Merge tag 'nfs-for-4.16-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

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

  Stable bugfixes:

   - Fix breakages in the nfsstat utility due to the inclusion of the
     NFSv4 LOOKUPP operation

   - Fix a NULL pointer dereference in nfs_idmap_prepare_pipe_upcall()
     due to nfs_idmap_legacy_upcall() being called without an 'aux'
     parameter

   - Fix a refcount leak in the standard O_DIRECT error path

   - Fix a refcount leak in the pNFS O_DIRECT fallback to MDS path

   - Fix CPU latency issues with nfs_commit_release_pages()

   - Fix the LAYOUTUNAVAILABLE error case in the file layout type

   - NFS: Fix a race between mmap() and O_DIRECT

  Features:

   - Support the statx() mask and query flags to enable optimisations
     when the user is requesting only attributes that are already up to
     date in the inode cache, or is specifying the AT_STATX_DONT_SYNC
     flag

   - Add a module alias for the SCSI pNFS layout type

  Bugfixes:

   - Automounting when resolving a NFSv4 referral should preserve the
     RDMA transport protocol settings

   - Various other RDMA bugfixes from Chuck

   - pNFS block layout fixes

   - Always set NFS_LOCK_LOST when a lock is lost"

* tag 'nfs-for-4.16-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (69 commits)
  NFS: Fix a race between mmap() and O_DIRECT
  NFS: Remove a redundant call to unmap_mapping_range()
  pnfs/blocklayout: Ensure disk address in block device map
  pnfs/blocklayout: pnfs_block_dev_map uses bytes, not sectors
  lockd: Fix server refcounting
  SUNRPC: Fix null rpc_clnt dereference in rpc_task_queued tracepoint
  SUNRPC: Micro-optimize __rpc_execute
  SUNRPC: task_run_action should display tk_callback
  sunrpc: Format RPC events consistently for display
  SUNRPC: Trace xprt_timer events
  xprtrdma: Correct some documenting comments
  xprtrdma: Fix "bytes registered" accounting
  xprtrdma: Instrument allocation/release of rpcrdma_req/rep objects
  xprtrdma: Add trace points to instrument QP and CQ access upcalls
  xprtrdma: Add trace points in the client-side backchannel code paths
  xprtrdma: Add trace points for connect events
  xprtrdma: Add trace points to instrument MR allocation and recovery
  xprtrdma: Add trace points to instrument memory invalidation
  xprtrdma: Add trace points in reply decoder path
  xprtrdma: Add trace points to instrument memory registration
  ..

1 files changed, 87 insertions, 7 deletions

diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
index 995d707537da..7cb5c38c19e4 100644
--- a/fs/nfs/blocklayout/blocklayout.c
+++ b/fs/nfs/blocklayout/blocklayout.c
@@ -137,6 +137,11 @@ bl_alloc_init_bio(int npg, struct block_device *bdev, sector_t disk_sector,
 	return bio;
 }
 
+static bool offset_in_map(u64 offset, struct pnfs_block_dev_map *map)
+{
+	return offset >= map->start && offset < map->start + map->len;
+}
+
 static struct bio *
 do_add_page_to_bio(struct bio *bio, int npg, int rw, sector_t isect,
 		struct page *page, struct pnfs_block_dev_map *map,
@@ -156,8 +161,8 @@ do_add_page_to_bio(struct bio *bio, int npg, int rw, sector_t isect,
 
 	/* translate to physical disk offset */
 	disk_addr = (u64)isect << SECTOR_SHIFT;
-	if (disk_addr < map->start || disk_addr >= map->start + map->len) {
-		if (!dev->map(dev, disk_addr, map))
+	if (!offset_in_map(disk_addr, map)) {
+		if (!dev->map(dev, disk_addr, map) || !offset_in_map(disk_addr, map))
 			return ERR_PTR(-EIO);
 		bio = bl_submit_bio(bio);
 	}
@@ -184,6 +189,29 @@ retry:
 	return bio;
 }
 
+static void bl_mark_devices_unavailable(struct nfs_pgio_header *header, bool rw)
+{
+	struct pnfs_block_layout *bl = BLK_LSEG2EXT(header->lseg);
+	size_t bytes_left = header->args.count;
+	sector_t isect, extent_length = 0;
+	struct pnfs_block_extent be;
+
+	isect = header->args.offset >> SECTOR_SHIFT;
+	bytes_left += header->args.offset - (isect << SECTOR_SHIFT);
+
+	while (bytes_left > 0) {
+		if (!ext_tree_lookup(bl, isect, &be, rw))
+				return;
+		extent_length = be.be_length - (isect - be.be_f_offset);
+		nfs4_mark_deviceid_unavailable(be.be_device);
+		isect += extent_length;
+		if (bytes_left > extent_length << SECTOR_SHIFT)
+			bytes_left -= extent_length << SECTOR_SHIFT;
+		else
+			bytes_left = 0;
+	}
+}
+
 static void bl_end_io_read(struct bio *bio)
 {
 	struct parallel_io *par = bio->bi_private;
@@ -194,6 +222,7 @@ static void bl_end_io_read(struct bio *bio)
 		if (!header->pnfs_error)
 			header->pnfs_error = -EIO;
 		pnfs_set_lo_fail(header->lseg);
+		bl_mark_devices_unavailable(header, false);
 	}
 
 	bio_put(bio);
@@ -323,6 +352,7 @@ static void bl_end_io_write(struct bio *bio)
 		if (!header->pnfs_error)
 			header->pnfs_error = -EIO;
 		pnfs_set_lo_fail(header->lseg);
+		bl_mark_devices_unavailable(header, true);
 	}
 	bio_put(bio);
 	put_parallel(par);
@@ -552,6 +582,31 @@ static int decode_sector_number(__be32 **rp, sector_t *sp)
 	return 0;
 }
 
+static struct nfs4_deviceid_node *
+bl_find_get_deviceid(struct nfs_server *server,
+		const struct nfs4_deviceid *id, struct rpc_cred *cred,
+		gfp_t gfp_mask)
+{
+	struct nfs4_deviceid_node *node;
+	unsigned long start, end;
+
+retry:
+	node = nfs4_find_get_deviceid(server, id, cred, gfp_mask);
+	if (!node)
+		return ERR_PTR(-ENODEV);
+
+	if (test_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags) == 0)
+		return node;
+
+	end = jiffies;
+	start = end - PNFS_DEVICE_RETRY_TIMEOUT;
+	if (!time_in_range(node->timestamp_unavailable, start, end)) {
+		nfs4_delete_deviceid(node->ld, node->nfs_client, id);
+		goto retry;
+	}
+	return ERR_PTR(-ENODEV);
+}
+
 static int
 bl_alloc_extent(struct xdr_stream *xdr, struct pnfs_layout_hdr *lo,
 		struct layout_verification *lv, struct list_head *extents,
@@ -573,16 +628,18 @@ bl_alloc_extent(struct xdr_stream *xdr, struct pnfs_layout_hdr *lo,
 	memcpy(&id, p, NFS4_DEVICEID4_SIZE);
 	p += XDR_QUADLEN(NFS4_DEVICEID4_SIZE);
 
-	error = -EIO;
-	be->be_device = nfs4_find_get_deviceid(NFS_SERVER(lo->plh_inode), &id,
+	be->be_device = bl_find_get_deviceid(NFS_SERVER(lo->plh_inode), &id,
 						lo->plh_lc_cred, gfp_mask);
-	if (!be->be_device)
+	if (IS_ERR(be->be_device)) {
+		error = PTR_ERR(be->be_device);
 		goto out_free_be;
+	}
 
 	/*
 	 * The next three values are read in as bytes, but stored in the
 	 * extent structure in 512-byte granularity.
 	 */
+	error = -EIO;
 	if (decode_sector_number(&p, &be->be_f_offset) < 0)
 		goto out_put_deviceid;
 	if (decode_sector_number(&p, &be->be_length) < 0)
@@ -692,11 +749,16 @@ out_free_scratch:
 	__free_page(scratch);
 out:
 	dprintk("%s returns %d\n", __func__, status);
-	if (status) {
+	switch (status) {
+	case -ENODEV:
+		/* Our extent block devices are unavailable */
+		set_bit(NFS_LSEG_UNAVAILABLE, &lseg->pls_flags);
+	case 0:
+		return lseg;
+	default:
 		kfree(lseg);
 		return ERR_PTR(status);
 	}
-	return lseg;
 }
 
 static void
@@ -798,6 +860,13 @@ bl_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
 	}
 
 	pnfs_generic_pg_init_read(pgio, req);
+
+	if (pgio->pg_lseg &&
+		test_bit(NFS_LSEG_UNAVAILABLE, &pgio->pg_lseg->pls_flags)) {
+		pnfs_error_mark_layout_for_return(pgio->pg_inode, pgio->pg_lseg);
+		pnfs_set_lo_fail(pgio->pg_lseg);
+		nfs_pageio_reset_read_mds(pgio);
+	}
 }
 
 /*
@@ -853,6 +922,14 @@ bl_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
 		wb_size = nfs_dreq_bytes_left(pgio->pg_dreq);
 
 	pnfs_generic_pg_init_write(pgio, req, wb_size);
+
+	if (pgio->pg_lseg &&
+		test_bit(NFS_LSEG_UNAVAILABLE, &pgio->pg_lseg->pls_flags)) {
+
+		pnfs_error_mark_layout_for_return(pgio->pg_inode, pgio->pg_lseg);
+		pnfs_set_lo_fail(pgio->pg_lseg);
+		nfs_pageio_reset_write_mds(pgio);
+	}
 }
 
 /*
@@ -887,6 +964,7 @@ static struct pnfs_layoutdriver_type blocklayout_type = {
 	.name				= "LAYOUT_BLOCK_VOLUME",
 	.owner				= THIS_MODULE,
 	.flags				= PNFS_LAYOUTRET_ON_SETATTR |
+					  PNFS_LAYOUTRET_ON_ERROR |
 					  PNFS_READ_WHOLE_PAGE,
 	.read_pagelist			= bl_read_pagelist,
 	.write_pagelist			= bl_write_pagelist,
@@ -910,6 +988,7 @@ static struct pnfs_layoutdriver_type scsilayout_type = {
 	.name				= "LAYOUT_SCSI",
 	.owner				= THIS_MODULE,
 	.flags				= PNFS_LAYOUTRET_ON_SETATTR |
+					  PNFS_LAYOUTRET_ON_ERROR |
 					  PNFS_READ_WHOLE_PAGE,
 	.read_pagelist			= bl_read_pagelist,
 	.write_pagelist			= bl_write_pagelist,
@@ -967,6 +1046,7 @@ static void __exit nfs4blocklayout_exit(void)
 }
 
 MODULE_ALIAS("nfs-layouttype4-3");
+MODULE_ALIAS("nfs-layouttype4-5");
 
 module_init(nfs4blocklayout_init);
 module_exit(nfs4blocklayout_exit);