Merge tag 'nfs-for-4.3-1' of git://git.linux-nfs.org/projects/trondmy…

…/linux-nfs

Pull NFS client updates from Trond Myklebust:
 "Highlights include:

  Stable patches:
   - Fix atomicity of pNFS commit list updates
   - Fix NFSv4 handling of open(O_CREAT|O_EXCL|O_RDONLY)
   - nfs_set_pgio_error sometimes misses errors
   - Fix a thinko in xs_connect()
   - Fix borkage in _same_data_server_addrs_locked()
   - Fix a NULL pointer dereference of migration recovery ops for v4.2
     client
   - Don't let the ctime override attribute barriers.
   - Revert "NFSv4: Remove incorrect check in can_open_delegated()"
   - Ensure flexfiles pNFS driver updates the inode after write finishes
   - flexfiles must not pollute the attribute cache with attrbutes from
     the DS
   - Fix a protocol error in layoutreturn
   - Fix a protocol issue with NFSv4.1 CLOSE stateids

  Bugfixes + cleanups
   - pNFS blocks bugfixes from Christoph
   - Various cleanups from Anna
   - More fixes for delegation corner cases
   - Don't fsync twice for O_SYNC/IS_SYNC files
   - Fix pNFS and flexfiles layoutstats bugs
   - pnfs/flexfiles: avoid duplicate tracking of mirror data
   - pnfs: Fix layoutget/layoutreturn/return-on-close serialisation
     issues
   - pnfs/flexfiles: error handling retries a layoutget before fallback
     to MDS

  Features:
   - Full support for the OPEN NFS4_CREATE_EXCLUSIVE4_1 mode from
     Kinglong
   - More RDMA client transport improvements from Chuck
   - Removal of the deprecated ib_reg_phys_mr() and ib_rereg_phys_mr()
     verbs from the SUNRPC, Lustre and core infiniband tree.
   - Optimise away the close-to-open getattr if there is no cached data"

* tag 'nfs-for-4.3-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (108 commits)
  NFSv4: Respect the server imposed limit on how many changes we may cache
  NFSv4: Express delegation limit in units of pages
  Revert "NFS: Make close(2) asynchronous when closing NFS O_DIRECT files"
  NFS: Optimise away the close-to-open getattr if there is no cached data
  NFSv4.1/flexfiles: Clean up ff_layout_write_done_cb/ff_layout_commit_done_cb
  NFSv4.1/flexfiles: Mark the layout for return in ff_layout_io_track_ds_error()
  nfs: Remove unneeded checking of the return value from scnprintf
  nfs: Fix truncated client owner id without proto type
  NFSv4.1/flexfiles: Mark layout for return if the mirrors are invalid
  NFSv4.1/flexfiles: RW layouts are valid only if all mirrors are valid
  NFSv4.1/flexfiles: Fix incorrect usage of pnfs_generic_mark_devid_invalid()
  NFSv4.1/flexfiles: Fix freeing of mirrors
  NFSv4.1/pNFS: Don't request a minimal read layout beyond the end of file
  NFSv4.1/pnfs: Handle LAYOUTGET return values correctly
  NFSv4.1/pnfs: Don't ask for a read layout for an empty file.
  NFSv4.1: Fix a protocol issue with CLOSE stateids
  NFSv4.1/flexfiles: Don't mark the entire deviceid as bad for file errors
  SUNRPC: Prevent SYN+SYNACK+RST storms
  SUNRPC: xs_reset_transport must mark the connection as disconnected
  NFSv4.1/pnfs: Ensure layoutreturn reserves space for the opaque payload
  ...

Documentation/kernel-parameters.txt

@@ -2285,6 +2285,15 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 			The default parameter value of '0' causes the kernel
 			not to attempt recovery of lost locks.
 
+	nfs4.layoutstats_timer =
+			[NFSv4.2] Change the rate at which the kernel sends
+			layoutstats to the pNFS metadata server.
+
+			Setting this to value to 0 causes the kernel to use
+			whatever value is the default set by the layout
+			driver. A non-zero value sets the minimum interval
+			in seconds between layoutstats transmissions.
+
 	nfsd.nfs4_disable_idmapping=
 			[NFSv4] When set to the default of '1', the NFSv4
 			server will return only numeric uids and gids to

drivers/infiniband/core/verbs.c

@@ -1144,73 +1144,6 @@ struct ib_mr *ib_get_dma_mr(struct ib_pd *pd, int mr_access_flags)
 }
 EXPORT_SYMBOL(ib_get_dma_mr);
 
-struct ib_mr *ib_reg_phys_mr(struct ib_pd *pd,
-			     struct ib_phys_buf *phys_buf_array,
-			     int num_phys_buf,
-			     int mr_access_flags,
-			     u64 *iova_start)
-{
-	struct ib_mr *mr;
-	int err;
-
-	err = ib_check_mr_access(mr_access_flags);
-	if (err)
-		return ERR_PTR(err);
-
-	if (!pd->device->reg_phys_mr)
-		return ERR_PTR(-ENOSYS);
-
-	mr = pd->device->reg_phys_mr(pd, phys_buf_array, num_phys_buf,
-				     mr_access_flags, iova_start);
-
-	if (!IS_ERR(mr)) {
-		mr->device  = pd->device;
-		mr->pd      = pd;
-		mr->uobject = NULL;
-		atomic_inc(&pd->usecnt);
-		atomic_set(&mr->usecnt, 0);
-	}
-
-	return mr;
-}
-EXPORT_SYMBOL(ib_reg_phys_mr);
-
-int ib_rereg_phys_mr(struct ib_mr *mr,
-		     int mr_rereg_mask,
-		     struct ib_pd *pd,
-		     struct ib_phys_buf *phys_buf_array,
-		     int num_phys_buf,
-		     int mr_access_flags,
-		     u64 *iova_start)
-{
-	struct ib_pd *old_pd;
-	int ret;
-
-	ret = ib_check_mr_access(mr_access_flags);
-	if (ret)
-		return ret;
-
-	if (!mr->device->rereg_phys_mr)
-		return -ENOSYS;
-
-	if (atomic_read(&mr->usecnt))
-		return -EBUSY;
-
-	old_pd = mr->pd;
-
-	ret = mr->device->rereg_phys_mr(mr, mr_rereg_mask, pd,
-					phys_buf_array, num_phys_buf,
-					mr_access_flags, iova_start);
-
-	if (!ret && (mr_rereg_mask & IB_MR_REREG_PD)) {
-		atomic_dec(&old_pd->usecnt);
-		atomic_inc(&pd->usecnt);
-	}
-
-	return ret;
-}
-EXPORT_SYMBOL(ib_rereg_phys_mr);
-
 int ib_query_mr(struct ib_mr *mr, struct ib_mr_attr *mr_attr)
 {
 	return mr->device->query_mr ?

fs/nfs/blocklayout/blocklayout.h

@@ -46,13 +46,6 @@
 
 struct pnfs_block_dev;
 
-enum pnfs_block_volume_type {
-	PNFS_BLOCK_VOLUME_SIMPLE	= 0,
-	PNFS_BLOCK_VOLUME_SLICE		= 1,
-	PNFS_BLOCK_VOLUME_CONCAT	= 2,
-	PNFS_BLOCK_VOLUME_STRIPE	= 3,
-};
-
 #define PNFS_BLOCK_MAX_UUIDS	4
 #define PNFS_BLOCK_MAX_DEVICES	64
 
@@ -117,13 +110,6 @@ struct pnfs_block_dev {
 			struct pnfs_block_dev_map *map);
 };
 
-enum exstate4 {
-	PNFS_BLOCK_READWRITE_DATA	= 0,
-	PNFS_BLOCK_READ_DATA		= 1,
-	PNFS_BLOCK_INVALID_DATA		= 2, /* mapped, but data is invalid */
-	PNFS_BLOCK_NONE_DATA		= 3  /* unmapped, it's a hole */
-};
-
 /* sector_t fields are all in 512-byte sectors */
 struct pnfs_block_extent {
 	union {
@@ -134,15 +120,12 @@ struct pnfs_block_extent {
 	sector_t	be_f_offset;	/* the starting offset in the file */
 	sector_t	be_length;	/* the size of the extent */
 	sector_t	be_v_offset;	/* the starting offset in the volume */
-	enum exstate4	be_state;	/* the state of this extent */
+	enum pnfs_block_extent_state be_state;	/* the state of this extent */
 #define EXTENT_WRITTEN		1
 #define EXTENT_COMMITTING	2
 	unsigned int	be_tag;
 };
 
-/* on the wire size of the extent */
-#define BL_EXTENT_SIZE	(7 * sizeof(__be32) + NFS4_DEVICEID4_SIZE)
-
 struct pnfs_block_layout {
 	struct pnfs_layout_hdr	bl_layout;
 	struct rb_root		bl_ext_rw;

fs/nfs/blocklayout/dev.c

@@ -22,7 +22,7 @@ bl_free_device(struct pnfs_block_dev *dev)
 		kfree(dev->children);
 	} else {
 		if (dev->bdev)
-			blkdev_put(dev->bdev, FMODE_READ);
+			blkdev_put(dev->bdev, FMODE_READ | FMODE_WRITE);
 	}
 }
 
@@ -65,6 +65,11 @@ nfs4_block_decode_volume(struct xdr_stream *xdr, struct pnfs_block_volume *b)
 				return -EIO;
 			p = xdr_decode_hyper(p, &b->simple.sigs[i].offset);
 			b->simple.sigs[i].sig_len = be32_to_cpup(p++);
+			if (b->simple.sigs[i].sig_len > PNFS_BLOCK_UUID_LEN) {
+				pr_info("signature too long: %d\n",
+					b->simple.sigs[i].sig_len);
+				return -EIO;
+			}
 
 			p = xdr_inline_decode(xdr, b->simple.sigs[i].sig_len);
 			if (!p)
@@ -195,7 +200,7 @@ bl_parse_simple(struct nfs_server *server, struct pnfs_block_dev *d,
 	if (!dev)
 		return -EIO;
 
-	d->bdev = blkdev_get_by_dev(dev, FMODE_READ, NULL);
+	d->bdev = blkdev_get_by_dev(dev, FMODE_READ | FMODE_WRITE, NULL);
 	if (IS_ERR(d->bdev)) {
 		printk(KERN_WARNING "pNFS: failed to open device %d:%d (%ld)\n",
 			MAJOR(dev), MINOR(dev), PTR_ERR(d->bdev));

fs/nfs/blocklayout/extent_tree.c

@@ -462,6 +462,12 @@ ext_tree_mark_written(struct pnfs_block_layout *bl, sector_t start,
 	return err;
 }
 
+static size_t ext_tree_layoutupdate_size(size_t count)
+{
+	return sizeof(__be32) /* number of entries */ +
+		PNFS_BLOCK_EXTENT_SIZE * count;
+}
+
 static void ext_tree_free_commitdata(struct nfs4_layoutcommit_args *arg,
 		size_t buffer_size)
 {
@@ -489,7 +495,7 @@ static int ext_tree_encode_commit(struct pnfs_block_layout *bl, __be32 *p,
 			continue;
 
 		(*count)++;
-		if (*count * BL_EXTENT_SIZE > buffer_size) {
+		if (ext_tree_layoutupdate_size(*count) > buffer_size) {
 			/* keep counting.. */
 			ret = -ENOSPC;
 			continue;
@@ -530,7 +536,7 @@ ext_tree_prepare_commit(struct nfs4_layoutcommit_args *arg)
 	if (unlikely(ret)) {
 		ext_tree_free_commitdata(arg, buffer_size);
 
-		buffer_size = sizeof(__be32) + BL_EXTENT_SIZE * count;
+		buffer_size = ext_tree_layoutupdate_size(count);
 		count = 0;
 
 		arg->layoutupdate_pages =
@@ -549,17 +555,14 @@ ext_tree_prepare_commit(struct nfs4_layoutcommit_args *arg)
 	}
 
 	*start_p = cpu_to_be32(count);
-	arg->layoutupdate_len = sizeof(__be32) + BL_EXTENT_SIZE * count;
+	arg->layoutupdate_len = ext_tree_layoutupdate_size(count);
 
 	if (unlikely(arg->layoutupdate_pages != &arg->layoutupdate_page)) {
-		__be32 *p = start_p;
+		void *p = start_p, *end = p + arg->layoutupdate_len;
 		int i = 0;
 
-		for (p = start_p;
-		     p < start_p + arg->layoutupdate_len;
-		     p += PAGE_SIZE) {
+		for ( ; p < end; p += PAGE_SIZE)
 			arg->layoutupdate_pages[i++] = vmalloc_to_page(p);
-		}
 	}
 
 	dprintk("%s found %zu ranges\n", __func__, count);

fs/nfs/callback.c

@@ -162,10 +162,6 @@ nfs41_callback_up(struct svc_serv *serv)
 	spin_lock_init(&serv->sv_cb_lock);
 	init_waitqueue_head(&serv->sv_cb_waitq);
 	rqstp = svc_prepare_thread(serv, &serv->sv_pools[0], NUMA_NO_NODE);
-	if (IS_ERR(rqstp)) {
-		svc_xprt_put(serv->sv_bc_xprt);
-		serv->sv_bc_xprt = NULL;
-	}
 	dprintk("--> %s return %d\n", __func__, PTR_ERR_OR_ZERO(rqstp));
 	return rqstp;
 }

fs/nfs/callback_proc.c

@@ -40,8 +40,11 @@ __be32 nfs4_callback_getattr(struct cb_getattrargs *args,
 		rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR));
 
 	inode = nfs_delegation_find_inode(cps->clp, &args->fh);
-	if (inode == NULL)
+	if (inode == NULL) {
+		trace_nfs4_cb_getattr(cps->clp, &args->fh, NULL,
+				-ntohl(res->status));
 		goto out;
+	}
 	nfsi = NFS_I(inode);
 	rcu_read_lock();
 	delegation = rcu_dereference(nfsi->delegation);
@@ -60,6 +63,7 @@ __be32 nfs4_callback_getattr(struct cb_getattrargs *args,
 	res->status = 0;
 out_iput:
 	rcu_read_unlock();
+	trace_nfs4_cb_getattr(cps->clp, &args->fh, inode, -ntohl(res->status));
 	iput(inode);
 out:
 	dprintk("%s: exit with status = %d\n", __func__, ntohl(res->status));
@@ -194,6 +198,7 @@ static u32 initiate_file_draining(struct nfs_client *clp,
 	spin_unlock(&ino->i_lock);
 	pnfs_free_lseg_list(&free_me_list);
 	pnfs_put_layout_hdr(lo);
+	trace_nfs4_cb_layoutrecall_inode(clp, &args->cbl_fh, ino, -rv);
 	iput(ino);
 out:
 	return rv;
@@ -554,7 +559,7 @@ __be32 nfs4_callback_recallslot(struct cb_recallslotargs *args, void *dummy,
 	status = htonl(NFS4_OK);
 
 	nfs41_set_target_slotid(fc_tbl, args->crsa_target_highest_slotid);
-	nfs41_server_notify_target_slotid_update(cps->clp);
+	nfs41_notify_server(cps->clp);
 out:
 	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
 	return status;

fs/nfs/client.c

@@ -20,6 +20,7 @@
 #include <linux/stat.h>
 #include <linux/errno.h>
 #include <linux/unistd.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/sunrpc/clnt.h>
 #include <linux/sunrpc/stats.h>
 #include <linux/sunrpc/metrics.h>
@@ -285,116 +286,6 @@ void nfs_put_client(struct nfs_client *clp)
 }
 EXPORT_SYMBOL_GPL(nfs_put_client);
 
-#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
-/*
- * Test if two ip6 socket addresses refer to the same socket by
- * comparing relevant fields. The padding bytes specifically, are not
- * compared. sin6_flowinfo is not compared because it only affects QoS
- * and sin6_scope_id is only compared if the address is "link local"
- * because "link local" addresses need only be unique to a specific
- * link. Conversely, ordinary unicast addresses might have different
- * sin6_scope_id.
- *
- * The caller should ensure both socket addresses are AF_INET6.
- */
-static int nfs_sockaddr_match_ipaddr6(const struct sockaddr *sa1,
-				      const struct sockaddr *sa2)
-{
-	const struct sockaddr_in6 *sin1 = (const struct sockaddr_in6 *)sa1;
-	const struct sockaddr_in6 *sin2 = (const struct sockaddr_in6 *)sa2;
-
-	if (!ipv6_addr_equal(&sin1->sin6_addr, &sin2->sin6_addr))
-		return 0;
-	else if (ipv6_addr_type(&sin1->sin6_addr) & IPV6_ADDR_LINKLOCAL)
-		return sin1->sin6_scope_id == sin2->sin6_scope_id;
-
-	return 1;
-}
-#else	/* !defined(CONFIG_IPV6) && !defined(CONFIG_IPV6_MODULE) */
-static int nfs_sockaddr_match_ipaddr6(const struct sockaddr *sa1,
-				      const struct sockaddr *sa2)
-{
-	return 0;
-}
-#endif
-
-/*
- * Test if two ip4 socket addresses refer to the same socket, by
- * comparing relevant fields. The padding bytes specifically, are
- * not compared.
- *
- * The caller should ensure both socket addresses are AF_INET.
- */
-static int nfs_sockaddr_match_ipaddr4(const struct sockaddr *sa1,
-				      const struct sockaddr *sa2)
-{
-	const struct sockaddr_in *sin1 = (const struct sockaddr_in *)sa1;
-	const struct sockaddr_in *sin2 = (const struct sockaddr_in *)sa2;
-
-	return sin1->sin_addr.s_addr == sin2->sin_addr.s_addr;
-}
-
-static int nfs_sockaddr_cmp_ip6(const struct sockaddr *sa1,
-				const struct sockaddr *sa2)
-{
-	const struct sockaddr_in6 *sin1 = (const struct sockaddr_in6 *)sa1;
-	const struct sockaddr_in6 *sin2 = (const struct sockaddr_in6 *)sa2;
-
-	return nfs_sockaddr_match_ipaddr6(sa1, sa2) &&
-		(sin1->sin6_port == sin2->sin6_port);
-}
-
-static int nfs_sockaddr_cmp_ip4(const struct sockaddr *sa1,
-				const struct sockaddr *sa2)
-{
-	const struct sockaddr_in *sin1 = (const struct sockaddr_in *)sa1;
-	const struct sockaddr_in *sin2 = (const struct sockaddr_in *)sa2;
-
-	return nfs_sockaddr_match_ipaddr4(sa1, sa2) &&
-		(sin1->sin_port == sin2->sin_port);
-}
-
-#if defined(CONFIG_NFS_V4_1)
-/*
- * Test if two socket addresses represent the same actual socket,
- * by comparing (only) relevant fields, excluding the port number.
- */
-int nfs_sockaddr_match_ipaddr(const struct sockaddr *sa1,
-			      const struct sockaddr *sa2)
-{
-	if (sa1->sa_family != sa2->sa_family)
-		return 0;
-
-	switch (sa1->sa_family) {
-	case AF_INET:
-		return nfs_sockaddr_match_ipaddr4(sa1, sa2);
-	case AF_INET6:
-		return nfs_sockaddr_match_ipaddr6(sa1, sa2);
-	}
-	return 0;
-}
-EXPORT_SYMBOL_GPL(nfs_sockaddr_match_ipaddr);
-#endif /* CONFIG_NFS_V4_1 */
-
-/*
- * Test if two socket addresses represent the same actual socket,
- * by comparing (only) relevant fields, including the port number.
- */
-static int nfs_sockaddr_cmp(const struct sockaddr *sa1,
-			    const struct sockaddr *sa2)
-{
-	if (sa1->sa_family != sa2->sa_family)
-		return 0;
-
-	switch (sa1->sa_family) {
-	case AF_INET:
-		return nfs_sockaddr_cmp_ip4(sa1, sa2);
-	case AF_INET6:
-		return nfs_sockaddr_cmp_ip6(sa1, sa2);
-	}
-	return 0;
-}
-
 /*
  * Find an nfs_client on the list that matches the initialisation data
  * that is supplied.
@@ -421,7 +312,7 @@ static struct nfs_client *nfs_match_client(const struct nfs_client_initdata *dat
 		if (clp->cl_minorversion != data->minorversion)
 			continue;
 		/* Match the full socket address */
-		if (!nfs_sockaddr_cmp(sap, clap))
+		if (!rpc_cmp_addr_port(sap, clap))
 			continue;
 
 		atomic_inc(&clp->cl_count);