
+	op->nr_files	= 1;
+	op->ops		= &afs_fetch_status_operation;
+	ret = afs_do_sync_operation(op);
+	if (ret < 0)
+		goto error;
 
 	afs_get_inode_cache(vnode);
 
-	/* success */
 	clear_bit(AFS_VNODE_UNSET, &vnode->flags);
-	inode->i_flags |= S_NOATIME;
 	unlock_new_inode(inode);
 	_leave(" = %p", inode);
 	return inode;
 
-	/* failure */
-bad_inode:
+error:
 	iget_failed(inode);
 	_leave(" = %d [bad]", ret);
 	return ERR_PTR(ret);
@@ -568,21 +575,49 @@ bad_inode:
  * mark the data attached to an inode as obsolete due to a write on the server
  * - might also want to ditch all the outstanding writes and dirty pages
  */
-void afs_zap_data(struct afs_vnode *vnode)
+static void afs_zap_data(struct afs_vnode *vnode)
 {
 	_enter("{%llx:%llu}", vnode->fid.vid, vnode->fid.vnode);
 
-#ifdef CONFIG_AFS_FSCACHE
-	fscache_invalidate(vnode->cache);
-#endif
+	afs_invalidate_cache(vnode, 0);
 
 	/* nuke all the non-dirty pages that aren't locked, mapped or being
 	 * written back in a regular file and completely discard the pages in a
 	 * directory or symlink */
-	if (S_ISREG(vnode->vfs_inode.i_mode))
-		invalidate_remote_inode(&vnode->vfs_inode);
+	if (S_ISREG(vnode->netfs.inode.i_mode))
+		invalidate_remote_inode(&vnode->netfs.inode);
 	else
-		invalidate_inode_pages2(vnode->vfs_inode.i_mapping);
+		invalidate_inode_pages2(vnode->netfs.inode.i_mapping);
+}
+
+/*
+ * Check to see if we have a server currently serving this volume and that it
+ * hasn't been reinitialised or dropped from the list.
+ */
+static bool afs_check_server_good(struct afs_vnode *vnode)
+{
+	struct afs_server_list *slist;
+	struct afs_server *server;
+	bool good;
+	int i;
+
+	if (vnode->cb_fs_s_break == atomic_read(&vnode->volume->cell->fs_s_break))
+		return true;
+
+	rcu_read_lock();
+
+	slist = rcu_dereference(vnode->volume->servers);
+	for (i = 0; i < slist->nr_servers; i++) {
+		server = slist->servers[i].server;
+		if (server == vnode->cb_server) {
+			good = (vnode->cb_s_break == server->cb_s_break);
+			rcu_read_unlock();
+			return good;
+		}
+	}
+
+	rcu_read_unlock();
+	return false;
 }
 
 /*
@@ -590,62 +625,46 @@ void afs_zap_data(struct afs_vnode *vnode)
  */
 bool afs_check_validity(struct afs_vnode *vnode)
 {
-	struct afs_cb_interest *cbi;
-	struct afs_server *server;
-	struct afs_volume *volume = vnode->volume;
 	enum afs_cb_break_reason need_clear = afs_cb_break_no_break;
 	time64_t now = ktime_get_real_seconds();
-	bool valid;
-	unsigned int cb_break, cb_s_break, cb_v_break;
+	unsigned int cb_break;
 	int seq = 0;
 
 	do {
 		read_seqbegin_or_lock(&vnode->cb_lock, &seq);
-		cb_v_break = READ_ONCE(volume->cb_v_break);
 		cb_break = vnode->cb_break;
 
 		if (test_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) {
-			cbi = rcu_dereference(vnode->cb_interest);
-			server = rcu_dereference(cbi->server);
-			cb_s_break = READ_ONCE(server->cb_s_break);
-
-			if (vnode->cb_s_break != cb_s_break ||
-			    vnode->cb_v_break != cb_v_break) {
-				vnode->cb_s_break = cb_s_break;
-				vnode->cb_v_break = cb_v_break;
-				need_clear = afs_cb_break_for_vsbreak;
-				valid = false;
-			} else if (test_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) {
+			if (vnode->cb_v_break != vnode->volume->cb_v_break)
+				need_clear = afs_cb_break_for_v_break;
+			else if (!afs_check_server_good(vnode))
+				need_clear = afs_cb_break_for_s_reinit;
+			else if (test_bit(AFS_VNODE_ZAP_DATA, &vnode->flags))
 				need_clear = afs_cb_break_for_zap;
-				valid = false;
-			} else if (vnode->cb_expires_at - 10 <= now) {
+			else if (vnode->cb_expires_at - 10 <= now)
 				need_clear = afs_cb_break_for_lapsed;
-				valid = false;
-			} else {
-				valid = true;
-			}
 		} else if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) {
-			valid = true;
+			;
 		} else {
-			vnode->cb_v_break = cb_v_break;
-			valid = false;
+			need_clear = afs_cb_break_no_promise;
 		}
 
 	} while (need_seqretry(&vnode->cb_lock, seq));
 
 	done_seqretry(&vnode->cb_lock, seq);
 
-	if (need_clear != afs_cb_break_no_break) {
-		write_seqlock(&vnode->cb_lock);
-		if (cb_break == vnode->cb_break)
-			__afs_break_callback(vnode, need_clear);
-		else
-			trace_afs_cb_miss(&vnode->fid, need_clear);
-		write_sequnlock(&vnode->cb_lock);
-		valid = false;
-	}
+	if (need_clear == afs_cb_break_no_break)
+		return true;
 
-	return valid;
+	write_seqlock(&vnode->cb_lock);
+	if (need_clear == afs_cb_break_no_promise)
+		vnode->cb_v_break = vnode->volume->cb_v_break;
+	else if (cb_break == vnode->cb_break)
+		__afs_break_callback(vnode, need_clear);
+	else
+		trace_afs_cb_miss(&vnode->fid, need_clear);
+	write_sequnlock(&vnode->cb_lock);
+	return false;
 }
 
 /*
@@ -659,21 +678,20 @@ bool afs_check_validity(struct afs_vnode *vnode)
  */
 int afs_validate(struct afs_vnode *vnode, struct key *key)
 {
-	bool valid;
 	int ret;
 
 	_enter("{v={%llx:%llu} fl=%lx},%x",
 	       vnode->fid.vid, vnode->fid.vnode, vnode->flags,
 	       key_serial(key));
 
-	rcu_read_lock();
-	valid = afs_check_validity(vnode);
-	rcu_read_unlock();
-
-	if (test_bit(AFS_VNODE_DELETED, &vnode->flags))
-		clear_nlink(&vnode->vfs_inode);
+	if (unlikely(test_bit(AFS_VNODE_DELETED, &vnode->flags))) {
+		if (vnode->netfs.inode.i_nlink)
+			clear_nlink(&vnode->netfs.inode);
+		goto valid;
+	}
 
-	if (valid)
+	if (test_bit(AFS_VNODE_CB_PROMISED, &vnode->flags) &&
+	    afs_check_validity(vnode))
 		goto valid;
 
 	down_write(&vnode->validate_lock);
@@ -719,18 +737,34 @@ error_unlock:
 /*
  * read the attributes of an inode
  */
-int afs_getattr(const struct path *path, struct kstat *stat,
-		u32 request_mask, unsigned int query_flags)
+int afs_getattr(struct user_namespace *mnt_userns, const struct path *path,
+		struct kstat *stat, u32 request_mask, unsigned int query_flags)
 {
 	struct inode *inode = d_inode(path->dentry);
 	struct afs_vnode *vnode = AFS_FS_I(inode);
-	int seq = 0;
+	struct key *key;
+	int ret, seq = 0;
 
 	_enter("{ ino=%lu v=%u }", inode->i_ino, inode->i_generation);
 
+	if (vnode->volume &&
+	    !(query_flags & AT_STATX_DONT_SYNC) &&
+	    !test_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) {
+		key = afs_request_key(vnode->volume->cell);
+		if (IS_ERR(key))
+			return PTR_ERR(key);
+		ret = afs_validate(vnode, key);
+		key_put(key);
+		if (ret < 0)
+			return ret;
+	}
+
 	do {
 		read_seqbegin_or_lock(&vnode->cb_lock, &seq);
-		generic_fillattr(inode, stat);
+		generic_fillattr(&init_user_ns, inode, stat);
+		if (test_bit(AFS_VNODE_SILLY_DELETED, &vnode->flags) &&
+		    stat->nlink > 0)
+			stat->nlink -= 1;
 	} while (need_seqretry(&vnode->cb_lock, seq));
 
 	done_seqretry(&vnode->cb_lock, seq);
@@ -755,10 +789,8 @@ int afs_drop_inode(struct inode *inode)
  */
 void afs_evict_inode(struct inode *inode)
 {
-	struct afs_cb_interest *cbi;
-	struct afs_vnode *vnode;
-
-	vnode = AFS_FS_I(inode);
+	struct afs_vnode_cache_aux aux;
+	struct afs_vnode *vnode = AFS_FS_I(inode);
 
 	_enter("{%llx:%llu.%d}",
 	       vnode->fid.vid,
@@ -770,16 +802,10 @@ void afs_evict_inode(struct inode *inode)
 	ASSERTCMP(inode->i_ino, ==, vnode->fid.vnode);
 
 	truncate_inode_pages_final(&inode->i_data);
-	clear_inode(inode);
 
-	write_seqlock(&vnode->cb_lock);
-	cbi = rcu_dereference_protected(vnode->cb_interest,
-					lockdep_is_held(&vnode->cb_lock.lock));
-	if (cbi) {
-		afs_put_cb_interest(afs_i2net(inode), cbi);
-		rcu_assign_pointer(vnode->cb_interest, NULL);
-	}
-	write_sequnlock(&vnode->cb_lock);
+	afs_set_cache_aux(vnode, &aux);
+	fscache_clear_inode_writeback(afs_vnode_cache(vnode), inode, &aux);
+	clear_inode(inode);
 
 	while (!list_empty(&vnode->wb_keys)) {
 		struct afs_wb_key *wbk = list_entry(vnode->wb_keys.next,
@@ -788,16 +814,8 @@ void afs_evict_inode(struct inode *inode)
 		afs_put_wb_key(wbk);
 	}
 
-#ifdef CONFIG_AFS_FSCACHE
-	{
-		struct afs_vnode_cache_aux aux;
-
-		aux.data_version = vnode->status.data_version;
-		fscache_relinquish_cookie(vnode->cache, &aux,
-					  test_bit(AFS_VNODE_DELETED, &vnode->flags));
-		vnode->cache = NULL;
-	}
-#endif
+	fscache_relinquish_cookie(afs_vnode_cache(vnode),
+				  test_bit(AFS_VNODE_DELETED, &vnode->flags));
 
 	afs_prune_wb_keys(vnode);
 	afs_put_permits(rcu_access_pointer(vnode->permit_cache));
@@ -808,69 +826,139 @@ void afs_evict_inode(struct inode *inode)
 	_leave("");
 }
 
+static void afs_setattr_success(struct afs_operation *op)
+{
+	struct afs_vnode_param *vp = &op->file[0];
+	struct inode *inode = &vp->vnode->netfs.inode;
+	loff_t old_i_size = i_size_read(inode);
+
+	op->setattr.old_i_size = old_i_size;
+	afs_vnode_commit_status(op, vp);
+	/* inode->i_size has now been changed. */
+
+	if (op->setattr.attr->ia_valid & ATTR_SIZE) {
+		loff_t size = op->setattr.attr->ia_size;
+		if (size > old_i_size)
+			pagecache_isize_extended(inode, old_i_size, size);
+	}
+}
+
+static void afs_setattr_edit_file(struct afs_operation *op)
+{
+	struct afs_vnode_param *vp = &op->file[0];
+	struct inode *inode = &vp->vnode->netfs.inode;
+
+	if (op->setattr.attr->ia_valid & ATTR_SIZE) {
+		loff_t size = op->setattr.attr->ia_size;
+		loff_t i_size = op->setattr.old_i_size;
+
+		if (size < i_size)
+			truncate_pagecache(inode, size);
+		if (size != i_size)
+			fscache_resize_cookie(afs_vnode_cache(vp->vnode),
+					      vp->scb.status.size);
+	}
+}
+
+static const struct afs_operation_ops afs_setattr_operation = {
+	.issue_afs_rpc	= afs_fs_setattr,
+	.issue_yfs_rpc	= yfs_fs_setattr,
+	.success	= afs_setattr_success,
+	.edit_dir	= afs_setattr_edit_file,
+};
+
 /*
  * set the attributes of an inode
  */
-int afs_setattr(struct dentry *dentry, struct iattr *attr)
+int afs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry,
+		struct iattr *attr)
 {
-	struct afs_fs_cursor fc;
-	struct afs_status_cb *scb;
+	const unsigned int supported =
+		ATTR_SIZE | ATTR_MODE | ATTR_UID | ATTR_GID |
+		ATTR_MTIME | ATTR_MTIME_SET | ATTR_TIMES_SET | ATTR_TOUCH;
+	struct afs_operation *op;
 	struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry));
-	struct key *key;
-	int ret = -ENOMEM;
+	struct inode *inode = &vnode->netfs.inode;
+	loff_t i_size;
+	int ret;
 
 	_enter("{%llx:%llu},{n=%pd},%x",
 	       vnode->fid.vid, vnode->fid.vnode, dentry,
 	       attr->ia_valid);
 
-	if (!(attr->ia_valid & (ATTR_SIZE | ATTR_MODE | ATTR_UID | ATTR_GID |
-				ATTR_MTIME))) {
+	if (!(attr->ia_valid & supported)) {
 		_leave(" = 0 [unsupported]");
 		return 0;
 	}
 
-	scb = kzalloc(sizeof(struct afs_status_cb), GFP_KERNEL);
-	if (!scb)
-		goto error;
+	i_size = i_size_read(inode);
+	if (attr->ia_valid & ATTR_SIZE) {
+		if (!S_ISREG(inode->i_mode))
+			return -EISDIR;
 
-	/* flush any dirty data outstanding on a regular file */
-	if (S_ISREG(vnode->vfs_inode.i_mode))
-		filemap_write_and_wait(vnode->vfs_inode.i_mapping);
+		ret = inode_newsize_ok(inode, attr->ia_size);
+		if (ret)
+			return ret;
 
-	if (attr->ia_valid & ATTR_FILE) {
-		key = afs_file_key(attr->ia_file);
-	} else {
-		key = afs_request_key(vnode->volume->cell);
-		if (IS_ERR(key)) {
-			ret = PTR_ERR(key);
-			goto error_scb;
-		}
+		if (attr->ia_size == i_size)
+			attr->ia_valid &= ~ATTR_SIZE;
 	}
 
-	ret = -ERESTARTSYS;
-	if (afs_begin_vnode_operation(&fc, vnode, key, false)) {
-		afs_dataversion_t data_version = vnode->status.data_version;
+	fscache_use_cookie(afs_vnode_cache(vnode), true);
 
-		if (attr->ia_valid & ATTR_SIZE)
-			data_version++;
+	/* Prevent any new writebacks from starting whilst we do this. */
+	down_write(&vnode->validate_lock);
+
+	if ((attr->ia_valid & ATTR_SIZE) && S_ISREG(inode->i_mode)) {
+		loff_t size = attr->ia_size;
+
+		/* Wait for any outstanding writes to the server to complete */
+		loff_t from = min(size, i_size);
+		loff_t to = max(size, i_size);
+		ret = filemap_fdatawait_range(inode->i_mapping, from, to);
+		if (ret < 0)
+			goto out_unlock;
 
-		while (afs_select_fileserver(&fc)) {
-			fc.cb_break = afs_calc_vnode_cb_break(vnode);
-			afs_fs_setattr(&fc, attr, scb);
+		/* Don't talk to the server if we're just shortening in-memory
+		 * writes that haven't gone to the server yet.
+		 */
+		if (!(attr->ia_valid & (supported & ~ATTR_SIZE & ~ATTR_MTIME)) &&
+		    attr->ia_size < i_size &&
+		    attr->ia_size > vnode->status.size) {
+			truncate_pagecache(inode, attr->ia_size);
+			fscache_resize_cookie(afs_vnode_cache(vnode),
+					      attr->ia_size);
+			i_size_write(inode, attr->ia_size);
+			ret = 0;
+			goto out_unlock;
 		}
+	}
 
-		afs_check_for_remote_deletion(&fc, vnode);
-		afs_vnode_commit_status(&fc, vnode, fc.cb_break,
-					&data_version, scb);
-		ret = afs_end_vnode_operation(&fc);
+	op = afs_alloc_operation(((attr->ia_valid & ATTR_FILE) ?
+				  afs_file_key(attr->ia_file) : NULL),
+				 vnode->volume);
+	if (IS_ERR(op)) {
+		ret = PTR_ERR(op);
+		goto out_unlock;
 	}
 
-	if (!(attr->ia_valid & ATTR_FILE))
-		key_put(key);
+	afs_op_set_vnode(op, 0, vnode);
+	op->setattr.attr = attr;
 
-error_scb:
-	kfree(scb);
-error:
+	if (attr->ia_valid & ATTR_SIZE) {
+		op->file[0].dv_delta = 1;
+		op->file[0].set_size = true;
+	}
+	op->ctime = attr->ia_ctime;
+	op->file[0].update_ctime = 1;
+	op->file[0].modification = true;
+
+	op->ops = &afs_setattr_operation;
+	ret = afs_do_sync_operation(op);
+
+out_unlock:
+	up_write(&vnode->validate_lock);
+	fscache_unuse_cookie(afs_vnode_cache(vnode), NULL, NULL);
 	_leave(" = %d", ret);
 	return ret;
 }