Merge tag 'ceph-for-4.19-rc1' of git://github.com/ceph/ceph-client

Pull ceph updates from Ilya Dryomov:
 "The main things are support for cephx v2 authentication protocol and
  basic support for rbd images within namespaces (myself).

  Also included are y2038 conversion patches from Arnd, a pile of
  miscellaneous fixes from Chengguang and Zheng's feature bit
  infrastructure for the filesystem"

* tag 'ceph-for-4.19-rc1' of git://github.com/ceph/ceph-client: (40 commits)
  ceph: don't drop message if it contains more data than expected
  ceph: support cephfs' own feature bits
  crush: fix using plain integer as NULL warning
  libceph: remove unnecessary non NULL check for request_key
  ceph: refactor error handling code in ceph_reserve_caps()
  ceph: refactor ceph_unreserve_caps()
  ceph: change to void return type for __do_request()
  ceph: compare fsc->max_file_size and inode->i_size for max file size limit
  ceph: add additional size check in ceph_setattr()
  ceph: add additional offset check in ceph_write_iter()
  ceph: add additional range check in ceph_fallocate()
  ceph: add new field max_file_size in ceph_fs_client
  libceph: weaken sizeof check in ceph_x_verify_authorizer_reply()
  libceph: check authorizer reply/challenge length before reading
  libceph: implement CEPHX_V2 calculation mode
  libceph: add authorizer challenge
  libceph: factor out encrypt_authorizer()
  libceph: factor out __ceph_x_decrypt()
  libceph: factor out __prepare_write_connect()
  libceph: store ceph_auth_handshake pointer in ceph_connection
  ...

drivers/block/rbd.c

@@ -181,6 +181,7 @@ struct rbd_image_header {
 struct rbd_spec {
 	u64		pool_id;
 	const char	*pool_name;
+	const char	*pool_ns;	/* NULL if default, never "" */
 
 	const char	*image_id;
 	const char	*image_name;
@@ -735,6 +736,7 @@ enum {
 	Opt_lock_timeout,
 	Opt_last_int,
 	/* int args above */
+	Opt_pool_ns,
 	Opt_last_string,
 	/* string args above */
 	Opt_read_only,
@@ -749,6 +751,7 @@ static match_table_t rbd_opts_tokens = {
 	{Opt_queue_depth, "queue_depth=%d"},
 	{Opt_lock_timeout, "lock_timeout=%d"},
 	/* int args above */
+	{Opt_pool_ns, "_pool_ns=%s"},
 	/* string args above */
 	{Opt_read_only, "read_only"},
 	{Opt_read_only, "ro"},		/* Alternate spelling */
@@ -776,9 +779,14 @@ struct rbd_options {
 #define RBD_EXCLUSIVE_DEFAULT	false
 #define RBD_TRIM_DEFAULT	true
 
+struct parse_rbd_opts_ctx {
+	struct rbd_spec		*spec;
+	struct rbd_options	*opts;
+};
+
 static int parse_rbd_opts_token(char *c, void *private)
 {
-	struct rbd_options *rbd_opts = private;
+	struct parse_rbd_opts_ctx *pctx = private;
 	substring_t argstr[MAX_OPT_ARGS];
 	int token, intval, ret;
 
@@ -786,7 +794,7 @@ static int parse_rbd_opts_token(char *c, void *private)
 	if (token < Opt_last_int) {
 		ret = match_int(&argstr[0], &intval);
 		if (ret < 0) {
-			pr_err("bad mount option arg (not int) at '%s'\n", c);
+			pr_err("bad option arg (not int) at '%s'\n", c);
 			return ret;
 		}
 		dout("got int token %d val %d\n", token, intval);
@@ -802,7 +810,7 @@ static int parse_rbd_opts_token(char *c, void *private)
 			pr_err("queue_depth out of range\n");
 			return -EINVAL;
 		}
-		rbd_opts->queue_depth = intval;
+		pctx->opts->queue_depth = intval;
 		break;
 	case Opt_lock_timeout:
 		/* 0 is "wait forever" (i.e. infinite timeout) */
@@ -810,22 +818,28 @@ static int parse_rbd_opts_token(char *c, void *private)
 			pr_err("lock_timeout out of range\n");
 			return -EINVAL;
 		}
-		rbd_opts->lock_timeout = msecs_to_jiffies(intval * 1000);
+		pctx->opts->lock_timeout = msecs_to_jiffies(intval * 1000);
+		break;
+	case Opt_pool_ns:
+		kfree(pctx->spec->pool_ns);
+		pctx->spec->pool_ns = match_strdup(argstr);
+		if (!pctx->spec->pool_ns)
+			return -ENOMEM;
 		break;
 	case Opt_read_only:
-		rbd_opts->read_only = true;
+		pctx->opts->read_only = true;
 		break;
 	case Opt_read_write:
-		rbd_opts->read_only = false;
+		pctx->opts->read_only = false;
 		break;
 	case Opt_lock_on_read:
-		rbd_opts->lock_on_read = true;
+		pctx->opts->lock_on_read = true;
 		break;
 	case Opt_exclusive:
-		rbd_opts->exclusive = true;
+		pctx->opts->exclusive = true;
 		break;
 	case Opt_notrim:
-		rbd_opts->trim = false;
+		pctx->opts->trim = false;
 		break;
 	default:
 		/* libceph prints "bad option" msg */
@@ -1452,7 +1466,7 @@ static void rbd_osd_req_format_write(struct rbd_obj_request *obj_request)
 	struct ceph_osd_request *osd_req = obj_request->osd_req;
 
 	osd_req->r_flags = CEPH_OSD_FLAG_WRITE;
-	ktime_get_real_ts(&osd_req->r_mtime);
+	ktime_get_real_ts64(&osd_req->r_mtime);
 	osd_req->r_data_offset = obj_request->ex.oe_off;
 }
 
@@ -1475,7 +1489,13 @@ rbd_osd_req_create(struct rbd_obj_request *obj_req, unsigned int num_ops)
 	req->r_callback = rbd_osd_req_callback;
 	req->r_priv = obj_req;
 
+	/*
+	 * Data objects may be stored in a separate pool, but always in
+	 * the same namespace in that pool as the header in its pool.
+	 */
+	ceph_oloc_copy(&req->r_base_oloc, &rbd_dev->header_oloc);
 	req->r_base_oloc.pool = rbd_dev->layout.pool_id;
+
 	if (ceph_oid_aprintf(&req->r_base_oid, GFP_NOIO, name_format,
 			rbd_dev->header.object_prefix, obj_req->ex.oe_objno))
 		goto err_req;
@@ -4119,6 +4139,14 @@ static ssize_t rbd_pool_id_show(struct device *dev,
 			(unsigned long long) rbd_dev->spec->pool_id);
 }
 
+static ssize_t rbd_pool_ns_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
+
+	return sprintf(buf, "%s\n", rbd_dev->spec->pool_ns ?: "");
+}
+
 static ssize_t rbd_name_show(struct device *dev,
 			     struct device_attribute *attr, char *buf)
 {
@@ -4217,6 +4245,7 @@ static DEVICE_ATTR(cluster_fsid, 0444, rbd_cluster_fsid_show, NULL);
 static DEVICE_ATTR(config_info, 0400, rbd_config_info_show, NULL);
 static DEVICE_ATTR(pool, 0444, rbd_pool_show, NULL);
 static DEVICE_ATTR(pool_id, 0444, rbd_pool_id_show, NULL);
+static DEVICE_ATTR(pool_ns, 0444, rbd_pool_ns_show, NULL);
 static DEVICE_ATTR(name, 0444, rbd_name_show, NULL);
 static DEVICE_ATTR(image_id, 0444, rbd_image_id_show, NULL);
 static DEVICE_ATTR(refresh, 0200, NULL, rbd_image_refresh);
@@ -4235,6 +4264,7 @@ static struct attribute *rbd_attrs[] = {
 	&dev_attr_config_info.attr,
 	&dev_attr_pool.attr,
 	&dev_attr_pool_id.attr,
+	&dev_attr_pool_ns.attr,
 	&dev_attr_name.attr,
 	&dev_attr_image_id.attr,
 	&dev_attr_current_snap.attr,
@@ -4295,6 +4325,7 @@ static void rbd_spec_free(struct kref *kref)
 	struct rbd_spec *spec = container_of(kref, struct rbd_spec, kref);
 
 	kfree(spec->pool_name);
+	kfree(spec->pool_ns);
 	kfree(spec->image_id);
 	kfree(spec->image_name);
 	kfree(spec->snap_name);
@@ -4353,6 +4384,12 @@ static struct rbd_device *__rbd_dev_create(struct rbd_client *rbdc,
 	rbd_dev->header.data_pool_id = CEPH_NOPOOL;
 	ceph_oid_init(&rbd_dev->header_oid);
 	rbd_dev->header_oloc.pool = spec->pool_id;
+	if (spec->pool_ns) {
+		WARN_ON(!*spec->pool_ns);
+		rbd_dev->header_oloc.pool_ns =
+		    ceph_find_or_create_string(spec->pool_ns,
+					       strlen(spec->pool_ns));
+	}
 
 	mutex_init(&rbd_dev->watch_mutex);
 	rbd_dev->watch_state = RBD_WATCH_STATE_UNREGISTERED;
@@ -4633,6 +4670,17 @@ static int rbd_dev_v2_parent_info(struct rbd_device *rbd_dev)
 		parent_spec->pool_id = pool_id;
 		parent_spec->image_id = image_id;
 		parent_spec->snap_id = snap_id;
+
+		/* TODO: support cloning across namespaces */
+		if (rbd_dev->spec->pool_ns) {
+			parent_spec->pool_ns = kstrdup(rbd_dev->spec->pool_ns,
+						       GFP_KERNEL);
+			if (!parent_spec->pool_ns) {
+				ret = -ENOMEM;
+				goto out_err;
+			}
+		}
+
 		rbd_dev->parent_spec = parent_spec;
 		parent_spec = NULL;	/* rbd_dev now owns this */
 	} else {
@@ -5146,8 +5194,7 @@ static int rbd_add_parse_args(const char *buf,
 	const char *mon_addrs;
 	char *snap_name;
 	size_t mon_addrs_size;
-	struct rbd_spec *spec = NULL;
-	struct rbd_options *rbd_opts = NULL;
+	struct parse_rbd_opts_ctx pctx = { 0 };
 	struct ceph_options *copts;
 	int ret;
 
@@ -5171,22 +5218,22 @@ static int rbd_add_parse_args(const char *buf,
 		goto out_err;
 	}
 
-	spec = rbd_spec_alloc();
-	if (!spec)
+	pctx.spec = rbd_spec_alloc();
+	if (!pctx.spec)
 		goto out_mem;
 
-	spec->pool_name = dup_token(&buf, NULL);
-	if (!spec->pool_name)
+	pctx.spec->pool_name = dup_token(&buf, NULL);
+	if (!pctx.spec->pool_name)
 		goto out_mem;
-	if (!*spec->pool_name) {
+	if (!*pctx.spec->pool_name) {
 		rbd_warn(NULL, "no pool name provided");
 		goto out_err;
 	}
 
-	spec->image_name = dup_token(&buf, NULL);
-	if (!spec->image_name)
+	pctx.spec->image_name = dup_token(&buf, NULL);
+	if (!pctx.spec->image_name)
 		goto out_mem;
-	if (!*spec->image_name) {
+	if (!*pctx.spec->image_name) {
 		rbd_warn(NULL, "no image name provided");
 		goto out_err;
 	}
@@ -5207,24 +5254,24 @@ static int rbd_add_parse_args(const char *buf,
 	if (!snap_name)
 		goto out_mem;
 	*(snap_name + len) = '\0';
-	spec->snap_name = snap_name;
+	pctx.spec->snap_name = snap_name;
 
 	/* Initialize all rbd options to the defaults */
 
-	rbd_opts = kzalloc(sizeof (*rbd_opts), GFP_KERNEL);
-	if (!rbd_opts)
+	pctx.opts = kzalloc(sizeof(*pctx.opts), GFP_KERNEL);
+	if (!pctx.opts)
 		goto out_mem;
 
-	rbd_opts->read_only = RBD_READ_ONLY_DEFAULT;
-	rbd_opts->queue_depth = RBD_QUEUE_DEPTH_DEFAULT;
-	rbd_opts->lock_timeout = RBD_LOCK_TIMEOUT_DEFAULT;
-	rbd_opts->lock_on_read = RBD_LOCK_ON_READ_DEFAULT;
-	rbd_opts->exclusive = RBD_EXCLUSIVE_DEFAULT;
-	rbd_opts->trim = RBD_TRIM_DEFAULT;
+	pctx.opts->read_only = RBD_READ_ONLY_DEFAULT;
+	pctx.opts->queue_depth = RBD_QUEUE_DEPTH_DEFAULT;
+	pctx.opts->lock_timeout = RBD_LOCK_TIMEOUT_DEFAULT;
+	pctx.opts->lock_on_read = RBD_LOCK_ON_READ_DEFAULT;
+	pctx.opts->exclusive = RBD_EXCLUSIVE_DEFAULT;
+	pctx.opts->trim = RBD_TRIM_DEFAULT;
 
 	copts = ceph_parse_options(options, mon_addrs,
-					mon_addrs + mon_addrs_size - 1,
-					parse_rbd_opts_token, rbd_opts);
+				   mon_addrs + mon_addrs_size - 1,
+				   parse_rbd_opts_token, &pctx);
 	if (IS_ERR(copts)) {
 		ret = PTR_ERR(copts);
 		goto out_err;
@@ -5232,15 +5279,15 @@ static int rbd_add_parse_args(const char *buf,
 	kfree(options);
 
 	*ceph_opts = copts;
-	*opts = rbd_opts;
-	*rbd_spec = spec;
+	*opts = pctx.opts;
+	*rbd_spec = pctx.spec;
 
 	return 0;
 out_mem:
 	ret = -ENOMEM;
 out_err:
-	kfree(rbd_opts);
-	rbd_spec_put(spec);
+	kfree(pctx.opts);
+	rbd_spec_put(pctx.spec);
 	kfree(options);
 
 	return ret;
@@ -5586,8 +5633,10 @@ static int rbd_dev_image_probe(struct rbd_device *rbd_dev, int depth)
 		ret = rbd_register_watch(rbd_dev);
 		if (ret) {
 			if (ret == -ENOENT)
-				pr_info("image %s/%s does not exist\n",
+				pr_info("image %s/%s%s%s does not exist\n",
 					rbd_dev->spec->pool_name,
+					rbd_dev->spec->pool_ns ?: "",
+					rbd_dev->spec->pool_ns ? "/" : "",
 					rbd_dev->spec->image_name);
 			goto err_out_format;
 		}
@@ -5609,8 +5658,10 @@ static int rbd_dev_image_probe(struct rbd_device *rbd_dev, int depth)
 		ret = rbd_spec_fill_names(rbd_dev);
 	if (ret) {
 		if (ret == -ENOENT)
-			pr_info("snap %s/%s@%s does not exist\n",
+			pr_info("snap %s/%s%s%s@%s does not exist\n",
 				rbd_dev->spec->pool_name,
+				rbd_dev->spec->pool_ns ?: "",
+				rbd_dev->spec->pool_ns ? "/" : "",
 				rbd_dev->spec->image_name,
 				rbd_dev->spec->snap_name);
 		goto err_out_probe;

fs/ceph/acl.c

@@ -45,6 +45,7 @@ static inline void ceph_set_cached_acl(struct inode *inode,
 struct posix_acl *ceph_get_acl(struct inode *inode, int type)
 {
 	int size;
+	unsigned int retry_cnt = 0;
 	const char *name;
 	char *value = NULL;
 	struct posix_acl *acl;
@@ -60,6 +61,7 @@ struct posix_acl *ceph_get_acl(struct inode *inode, int type)
 		BUG();
 	}
 
+retry:
 	size = __ceph_getxattr(inode, name, "", 0);
 	if (size > 0) {
 		value = kzalloc(size, GFP_NOFS);
@@ -68,12 +70,22 @@ struct posix_acl *ceph_get_acl(struct inode *inode, int type)
 		size = __ceph_getxattr(inode, name, value, size);
 	}
 
-	if (size > 0)
+	if (size == -ERANGE && retry_cnt < 10) {
+		retry_cnt++;
+		kfree(value);
+		value = NULL;
+		goto retry;
+	}
+
+	if (size > 0) {
 		acl = posix_acl_from_xattr(&init_user_ns, value, size);
-	else if (size == -ERANGE || size == -ENODATA || size == 0)
+	} else if (size == -ENODATA || size == 0) {
 		acl = NULL;
-	else
+	} else {
+		pr_err_ratelimited("get acl %llx.%llx failed, err=%d\n",
+				   ceph_vinop(inode), size);
 		acl = ERR_PTR(-EIO);
+	}
 
 	kfree(value);
 
@@ -89,6 +101,7 @@ int ceph_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 	const char *name = NULL;
 	char *value = NULL;
 	struct iattr newattrs;
+	struct timespec64 old_ctime = inode->i_ctime;
 	umode_t new_mode = inode->i_mode, old_mode = inode->i_mode;
 
 	switch (type) {
@@ -133,7 +146,7 @@ int ceph_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 	if (new_mode != old_mode) {
 		newattrs.ia_ctime = current_time(inode);
 		newattrs.ia_mode = new_mode;
-		newattrs.ia_valid = ATTR_MODE;
+		newattrs.ia_valid = ATTR_MODE | ATTR_CTIME;
 		ret = __ceph_setattr(inode, &newattrs);
 		if (ret)
 			goto out_free;
@@ -142,8 +155,9 @@ int ceph_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 	ret = __ceph_setxattr(inode, name, value, size, 0);
 	if (ret) {
 		if (new_mode != old_mode) {
+			newattrs.ia_ctime = old_ctime;
 			newattrs.ia_mode = old_mode;
-			newattrs.ia_valid = ATTR_MODE;
+			newattrs.ia_valid = ATTR_MODE | ATTR_CTIME;
 			__ceph_setattr(inode, &newattrs);
 		}
 		goto out_free;
@@ -171,10 +185,10 @@ int ceph_pre_init_acls(struct inode *dir, umode_t *mode,
 		return err;
 
 	if (acl) {
-		int ret = posix_acl_equiv_mode(acl, mode);
-		if (ret < 0)
+		err = posix_acl_equiv_mode(acl, mode);
+		if (err < 0)
 			goto out_err;
-		if (ret == 0) {
+		if (err == 0) {
 			posix_acl_release(acl);
 			acl = NULL;
 		}

fs/ceph/addr.c

@@ -574,7 +574,6 @@ static u64 get_writepages_data_length(struct inode *inode,
  */
 static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 {
-	struct timespec ts;
 	struct inode *inode;
 	struct ceph_inode_info *ci;
 	struct ceph_fs_client *fsc;
@@ -625,12 +624,11 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 		set_bdi_congested(inode_to_bdi(inode), BLK_RW_ASYNC);
 
 	set_page_writeback(page);
-	ts = timespec64_to_timespec(inode->i_mtime);
 	err = ceph_osdc_writepages(&fsc->client->osdc, ceph_vino(inode),
 				   &ci->i_layout, snapc, page_off, len,
 				   ceph_wbc.truncate_seq,
 				   ceph_wbc.truncate_size,
-				   &ts, &page, 1);
+				   &inode->i_mtime, &page, 1);
 	if (err < 0) {
 		struct writeback_control tmp_wbc;
 		if (!wbc)
@@ -1134,7 +1132,7 @@ new_request:
 			pages = NULL;
 		}
 
-		req->r_mtime = timespec64_to_timespec(inode->i_mtime);
+		req->r_mtime = inode->i_mtime;
 		rc = ceph_osdc_start_request(&fsc->client->osdc, req, true);
 		BUG_ON(rc);
 		req = NULL;
@@ -1431,7 +1429,7 @@ static void ceph_restore_sigs(sigset_t *oldset)
 /*
  * vm ops
  */
-static int ceph_filemap_fault(struct vm_fault *vmf)
+static vm_fault_t ceph_filemap_fault(struct vm_fault *vmf)
 {
 	struct vm_area_struct *vma = vmf->vma;
 	struct inode *inode = file_inode(vma->vm_file);
@@ -1439,8 +1437,9 @@ static int ceph_filemap_fault(struct vm_fault *vmf)
 	struct ceph_file_info *fi = vma->vm_file->private_data;
 	struct page *pinned_page = NULL;
 	loff_t off = vmf->pgoff << PAGE_SHIFT;
-	int want, got, ret;
+	int want, got, err;
 	sigset_t oldset;
+	vm_fault_t ret = VM_FAULT_SIGBUS;
 
 	ceph_block_sigs(&oldset);
 
@@ -1452,8 +1451,8 @@ static int ceph_filemap_fault(struct vm_fault *vmf)
 		want = CEPH_CAP_FILE_CACHE;
 
 	got = 0;
-	ret = ceph_get_caps(ci, CEPH_CAP_FILE_RD, want, -1, &got, &pinned_page);
-	if (ret < 0)
+	err = ceph_get_caps(ci, CEPH_CAP_FILE_RD, want, -1, &got, &pinned_page);
+	if (err < 0)
 		goto out_restore;
 
 	dout("filemap_fault %p %llu~%zd got cap refs on %s\n",
@@ -1465,16 +1464,17 @@ static int ceph_filemap_fault(struct vm_fault *vmf)
 		ceph_add_rw_context(fi, &rw_ctx);
 		ret = filemap_fault(vmf);
 		ceph_del_rw_context(fi, &rw_ctx);
+		dout("filemap_fault %p %llu~%zd drop cap refs %s ret %x\n",
+			inode, off, (size_t)PAGE_SIZE,
+				ceph_cap_string(got), ret);
 	} else
-		ret = -EAGAIN;
+		err = -EAGAIN;
 
-	dout("filemap_fault %p %llu~%zd dropping cap refs on %s ret %d\n",
-	     inode, off, (size_t)PAGE_SIZE, ceph_cap_string(got), ret);
 	if (pinned_page)
 		put_page(pinned_page);
 	ceph_put_cap_refs(ci, got);
 
-	if (ret != -EAGAIN)
+	if (err != -EAGAIN)
 		goto out_restore;
 
 	/* read inline data */
@@ -1482,7 +1482,6 @@ static int ceph_filemap_fault(struct vm_fault *vmf)
 		/* does not support inline data > PAGE_SIZE */
 		ret = VM_FAULT_SIGBUS;
 	} else {
-		int ret1;
 		struct address_space *mapping = inode->i_mapping;
 		struct page *page = find_or_create_page(mapping, 0,
 						mapping_gfp_constraint(mapping,
@@ -1491,32 +1490,32 @@ static int ceph_filemap_fault(struct vm_fault *vmf)
 			ret = VM_FAULT_OOM;
 			goto out_inline;
 		}
-		ret1 = __ceph_do_getattr(inode, page,
+		err = __ceph_do_getattr(inode, page,
 					 CEPH_STAT_CAP_INLINE_DATA, true);
-		if (ret1 < 0 || off >= i_size_read(inode)) {
+		if (err < 0 || off >= i_size_read(inode)) {
 			unlock_page(page);
 			put_page(page);
-			if (ret1 < 0)
-				ret = ret1;
+			if (err == -ENOMEM)
+				ret = VM_FAULT_OOM;
 			else
 				ret = VM_FAULT_SIGBUS;
 			goto out_inline;
 		}
-		if (ret1 < PAGE_SIZE)
-			zero_user_segment(page, ret1, PAGE_SIZE);
+		if (err < PAGE_SIZE)
+			zero_user_segment(page, err, PAGE_SIZE);
 		else
 			flush_dcache_page(page);
 		SetPageUptodate(page);
 		vmf->page = page;
 		ret = VM_FAULT_MAJOR | VM_FAULT_LOCKED;
 out_inline:
-		dout("filemap_fault %p %llu~%zd read inline data ret %d\n",
+		dout("filemap_fault %p %llu~%zd read inline data ret %x\n",
 		     inode, off, (size_t)PAGE_SIZE, ret);
 	}
 out_restore:
 	ceph_restore_sigs(&oldset);
-	if (ret < 0)
-		ret = (ret == -ENOMEM) ? VM_FAULT_OOM : VM_FAULT_SIGBUS;
+	if (err < 0)
+		ret = vmf_error(err);
 
 	return ret;
 }
@@ -1524,7 +1523,7 @@ out_restore:
 /*
  * Reuse write_begin here for simplicity.
  */
-static int ceph_page_mkwrite(struct vm_fault *vmf)
+static vm_fault_t ceph_page_mkwrite(struct vm_fault *vmf)
 {
 	struct vm_area_struct *vma = vmf->vma;
 	struct inode *inode = file_inode(vma->vm_file);
@@ -1535,8 +1534,9 @@ static int ceph_page_mkwrite(struct vm_fault *vmf)
 	loff_t off = page_offset(page);
 	loff_t size = i_size_read(inode);
 	size_t len;
-	int want, got, ret;
+	int want, got, err;
 	sigset_t oldset;
+	vm_fault_t ret = VM_FAULT_SIGBUS;
 
 	prealloc_cf = ceph_alloc_cap_flush();
 	if (!prealloc_cf)
@@ -1550,10 +1550,10 @@ static int ceph_page_mkwrite(struct vm_fault *vmf)
 			lock_page(page);
 			locked_page = page;
 		}
-		ret = ceph_uninline_data(vma->vm_file, locked_page);
+		err = ceph_uninline_data(vma->vm_file, locked_page);
 		if (locked_page)
 			unlock_page(locked_page);
-		if (ret < 0)
+		if (err < 0)
 			goto out_free;
 	}
 
@@ -1570,9 +1570,9 @@ static int ceph_page_mkwrite(struct vm_fault *vmf)
 		want = CEPH_CAP_FILE_BUFFER;
 
 	got = 0;
-	ret = ceph_get_caps(ci, CEPH_CAP_FILE_WR, want, off + len,
+	err = ceph_get_caps(ci, CEPH_CAP_FILE_WR, want, off + len,
 			    &got, NULL);
-	if (ret < 0)
+	if (err < 0)
 		goto out_free;
 
 	dout("page_mkwrite %p %llu~%zd got cap refs on %s\n",
@@ -1590,13 +1590,13 @@ static int ceph_page_mkwrite(struct vm_fault *vmf)
 			break;
 		}
 
-		ret = ceph_update_writeable_page(vma->vm_file, off, len, page);
-		if (ret >= 0) {
+		err = ceph_update_writeable_page(vma->vm_file, off, len, page);
+		if (err >= 0) {
 			/* success.  we'll keep the page locked. */
 			set_page_dirty(page);
 			ret = VM_FAULT_LOCKED;
 		}
-	} while (ret == -EAGAIN);
+	} while (err == -EAGAIN);
 
 	if (ret == VM_FAULT_LOCKED ||
 	    ci->i_inline_version != CEPH_INLINE_NONE) {
@@ -1610,14 +1610,14 @@ static int ceph_page_mkwrite(struct vm_fault *vmf)
 			__mark_inode_dirty(inode, dirty);
 	}
 
-	dout("page_mkwrite %p %llu~%zd dropping cap refs on %s ret %d\n",
+	dout("page_mkwrite %p %llu~%zd dropping cap refs on %s ret %x\n",
 	     inode, off, len, ceph_cap_string(got), ret);
 	ceph_put_cap_refs(ci, got);
 out_free:
 	ceph_restore_sigs(&oldset);
 	ceph_free_cap_flush(prealloc_cf);
-	if (ret < 0)
-		ret = (ret == -ENOMEM) ? VM_FAULT_OOM : VM_FAULT_SIGBUS;
+	if (err < 0)
+		ret = vmf_error(err);
 	return ret;
 }
 
@@ -1734,7 +1734,7 @@ int ceph_uninline_data(struct file *filp, struct page *locked_page)
 		goto out;
 	}
 
-	req->r_mtime = timespec64_to_timespec(inode->i_mtime);
+	req->r_mtime = inode->i_mtime;
 	err = ceph_osdc_start_request(&fsc->client->osdc, req, false);
 	if (!err)
 		err = ceph_osdc_wait_request(&fsc->client->osdc, req);
@@ -1776,7 +1776,7 @@ int ceph_uninline_data(struct file *filp, struct page *locked_page)
 			goto out_put;
 	}
 
-	req->r_mtime = timespec64_to_timespec(inode->i_mtime);
+	req->r_mtime = inode->i_mtime;
 	err = ceph_osdc_start_request(&fsc->client->osdc, req, false);
 	if (!err)
 		err = ceph_osdc_wait_request(&fsc->client->osdc, req);
@@ -1937,7 +1937,7 @@ static int __ceph_pool_perm_get(struct ceph_inode_info *ci,
 				     0, false, true);
 	err = ceph_osdc_start_request(&fsc->client->osdc, rd_req, false);
 
-	wr_req->r_mtime = timespec64_to_timespec(ci->vfs_inode.i_mtime);
+	wr_req->r_mtime = ci->vfs_inode.i_mtime;
 	err2 = ceph_osdc_start_request(&fsc->client->osdc, wr_req, false);
 
 	if (!err)

fs/ceph/cache.c

@@ -25,8 +25,9 @@
 #include "cache.h"
 
 struct ceph_aux_inode {
-	u64 		version;
-	struct timespec	mtime;
+	u64 	version;
+	u64	mtime_sec;
+	u64	mtime_nsec;
 };
 
 struct fscache_netfs ceph_cache_netfs = {
@@ -130,7 +131,8 @@ static enum fscache_checkaux ceph_fscache_inode_check_aux(
 
 	memset(&aux, 0, sizeof(aux));
 	aux.version = ci->i_version;
-	aux.mtime = timespec64_to_timespec(inode->i_mtime);
+	aux.mtime_sec = inode->i_mtime.tv_sec;
+	aux.mtime_nsec = inode->i_mtime.tv_nsec;
 
 	if (memcmp(data, &aux, sizeof(aux)) != 0)
 		return FSCACHE_CHECKAUX_OBSOLETE;
@@ -163,7 +165,8 @@ void ceph_fscache_register_inode_cookie(struct inode *inode)
 	if (!ci->fscache) {
 		memset(&aux, 0, sizeof(aux));
 		aux.version = ci->i_version;
-		aux.mtime = timespec64_to_timespec(inode->i_mtime);
+		aux.mtime_sec = inode->i_mtime.tv_sec;
+		aux.mtime_nsec = inode->i_mtime.tv_nsec;
 		ci->fscache = fscache_acquire_cookie(fsc->fscache,
 						     &ceph_fscache_inode_object_def,
 						     &ci->i_vino, sizeof(ci->i_vino),

fs/ceph/caps.c

@@ -156,6 +156,37 @@ void ceph_adjust_min_caps(struct ceph_mds_client *mdsc, int delta)
 	spin_unlock(&mdsc->caps_list_lock);
 }
 
+static void __ceph_unreserve_caps(struct ceph_mds_client *mdsc, int nr_caps)
+{
+	struct ceph_cap *cap;
+	int i;
+
+	if (nr_caps) {
+		BUG_ON(mdsc->caps_reserve_count < nr_caps);
+		mdsc->caps_reserve_count -= nr_caps;
+		if (mdsc->caps_avail_count >=
+		    mdsc->caps_reserve_count + mdsc->caps_min_count) {
+			mdsc->caps_total_count -= nr_caps;
+			for (i = 0; i < nr_caps; i++) {
+				cap = list_first_entry(&mdsc->caps_list,
+					struct ceph_cap, caps_item);
+				list_del(&cap->caps_item);
+				kmem_cache_free(ceph_cap_cachep, cap);
+			}
+		} else {
+			mdsc->caps_avail_count += nr_caps;
+		}
+
+		dout("%s: caps %d = %d used + %d resv + %d avail\n",
+		     __func__,
+		     mdsc->caps_total_count, mdsc->caps_use_count,
+		     mdsc->caps_reserve_count, mdsc->caps_avail_count);
+		BUG_ON(mdsc->caps_total_count != mdsc->caps_use_count +
+						 mdsc->caps_reserve_count +
+						 mdsc->caps_avail_count);
+	}
+}
+
 /*
  * Called under mdsc->mutex.
  */
@@ -167,6 +198,7 @@ int ceph_reserve_caps(struct ceph_mds_client *mdsc,
 	int have;
 	int alloc = 0;
 	int max_caps;
+	int err = 0;
 	bool trimmed = false;
 	struct ceph_mds_session *s;
 	LIST_HEAD(newcaps);
@@ -233,9 +265,14 @@ int ceph_reserve_caps(struct ceph_mds_client *mdsc,
 
 		pr_warn("reserve caps ctx=%p ENOMEM need=%d got=%d\n",
 			ctx, need, have + alloc);
-		goto out_nomem;
+		err = -ENOMEM;
+		break;
+	}
+
+	if (!err) {
+		BUG_ON(have + alloc != need);
+		ctx->count = need;
 	}
-	BUG_ON(have + alloc != need);
 
 	spin_lock(&mdsc->caps_list_lock);
 	mdsc->caps_total_count += alloc;
@@ -245,77 +282,26 @@ int ceph_reserve_caps(struct ceph_mds_client *mdsc,
 	BUG_ON(mdsc->caps_total_count != mdsc->caps_use_count +
 					 mdsc->caps_reserve_count +
 					 mdsc->caps_avail_count);
+
+	if (err)
+		__ceph_unreserve_caps(mdsc, have + alloc);
+
 	spin_unlock(&mdsc->caps_list_lock);
 
-	ctx->count = need;
 	dout("reserve caps ctx=%p %d = %d used + %d resv + %d avail\n",
 	     ctx, mdsc->caps_total_count, mdsc->caps_use_count,
 	     mdsc->caps_reserve_count, mdsc->caps_avail_count);
-	return 0;
-
-out_nomem:
-
-	spin_lock(&mdsc->caps_list_lock);
-	mdsc->caps_avail_count += have;
-	mdsc->caps_reserve_count -= have;
-
-	while (!list_empty(&newcaps)) {
-		cap = list_first_entry(&newcaps,
-				struct ceph_cap, caps_item);
-		list_del(&cap->caps_item);
-
-		/* Keep some preallocated caps around (ceph_min_count), to
-		 * avoid lots of free/alloc churn. */
-		if (mdsc->caps_avail_count >=
-		    mdsc->caps_reserve_count + mdsc->caps_min_count) {
-			kmem_cache_free(ceph_cap_cachep, cap);
-		} else {
-			mdsc->caps_avail_count++;
-			mdsc->caps_total_count++;
-			list_add(&cap->caps_item, &mdsc->caps_list);
-		}
-	}
-
-	BUG_ON(mdsc->caps_total_count != mdsc->caps_use_count +
-					 mdsc->caps_reserve_count +
-					 mdsc->caps_avail_count);
-	spin_unlock(&mdsc->caps_list_lock);
-	return -ENOMEM;
+	return err;
 }
 
-int ceph_unreserve_caps(struct ceph_mds_client *mdsc,
+void ceph_unreserve_caps(struct ceph_mds_client *mdsc,
 			struct ceph_cap_reservation *ctx)
 {
-	int i;
-	struct ceph_cap *cap;
-
 	dout("unreserve caps ctx=%p count=%d\n", ctx, ctx->count);
-	if (ctx->count) {
-		spin_lock(&mdsc->caps_list_lock);
-		BUG_ON(mdsc->caps_reserve_count < ctx->count);
-		mdsc->caps_reserve_count -= ctx->count;
-		if (mdsc->caps_avail_count >=
-		    mdsc->caps_reserve_count + mdsc->caps_min_count) {
-			mdsc->caps_total_count -= ctx->count;
-			for (i = 0; i < ctx->count; i++) {
-				cap = list_first_entry(&mdsc->caps_list,
-					struct ceph_cap, caps_item);
-				list_del(&cap->caps_item);
-				kmem_cache_free(ceph_cap_cachep, cap);
-			}
-		} else {
-			mdsc->caps_avail_count += ctx->count;
-		}
-		ctx->count = 0;
-		dout("unreserve caps %d = %d used + %d resv + %d avail\n",
-		     mdsc->caps_total_count, mdsc->caps_use_count,
-		     mdsc->caps_reserve_count, mdsc->caps_avail_count);
-		BUG_ON(mdsc->caps_total_count != mdsc->caps_use_count +
-						 mdsc->caps_reserve_count +
-						 mdsc->caps_avail_count);
-		spin_unlock(&mdsc->caps_list_lock);
-	}
-	return 0;
+	spin_lock(&mdsc->caps_list_lock);
+	__ceph_unreserve_caps(mdsc, ctx->count);
+	ctx->count = 0;
+	spin_unlock(&mdsc->caps_list_lock);
 }
 
 struct ceph_cap *ceph_get_cap(struct ceph_mds_client *mdsc,
@@ -1125,7 +1111,7 @@ struct cap_msg_args {
 	u64			flush_tid, oldest_flush_tid, size, max_size;
 	u64			xattr_version;
 	struct ceph_buffer	*xattr_buf;
-	struct timespec		atime, mtime, ctime;
+	struct timespec64	atime, mtime, ctime;
 	int			op, caps, wanted, dirty;
 	u32			seq, issue_seq, mseq, time_warp_seq;
 	u32			flags;
@@ -1146,7 +1132,7 @@ static int send_cap_msg(struct cap_msg_args *arg)
 	struct ceph_msg *msg;
 	void *p;
 	size_t extra_len;
-	struct timespec zerotime = {0};
+	struct timespec64 zerotime = {0};
 	struct ceph_osd_client *osdc = &arg->session->s_mdsc->fsc->client->osdc;
 
 	dout("send_cap_msg %s %llx %llx caps %s wanted %s dirty %s"
@@ -1186,9 +1172,9 @@ static int send_cap_msg(struct cap_msg_args *arg)
 
 	fc->size = cpu_to_le64(arg->size);
 	fc->max_size = cpu_to_le64(arg->max_size);
-	ceph_encode_timespec(&fc->mtime, &arg->mtime);
-	ceph_encode_timespec(&fc->atime, &arg->atime);
-	ceph_encode_timespec(&fc->ctime, &arg->ctime);
+	ceph_encode_timespec64(&fc->mtime, &arg->mtime);
+	ceph_encode_timespec64(&fc->atime, &arg->atime);
+	ceph_encode_timespec64(&fc->ctime, &arg->ctime);
 	fc->time_warp_seq = cpu_to_le32(arg->time_warp_seq);
 
 	fc->uid = cpu_to_le32(from_kuid(&init_user_ns, arg->uid));
@@ -1237,7 +1223,7 @@ static int send_cap_msg(struct cap_msg_args *arg)
 	 * We just zero these out for now, as the MDS ignores them unless
 	 * the requisite feature flags are set (which we don't do yet).
 	 */
-	ceph_encode_timespec(p, &zerotime);
+	ceph_encode_timespec64(p, &zerotime);
 	p += sizeof(struct ceph_timespec);
 	ceph_encode_64(&p, 0);
 
@@ -1360,9 +1346,9 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
 		arg.xattr_buf = NULL;
 	}
 
-	arg.mtime = timespec64_to_timespec(inode->i_mtime);
-	arg.atime = timespec64_to_timespec(inode->i_atime);
-	arg.ctime = timespec64_to_timespec(inode->i_ctime);
+	arg.mtime = inode->i_mtime;
+	arg.atime = inode->i_atime;
+	arg.ctime = inode->i_ctime;
 
 	arg.op = op;
 	arg.caps = cap->implemented;
@@ -3148,11 +3134,11 @@ static void handle_cap_grant(struct inode *inode,
 	}
 
 	if (newcaps & CEPH_CAP_ANY_RD) {
-		struct timespec mtime, atime, ctime;
+		struct timespec64 mtime, atime, ctime;
 		/* ctime/mtime/atime? */
-		ceph_decode_timespec(&mtime, &grant->mtime);
-		ceph_decode_timespec(&atime, &grant->atime);
-		ceph_decode_timespec(&ctime, &grant->ctime);
+		ceph_decode_timespec64(&mtime, &grant->mtime);
+		ceph_decode_timespec64(&atime, &grant->atime);
+		ceph_decode_timespec64(&ctime, &grant->ctime);
 		ceph_fill_file_time(inode, extra_info->issued,
 				    le32_to_cpu(grant->time_warp_seq),
 				    &ctime, &mtime, &atime);

fs/ceph/dir.c

@@ -827,12 +827,14 @@ static int ceph_mknod(struct inode *dir, struct dentry *dentry,
 	if (ceph_snap(dir) != CEPH_NOSNAP)
 		return -EROFS;
 
-	if (ceph_quota_is_max_files_exceeded(dir))
-		return -EDQUOT;
+	if (ceph_quota_is_max_files_exceeded(dir)) {
+		err = -EDQUOT;
+		goto out;
+	}
 
 	err = ceph_pre_init_acls(dir, &mode, &acls);
 	if (err < 0)
-		return err;
+		goto out;
 
 	dout("mknod in dir %p dentry %p mode 0%ho rdev %d\n",
 	     dir, dentry, mode, rdev);
@@ -883,8 +885,10 @@ static int ceph_symlink(struct inode *dir, struct dentry *dentry,
 	if (ceph_snap(dir) != CEPH_NOSNAP)
 		return -EROFS;
 
-	if (ceph_quota_is_max_files_exceeded(dir))
-		return -EDQUOT;
+	if (ceph_quota_is_max_files_exceeded(dir)) {
+		err = -EDQUOT;
+		goto out;
+	}
 
 	dout("symlink in dir %p dentry %p to '%s'\n", dir, dentry, dest);
 	req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_SYMLINK, USE_AUTH_MDS);
@@ -1393,7 +1397,7 @@ static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size,
 				" rfiles:   %20lld\n"
 				" rsubdirs: %20lld\n"
 				"rbytes:    %20lld\n"
-				"rctime:    %10ld.%09ld\n",
+				"rctime:    %10lld.%09ld\n",
 				ci->i_files + ci->i_subdirs,
 				ci->i_files,
 				ci->i_subdirs,
@@ -1401,8 +1405,8 @@ static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size,
 				ci->i_rfiles,
 				ci->i_rsubdirs,
 				ci->i_rbytes,
-				(long)ci->i_rctime.tv_sec,
-				(long)ci->i_rctime.tv_nsec);
+				ci->i_rctime.tv_sec,
+				ci->i_rctime.tv_nsec);
 	}
 
 	if (*ppos >= dfi->dir_info_len)

fs/ceph/file.c

@@ -720,7 +720,7 @@ struct ceph_aio_request {
 	struct list_head osd_reqs;
 	unsigned num_reqs;
 	atomic_t pending_reqs;
-	struct timespec mtime;
+	struct timespec64 mtime;
 	struct ceph_cap_flush *prealloc_cf;
 };
 
@@ -922,7 +922,7 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter,
 	int num_pages = 0;
 	int flags;
 	int ret;
-	struct timespec mtime = timespec64_to_timespec(current_time(inode));
+	struct timespec64 mtime = current_time(inode);
 	size_t count = iov_iter_count(iter);
 	loff_t pos = iocb->ki_pos;
 	bool write = iov_iter_rw(iter) == WRITE;
@@ -1130,7 +1130,7 @@ ceph_sync_write(struct kiocb *iocb, struct iov_iter *from, loff_t pos,
 	int flags;
 	int ret;
 	bool check_caps = false;
-	struct timespec mtime = timespec64_to_timespec(current_time(inode));
+	struct timespec64 mtime = current_time(inode);
 	size_t count = iov_iter_count(from);
 
 	if (ceph_snap(file_inode(file)) != CEPH_NOSNAP)
@@ -1383,12 +1383,12 @@ static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	struct ceph_file_info *fi = file->private_data;
 	struct inode *inode = file_inode(file);
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_osd_client *osdc =
-		&ceph_sb_to_client(inode->i_sb)->client->osdc;
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
 	struct ceph_cap_flush *prealloc_cf;
 	ssize_t count, written = 0;
 	int err, want, got;
 	loff_t pos;
+	loff_t limit = max(i_size_read(inode), fsc->max_file_size);
 
 	if (ceph_snap(inode) != CEPH_NOSNAP)
 		return -EROFS;
@@ -1414,6 +1414,13 @@ retry_snap:
 		goto out;
 
 	pos = iocb->ki_pos;
+	if (unlikely(pos >= limit)) {
+		err = -EFBIG;
+		goto out;
+	} else {
+		iov_iter_truncate(from, limit - pos);
+	}
+
 	count = iov_iter_count(from);
 	if (ceph_quota_is_max_bytes_exceeded(inode, pos + count)) {
 		err = -EDQUOT;
@@ -1435,7 +1442,7 @@ retry_snap:
 	}
 
 	/* FIXME: not complete since it doesn't account for being at quota */
-	if (ceph_osdmap_flag(osdc, CEPH_OSDMAP_FULL)) {
+	if (ceph_osdmap_flag(&fsc->client->osdc, CEPH_OSDMAP_FULL)) {
 		err = -ENOSPC;
 		goto out;
 	}
@@ -1525,7 +1532,7 @@ retry_snap:
 	}
 
 	if (written >= 0) {
-		if (ceph_osdmap_flag(osdc, CEPH_OSDMAP_NEARFULL))
+		if (ceph_osdmap_flag(&fsc->client->osdc, CEPH_OSDMAP_NEARFULL))
 			iocb->ki_flags |= IOCB_DSYNC;
 		written = generic_write_sync(iocb, written);
 	}
@@ -1546,6 +1553,7 @@ out_unlocked:
 static loff_t ceph_llseek(struct file *file, loff_t offset, int whence)
 {
 	struct inode *inode = file->f_mapping->host;
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
 	loff_t i_size;
 	loff_t ret;
 
@@ -1590,7 +1598,7 @@ static loff_t ceph_llseek(struct file *file, loff_t offset, int whence)
 		break;
 	}
 
-	ret = vfs_setpos(file, offset, inode->i_sb->s_maxbytes);
+	ret = vfs_setpos(file, offset, max(i_size, fsc->max_file_size));
 
 out:
 	inode_unlock(inode);
@@ -1662,7 +1670,7 @@ static int ceph_zero_partial_object(struct inode *inode,
 		goto out;
 	}
 
-	req->r_mtime = timespec64_to_timespec(inode->i_mtime);
+	req->r_mtime = inode->i_mtime;
 	ret = ceph_osdc_start_request(&fsc->client->osdc, req, false);
 	if (!ret) {
 		ret = ceph_osdc_wait_request(&fsc->client->osdc, req);
@@ -1727,8 +1735,7 @@ static long ceph_fallocate(struct file *file, int mode,
 	struct ceph_file_info *fi = file->private_data;
 	struct inode *inode = file_inode(file);
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_osd_client *osdc =
-		&ceph_inode_to_client(inode)->client->osdc;
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
 	struct ceph_cap_flush *prealloc_cf;
 	int want, got = 0;
 	int dirty;
@@ -1736,6 +1743,9 @@ static long ceph_fallocate(struct file *file, int mode,
 	loff_t endoff = 0;
 	loff_t size;
 
+	if ((offset + length) > max(i_size_read(inode), fsc->max_file_size))
+		return -EFBIG;
+
 	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
 		return -EOPNOTSUPP;
 
@@ -1759,7 +1769,7 @@ static long ceph_fallocate(struct file *file, int mode,
 		goto unlock;
 	}
 
-	if (ceph_osdmap_flag(osdc, CEPH_OSDMAP_FULL) &&
+	if (ceph_osdmap_flag(&fsc->client->osdc, CEPH_OSDMAP_FULL) &&
 	    !(mode & FALLOC_FL_PUNCH_HOLE)) {
 		ret = -ENOSPC;
 		goto unlock;

fs/ceph/inode.c

@@ -658,13 +658,10 @@ int ceph_fill_file_size(struct inode *inode, int issued,
 }
 
 void ceph_fill_file_time(struct inode *inode, int issued,
-			 u64 time_warp_seq, struct timespec *ctime,
-			 struct timespec *mtime, struct timespec *atime)
+			 u64 time_warp_seq, struct timespec64 *ctime,
+			 struct timespec64 *mtime, struct timespec64 *atime)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct timespec64 ctime64 = timespec_to_timespec64(*ctime);
-	struct timespec64 mtime64 = timespec_to_timespec64(*mtime);
-	struct timespec64 atime64 = timespec_to_timespec64(*atime);
 	int warn = 0;
 
 	if (issued & (CEPH_CAP_FILE_EXCL|
@@ -673,39 +670,39 @@ void ceph_fill_file_time(struct inode *inode, int issued,
 		      CEPH_CAP_AUTH_EXCL|
 		      CEPH_CAP_XATTR_EXCL)) {
 		if (ci->i_version == 0 ||
-		    timespec64_compare(&ctime64, &inode->i_ctime) > 0) {
+		    timespec64_compare(ctime, &inode->i_ctime) > 0) {
 			dout("ctime %lld.%09ld -> %lld.%09ld inc w/ cap\n",
-			     (long long)inode->i_ctime.tv_sec, inode->i_ctime.tv_nsec,
-			     (long long)ctime->tv_sec, ctime->tv_nsec);
-			inode->i_ctime = ctime64;
+			     inode->i_ctime.tv_sec, inode->i_ctime.tv_nsec,
+			     ctime->tv_sec, ctime->tv_nsec);
+			inode->i_ctime = *ctime;
 		}
 		if (ci->i_version == 0 ||
 		    ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) > 0) {
 			/* the MDS did a utimes() */
 			dout("mtime %lld.%09ld -> %lld.%09ld "
 			     "tw %d -> %d\n",
-			     (long long)inode->i_mtime.tv_sec, inode->i_mtime.tv_nsec,
-			     (long long)mtime->tv_sec, mtime->tv_nsec,
+			     inode->i_mtime.tv_sec, inode->i_mtime.tv_nsec,
+			     mtime->tv_sec, mtime->tv_nsec,
 			     ci->i_time_warp_seq, (int)time_warp_seq);
 
-			inode->i_mtime = mtime64;
-			inode->i_atime = atime64;
+			inode->i_mtime = *mtime;
+			inode->i_atime = *atime;
 			ci->i_time_warp_seq = time_warp_seq;
 		} else if (time_warp_seq == ci->i_time_warp_seq) {
 			/* nobody did utimes(); take the max */
-			if (timespec64_compare(&mtime64, &inode->i_mtime) > 0) {
+			if (timespec64_compare(mtime, &inode->i_mtime) > 0) {
 				dout("mtime %lld.%09ld -> %lld.%09ld inc\n",
-				     (long long)inode->i_mtime.tv_sec,
+				     inode->i_mtime.tv_sec,
 				     inode->i_mtime.tv_nsec,
-				     (long long)mtime->tv_sec, mtime->tv_nsec);
-				inode->i_mtime = mtime64;
+				     mtime->tv_sec, mtime->tv_nsec);
+				inode->i_mtime = *mtime;
 			}
-			if (timespec64_compare(&atime64, &inode->i_atime) > 0) {
+			if (timespec64_compare(atime, &inode->i_atime) > 0) {
 				dout("atime %lld.%09ld -> %lld.%09ld inc\n",
-				     (long long)inode->i_atime.tv_sec,
+				     inode->i_atime.tv_sec,
 				     inode->i_atime.tv_nsec,
-				     (long long)atime->tv_sec, atime->tv_nsec);
-				inode->i_atime = atime64;
+				     atime->tv_sec, atime->tv_nsec);
+				inode->i_atime = *atime;
 			}
 		} else if (issued & CEPH_CAP_FILE_EXCL) {
 			/* we did a utimes(); ignore mds values */
@@ -715,9 +712,9 @@ void ceph_fill_file_time(struct inode *inode, int issued,
 	} else {
 		/* we have no write|excl caps; whatever the MDS says is true */
 		if (ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) >= 0) {
-			inode->i_ctime = ctime64;
-			inode->i_mtime = mtime64;
-			inode->i_atime = atime64;
+			inode->i_ctime = *ctime;
+			inode->i_mtime = *mtime;
+			inode->i_atime = *atime;
 			ci->i_time_warp_seq = time_warp_seq;
 		} else {
 			warn = 1;
@@ -743,7 +740,7 @@ static int fill_inode(struct inode *inode, struct page *locked_page,
 	struct ceph_mds_reply_inode *info = iinfo->in;
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int issued, new_issued, info_caps;
-	struct timespec mtime, atime, ctime;
+	struct timespec64 mtime, atime, ctime;
 	struct ceph_buffer *xattr_blob = NULL;
 	struct ceph_string *pool_ns = NULL;
 	struct ceph_cap *new_cap = NULL;
@@ -823,9 +820,9 @@ static int fill_inode(struct inode *inode, struct page *locked_page,
 
 	if (new_version || (new_issued & CEPH_CAP_ANY_RD)) {
 		/* be careful with mtime, atime, size */
-		ceph_decode_timespec(&atime, &info->atime);
-		ceph_decode_timespec(&mtime, &info->mtime);
-		ceph_decode_timespec(&ctime, &info->ctime);
+		ceph_decode_timespec64(&atime, &info->atime);
+		ceph_decode_timespec64(&mtime, &info->mtime);
+		ceph_decode_timespec64(&ctime, &info->ctime);
 		ceph_fill_file_time(inode, issued,
 				le32_to_cpu(info->time_warp_seq),
 				&ctime, &mtime, &atime);
@@ -872,7 +869,7 @@ static int fill_inode(struct inode *inode, struct page *locked_page,
 			ci->i_rbytes = le64_to_cpu(info->rbytes);
 			ci->i_rfiles = le64_to_cpu(info->rfiles);
 			ci->i_rsubdirs = le64_to_cpu(info->rsubdirs);
-			ceph_decode_timespec(&ci->i_rctime, &info->rctime);
+			ceph_decode_timespec64(&ci->i_rctime, &info->rctime);
 		}
 	}
 
@@ -1954,7 +1951,6 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr)
 	int err = 0;
 	int inode_dirty_flags = 0;
 	bool lock_snap_rwsem = false;
-	struct timespec ts;
 
 	prealloc_cf = ceph_alloc_cap_flush();
 	if (!prealloc_cf)
@@ -2030,8 +2026,8 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr)
 
 	if (ia_valid & ATTR_ATIME) {
 		dout("setattr %p atime %lld.%ld -> %lld.%ld\n", inode,
-		     (long long)inode->i_atime.tv_sec, inode->i_atime.tv_nsec,
-		     (long long)attr->ia_atime.tv_sec, attr->ia_atime.tv_nsec);
+		     inode->i_atime.tv_sec, inode->i_atime.tv_nsec,
+		     attr->ia_atime.tv_sec, attr->ia_atime.tv_nsec);
 		if (issued & CEPH_CAP_FILE_EXCL) {
 			ci->i_time_warp_seq++;
 			inode->i_atime = attr->ia_atime;
@@ -2043,8 +2039,8 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr)
 			dirtied |= CEPH_CAP_FILE_WR;
 		} else if ((issued & CEPH_CAP_FILE_SHARED) == 0 ||
 			   !timespec64_equal(&inode->i_atime, &attr->ia_atime)) {
-			ts = timespec64_to_timespec(attr->ia_atime);
-			ceph_encode_timespec(&req->r_args.setattr.atime, &ts);
+			ceph_encode_timespec64(&req->r_args.setattr.atime,
+					       &attr->ia_atime);
 			mask |= CEPH_SETATTR_ATIME;
 			release |= CEPH_CAP_FILE_SHARED |
 				   CEPH_CAP_FILE_RD | CEPH_CAP_FILE_WR;
@@ -2052,8 +2048,8 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr)
 	}
 	if (ia_valid & ATTR_MTIME) {
 		dout("setattr %p mtime %lld.%ld -> %lld.%ld\n", inode,
-		     (long long)inode->i_mtime.tv_sec, inode->i_mtime.tv_nsec,
-		     (long long)attr->ia_mtime.tv_sec, attr->ia_mtime.tv_nsec);
+		     inode->i_mtime.tv_sec, inode->i_mtime.tv_nsec,
+		     attr->ia_mtime.tv_sec, attr->ia_mtime.tv_nsec);
 		if (issued & CEPH_CAP_FILE_EXCL) {
 			ci->i_time_warp_seq++;
 			inode->i_mtime = attr->ia_mtime;
@@ -2065,8 +2061,8 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr)
 			dirtied |= CEPH_CAP_FILE_WR;
 		} else if ((issued & CEPH_CAP_FILE_SHARED) == 0 ||
 			   !timespec64_equal(&inode->i_mtime, &attr->ia_mtime)) {
-			ts = timespec64_to_timespec(attr->ia_mtime);
-			ceph_encode_timespec(&req->r_args.setattr.mtime, &ts);
+			ceph_encode_timespec64(&req->r_args.setattr.mtime,
+					       &attr->ia_mtime);
 			mask |= CEPH_SETATTR_MTIME;
 			release |= CEPH_CAP_FILE_SHARED |
 				   CEPH_CAP_FILE_RD | CEPH_CAP_FILE_WR;
@@ -2097,8 +2093,8 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr)
 		bool only = (ia_valid & (ATTR_SIZE|ATTR_MTIME|ATTR_ATIME|
 					 ATTR_MODE|ATTR_UID|ATTR_GID)) == 0;
 		dout("setattr %p ctime %lld.%ld -> %lld.%ld (%s)\n", inode,
-		     (long long)inode->i_ctime.tv_sec, inode->i_ctime.tv_nsec,
-		     (long long)attr->ia_ctime.tv_sec, attr->ia_ctime.tv_nsec,
+		     inode->i_ctime.tv_sec, inode->i_ctime.tv_nsec,
+		     attr->ia_ctime.tv_sec, attr->ia_ctime.tv_nsec,
 		     only ? "ctime only" : "ignored");
 		if (only) {
 			/*
@@ -2140,7 +2136,7 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr)
 		req->r_inode_drop = release;
 		req->r_args.setattr.mask = cpu_to_le32(mask);
 		req->r_num_caps = 1;
-		req->r_stamp = timespec64_to_timespec(attr->ia_ctime);
+		req->r_stamp = attr->ia_ctime;
 		err = ceph_mdsc_do_request(mdsc, NULL, req);
 	}
 	dout("setattr %p result=%d (%s locally, %d remote)\n", inode, err,
@@ -2161,6 +2157,7 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr)
 int ceph_setattr(struct dentry *dentry, struct iattr *attr)
 {
 	struct inode *inode = d_inode(dentry);
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
 	int err;
 
 	if (ceph_snap(inode) != CEPH_NOSNAP)
@@ -2170,6 +2167,10 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr)
 	if (err != 0)
 		return err;
 
+	if ((attr->ia_valid & ATTR_SIZE) &&
+	    attr->ia_size > max(inode->i_size, fsc->max_file_size))
+		return -EFBIG;
+
 	if ((attr->ia_valid & ATTR_SIZE) &&
 	    ceph_quota_is_max_bytes_exceeded(inode, attr->ia_size))
 		return -EDQUOT;

fs/ceph/mds_client.c

@@ -902,6 +902,27 @@ static struct ceph_msg *create_session_msg(u32 op, u64 seq)
 	return msg;
 }
 
+static void encode_supported_features(void **p, void *end)
+{
+	static const unsigned char bits[] = CEPHFS_FEATURES_CLIENT_SUPPORTED;
+	static const size_t count = ARRAY_SIZE(bits);
+
+	if (count > 0) {
+		size_t i;
+		size_t size = ((size_t)bits[count - 1] + 64) / 64 * 8;
+
+		BUG_ON(*p + 4 + size > end);
+		ceph_encode_32(p, size);
+		memset(*p, 0, size);
+		for (i = 0; i < count; i++)
+			((unsigned char*)(*p))[i / 8] |= 1 << (bits[i] % 8);
+		*p += size;
+	} else {
+		BUG_ON(*p + 4 > end);
+		ceph_encode_32(p, 0);
+	}
+}
+
 /*
  * session message, specialization for CEPH_SESSION_REQUEST_OPEN
  * to include additional client metadata fields.
@@ -911,11 +932,11 @@ static struct ceph_msg *create_session_open_msg(struct ceph_mds_client *mdsc, u6
 	struct ceph_msg *msg;
 	struct ceph_mds_session_head *h;
 	int i = -1;
-	int metadata_bytes = 0;
+	int extra_bytes = 0;
 	int metadata_key_count = 0;
 	struct ceph_options *opt = mdsc->fsc->client->options;
 	struct ceph_mount_options *fsopt = mdsc->fsc->mount_options;
-	void *p;
+	void *p, *end;
 
 	const char* metadata[][2] = {
 		{"hostname", mdsc->nodename},
@@ -926,21 +947,26 @@ static struct ceph_msg *create_session_open_msg(struct ceph_mds_client *mdsc, u6
 	};
 
 	/* Calculate serialized length of metadata */
-	metadata_bytes = 4;  /* map length */
+	extra_bytes = 4;  /* map length */
 	for (i = 0; metadata[i][0]; ++i) {
-		metadata_bytes += 8 + strlen(metadata[i][0]) +
+		extra_bytes += 8 + strlen(metadata[i][0]) +
 			strlen(metadata[i][1]);
 		metadata_key_count++;
 	}
+	/* supported feature */
+	extra_bytes += 4 + 8;
 
 	/* Allocate the message */
-	msg = ceph_msg_new(CEPH_MSG_CLIENT_SESSION, sizeof(*h) + metadata_bytes,
+	msg = ceph_msg_new(CEPH_MSG_CLIENT_SESSION, sizeof(*h) + extra_bytes,
 			   GFP_NOFS, false);
 	if (!msg) {
 		pr_err("create_session_msg ENOMEM creating msg\n");
 		return NULL;
 	}
-	h = msg->front.iov_base;
+	p = msg->front.iov_base;
+	end = p + msg->front.iov_len;
+
+	h = p;
 	h->op = cpu_to_le32(CEPH_SESSION_REQUEST_OPEN);
 	h->seq = cpu_to_le64(seq);
 
@@ -950,11 +976,11 @@ static struct ceph_msg *create_session_open_msg(struct ceph_mds_client *mdsc, u6
 	 *
 	 * ClientSession messages with metadata are v2
 	 */
-	msg->hdr.version = cpu_to_le16(2);
+	msg->hdr.version = cpu_to_le16(3);
 	msg->hdr.compat_version = cpu_to_le16(1);
 
 	/* The write pointer, following the session_head structure */
-	p = msg->front.iov_base + sizeof(*h);
+	p += sizeof(*h);
 
 	/* Number of entries in the map */
 	ceph_encode_32(&p, metadata_key_count);
@@ -972,6 +998,10 @@ static struct ceph_msg *create_session_open_msg(struct ceph_mds_client *mdsc, u6
 		p += val_len;
 	}
 
+	encode_supported_features(&p, end);
+	msg->front.iov_len = p - msg->front.iov_base;
+	msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
+
 	return msg;
 }
 
@@ -1779,6 +1809,7 @@ struct ceph_mds_request *
 ceph_mdsc_create_request(struct ceph_mds_client *mdsc, int op, int mode)
 {
 	struct ceph_mds_request *req = kzalloc(sizeof(*req), GFP_NOFS);
+	struct timespec64 ts;
 
 	if (!req)
 		return ERR_PTR(-ENOMEM);
@@ -1797,7 +1828,8 @@ ceph_mdsc_create_request(struct ceph_mds_client *mdsc, int op, int mode)
 	init_completion(&req->r_safe_completion);
 	INIT_LIST_HEAD(&req->r_unsafe_item);
 
-	req->r_stamp = timespec_trunc(current_kernel_time(), mdsc->fsc->sb->s_time_gran);
+	ktime_get_coarse_real_ts64(&ts);
+	req->r_stamp = timespec64_trunc(ts, mdsc->fsc->sb->s_time_gran);
 
 	req->r_op = op;
 	req->r_direct_mode = mode;
@@ -2094,7 +2126,7 @@ static struct ceph_msg *create_request_message(struct ceph_mds_client *mdsc,
 	/* time stamp */
 	{
 		struct ceph_timespec ts;
-		ceph_encode_timespec(&ts, &req->r_stamp);
+		ceph_encode_timespec64(&ts, &req->r_stamp);
 		ceph_encode_copy(&p, &ts, sizeof(ts));
 	}
 
@@ -2187,7 +2219,7 @@ static int __prepare_send_request(struct ceph_mds_client *mdsc,
 		p = msg->front.iov_base + req->r_request_release_offset;
 		{
 			struct ceph_timespec ts;
-			ceph_encode_timespec(&ts, &req->r_stamp);
+			ceph_encode_timespec64(&ts, &req->r_stamp);
 			ceph_encode_copy(&p, &ts, sizeof(ts));
 		}
 
@@ -2225,7 +2257,7 @@ static int __prepare_send_request(struct ceph_mds_client *mdsc,
 /*
  * send request, or put it on the appropriate wait list.
  */
-static int __do_request(struct ceph_mds_client *mdsc,
+static void __do_request(struct ceph_mds_client *mdsc,
 			struct ceph_mds_request *req)
 {
 	struct ceph_mds_session *session = NULL;
@@ -2235,7 +2267,7 @@ static int __do_request(struct ceph_mds_client *mdsc,
 	if (req->r_err || test_bit(CEPH_MDS_R_GOT_RESULT, &req->r_req_flags)) {
 		if (test_bit(CEPH_MDS_R_ABORTED, &req->r_req_flags))
 			__unregister_request(mdsc, req);
-		goto out;
+		return;
 	}
 
 	if (req->r_timeout &&
@@ -2258,7 +2290,7 @@ static int __do_request(struct ceph_mds_client *mdsc,
 		if (mdsc->mdsmap->m_epoch == 0) {
 			dout("do_request no mdsmap, waiting for map\n");
 			list_add(&req->r_wait, &mdsc->waiting_for_map);
-			goto finish;
+			return;
 		}
 		if (!(mdsc->fsc->mount_options->flags &
 		      CEPH_MOUNT_OPT_MOUNTWAIT) &&
@@ -2276,7 +2308,7 @@ static int __do_request(struct ceph_mds_client *mdsc,
 	    ceph_mdsmap_get_state(mdsc->mdsmap, mds) < CEPH_MDS_STATE_ACTIVE) {
 		dout("do_request no mds or not active, waiting for map\n");
 		list_add(&req->r_wait, &mdsc->waiting_for_map);
-		goto out;
+		return;
 	}
 
 	/* get, open session */
@@ -2326,8 +2358,7 @@ finish:
 		complete_request(mdsc, req);
 		__unregister_request(mdsc, req);
 	}
-out:
-	return err;
+	return;
 }
 
 /*
@@ -2748,7 +2779,7 @@ static void handle_session(struct ceph_mds_session *session,
 	int wake = 0;
 
 	/* decode */
-	if (msg->front.iov_len != sizeof(*h))
+	if (msg->front.iov_len < sizeof(*h))
 		goto bad;
 	op = le32_to_cpu(h->op);
 	seq = le64_to_cpu(h->seq);
@@ -2958,15 +2989,12 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap,
 		rec.v2.flock_len = (__force __le32)
 			((ci->i_ceph_flags & CEPH_I_ERROR_FILELOCK) ? 0 : 1);
 	} else {
-		struct timespec ts;
 		rec.v1.cap_id = cpu_to_le64(cap->cap_id);
 		rec.v1.wanted = cpu_to_le32(__ceph_caps_wanted(ci));
 		rec.v1.issued = cpu_to_le32(cap->issued);
 		rec.v1.size = cpu_to_le64(inode->i_size);
-		ts = timespec64_to_timespec(inode->i_mtime);
-		ceph_encode_timespec(&rec.v1.mtime, &ts);
-		ts = timespec64_to_timespec(inode->i_atime);
-		ceph_encode_timespec(&rec.v1.atime, &ts);
+		ceph_encode_timespec64(&rec.v1.mtime, &inode->i_mtime);
+		ceph_encode_timespec64(&rec.v1.atime, &inode->i_atime);
 		rec.v1.snaprealm = cpu_to_le64(ci->i_snap_realm->ino);
 		rec.v1.pathbase = cpu_to_le64(pathbase);
 	}
@@ -3378,10 +3406,10 @@ static void handle_lease(struct ceph_mds_client *mdsc,
 	vino.ino = le64_to_cpu(h->ino);
 	vino.snap = CEPH_NOSNAP;
 	seq = le32_to_cpu(h->seq);
-	dname.name = (void *)h + sizeof(*h) + sizeof(u32);
-	dname.len = msg->front.iov_len - sizeof(*h) - sizeof(u32);
-	if (dname.len != get_unaligned_le32(h+1))
+	dname.len = get_unaligned_le32(h + 1);
+	if (msg->front.iov_len < sizeof(*h) + sizeof(u32) + dname.len)
 		goto bad;
+	dname.name = (void *)(h + 1) + sizeof(u32);
 
 	/* lookup inode */
 	inode = ceph_find_inode(sb, vino);
@@ -3644,8 +3672,8 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc)
 	init_rwsem(&mdsc->pool_perm_rwsem);
 	mdsc->pool_perm_tree = RB_ROOT;
 
-	strncpy(mdsc->nodename, utsname()->nodename,
-		sizeof(mdsc->nodename) - 1);
+	strscpy(mdsc->nodename, utsname()->nodename,
+		sizeof(mdsc->nodename));
 	return 0;
 }
 
@@ -4019,7 +4047,8 @@ void ceph_mdsc_handle_mdsmap(struct ceph_mds_client *mdsc, struct ceph_msg *msg)
 	} else {
 		mdsc->mdsmap = newmap;  /* first mds map */
 	}
-	mdsc->fsc->sb->s_maxbytes = mdsc->mdsmap->m_max_file_size;
+	mdsc->fsc->max_file_size = min((loff_t)mdsc->mdsmap->m_max_file_size,
+					MAX_LFS_FILESIZE);
 
 	__wake_requests(mdsc, &mdsc->waiting_for_map);
 	ceph_monc_got_map(&mdsc->fsc->client->monc, CEPH_SUB_MDSMAP,
@@ -4155,6 +4184,16 @@ static struct ceph_auth_handshake *get_authorizer(struct ceph_connection *con,
 	return auth;
 }
 
+static int add_authorizer_challenge(struct ceph_connection *con,
+				    void *challenge_buf, int challenge_buf_len)
+{
+	struct ceph_mds_session *s = con->private;
+	struct ceph_mds_client *mdsc = s->s_mdsc;
+	struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
+
+	return ceph_auth_add_authorizer_challenge(ac, s->s_auth.authorizer,
+					    challenge_buf, challenge_buf_len);
+}
 
 static int verify_authorizer_reply(struct ceph_connection *con)
 {
@@ -4218,6 +4257,7 @@ static const struct ceph_connection_operations mds_con_ops = {
 	.put = con_put,
 	.dispatch = dispatch,
 	.get_authorizer = get_authorizer,
+	.add_authorizer_challenge = add_authorizer_challenge,
 	.verify_authorizer_reply = verify_authorizer_reply,
 	.invalidate_authorizer = invalidate_authorizer,
 	.peer_reset = peer_reset,

fs/ceph/mds_client.h

@@ -16,6 +16,18 @@
 #include <linux/ceph/mdsmap.h>
 #include <linux/ceph/auth.h>
 
+/* The first 8 bits are reserved for old ceph releases */
+#define CEPHFS_FEATURE_MIMIC    8
+
+#define CEPHFS_FEATURES_ALL {           \
+  0, 1, 2, 3, 4, 5, 6, 7,		\
+  CEPHFS_FEATURE_MIMIC,                 \
+}
+
+#define CEPHFS_FEATURES_CLIENT_SUPPORTED CEPHFS_FEATURES_ALL
+#define CEPHFS_FEATURES_CLIENT_REQUIRED {}
+
+
 /*
  * Some lock dependencies:
  *
@@ -229,7 +241,7 @@ struct ceph_mds_request {
 	int r_fmode;        /* file mode, if expecting cap */
 	kuid_t r_uid;
 	kgid_t r_gid;
-	struct timespec r_stamp;
+	struct timespec64 r_stamp;
 
 	/* for choosing which mds to send this request to */
 	int r_direct_mode;

fs/ceph/quota.c

@@ -48,7 +48,7 @@ void ceph_handle_quota(struct ceph_mds_client *mdsc,
 	struct inode *inode;
 	struct ceph_inode_info *ci;
 
-	if (msg->front.iov_len != sizeof(*h)) {
+	if (msg->front.iov_len < sizeof(*h)) {
 		pr_err("%s corrupt message mds%d len %d\n", __func__,
 		       session->s_mds, (int)msg->front.iov_len);
 		ceph_msg_dump(msg);

fs/ceph/snap.c

@@ -594,9 +594,9 @@ int __ceph_finish_cap_snap(struct ceph_inode_info *ci,
 
 	BUG_ON(capsnap->writing);
 	capsnap->size = inode->i_size;
-	capsnap->mtime = timespec64_to_timespec(inode->i_mtime);
-	capsnap->atime = timespec64_to_timespec(inode->i_atime);
-	capsnap->ctime = timespec64_to_timespec(inode->i_ctime);
+	capsnap->mtime = inode->i_mtime;
+	capsnap->atime = inode->i_atime;
+	capsnap->ctime = inode->i_ctime;
 	capsnap->time_warp_seq = ci->i_time_warp_seq;
 	capsnap->truncate_size = ci->i_truncate_size;
 	capsnap->truncate_seq = ci->i_truncate_seq;

fs/ceph/super.c

@@ -219,8 +219,7 @@ static int parse_fsopt_token(char *c, void *private)
 	if (token < Opt_last_int) {
 		ret = match_int(&argstr[0], &intval);
 		if (ret < 0) {
-			pr_err("bad mount option arg (not int) "
-			       "at '%s'\n", c);
+			pr_err("bad option arg (not int) at '%s'\n", c);
 			return ret;
 		}
 		dout("got int token %d val %d\n", token, intval);
@@ -941,11 +940,12 @@ static int ceph_set_super(struct super_block *s, void *data)
 	dout("set_super %p data %p\n", s, data);
 
 	s->s_flags = fsc->mount_options->sb_flags;
-	s->s_maxbytes = 1ULL << 40;  /* temp value until we get mdsmap */
+	s->s_maxbytes = MAX_LFS_FILESIZE;
 
 	s->s_xattr = ceph_xattr_handlers;
 	s->s_fs_info = fsc;
 	fsc->sb = s;
+	fsc->max_file_size = 1ULL << 40; /* temp value until we get mdsmap */
 
 	s->s_op = &ceph_super_ops;
 	s->s_d_op = &ceph_dentry_ops;

fs/ceph/super.h

@@ -98,6 +98,7 @@ struct ceph_fs_client {
 
 	unsigned long mount_state;
 	int min_caps;                  /* min caps i added */
+	loff_t max_file_size;
 
 	struct ceph_mds_client *mdsc;
 
@@ -193,7 +194,7 @@ struct ceph_cap_snap {
 	u64 xattr_version;
 
 	u64 size;
-	struct timespec mtime, atime, ctime;
+	struct timespec64 mtime, atime, ctime;
 	u64 time_warp_seq;
 	u64 truncate_size;
 	u32 truncate_seq;
@@ -307,7 +308,7 @@ struct ceph_inode_info {
 	char *i_symlink;
 
 	/* for dirs */
-	struct timespec i_rctime;
+	struct timespec64 i_rctime;
 	u64 i_rbytes, i_rfiles, i_rsubdirs;
 	u64 i_files, i_subdirs;
 
@@ -655,7 +656,7 @@ extern void ceph_caps_finalize(struct ceph_mds_client *mdsc);
 extern void ceph_adjust_min_caps(struct ceph_mds_client *mdsc, int delta);
 extern int ceph_reserve_caps(struct ceph_mds_client *mdsc,
 			     struct ceph_cap_reservation *ctx, int need);
-extern int ceph_unreserve_caps(struct ceph_mds_client *mdsc,
+extern void ceph_unreserve_caps(struct ceph_mds_client *mdsc,
 			       struct ceph_cap_reservation *ctx);
 extern void ceph_reservation_status(struct ceph_fs_client *client,
 				    int *total, int *avail, int *used,
@@ -857,8 +858,9 @@ extern struct inode *ceph_get_snapdir(struct inode *parent);
 extern int ceph_fill_file_size(struct inode *inode, int issued,
 			       u32 truncate_seq, u64 truncate_size, u64 size);
 extern void ceph_fill_file_time(struct inode *inode, int issued,
-				u64 time_warp_seq, struct timespec *ctime,
-				struct timespec *mtime, struct timespec *atime);
+				u64 time_warp_seq, struct timespec64 *ctime,
+				struct timespec64 *mtime,
+				struct timespec64 *atime);
 extern int ceph_fill_trace(struct super_block *sb,
 			   struct ceph_mds_request *req);
 extern int ceph_readdir_prepopulate(struct ceph_mds_request *req,

fs/ceph/xattr.c

@@ -224,8 +224,8 @@ static size_t ceph_vxattrcb_dir_rbytes(struct ceph_inode_info *ci, char *val,
 static size_t ceph_vxattrcb_dir_rctime(struct ceph_inode_info *ci, char *val,
 				       size_t size)
 {
-	return snprintf(val, size, "%ld.09%ld", (long)ci->i_rctime.tv_sec,
-			(long)ci->i_rctime.tv_nsec);
+	return snprintf(val, size, "%lld.09%ld", ci->i_rctime.tv_sec,
+			ci->i_rctime.tv_nsec);
 }
 
 /* quotas */

include/linux/ceph/auth.h

@@ -64,6 +64,10 @@ struct ceph_auth_client_ops {
 	/* ensure that an existing authorizer is up to date */
 	int (*update_authorizer)(struct ceph_auth_client *ac, int peer_type,
 				 struct ceph_auth_handshake *auth);
+	int (*add_authorizer_challenge)(struct ceph_auth_client *ac,
+					struct ceph_authorizer *a,
+					void *challenge_buf,
+					int challenge_buf_len);
 	int (*verify_authorizer_reply)(struct ceph_auth_client *ac,
 				       struct ceph_authorizer *a);
 	void (*invalidate_authorizer)(struct ceph_auth_client *ac,
@@ -118,6 +122,10 @@ void ceph_auth_destroy_authorizer(struct ceph_authorizer *a);
 extern int ceph_auth_update_authorizer(struct ceph_auth_client *ac,
 				       int peer_type,
 				       struct ceph_auth_handshake *a);
+int ceph_auth_add_authorizer_challenge(struct ceph_auth_client *ac,
+				       struct ceph_authorizer *a,
+				       void *challenge_buf,
+				       int challenge_buf_len);
 extern int ceph_auth_verify_authorizer_reply(struct ceph_auth_client *ac,
 					     struct ceph_authorizer *a);
 extern void ceph_auth_invalidate_authorizer(struct ceph_auth_client *ac,

include/linux/ceph/ceph_features.h

@@ -165,9 +165,9 @@ DEFINE_CEPH_FEATURE(58, 1, FS_FILE_LAYOUT_V2) // overlap
 DEFINE_CEPH_FEATURE(59, 1, FS_BTIME)
 DEFINE_CEPH_FEATURE(59, 1, FS_CHANGE_ATTR) // overlap
 DEFINE_CEPH_FEATURE(59, 1, MSG_ADDR2) // overlap
-DEFINE_CEPH_FEATURE(60, 1, BLKIN_TRACING)  // *do not share this bit*
+DEFINE_CEPH_FEATURE(60, 1, OSD_RECOVERY_DELETES) // *do not share this bit*
+DEFINE_CEPH_FEATURE(61, 1, CEPHX_V2)             // *do not share this bit*
 
-DEFINE_CEPH_FEATURE(61, 1, RESERVED2)          // unused, but slow down!
 DEFINE_CEPH_FEATURE(62, 1, RESERVED)           // do not use; used as a sentinal
 DEFINE_CEPH_FEATURE_DEPRECATED(63, 1, RESERVED_BROKEN, LUMINOUS) // client-facing
 
@@ -210,7 +210,8 @@ DEFINE_CEPH_FEATURE_DEPRECATED(63, 1, RESERVED_BROKEN, LUMINOUS) // client-facin
 	 CEPH_FEATURE_SERVER_JEWEL |		\
 	 CEPH_FEATURE_MON_STATEFUL_SUB |	\
 	 CEPH_FEATURE_CRUSH_TUNABLES5 |		\
-	 CEPH_FEATURE_NEW_OSDOPREPLY_ENCODING)
+	 CEPH_FEATURE_NEW_OSDOPREPLY_ENCODING |	\
+	 CEPH_FEATURE_CEPHX_V2)
 
 #define CEPH_FEATURES_REQUIRED_DEFAULT   \
 	(CEPH_FEATURE_NOSRCADDR |	 \

include/linux/ceph/decode.h

@@ -194,16 +194,22 @@ ceph_decode_skip_n(p, end, sizeof(u8), bad)
 	} while (0)
 
 /*
- * struct ceph_timespec <-> struct timespec
+ * struct ceph_timespec <-> struct timespec64
  */
-static inline void ceph_decode_timespec(struct timespec *ts,
-					const struct ceph_timespec *tv)
+static inline void ceph_decode_timespec64(struct timespec64 *ts,
+					  const struct ceph_timespec *tv)
 {
-	ts->tv_sec = (__kernel_time_t)le32_to_cpu(tv->tv_sec);
+	/*
+	 * This will still overflow in year 2106.  We could extend
+	 * the protocol to steal two more bits from tv_nsec to
+	 * add three more 136 year epochs after that the way ext4
+	 * does if necessary.
+	 */
+	ts->tv_sec = (time64_t)le32_to_cpu(tv->tv_sec);
 	ts->tv_nsec = (long)le32_to_cpu(tv->tv_nsec);
 }
-static inline void ceph_encode_timespec(struct ceph_timespec *tv,
-					const struct timespec *ts)
+static inline void ceph_encode_timespec64(struct ceph_timespec *tv,
+					  const struct timespec64 *ts)
 {
 	tv->tv_sec = cpu_to_le32((u32)ts->tv_sec);
 	tv->tv_nsec = cpu_to_le32((u32)ts->tv_nsec);

include/linux/ceph/messenger.h

@@ -31,6 +31,9 @@ struct ceph_connection_operations {
 	struct ceph_auth_handshake *(*get_authorizer) (
 				struct ceph_connection *con,
 			       int *proto, int force_new);
+	int (*add_authorizer_challenge)(struct ceph_connection *con,
+					void *challenge_buf,
+					int challenge_buf_len);
 	int (*verify_authorizer_reply) (struct ceph_connection *con);
 	int (*invalidate_authorizer)(struct ceph_connection *con);
 
@@ -286,9 +289,8 @@ struct ceph_connection {
 				 attempt for this connection, client */
 	u32 peer_global_seq;  /* peer's global seq for this connection */
 
+	struct ceph_auth_handshake *auth;
 	int auth_retry;       /* true if we need a newer authorizer */
-	void *auth_reply_buf;   /* where to put the authorizer reply */
-	int auth_reply_buf_len;
 
 	struct mutex mutex;
 
@@ -330,7 +332,7 @@ struct ceph_connection {
 	int in_base_pos;     /* bytes read */
 	__le64 in_temp_ack;  /* for reading an ack */
 
-	struct timespec last_keepalive_ack; /* keepalive2 ack stamp */
+	struct timespec64 last_keepalive_ack; /* keepalive2 ack stamp */
 
 	struct delayed_work work;	    /* send|recv work */
 	unsigned long       delay;          /* current delay interval */

include/linux/ceph/msgr.h

@@ -91,7 +91,7 @@ struct ceph_entity_inst {
 #define CEPH_MSGR_TAG_SEQ           13 /* 64-bit int follows with seen seq number */
 #define CEPH_MSGR_TAG_KEEPALIVE2    14 /* keepalive2 byte + ceph_timespec */
 #define CEPH_MSGR_TAG_KEEPALIVE2_ACK 15 /* keepalive2 reply */
-
+#define CEPH_MSGR_TAG_CHALLENGE_AUTHORIZER 16  /* cephx v2 doing server challenge */
 
 /*
  * connection negotiation

include/linux/ceph/osd_client.h

@@ -199,7 +199,7 @@ struct ceph_osd_request {
 	/* set by submitter */
 	u64 r_snapid;                         /* for reads, CEPH_NOSNAP o/w */
 	struct ceph_snap_context *r_snapc;    /* for writes */
-	struct timespec r_mtime;              /* ditto */
+	struct timespec64 r_mtime;            /* ditto */
 	u64 r_data_offset;                    /* ditto */
 	bool r_linger;                        /* don't resend on failure */
 
@@ -253,7 +253,7 @@ struct ceph_osd_linger_request {
 	struct ceph_osd_request_target t;
 	u32 map_dne_bound;
 
-	struct timespec mtime;
+	struct timespec64 mtime;
 
 	struct kref kref;
 	struct mutex lock;
@@ -508,7 +508,7 @@ extern int ceph_osdc_writepages(struct ceph_osd_client *osdc,
 				struct ceph_snap_context *sc,
 				u64 off, u64 len,
 				u32 truncate_seq, u64 truncate_size,
-				struct timespec *mtime,
+				struct timespec64 *mtime,
 				struct page **pages, int nr_pages);
 
 /* watch/notify */
@@ -528,12 +528,12 @@ int ceph_osdc_notify_ack(struct ceph_osd_client *osdc,
 			 u64 notify_id,
 			 u64 cookie,
 			 void *payload,
-			 size_t payload_len);
+			 u32 payload_len);
 int ceph_osdc_notify(struct ceph_osd_client *osdc,
 		     struct ceph_object_id *oid,
 		     struct ceph_object_locator *oloc,
 		     void *payload,
-		     size_t payload_len,
+		     u32 payload_len,
 		     u32 timeout,
 		     struct page ***preply_pages,
 		     size_t *preply_len);

include/linux/ceph/pagelist.h

@@ -68,7 +68,7 @@ static inline int ceph_pagelist_encode_8(struct ceph_pagelist *pl, u8 v)
 	return ceph_pagelist_append(pl, &v, 1);
 }
 static inline int ceph_pagelist_encode_string(struct ceph_pagelist *pl,
-					      char *s, size_t len)
+					      char *s, u32 len)
 {
 	int ret = ceph_pagelist_encode_32(pl, len);
 	if (ret)

net/ceph/Kconfig

@@ -41,4 +41,3 @@ config CEPH_LIB_USE_DNS_RESOLVER
 	  Documentation/networking/dns_resolver.txt
 
 	  If unsure, say N.
-

net/ceph/Makefile

@@ -15,4 +15,3 @@ libceph-y := ceph_common.o messenger.o msgpool.o buffer.o pagelist.o \
 	auth_x.o \
 	ceph_fs.o ceph_strings.o ceph_hash.o \
 	pagevec.o snapshot.o string_table.o
-

net/ceph/auth.c

@@ -315,6 +315,22 @@ int ceph_auth_update_authorizer(struct ceph_auth_client *ac,
 }
 EXPORT_SYMBOL(ceph_auth_update_authorizer);
 
+int ceph_auth_add_authorizer_challenge(struct ceph_auth_client *ac,
+				       struct ceph_authorizer *a,
+				       void *challenge_buf,
+				       int challenge_buf_len)
+{
+	int ret = 0;
+
+	mutex_lock(&ac->mutex);
+	if (ac->ops && ac->ops->add_authorizer_challenge)
+		ret = ac->ops->add_authorizer_challenge(ac, a, challenge_buf,
+							challenge_buf_len);
+	mutex_unlock(&ac->mutex);
+	return ret;
+}
+EXPORT_SYMBOL(ceph_auth_add_authorizer_challenge);
+
 int ceph_auth_verify_authorizer_reply(struct ceph_auth_client *ac,
 				      struct ceph_authorizer *a)
 {

net/ceph/auth_none.c

@@ -142,4 +142,3 @@ int ceph_auth_none_init(struct ceph_auth_client *ac)
 	ac->ops = &ceph_auth_none_ops;
 	return 0;
 }
-

net/ceph/auth_none.h

@@ -26,4 +26,3 @@ struct ceph_auth_none_info {
 int ceph_auth_none_init(struct ceph_auth_client *ac);
 
 #endif
-

net/ceph/auth_x.c

@@ -9,6 +9,7 @@
 
 #include <linux/ceph/decode.h>
 #include <linux/ceph/auth.h>
+#include <linux/ceph/ceph_features.h>
 #include <linux/ceph/libceph.h>
 #include <linux/ceph/messenger.h>
 
@@ -70,25 +71,40 @@ static int ceph_x_encrypt(struct ceph_crypto_key *secret, void *buf,
 	return sizeof(u32) + ciphertext_len;
 }
 
+static int __ceph_x_decrypt(struct ceph_crypto_key *secret, void *p,
+			    int ciphertext_len)
+{
+	struct ceph_x_encrypt_header *hdr = p;
+	int plaintext_len;
+	int ret;
+
+	ret = ceph_crypt(secret, false, p, ciphertext_len, ciphertext_len,
+			 &plaintext_len);
+	if (ret)
+		return ret;
+
+	if (le64_to_cpu(hdr->magic) != CEPHX_ENC_MAGIC) {
+		pr_err("%s bad magic\n", __func__);
+		return -EINVAL;
+	}
+
+	return plaintext_len - sizeof(*hdr);
+}
+
 static int ceph_x_decrypt(struct ceph_crypto_key *secret, void **p, void *end)
 {
-	struct ceph_x_encrypt_header *hdr = *p + sizeof(u32);
-	int ciphertext_len, plaintext_len;
+	int ciphertext_len;
 	int ret;
 
 	ceph_decode_32_safe(p, end, ciphertext_len, e_inval);
 	ceph_decode_need(p, end, ciphertext_len, e_inval);
 
-	ret = ceph_crypt(secret, false, *p, end - *p, ciphertext_len,
-			 &plaintext_len);
-	if (ret)
+	ret = __ceph_x_decrypt(secret, *p, ciphertext_len);
+	if (ret < 0)
 		return ret;
 
-	if (hdr->struct_v != 1 || le64_to_cpu(hdr->magic) != CEPHX_ENC_MAGIC)
-		return -EPERM;
-
 	*p += ciphertext_len;
-	return plaintext_len - sizeof(struct ceph_x_encrypt_header);
+	return ret;
 
 e_inval:
 	return -EINVAL;
@@ -149,12 +165,12 @@ static int process_one_ticket(struct ceph_auth_client *ac,
 	void *dp, *dend;
 	int dlen;
 	char is_enc;
-	struct timespec validity;
+	struct timespec64 validity;
 	void *tp, *tpend;
 	void **ptp;
 	struct ceph_crypto_key new_session_key = { 0 };
 	struct ceph_buffer *new_ticket_blob;
-	unsigned long new_expires, new_renew_after;
+	time64_t new_expires, new_renew_after;
 	u64 new_secret_id;
 	int ret;
 
@@ -189,11 +205,11 @@ static int process_one_ticket(struct ceph_auth_client *ac,
 	if (ret)
 		goto out;
 
-	ceph_decode_timespec(&validity, dp);
+	ceph_decode_timespec64(&validity, dp);
 	dp += sizeof(struct ceph_timespec);
-	new_expires = get_seconds() + validity.tv_sec;
+	new_expires = ktime_get_real_seconds() + validity.tv_sec;
 	new_renew_after = new_expires - (validity.tv_sec / 4);
-	dout(" expires=%lu renew_after=%lu\n", new_expires,
+	dout(" expires=%llu renew_after=%llu\n", new_expires,
 	     new_renew_after);
 
 	/* ticket blob for service */
@@ -275,6 +291,51 @@ bad:
 	return -EINVAL;
 }
 
+/*
+ * Encode and encrypt the second part (ceph_x_authorize_b) of the
+ * authorizer.  The first part (ceph_x_authorize_a) should already be
+ * encoded.
+ */
+static int encrypt_authorizer(struct ceph_x_authorizer *au,
+			      u64 *server_challenge)
+{
+	struct ceph_x_authorize_a *msg_a;
+	struct ceph_x_authorize_b *msg_b;
+	void *p, *end;
+	int ret;
+
+	msg_a = au->buf->vec.iov_base;
+	WARN_ON(msg_a->ticket_blob.secret_id != cpu_to_le64(au->secret_id));
+	p = (void *)(msg_a + 1) + le32_to_cpu(msg_a->ticket_blob.blob_len);
+	end = au->buf->vec.iov_base + au->buf->vec.iov_len;
+
+	msg_b = p + ceph_x_encrypt_offset();
+	msg_b->struct_v = 2;
+	msg_b->nonce = cpu_to_le64(au->nonce);
+	if (server_challenge) {
+		msg_b->have_challenge = 1;
+		msg_b->server_challenge_plus_one =
+		    cpu_to_le64(*server_challenge + 1);
+	} else {
+		msg_b->have_challenge = 0;
+		msg_b->server_challenge_plus_one = 0;
+	}
+
+	ret = ceph_x_encrypt(&au->session_key, p, end - p, sizeof(*msg_b));
+	if (ret < 0)
+		return ret;
+
+	p += ret;
+	if (server_challenge) {
+		WARN_ON(p != end);
+	} else {
+		WARN_ON(p > end);
+		au->buf->vec.iov_len = p - au->buf->vec.iov_base;
+	}
+
+	return 0;
+}
+
 static void ceph_x_authorizer_cleanup(struct ceph_x_authorizer *au)
 {
 	ceph_crypto_key_destroy(&au->session_key);
@@ -291,7 +352,6 @@ static int ceph_x_build_authorizer(struct ceph_auth_client *ac,
 	int maxlen;
 	struct ceph_x_authorize_a *msg_a;
 	struct ceph_x_authorize_b *msg_b;
-	void *p, *end;
 	int ret;
 	int ticket_blob_len =
 		(th->ticket_blob ? th->ticket_blob->vec.iov_len : 0);
@@ -335,21 +395,13 @@ static int ceph_x_build_authorizer(struct ceph_auth_client *ac,
 	dout(" th %p secret_id %lld %lld\n", th, th->secret_id,
 	     le64_to_cpu(msg_a->ticket_blob.secret_id));
 
-	p = msg_a + 1;
-	p += ticket_blob_len;
-	end = au->buf->vec.iov_base + au->buf->vec.iov_len;
-
-	msg_b = p + ceph_x_encrypt_offset();
-	msg_b->struct_v = 1;
 	get_random_bytes(&au->nonce, sizeof(au->nonce));
-	msg_b->nonce = cpu_to_le64(au->nonce);
-	ret = ceph_x_encrypt(&au->session_key, p, end - p, sizeof(*msg_b));
-	if (ret < 0)
+	ret = encrypt_authorizer(au, NULL);
+	if (ret) {
+		pr_err("failed to encrypt authorizer: %d", ret);
 		goto out_au;
+	}
 
-	p += ret;
-	WARN_ON(p > end);
-	au->buf->vec.iov_len = p - au->buf->vec.iov_base;
 	dout(" built authorizer nonce %llx len %d\n", au->nonce,
 	     (int)au->buf->vec.iov_len);
 	return 0;
@@ -385,13 +437,13 @@ static bool need_key(struct ceph_x_ticket_handler *th)
 	if (!th->have_key)
 		return true;
 
-	return get_seconds() >= th->renew_after;
+	return ktime_get_real_seconds() >= th->renew_after;
 }
 
 static bool have_key(struct ceph_x_ticket_handler *th)
 {
 	if (th->have_key) {
-		if (get_seconds() >= th->expires)
+		if (ktime_get_real_seconds() >= th->expires)
 			th->have_key = false;
 	}
 
@@ -626,6 +678,54 @@ static int ceph_x_update_authorizer(
 	return 0;
 }
 
+static int decrypt_authorize_challenge(struct ceph_x_authorizer *au,
+				       void *challenge_buf,
+				       int challenge_buf_len,
+				       u64 *server_challenge)
+{
+	struct ceph_x_authorize_challenge *ch =
+	    challenge_buf + sizeof(struct ceph_x_encrypt_header);
+	int ret;
+
+	/* no leading len */
+	ret = __ceph_x_decrypt(&au->session_key, challenge_buf,
+			       challenge_buf_len);
+	if (ret < 0)
+		return ret;
+	if (ret < sizeof(*ch)) {
+		pr_err("bad size %d for ceph_x_authorize_challenge\n", ret);
+		return -EINVAL;
+	}
+
+	*server_challenge = le64_to_cpu(ch->server_challenge);
+	return 0;
+}
+
+static int ceph_x_add_authorizer_challenge(struct ceph_auth_client *ac,
+					   struct ceph_authorizer *a,
+					   void *challenge_buf,
+					   int challenge_buf_len)
+{
+	struct ceph_x_authorizer *au = (void *)a;
+	u64 server_challenge;
+	int ret;
+
+	ret = decrypt_authorize_challenge(au, challenge_buf, challenge_buf_len,
+					  &server_challenge);
+	if (ret) {
+		pr_err("failed to decrypt authorize challenge: %d", ret);
+		return ret;
+	}
+
+	ret = encrypt_authorizer(au, &server_challenge);
+	if (ret) {
+		pr_err("failed to encrypt authorizer w/ challenge: %d", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
 static int ceph_x_verify_authorizer_reply(struct ceph_auth_client *ac,
 					  struct ceph_authorizer *a)
 {
@@ -637,8 +737,10 @@ static int ceph_x_verify_authorizer_reply(struct ceph_auth_client *ac,
 	ret = ceph_x_decrypt(&au->session_key, &p, p + CEPHX_AU_ENC_BUF_LEN);
 	if (ret < 0)
 		return ret;
-	if (ret != sizeof(*reply))
-		return -EPERM;
+	if (ret < sizeof(*reply)) {
+		pr_err("bad size %d for ceph_x_authorize_reply\n", ret);
+		return -EINVAL;
+	}
 
 	if (au->nonce + 1 != le64_to_cpu(reply->nonce_plus_one))
 		ret = -EPERM;
@@ -704,26 +806,64 @@ static int calc_signature(struct ceph_x_authorizer *au, struct ceph_msg *msg,
 			  __le64 *psig)
 {
 	void *enc_buf = au->enc_buf;
-	struct {
-		__le32 len;
-		__le32 header_crc;
-		__le32 front_crc;
-		__le32 middle_crc;
-		__le32 data_crc;
-	} __packed *sigblock = enc_buf + ceph_x_encrypt_offset();
 	int ret;
 
-	sigblock->len = cpu_to_le32(4*sizeof(u32));
-	sigblock->header_crc = msg->hdr.crc;
-	sigblock->front_crc = msg->footer.front_crc;
-	sigblock->middle_crc = msg->footer.middle_crc;
-	sigblock->data_crc =  msg->footer.data_crc;
-	ret = ceph_x_encrypt(&au->session_key, enc_buf, CEPHX_AU_ENC_BUF_LEN,
-			     sizeof(*sigblock));
-	if (ret < 0)
-		return ret;
+	if (!CEPH_HAVE_FEATURE(msg->con->peer_features, CEPHX_V2)) {
+		struct {
+			__le32 len;
+			__le32 header_crc;
+			__le32 front_crc;
+			__le32 middle_crc;
+			__le32 data_crc;
+		} __packed *sigblock = enc_buf + ceph_x_encrypt_offset();
+
+		sigblock->len = cpu_to_le32(4*sizeof(u32));
+		sigblock->header_crc = msg->hdr.crc;
+		sigblock->front_crc = msg->footer.front_crc;
+		sigblock->middle_crc = msg->footer.middle_crc;
+		sigblock->data_crc =  msg->footer.data_crc;
+
+		ret = ceph_x_encrypt(&au->session_key, enc_buf,
+				     CEPHX_AU_ENC_BUF_LEN, sizeof(*sigblock));
+		if (ret < 0)
+			return ret;
+
+		*psig = *(__le64 *)(enc_buf + sizeof(u32));
+	} else {
+		struct {
+			__le32 header_crc;
+			__le32 front_crc;
+			__le32 front_len;
+			__le32 middle_crc;
+			__le32 middle_len;
+			__le32 data_crc;
+			__le32 data_len;
+			__le32 seq_lower_word;
+		} __packed *sigblock = enc_buf;
+		struct {
+			__le64 a, b, c, d;
+		} __packed *penc = enc_buf;
+		int ciphertext_len;
+
+		sigblock->header_crc = msg->hdr.crc;
+		sigblock->front_crc = msg->footer.front_crc;
+		sigblock->front_len = msg->hdr.front_len;
+		sigblock->middle_crc = msg->footer.middle_crc;
+		sigblock->middle_len = msg->hdr.middle_len;
+		sigblock->data_crc =  msg->footer.data_crc;
+		sigblock->data_len = msg->hdr.data_len;
+		sigblock->seq_lower_word = *(__le32 *)&msg->hdr.seq;
+
+		/* no leading len, no ceph_x_encrypt_header */
+		ret = ceph_crypt(&au->session_key, true, enc_buf,
+				 CEPHX_AU_ENC_BUF_LEN, sizeof(*sigblock),
+				 &ciphertext_len);
+		if (ret)
+			return ret;
+
+		*psig = penc->a ^ penc->b ^ penc->c ^ penc->d;
+	}
 
-	*psig = *(__le64 *)(enc_buf + sizeof(u32));
 	return 0;
 }
 
@@ -778,6 +918,7 @@ static const struct ceph_auth_client_ops ceph_x_ops = {
 	.handle_reply = ceph_x_handle_reply,
 	.create_authorizer = ceph_x_create_authorizer,
 	.update_authorizer = ceph_x_update_authorizer,
+	.add_authorizer_challenge = ceph_x_add_authorizer_challenge,
 	.verify_authorizer_reply = ceph_x_verify_authorizer_reply,
 	.invalidate_authorizer = ceph_x_invalidate_authorizer,
 	.reset =  ceph_x_reset,
@@ -823,5 +964,3 @@ out_nomem:
 out:
 	return ret;
 }
-
-

net/ceph/auth_x.h

@@ -22,7 +22,7 @@ struct ceph_x_ticket_handler {
 	u64 secret_id;
 	struct ceph_buffer *ticket_blob;
 
-	unsigned long renew_after, expires;
+	time64_t renew_after, expires;
 };
 
 #define CEPHX_AU_ENC_BUF_LEN	128  /* big enough for encrypted blob */
@@ -52,4 +52,3 @@ struct ceph_x_info {
 int ceph_x_init(struct ceph_auth_client *ac);
 
 #endif
-

net/ceph/auth_x_protocol.h

@@ -70,6 +70,13 @@ struct ceph_x_authorize_a {
 struct ceph_x_authorize_b {
 	__u8 struct_v;
 	__le64 nonce;
+	__u8 have_challenge;
+	__le64 server_challenge_plus_one;
+} __attribute__ ((packed));
+
+struct ceph_x_authorize_challenge {
+	__u8 struct_v;
+	__le64 server_challenge;
 } __attribute__ ((packed));
 
 struct ceph_x_authorize_reply {

net/ceph/ceph_common.c

@@ -304,7 +304,7 @@ static int get_secret(struct ceph_crypto_key *dst, const char *name) {
 	struct ceph_crypto_key *ckey;
 
 	ukey = request_key(&key_type_ceph, name, NULL);
-	if (!ukey || IS_ERR(ukey)) {
+	if (IS_ERR(ukey)) {
 		/* request_key errors don't map nicely to mount(2)
 		   errors; don't even try, but still printk */
 		key_err = PTR_ERR(ukey);
@@ -379,7 +379,7 @@ ceph_parse_options(char *options, const char *dev_name,
 
 	/* parse mount options */
 	while ((c = strsep(&options, ",")) != NULL) {
-		int token, intval, ret;
+		int token, intval;
 		if (!*c)
 			continue;
 		err = -EINVAL;
@@ -394,11 +394,10 @@ ceph_parse_options(char *options, const char *dev_name,
 			continue;
 		}
 		if (token < Opt_last_int) {
-			ret = match_int(&argstr[0], &intval);
-			if (ret < 0) {
-				pr_err("bad mount option arg (not int) "
-				       "at '%s'\n", c);
-				continue;
+			err = match_int(&argstr[0], &intval);
+			if (err < 0) {
+				pr_err("bad option arg (not int) at '%s'\n", c);
+				goto out;
 			}
 			dout("got int token %d val %d\n", token, intval);
 		} else if (token > Opt_last_int && token < Opt_last_string) {

net/ceph/cls_lock_client.c

@@ -32,7 +32,7 @@ int ceph_cls_lock(struct ceph_osd_client *osdc,
 	int desc_len = strlen(desc);
 	void *p, *end;
 	struct page *lock_op_page;
-	struct timespec mtime;
+	struct timespec64 mtime;
 	int ret;
 
 	lock_op_buf_size = name_len + sizeof(__le32) +
@@ -63,7 +63,7 @@ int ceph_cls_lock(struct ceph_osd_client *osdc,
 	ceph_encode_string(&p, end, desc, desc_len);
 	/* only support infinite duration */
 	memset(&mtime, 0, sizeof(mtime));
-	ceph_encode_timespec(p, &mtime);
+	ceph_encode_timespec64(p, &mtime);
 	p += sizeof(struct ceph_timespec);
 	ceph_encode_8(&p, flags);
 

net/ceph/crush/mapper.c

@@ -514,7 +514,7 @@ static int crush_choose_firstn(const struct crush_map *map,
 						in, work->work[-1-in->id],
 						x, r,
 						(choose_args ?
-						 &choose_args[-1-in->id] : 0),
+						 &choose_args[-1-in->id] : NULL),
 						outpos);
 				if (item >= map->max_devices) {
 					dprintk("   bad item %d\n", item);
@@ -725,7 +725,7 @@ static void crush_choose_indep(const struct crush_map *map,
 					in, work->work[-1-in->id],
 					x, r,
 					(choose_args ?
-					 &choose_args[-1-in->id] : 0),
+					 &choose_args[-1-in->id] : NULL),
 					outpos);
 				if (item >= map->max_devices) {
 					dprintk("   bad item %d\n", item);

net/ceph/messenger.c

@@ -1417,11 +1417,11 @@ static void prepare_write_keepalive(struct ceph_connection *con)
 	dout("prepare_write_keepalive %p\n", con);
 	con_out_kvec_reset(con);
 	if (con->peer_features & CEPH_FEATURE_MSGR_KEEPALIVE2) {
-		struct timespec now;
+		struct timespec64 now;
 
-		ktime_get_real_ts(&now);
+		ktime_get_real_ts64(&now);
 		con_out_kvec_add(con, sizeof(tag_keepalive2), &tag_keepalive2);
-		ceph_encode_timespec(&con->out_temp_keepalive2, &now);
+		ceph_encode_timespec64(&con->out_temp_keepalive2, &now);
 		con_out_kvec_add(con, sizeof(con->out_temp_keepalive2),
 				 &con->out_temp_keepalive2);
 	} else {
@@ -1434,24 +1434,26 @@ static void prepare_write_keepalive(struct ceph_connection *con)
  * Connection negotiation.
  */
 
-static struct ceph_auth_handshake *get_connect_authorizer(struct ceph_connection *con,
-						int *auth_proto)
+static int get_connect_authorizer(struct ceph_connection *con)
 {
 	struct ceph_auth_handshake *auth;
+	int auth_proto;
 
 	if (!con->ops->get_authorizer) {
+		con->auth = NULL;
 		con->out_connect.authorizer_protocol = CEPH_AUTH_UNKNOWN;
 		con->out_connect.authorizer_len = 0;
-		return NULL;
+		return 0;
 	}
 
-	auth = con->ops->get_authorizer(con, auth_proto, con->auth_retry);
+	auth = con->ops->get_authorizer(con, &auth_proto, con->auth_retry);
 	if (IS_ERR(auth))
-		return auth;
+		return PTR_ERR(auth);
 
-	con->auth_reply_buf = auth->authorizer_reply_buf;
-	con->auth_reply_buf_len = auth->authorizer_reply_buf_len;
-	return auth;
+	con->auth = auth;
+	con->out_connect.authorizer_protocol = cpu_to_le32(auth_proto);
+	con->out_connect.authorizer_len = cpu_to_le32(auth->authorizer_buf_len);
+	return 0;
 }
 
 /*
@@ -1467,12 +1469,22 @@ static void prepare_write_banner(struct ceph_connection *con)
 	con_flag_set(con, CON_FLAG_WRITE_PENDING);
 }
 
+static void __prepare_write_connect(struct ceph_connection *con)
+{
+	con_out_kvec_add(con, sizeof(con->out_connect), &con->out_connect);
+	if (con->auth)
+		con_out_kvec_add(con, con->auth->authorizer_buf_len,
+				 con->auth->authorizer_buf);
+
+	con->out_more = 0;
+	con_flag_set(con, CON_FLAG_WRITE_PENDING);
+}
+
 static int prepare_write_connect(struct ceph_connection *con)
 {
 	unsigned int global_seq = get_global_seq(con->msgr, 0);
 	int proto;
-	int auth_proto;
-	struct ceph_auth_handshake *auth;
+	int ret;
 
 	switch (con->peer_name.type) {
 	case CEPH_ENTITY_TYPE_MON:
@@ -1499,24 +1511,11 @@ static int prepare_write_connect(struct ceph_connection *con)
 	con->out_connect.protocol_version = cpu_to_le32(proto);
 	con->out_connect.flags = 0;
 
-	auth_proto = CEPH_AUTH_UNKNOWN;
-	auth = get_connect_authorizer(con, &auth_proto);
-	if (IS_ERR(auth))
-		return PTR_ERR(auth);
-
-	con->out_connect.authorizer_protocol = cpu_to_le32(auth_proto);
-	con->out_connect.authorizer_len = auth ?
-		cpu_to_le32(auth->authorizer_buf_len) : 0;
-
-	con_out_kvec_add(con, sizeof (con->out_connect),
-					&con->out_connect);
-	if (auth && auth->authorizer_buf_len)
-		con_out_kvec_add(con, auth->authorizer_buf_len,
-					auth->authorizer_buf);
-
-	con->out_more = 0;
-	con_flag_set(con, CON_FLAG_WRITE_PENDING);
+	ret = get_connect_authorizer(con);
+	if (ret)
+		return ret;
 
+	__prepare_write_connect(con);
 	return 0;
 }
 
@@ -1781,11 +1780,21 @@ static int read_partial_connect(struct ceph_connection *con)
 	if (ret <= 0)
 		goto out;
 
-	size = le32_to_cpu(con->in_reply.authorizer_len);
-	end += size;
-	ret = read_partial(con, end, size, con->auth_reply_buf);
-	if (ret <= 0)
-		goto out;
+	if (con->auth) {
+		size = le32_to_cpu(con->in_reply.authorizer_len);
+		if (size > con->auth->authorizer_reply_buf_len) {
+			pr_err("authorizer reply too big: %d > %zu\n", size,
+			       con->auth->authorizer_reply_buf_len);
+			ret = -EINVAL;
+			goto out;
+		}
+
+		end += size;
+		ret = read_partial(con, end, size,
+				   con->auth->authorizer_reply_buf);
+		if (ret <= 0)
+			goto out;
+	}
 
 	dout("read_partial_connect %p tag %d, con_seq = %u, g_seq = %u\n",
 	     con, (int)con->in_reply.tag,
@@ -1793,7 +1802,6 @@ static int read_partial_connect(struct ceph_connection *con)
 	     le32_to_cpu(con->in_reply.global_seq));
 out:
 	return ret;
-
 }
 
 /*
@@ -2076,12 +2084,27 @@ static int process_connect(struct ceph_connection *con)
 
 	dout("process_connect on %p tag %d\n", con, (int)con->in_tag);
 
-	if (con->auth_reply_buf) {
+	if (con->auth) {
 		/*
 		 * Any connection that defines ->get_authorizer()
-		 * should also define ->verify_authorizer_reply().
+		 * should also define ->add_authorizer_challenge() and
+		 * ->verify_authorizer_reply().
+		 *
 		 * See get_connect_authorizer().
 		 */
+		if (con->in_reply.tag == CEPH_MSGR_TAG_CHALLENGE_AUTHORIZER) {
+			ret = con->ops->add_authorizer_challenge(
+				    con, con->auth->authorizer_reply_buf,
+				    le32_to_cpu(con->in_reply.authorizer_len));
+			if (ret < 0)
+				return ret;
+
+			con_out_kvec_reset(con);
+			__prepare_write_connect(con);
+			prepare_read_connect(con);
+			return 0;
+		}
+
 		ret = con->ops->verify_authorizer_reply(con);
 		if (ret < 0) {
 			con->error_msg = "bad authorize reply";
@@ -2555,7 +2578,7 @@ static int read_keepalive_ack(struct ceph_connection *con)
 	int ret = read_partial(con, size, size, &ceph_ts);
 	if (ret <= 0)
 		return ret;
-	ceph_decode_timespec(&con->last_keepalive_ack, &ceph_ts);
+	ceph_decode_timespec64(&con->last_keepalive_ack, &ceph_ts);
 	prepare_read_tag(con);
 	return 1;
 }
@@ -3223,12 +3246,12 @@ bool ceph_con_keepalive_expired(struct ceph_connection *con,
 {
 	if (interval > 0 &&
 	    (con->peer_features & CEPH_FEATURE_MSGR_KEEPALIVE2)) {
-		struct timespec now;
-		struct timespec ts;
-		ktime_get_real_ts(&now);
-		jiffies_to_timespec(interval, &ts);
-		ts = timespec_add(con->last_keepalive_ack, ts);
-		return timespec_compare(&now, &ts) >= 0;
+		struct timespec64 now;
+		struct timespec64 ts;
+		ktime_get_real_ts64(&now);
+		jiffies_to_timespec64(interval, &ts);
+		ts = timespec64_add(con->last_keepalive_ack, ts);
+		return timespec64_compare(&now, &ts) >= 0;
 	}
 	return false;
 }

net/ceph/mon_client.c

@@ -1249,7 +1249,7 @@ static void dispatch(struct ceph_connection *con, struct ceph_msg *msg)
 		if (monc->client->extra_mon_dispatch &&
 		    monc->client->extra_mon_dispatch(monc->client, msg) == 0)
 			break;
-			
+
 		pr_err("received unknown message type %d %s\n", type,
 		       ceph_msg_type_name(type));
 	}

net/ceph/osd_client.c

@@ -1978,7 +1978,7 @@ static void encode_request_partial(struct ceph_osd_request *req,
 	p += sizeof(struct ceph_blkin_trace_info);
 
 	ceph_encode_32(&p, 0); /* client_inc, always 0 */
-	ceph_encode_timespec(p, &req->r_mtime);
+	ceph_encode_timespec64(p, &req->r_mtime);
 	p += sizeof(struct ceph_timespec);
 
 	encode_oloc(&p, end, &req->r_t.target_oloc);
@@ -4512,7 +4512,7 @@ ceph_osdc_watch(struct ceph_osd_client *osdc,
 	ceph_oid_copy(&lreq->t.base_oid, oid);
 	ceph_oloc_copy(&lreq->t.base_oloc, oloc);
 	lreq->t.flags = CEPH_OSD_FLAG_WRITE;
-	ktime_get_real_ts(&lreq->mtime);
+	ktime_get_real_ts64(&lreq->mtime);
 
 	lreq->reg_req = alloc_linger_request(lreq);
 	if (!lreq->reg_req) {
@@ -4570,7 +4570,7 @@ int ceph_osdc_unwatch(struct ceph_osd_client *osdc,
 	ceph_oid_copy(&req->r_base_oid, &lreq->t.base_oid);
 	ceph_oloc_copy(&req->r_base_oloc, &lreq->t.base_oloc);
 	req->r_flags = CEPH_OSD_FLAG_WRITE;
-	ktime_get_real_ts(&req->r_mtime);
+	ktime_get_real_ts64(&req->r_mtime);
 	osd_req_op_watch_init(req, 0, lreq->linger_id,
 			      CEPH_OSD_WATCH_OP_UNWATCH);
 
@@ -4591,7 +4591,7 @@ EXPORT_SYMBOL(ceph_osdc_unwatch);
 
 static int osd_req_op_notify_ack_init(struct ceph_osd_request *req, int which,
 				      u64 notify_id, u64 cookie, void *payload,
-				      size_t payload_len)
+				      u32 payload_len)
 {
 	struct ceph_osd_req_op *op;
 	struct ceph_pagelist *pl;
@@ -4628,7 +4628,7 @@ int ceph_osdc_notify_ack(struct ceph_osd_client *osdc,
 			 u64 notify_id,
 			 u64 cookie,
 			 void *payload,
-			 size_t payload_len)
+			 u32 payload_len)
 {
 	struct ceph_osd_request *req;
 	int ret;
@@ -4661,7 +4661,7 @@ EXPORT_SYMBOL(ceph_osdc_notify_ack);
 
 static int osd_req_op_notify_init(struct ceph_osd_request *req, int which,
 				  u64 cookie, u32 prot_ver, u32 timeout,
-				  void *payload, size_t payload_len)
+				  void *payload, u32 payload_len)
 {
 	struct ceph_osd_req_op *op;
 	struct ceph_pagelist *pl;
@@ -4701,7 +4701,7 @@ int ceph_osdc_notify(struct ceph_osd_client *osdc,
 		     struct ceph_object_id *oid,
 		     struct ceph_object_locator *oloc,
 		     void *payload,
-		     size_t payload_len,
+		     u32 payload_len,
 		     u32 timeout,
 		     struct page ***preply_pages,
 		     size_t *preply_len)
@@ -5136,7 +5136,7 @@ int ceph_osdc_writepages(struct ceph_osd_client *osdc, struct ceph_vino vino,
 			 struct ceph_snap_context *snapc,
 			 u64 off, u64 len,
 			 u32 truncate_seq, u64 truncate_size,
-			 struct timespec *mtime,
+			 struct timespec64 *mtime,
 			 struct page **pages, int num_pages)
 {
 	struct ceph_osd_request *req;
@@ -5393,6 +5393,16 @@ static struct ceph_auth_handshake *get_authorizer(struct ceph_connection *con,
 	return auth;
 }
 
+static int add_authorizer_challenge(struct ceph_connection *con,
+				    void *challenge_buf, int challenge_buf_len)
+{
+	struct ceph_osd *o = con->private;
+	struct ceph_osd_client *osdc = o->o_osdc;
+	struct ceph_auth_client *ac = osdc->client->monc.auth;
+
+	return ceph_auth_add_authorizer_challenge(ac, o->o_auth.authorizer,
+					    challenge_buf, challenge_buf_len);
+}
 
 static int verify_authorizer_reply(struct ceph_connection *con)
 {
@@ -5442,6 +5452,7 @@ static const struct ceph_connection_operations osd_con_ops = {
 	.put = put_osd_con,
 	.dispatch = dispatch,
 	.get_authorizer = get_authorizer,
+	.add_authorizer_challenge = add_authorizer_challenge,
 	.verify_authorizer_reply = verify_authorizer_reply,
 	.invalidate_authorizer = invalidate_authorizer,
 	.alloc_msg = alloc_msg,

net/ceph/pagevec.c

@@ -197,4 +197,3 @@ void ceph_zero_page_vector_range(int off, int len, struct page **pages)
 	}
 }
 EXPORT_SYMBOL(ceph_zero_page_vector_range);
-