1 files changed, 102 insertions, 83 deletions

diff --git a/fs/nfs/pnfs_nfs.c b/fs/nfs/pnfs_nfs.c
index a0112ad4937a..9976cc16b689 100644
--- a/fs/nfs/pnfs_nfs.c
+++ b/fs/nfs/pnfs_nfs.c
@@ -16,6 +16,8 @@
 #include "nfs4session.h"
 #include "internal.h"
 #include "pnfs.h"
+#include "netns.h"
+#include "nfs4trace.h"
 
 #define NFSDBG_FACILITY		NFSDBG_PNFS
 
@@ -351,53 +353,6 @@ void pnfs_generic_recover_commit_reqs(struct list_head *dst,
 }
 EXPORT_SYMBOL_GPL(pnfs_generic_recover_commit_reqs);
 
-static struct nfs_page *
-pnfs_bucket_search_commit_reqs(struct pnfs_commit_bucket *buckets,
-			       unsigned int nbuckets, struct folio *folio)
-{
-	struct nfs_page *req;
-	struct pnfs_commit_bucket *b;
-	unsigned int i;
-
-	/* Linearly search the commit lists for each bucket until a matching
-	 * request is found */
-	for (i = 0, b = buckets; i < nbuckets; i++, b++) {
-		list_for_each_entry(req, &b->written, wb_list) {
-			if (nfs_page_to_folio(req) == folio)
-				return req->wb_head;
-		}
-		list_for_each_entry(req, &b->committing, wb_list) {
-			if (nfs_page_to_folio(req) == folio)
-				return req->wb_head;
-		}
-	}
-	return NULL;
-}
-
-/* pnfs_generic_search_commit_reqs - Search lists in @cinfo for the head request
- *				   for @folio
- * @cinfo - commit info for current inode
- * @folio - page to search for matching head request
- *
- * Return: the head request if one is found, otherwise %NULL.
- */
-struct nfs_page *pnfs_generic_search_commit_reqs(struct nfs_commit_info *cinfo,
-						 struct folio *folio)
-{
-	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
-	struct pnfs_commit_array *array;
-	struct nfs_page *req;
-
-	list_for_each_entry(array, &fl_cinfo->commits, cinfo_list) {
-		req = pnfs_bucket_search_commit_reqs(array->buckets,
-						     array->nbuckets, folio);
-		if (req)
-			return req;
-	}
-	return NULL;
-}
-EXPORT_SYMBOL_GPL(pnfs_generic_search_commit_reqs);
-
 static struct pnfs_layout_segment *
 pnfs_bucket_get_committing(struct list_head *head,
 			   struct pnfs_commit_bucket *bucket,
@@ -537,7 +492,7 @@ pnfs_generic_commit_pagelist(struct inode *inode, struct list_head *mds_pages,
 			nfs_initiate_commit(NFS_CLIENT(inode), data,
 					    NFS_PROTO(data->inode),
 					    data->mds_ops, how,
-					    RPC_TASK_CRED_NOREF);
+					    RPC_TASK_CRED_NOREF, NULL);
 		} else {
 			nfs_init_commit(data, NULL, data->lseg, cinfo);
 			initiate_commit(data, how);
@@ -551,14 +506,14 @@ EXPORT_SYMBOL_GPL(pnfs_generic_commit_pagelist);
 /*
  * Data server cache
  *
- * Data servers can be mapped to different device ids.
- * nfs4_pnfs_ds reference counting
+ * Data servers can be mapped to different device ids, but should
+ * never be shared between net namespaces.
+ *
+ * nfs4_pnfs_ds reference counting:
  *   - set to 1 on allocation
  *   - incremented when a device id maps a data server already in the cache.
  *   - decremented when deviceid is removed from the cache.
  */
-static DEFINE_SPINLOCK(nfs4_ds_cache_lock);
-static LIST_HEAD(nfs4_data_server_cache);
 
 /* Debug routines */
 static void
@@ -651,11 +606,11 @@ _same_data_server_addrs_locked(const struct list_head *dsaddrs1,
  * Lookup DS by addresses.  nfs4_ds_cache_lock is held
  */
 static struct nfs4_pnfs_ds *
-_data_server_lookup_locked(const struct list_head *dsaddrs)
+_data_server_lookup_locked(const struct nfs_net *nn, const struct list_head *dsaddrs)
 {
 	struct nfs4_pnfs_ds *ds;
 
-	list_for_each_entry(ds, &nfs4_data_server_cache, ds_node)
+	list_for_each_entry(ds, &nn->nfs4_data_server_cache, ds_node)
 		if (_same_data_server_addrs_locked(&ds->ds_addrs, dsaddrs))
 			return ds;
 	return NULL;
@@ -700,10 +655,11 @@ static void destroy_ds(struct nfs4_pnfs_ds *ds)
 
 void nfs4_pnfs_ds_put(struct nfs4_pnfs_ds *ds)
 {
-	if (refcount_dec_and_lock(&ds->ds_count,
-				&nfs4_ds_cache_lock)) {
+	struct nfs_net *nn = net_generic(ds->ds_net, nfs_net_id);
+
+	if (refcount_dec_and_lock(&ds->ds_count, &nn->nfs4_data_server_lock)) {
 		list_del_init(&ds->ds_node);
-		spin_unlock(&nfs4_ds_cache_lock);
+		spin_unlock(&nn->nfs4_data_server_lock);
 		destroy_ds(ds);
 	}
 }
@@ -763,8 +719,9 @@ out_err:
  * uncached and return cached struct nfs4_pnfs_ds.
  */
 struct nfs4_pnfs_ds *
-nfs4_pnfs_ds_add(struct list_head *dsaddrs, gfp_t gfp_flags)
+nfs4_pnfs_ds_add(const struct net *net, struct list_head *dsaddrs, gfp_t gfp_flags)
 {
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
 	struct nfs4_pnfs_ds *tmp_ds, *ds = NULL;
 	char *remotestr;
 
@@ -780,16 +737,17 @@ nfs4_pnfs_ds_add(struct list_head *dsaddrs, gfp_t gfp_flags)
 	/* this is only used for debugging, so it's ok if its NULL */
 	remotestr = nfs4_pnfs_remotestr(dsaddrs, gfp_flags);
 
-	spin_lock(&nfs4_ds_cache_lock);
-	tmp_ds = _data_server_lookup_locked(dsaddrs);
+	spin_lock(&nn->nfs4_data_server_lock);
+	tmp_ds = _data_server_lookup_locked(nn, dsaddrs);
 	if (tmp_ds == NULL) {
 		INIT_LIST_HEAD(&ds->ds_addrs);
 		list_splice_init(dsaddrs, &ds->ds_addrs);
 		ds->ds_remotestr = remotestr;
 		refcount_set(&ds->ds_count, 1);
 		INIT_LIST_HEAD(&ds->ds_node);
+		ds->ds_net = net;
 		ds->ds_clp = NULL;
-		list_add(&ds->ds_node, &nfs4_data_server_cache);
+		list_add(&ds->ds_node, &nn->nfs4_data_server_cache);
 		dprintk("%s add new data server %s\n", __func__,
 			ds->ds_remotestr);
 	} else {
@@ -801,7 +759,7 @@ nfs4_pnfs_ds_add(struct list_head *dsaddrs, gfp_t gfp_flags)
 			refcount_read(&tmp_ds->ds_count));
 		ds = tmp_ds;
 	}
-	spin_unlock(&nfs4_ds_cache_lock);
+	spin_unlock(&nn->nfs4_data_server_lock);
 out:
 	return ds;
 }
@@ -851,7 +809,11 @@ static int _nfs4_pnfs_v3_ds_connect(struct nfs_server *mds_srv,
 				 unsigned int retrans)
 {
 	struct nfs_client *clp = ERR_PTR(-EIO);
+	struct nfs_client *mds_clp = mds_srv->nfs_client;
+	enum xprtsec_policies xprtsec_policy = mds_clp->cl_xprtsec.policy;
 	struct nfs4_pnfs_ds_addr *da;
+	unsigned long connect_timeout = timeo * (retrans + 1) * HZ / 10;
+	int ds_proto;
 	int status = 0;
 
 	dprintk("--> %s DS %s\n", __func__, ds->ds_remotestr);
@@ -870,21 +832,33 @@ static int _nfs4_pnfs_v3_ds_connect(struct nfs_server *mds_srv,
 				.dstaddr = (struct sockaddr *)&da->da_addr,
 				.addrlen = da->da_addrlen,
 				.servername = clp->cl_hostname,
+				.connect_timeout = connect_timeout,
+				.reconnect_timeout = connect_timeout,
+				.xprtsec = clp->cl_xprtsec,
 			};
 
-			if (da->da_transport != clp->cl_proto)
+			if (xprt_args.ident == XPRT_TRANSPORT_TCP &&
+			    clp->cl_proto == XPRT_TRANSPORT_TCP_TLS)
+				xprt_args.ident = XPRT_TRANSPORT_TCP_TLS;
+
+			if (xprt_args.ident != clp->cl_proto)
 				continue;
-			if (da->da_addr.ss_family != clp->cl_addr.ss_family)
+			if (xprt_args.dstaddr->sa_family !=
+			    clp->cl_addr.ss_family)
 				continue;
 			/* Add this address as an alias */
 			rpc_clnt_add_xprt(clp->cl_rpcclient, &xprt_args,
-					rpc_clnt_test_and_add_xprt, NULL);
+					  rpc_clnt_test_and_add_xprt, NULL);
 			continue;
 		}
-		clp = get_v3_ds_connect(mds_srv,
-				&da->da_addr,
-				da->da_addrlen, da->da_transport,
-				timeo, retrans);
+
+		ds_proto = da->da_transport;
+		if (ds_proto == XPRT_TRANSPORT_TCP &&
+		    xprtsec_policy != RPC_XPRTSEC_NONE)
+			ds_proto = XPRT_TRANSPORT_TCP_TLS;
+
+		clp = get_v3_ds_connect(mds_srv, &da->da_addr, da->da_addrlen,
+					ds_proto, timeo, retrans);
 		if (IS_ERR(clp))
 			continue;
 		clp->cl_rpcclient->cl_softerr = 0;
@@ -910,12 +884,17 @@ static int _nfs4_pnfs_v4_ds_connect(struct nfs_server *mds_srv,
 				 u32 minor_version)
 {
 	struct nfs_client *clp = ERR_PTR(-EIO);
+	struct nfs_client *mds_clp = mds_srv->nfs_client;
+	enum xprtsec_policies xprtsec_policy = mds_clp->cl_xprtsec.policy;
 	struct nfs4_pnfs_ds_addr *da;
+	int ds_proto;
 	int status = 0;
 
 	dprintk("--> %s DS %s\n", __func__, ds->ds_remotestr);
 
 	list_for_each_entry(da, &ds->ds_addrs, da_node) {
+		char servername[48];
+
 		dprintk("%s: DS %s: trying address %s\n",
 			__func__, ds->ds_remotestr, da->da_remotestr);
 
@@ -926,6 +905,7 @@ static int _nfs4_pnfs_v4_ds_connect(struct nfs_server *mds_srv,
 				.dstaddr = (struct sockaddr *)&da->da_addr,
 				.addrlen = da->da_addrlen,
 				.servername = clp->cl_hostname,
+				.xprtsec = clp->cl_xprtsec,
 			};
 			struct nfs4_add_xprt_data xprtdata = {
 				.clp = clp,
@@ -935,26 +915,63 @@ static int _nfs4_pnfs_v4_ds_connect(struct nfs_server *mds_srv,
 				.data = &xprtdata,
 			};
 
-			if (da->da_transport != clp->cl_proto)
+			if (xprt_args.ident == XPRT_TRANSPORT_TCP &&
+			    clp->cl_proto == XPRT_TRANSPORT_TCP_TLS) {
+				struct sockaddr *addr =
+					(struct sockaddr *)&da->da_addr;
+				struct sockaddr_in *sin =
+					(struct sockaddr_in *)&da->da_addr;
+				struct sockaddr_in6 *sin6 =
+					(struct sockaddr_in6 *)&da->da_addr;
+
+				/* for NFS with TLS we need to supply a correct
+				 * servername of the trunked transport, not the
+				 * servername of the main transport stored in
+				 * clp->cl_hostname. And set the protocol to
+				 * indicate to use TLS
+				 */
+				servername[0] = '\0';
+				switch(addr->sa_family) {
+				case AF_INET:
+					snprintf(servername, sizeof(servername),
+						"%pI4", &sin->sin_addr.s_addr);
+					break;
+				case AF_INET6:
+					snprintf(servername, sizeof(servername),
+						"%pI6", &sin6->sin6_addr);
+					break;
+				default:
+					/* do not consider this address */
+					continue;
+				}
+				xprt_args.ident = XPRT_TRANSPORT_TCP_TLS;
+				xprt_args.servername = servername;
+			}
+			if (xprt_args.ident != clp->cl_proto)
 				continue;
-			if (da->da_addr.ss_family != clp->cl_addr.ss_family)
+			if (xprt_args.dstaddr->sa_family !=
+			    clp->cl_addr.ss_family)
 				continue;
+
 			/**
 			* Test this address for session trunking and
 			* add as an alias
 			*/
-			xprtdata.cred = nfs4_get_clid_cred(clp),
+			xprtdata.cred = nfs4_get_clid_cred(clp);
 			rpc_clnt_add_xprt(clp->cl_rpcclient, &xprt_args,
 					  rpc_clnt_setup_test_and_add_xprt,
 					  &rpcdata);
 			if (xprtdata.cred)
 				put_cred(xprtdata.cred);
 		} else {
-			clp = nfs4_set_ds_client(mds_srv,
-						&da->da_addr,
-						da->da_addrlen,
-						da->da_transport, timeo,
-						retrans, minor_version);
+			ds_proto = da->da_transport;
+			if (ds_proto == XPRT_TRANSPORT_TCP &&
+			    xprtsec_policy != RPC_XPRTSEC_NONE)
+				ds_proto = XPRT_TRANSPORT_TCP_TLS;
+
+			clp = nfs4_set_ds_client(mds_srv, &da->da_addr,
+						 da->da_addrlen, ds_proto,
+						 timeo, retrans, minor_version);
 			if (IS_ERR(clp))
 				continue;
 
@@ -965,7 +982,6 @@ static int _nfs4_pnfs_v4_ds_connect(struct nfs_server *mds_srv,
 				clp = ERR_PTR(-EIO);
 				continue;
 			}
-
 		}
 	}
 
@@ -996,8 +1012,10 @@ int nfs4_pnfs_ds_connect(struct nfs_server *mds_srv, struct nfs4_pnfs_ds *ds,
 		err = nfs4_wait_ds_connect(ds);
 		if (err || ds->ds_clp)
 			goto out;
-		if (nfs4_test_deviceid_unavailable(devid))
-			return -ENODEV;
+		if (nfs4_test_deviceid_unavailable(devid)) {
+			err = -ENODEV;
+			goto out;
+		}
 	} while (test_and_set_bit(NFS4DS_CONNECTING, &ds->ds_state) != 0);
 
 	if (ds->ds_clp)
@@ -1027,11 +1045,12 @@ out:
 		if (!ds->ds_clp || !nfs_client_init_is_complete(ds->ds_clp)) {
 			WARN_ON_ONCE(ds->ds_clp ||
 				!nfs4_test_deviceid_unavailable(devid));
-			return -EINVAL;
-		}
-		err = nfs_client_init_status(ds->ds_clp);
+			err = -EINVAL;
+		} else
+			err = nfs_client_init_status(ds->ds_clp);
 	}
 
+	trace_pnfs_ds_connect(ds->ds_remotestr, err);
 	return err;
 }
 EXPORT_SYMBOL_GPL(nfs4_pnfs_ds_connect);