5 files changed, 394 insertions, 0 deletions

diff --git a/fs/nfs_common/Makefile b/fs/nfs_common/Makefile
index 119c75ab9fd0..a5e54809701e 100644
--- a/fs/nfs_common/Makefile
+++ b/fs/nfs_common/Makefile
@@ -6,5 +6,10 @@
 obj-$(CONFIG_NFS_ACL_SUPPORT) += nfs_acl.o
 nfs_acl-objs := nfsacl.o
 
+obj-$(CONFIG_NFS_COMMON_LOCALIO_SUPPORT) += nfs_localio.o
+nfs_localio-objs := nfslocalio.o
+
 obj-$(CONFIG_GRACE_PERIOD) += grace.o
 obj-$(CONFIG_NFS_V4_2_SSC_HELPER) += nfs_ssc.o
+
+obj-$(CONFIG_NFS_COMMON) += common.o
diff --git a/fs/nfs_common/common.c b/fs/nfs_common/common.c
new file mode 100644
index 000000000000..af09aed09fd2
--- /dev/null
+++ b/fs/nfs_common/common.c
@@ -0,0 +1,201 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/module.h>
+#include <linux/nfs_common.h>
+#include <linux/nfs4.h>
+
+/*
+ * We need to translate between nfs status return values and
+ * the local errno values which may not be the same.
+ */
+static const struct {
+	int stat;
+	int errno;
+} nfs_errtbl[] = {
+	{ NFS_OK,		0		},
+	{ NFSERR_PERM,		-EPERM		},
+	{ NFSERR_NOENT,		-ENOENT		},
+	{ NFSERR_IO,		-EIO		},
+	{ NFSERR_NXIO,		-ENXIO		},
+/*	{ NFSERR_EAGAIN,	-EAGAIN		}, */
+	{ NFSERR_ACCES,		-EACCES		},
+	{ NFSERR_EXIST,		-EEXIST		},
+	{ NFSERR_XDEV,		-EXDEV		},
+	{ NFSERR_NODEV,		-ENODEV		},
+	{ NFSERR_NOTDIR,	-ENOTDIR	},
+	{ NFSERR_ISDIR,		-EISDIR		},
+	{ NFSERR_INVAL,		-EINVAL		},
+	{ NFSERR_FBIG,		-EFBIG		},
+	{ NFSERR_NOSPC,		-ENOSPC		},
+	{ NFSERR_ROFS,		-EROFS		},
+	{ NFSERR_MLINK,		-EMLINK		},
+	{ NFSERR_NAMETOOLONG,	-ENAMETOOLONG	},
+	{ NFSERR_NOTEMPTY,	-ENOTEMPTY	},
+	{ NFSERR_DQUOT,		-EDQUOT		},
+	{ NFSERR_STALE,		-ESTALE		},
+	{ NFSERR_REMOTE,	-EREMOTE	},
+#ifdef EWFLUSH
+	{ NFSERR_WFLUSH,	-EWFLUSH	},
+#endif
+	{ NFSERR_BADHANDLE,	-EBADHANDLE	},
+	{ NFSERR_NOT_SYNC,	-ENOTSYNC	},
+	{ NFSERR_BAD_COOKIE,	-EBADCOOKIE	},
+	{ NFSERR_NOTSUPP,	-ENOTSUPP	},
+	{ NFSERR_TOOSMALL,	-ETOOSMALL	},
+	{ NFSERR_SERVERFAULT,	-EREMOTEIO	},
+	{ NFSERR_BADTYPE,	-EBADTYPE	},
+	{ NFSERR_JUKEBOX,	-EJUKEBOX	},
+};
+
+/**
+ * nfs_stat_to_errno - convert an NFS status code to a local errno
+ * @status: NFS status code to convert
+ *
+ * Returns a local errno value, or -EIO if the NFS status code is
+ * not recognized.  This function is used jointly by NFSv2 and NFSv3.
+ */
+int nfs_stat_to_errno(enum nfs_stat status)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(nfs_errtbl); i++) {
+		if (nfs_errtbl[i].stat == (int)status)
+			return nfs_errtbl[i].errno;
+	}
+	return -EIO;
+}
+EXPORT_SYMBOL_GPL(nfs_stat_to_errno);
+
+/*
+ * We need to translate between nfs v4 status return values and
+ * the local errno values which may not be the same.
+ *
+ * nfs4_errtbl_common[] is used before more specialized mappings
+ * available in nfs4_errtbl[] or nfs4_errtbl_localio[].
+ */
+static const struct {
+	int stat;
+	int errno;
+} nfs4_errtbl_common[] = {
+	{ NFS4_OK,		0		},
+	{ NFS4ERR_PERM,		-EPERM		},
+	{ NFS4ERR_NOENT,	-ENOENT		},
+	{ NFS4ERR_IO,		-EIO		},
+	{ NFS4ERR_NXIO,		-ENXIO		},
+	{ NFS4ERR_ACCESS,	-EACCES		},
+	{ NFS4ERR_EXIST,	-EEXIST		},
+	{ NFS4ERR_XDEV,		-EXDEV		},
+	{ NFS4ERR_NOTDIR,	-ENOTDIR	},
+	{ NFS4ERR_ISDIR,	-EISDIR		},
+	{ NFS4ERR_INVAL,	-EINVAL		},
+	{ NFS4ERR_FBIG,		-EFBIG		},
+	{ NFS4ERR_NOSPC,	-ENOSPC		},
+	{ NFS4ERR_ROFS,		-EROFS		},
+	{ NFS4ERR_MLINK,	-EMLINK		},
+	{ NFS4ERR_NAMETOOLONG,	-ENAMETOOLONG	},
+	{ NFS4ERR_NOTEMPTY,	-ENOTEMPTY	},
+	{ NFS4ERR_DQUOT,	-EDQUOT		},
+	{ NFS4ERR_STALE,	-ESTALE		},
+	{ NFS4ERR_BADHANDLE,	-EBADHANDLE	},
+	{ NFS4ERR_BAD_COOKIE,	-EBADCOOKIE	},
+	{ NFS4ERR_NOTSUPP,	-ENOTSUPP	},
+	{ NFS4ERR_TOOSMALL,	-ETOOSMALL	},
+	{ NFS4ERR_BADTYPE,	-EBADTYPE	},
+	{ NFS4ERR_SYMLINK,	-ELOOP		},
+	{ NFS4ERR_DEADLOCK,	-EDEADLK	},
+};
+
+static const struct {
+	int stat;
+	int errno;
+} nfs4_errtbl[] = {
+	{ NFS4ERR_SERVERFAULT,	-EREMOTEIO	},
+	{ NFS4ERR_LOCKED,	-EAGAIN		},
+	{ NFS4ERR_OP_ILLEGAL,	-EOPNOTSUPP	},
+	{ NFS4ERR_NOXATTR,	-ENODATA	},
+	{ NFS4ERR_XATTR2BIG,	-E2BIG		},
+};
+
+/*
+ * Convert an NFS error code to a local one.
+ * This one is used by NFSv4.
+ */
+int nfs4_stat_to_errno(int stat)
+{
+	int i;
+
+	/* First check nfs4_errtbl_common */
+	for (i = 0; i < ARRAY_SIZE(nfs4_errtbl_common); i++) {
+		if (nfs4_errtbl_common[i].stat == stat)
+			return nfs4_errtbl_common[i].errno;
+	}
+	/* Then check nfs4_errtbl */
+	for (i = 0; i < ARRAY_SIZE(nfs4_errtbl); i++) {
+		if (nfs4_errtbl[i].stat == stat)
+			return nfs4_errtbl[i].errno;
+	}
+	if (stat <= 10000 || stat > 10100) {
+		/* The server is looney tunes. */
+		return -EREMOTEIO;
+	}
+	/* If we cannot translate the error, the recovery routines should
+	 * handle it.
+	 * Note: remaining NFSv4 error codes have values > 10000, so should
+	 * not conflict with native Linux error codes.
+	 */
+	return -stat;
+}
+EXPORT_SYMBOL_GPL(nfs4_stat_to_errno);
+
+/*
+ * This table is useful for conversion from local errno to NFS error.
+ * It provides more logically correct mappings for use with LOCALIO
+ * (which is focused on converting from errno to NFS status).
+ */
+static const struct {
+	int stat;
+	int errno;
+} nfs4_errtbl_localio[] = {
+	/* Map errors differently than nfs4_errtbl */
+	{ NFS4ERR_IO,		-EREMOTEIO	},
+	{ NFS4ERR_DELAY,	-EAGAIN		},
+	{ NFS4ERR_FBIG,		-E2BIG		},
+	/* Map errors not handled by nfs4_errtbl */
+	{ NFS4ERR_STALE,	-EBADF		},
+	{ NFS4ERR_STALE,	-EOPENSTALE	},
+	{ NFS4ERR_DELAY,	-ETIMEDOUT	},
+	{ NFS4ERR_DELAY,	-ERESTARTSYS	},
+	{ NFS4ERR_DELAY,	-ENOMEM		},
+	{ NFS4ERR_IO,		-ETXTBSY	},
+	{ NFS4ERR_IO,		-EBUSY		},
+	{ NFS4ERR_SERVERFAULT,	-ESERVERFAULT	},
+	{ NFS4ERR_SERVERFAULT,	-ENFILE		},
+	{ NFS4ERR_IO,		-EUCLEAN	},
+	{ NFS4ERR_PERM,		-ENOKEY		},
+};
+
+/*
+ * Convert an errno to an NFS error code for LOCALIO.
+ */
+__u32 nfs_localio_errno_to_nfs4_stat(int errno)
+{
+	int i;
+
+	/* First check nfs4_errtbl_common */
+	for (i = 0; i < ARRAY_SIZE(nfs4_errtbl_common); i++) {
+		if (nfs4_errtbl_common[i].errno == errno)
+			return nfs4_errtbl_common[i].stat;
+	}
+	/* Then check nfs4_errtbl_localio */
+	for (i = 0; i < ARRAY_SIZE(nfs4_errtbl_localio); i++) {
+		if (nfs4_errtbl_localio[i].errno == errno)
+			return nfs4_errtbl_localio[i].stat;
+	}
+	/* If we cannot translate the error, the recovery routines should
+	 * handle it.
+	 * Note: remaining NFSv4 error codes have values > 10000, so should
+	 * not conflict with native Linux error codes.
+	 */
+	return NFS4ERR_SERVERFAULT;
+}
+EXPORT_SYMBOL_GPL(nfs_localio_errno_to_nfs4_stat);
diff --git a/fs/nfs_common/grace.c b/fs/nfs_common/grace.c
index 1479583fbb62..27cd0d13143b 100644
--- a/fs/nfs_common/grace.c
+++ b/fs/nfs_common/grace.c
@@ -139,6 +139,7 @@ exit_grace(void)
 }
 
 MODULE_AUTHOR("Jeff Layton <jlayton@primarydata.com>");
+MODULE_DESCRIPTION("NFS client and server infrastructure");
 MODULE_LICENSE("GPL");
 module_init(init_grace)
 module_exit(exit_grace)
diff --git a/fs/nfs_common/nfsacl.c b/fs/nfs_common/nfsacl.c
index 5a5bd85d08f8..ea382b75b26c 100644
--- a/fs/nfs_common/nfsacl.c
+++ b/fs/nfs_common/nfsacl.c
@@ -29,6 +29,7 @@
 #include <linux/nfs3.h>
 #include <linux/sort.h>
 
+MODULE_DESCRIPTION("NFS ACL support");
 MODULE_LICENSE("GPL");
 
 struct nfsacl_encode_desc {
diff --git a/fs/nfs_common/nfslocalio.c b/fs/nfs_common/nfslocalio.c
new file mode 100644
index 000000000000..a74ec08f6c96
--- /dev/null
+++ b/fs/nfs_common/nfslocalio.c
@@ -0,0 +1,186 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2024 Mike Snitzer <snitzer@hammerspace.com>
+ * Copyright (C) 2024 NeilBrown <neilb@suse.de>
+ */
+
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/nfslocalio.h>
+#include <net/netns/generic.h>
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("NFS localio protocol bypass support");
+
+static DEFINE_SPINLOCK(nfs_uuid_lock);
+
+/*
+ * Global list of nfs_uuid_t instances
+ * that is protected by nfs_uuid_lock.
+ */
+static LIST_HEAD(nfs_uuids);
+
+void nfs_uuid_init(nfs_uuid_t *nfs_uuid)
+{
+	nfs_uuid->net = NULL;
+	nfs_uuid->dom = NULL;
+	INIT_LIST_HEAD(&nfs_uuid->list);
+}
+EXPORT_SYMBOL_GPL(nfs_uuid_init);
+
+bool nfs_uuid_begin(nfs_uuid_t *nfs_uuid)
+{
+	spin_lock(&nfs_uuid_lock);
+	/* Is this nfs_uuid already in use? */
+	if (!list_empty(&nfs_uuid->list)) {
+		spin_unlock(&nfs_uuid_lock);
+		return false;
+	}
+	uuid_gen(&nfs_uuid->uuid);
+	list_add_tail(&nfs_uuid->list, &nfs_uuids);
+	spin_unlock(&nfs_uuid_lock);
+
+	return true;
+}
+EXPORT_SYMBOL_GPL(nfs_uuid_begin);
+
+void nfs_uuid_end(nfs_uuid_t *nfs_uuid)
+{
+	if (nfs_uuid->net == NULL) {
+		spin_lock(&nfs_uuid_lock);
+		if (nfs_uuid->net == NULL)
+			list_del_init(&nfs_uuid->list);
+		spin_unlock(&nfs_uuid_lock);
+	}
+}
+EXPORT_SYMBOL_GPL(nfs_uuid_end);
+
+static nfs_uuid_t * nfs_uuid_lookup_locked(const uuid_t *uuid)
+{
+	nfs_uuid_t *nfs_uuid;
+
+	list_for_each_entry(nfs_uuid, &nfs_uuids, list)
+		if (uuid_equal(&nfs_uuid->uuid, uuid))
+			return nfs_uuid;
+
+	return NULL;
+}
+
+static struct module *nfsd_mod;
+
+void nfs_uuid_is_local(const uuid_t *uuid, struct list_head *list,
+		       struct net *net, struct auth_domain *dom,
+		       struct module *mod)
+{
+	nfs_uuid_t *nfs_uuid;
+
+	spin_lock(&nfs_uuid_lock);
+	nfs_uuid = nfs_uuid_lookup_locked(uuid);
+	if (nfs_uuid) {
+		kref_get(&dom->ref);
+		nfs_uuid->dom = dom;
+		/*
+		 * We don't hold a ref on the net, but instead put
+		 * ourselves on a list so the net pointer can be
+		 * invalidated.
+		 */
+		list_move(&nfs_uuid->list, list);
+		rcu_assign_pointer(nfs_uuid->net, net);
+
+		__module_get(mod);
+		nfsd_mod = mod;
+	}
+	spin_unlock(&nfs_uuid_lock);
+}
+EXPORT_SYMBOL_GPL(nfs_uuid_is_local);
+
+static void nfs_uuid_put_locked(nfs_uuid_t *nfs_uuid)
+{
+	if (nfs_uuid->net) {
+		module_put(nfsd_mod);
+		nfs_uuid->net = NULL;
+	}
+	if (nfs_uuid->dom) {
+		auth_domain_put(nfs_uuid->dom);
+		nfs_uuid->dom = NULL;
+	}
+	list_del_init(&nfs_uuid->list);
+}
+
+void nfs_uuid_invalidate_clients(struct list_head *list)
+{
+	nfs_uuid_t *nfs_uuid, *tmp;
+
+	spin_lock(&nfs_uuid_lock);
+	list_for_each_entry_safe(nfs_uuid, tmp, list, list)
+		nfs_uuid_put_locked(nfs_uuid);
+	spin_unlock(&nfs_uuid_lock);
+}
+EXPORT_SYMBOL_GPL(nfs_uuid_invalidate_clients);
+
+void nfs_uuid_invalidate_one_client(nfs_uuid_t *nfs_uuid)
+{
+	if (nfs_uuid->net) {
+		spin_lock(&nfs_uuid_lock);
+		nfs_uuid_put_locked(nfs_uuid);
+		spin_unlock(&nfs_uuid_lock);
+	}
+}
+EXPORT_SYMBOL_GPL(nfs_uuid_invalidate_one_client);
+
+struct nfsd_file *nfs_open_local_fh(nfs_uuid_t *uuid,
+		   struct rpc_clnt *rpc_clnt, const struct cred *cred,
+		   const struct nfs_fh *nfs_fh, const fmode_t fmode)
+{
+	struct net *net;
+	struct nfsd_file *localio;
+
+	/*
+	 * Not running in nfsd context, so must safely get reference on nfsd_serv.
+	 * But the server may already be shutting down, if so disallow new localio.
+	 * uuid->net is NOT a counted reference, but rcu_read_lock() ensures that
+	 * if uuid->net is not NULL, then calling nfsd_serv_try_get() is safe
+	 * and if it succeeds we will have an implied reference to the net.
+	 *
+	 * Otherwise NFS may not have ref on NFSD and therefore cannot safely
+	 * make 'nfs_to' calls.
+	 */
+	rcu_read_lock();
+	net = rcu_dereference(uuid->net);
+	if (!net || !nfs_to->nfsd_serv_try_get(net)) {
+		rcu_read_unlock();
+		return ERR_PTR(-ENXIO);
+	}
+	rcu_read_unlock();
+	/* We have an implied reference to net thanks to nfsd_serv_try_get */
+	localio = nfs_to->nfsd_open_local_fh(net, uuid->dom, rpc_clnt,
+					     cred, nfs_fh, fmode);
+	if (IS_ERR(localio))
+		nfs_to_nfsd_net_put(net);
+
+	return localio;
+}
+EXPORT_SYMBOL_GPL(nfs_open_local_fh);
+
+/*
+ * The NFS LOCALIO code needs to call into NFSD using various symbols,
+ * but cannot be statically linked, because that will make the NFS
+ * module always depend on the NFSD module.
+ *
+ * 'nfs_to' provides NFS access to NFSD functions needed for LOCALIO,
+ * its lifetime is tightly coupled to the NFSD module and will always
+ * be available to NFS LOCALIO because any successful client<->server
+ * LOCALIO handshake results in a reference on the NFSD module (above),
+ * so NFS implicitly holds a reference to the NFSD module and its
+ * functions in the 'nfs_to' nfsd_localio_operations cannot disappear.
+ *
+ * If the last NFS client using LOCALIO disconnects (and its reference
+ * on NFSD dropped) then NFSD could be unloaded, resulting in 'nfs_to'
+ * functions being invalid pointers. But if NFSD isn't loaded then NFS
+ * will not be able to handshake with NFSD and will have no cause to
+ * try to call 'nfs_to' function pointers. If/when NFSD is reloaded it
+ * will reinitialize the 'nfs_to' function pointers and make LOCALIO
+ * possible.
+ */
+const struct nfsd_localio_operations *nfs_to;
+EXPORT_SYMBOL_GPL(nfs_to);