Linux-libre 4.6.2-gnu

14 files changed, 477 insertions, 92 deletions

diff --git a/fs/nfsd/Kconfig b/fs/nfsd/Kconfig
index a0b77fc1b..c9f583d7b 100644
--- a/fs/nfsd/Kconfig
+++ b/fs/nfsd/Kconfig
@@ -84,12 +84,30 @@ config NFSD_V4
 	  If unsure, say N.
 
 config NFSD_PNFS
-	bool "NFSv4.1 server support for Parallel NFS (pNFS)"
-	depends on NFSD_V4
+	bool
+
+config NFSD_BLOCKLAYOUT
+	bool "NFSv4.1 server support for pNFS block layouts"
+	depends on NFSD_V4 && BLOCK
+	select NFSD_PNFS
+	help
+	  This option enables support for the exporting pNFS block layouts
+	  in the kernel's NFS server. The pNFS block layout enables NFS
+	  clients to directly perform I/O to block devices accesible to both
+	  the server and the clients.  See RFC 5663 for more details.
+
+	  If unsure, say N.
+
+config NFSD_SCSILAYOUT
+	bool "NFSv4.1 server support for pNFS SCSI layouts"
+	depends on NFSD_V4 && BLOCK
+	select NFSD_PNFS
 	help
-	  This option enables support for the parallel NFS features of the
-	  minor version 1 of the NFSv4 protocol (RFC5661) in the kernel's NFS
-	  server.
+	  This option enables support for the exporting pNFS SCSI layouts
+	  in the kernel's NFS server. The pNFS SCSI layout enables NFS
+	  clients to directly perform I/O to SCSI devices accesible to both
+	  the server and the clients.  See draft-ietf-nfsv4-scsi-layout for
+	  more details.
 
 	  If unsure, say N.
 
diff --git a/fs/nfsd/Makefile b/fs/nfsd/Makefile
index 9a6028e12..3ae5f3c77 100644
--- a/fs/nfsd/Makefile
+++ b/fs/nfsd/Makefile
@@ -17,4 +17,6 @@ nfsd-$(CONFIG_NFSD_V3)	+= nfs3proc.o nfs3xdr.o
 nfsd-$(CONFIG_NFSD_V3_ACL) += nfs3acl.o
 nfsd-$(CONFIG_NFSD_V4)	+= nfs4proc.o nfs4xdr.o nfs4state.o nfs4idmap.o \
 			   nfs4acl.o nfs4callback.o nfs4recover.o
-nfsd-$(CONFIG_NFSD_PNFS) += nfs4layouts.o blocklayout.o blocklayoutxdr.o
+nfsd-$(CONFIG_NFSD_PNFS) += nfs4layouts.o
+nfsd-$(CONFIG_NFSD_BLOCKLAYOUT) += blocklayout.o blocklayoutxdr.o
+nfsd-$(CONFIG_NFSD_SCSILAYOUT) += blocklayout.o blocklayoutxdr.o
diff --git a/fs/nfsd/blocklayout.c b/fs/nfsd/blocklayout.c
index c29d9421b..e55b52426 100644
--- a/fs/nfsd/blocklayout.c
+++ b/fs/nfsd/blocklayout.c
@@ -1,11 +1,14 @@
 /*
- * Copyright (c) 2014 Christoph Hellwig.
+ * Copyright (c) 2014-2016 Christoph Hellwig.
  */
 #include <linux/exportfs.h>
 #include <linux/genhd.h>
 #include <linux/slab.h>
+#include <linux/pr.h>
 
 #include <linux/nfsd/debug.h>
+#include <scsi/scsi_proto.h>
+#include <scsi/scsi_common.h>
 
 #include "blocklayoutxdr.h"
 #include "pnfs.h"
@@ -13,37 +16,6 @@
 #define NFSDDBG_FACILITY	NFSDDBG_PNFS
 
 
-static int
-nfsd4_block_get_device_info_simple(struct super_block *sb,
-		struct nfsd4_getdeviceinfo *gdp)
-{
-	struct pnfs_block_deviceaddr *dev;
-	struct pnfs_block_volume *b;
-
-	dev = kzalloc(sizeof(struct pnfs_block_deviceaddr) +
-		      sizeof(struct pnfs_block_volume), GFP_KERNEL);
-	if (!dev)
-		return -ENOMEM;
-	gdp->gd_device = dev;
-
-	dev->nr_volumes = 1;
-	b = &dev->volumes[0];
-
-	b->type = PNFS_BLOCK_VOLUME_SIMPLE;
-	b->simple.sig_len = PNFS_BLOCK_UUID_LEN;
-	return sb->s_export_op->get_uuid(sb, b->simple.sig, &b->simple.sig_len,
-			&b->simple.offset);
-}
-
-static __be32
-nfsd4_block_proc_getdeviceinfo(struct super_block *sb,
-		struct nfsd4_getdeviceinfo *gdp)
-{
-	if (sb->s_bdev != sb->s_bdev->bd_contains)
-		return nfserr_inval;
-	return nfserrno(nfsd4_block_get_device_info_simple(sb, gdp));
-}
-
 static __be32
 nfsd4_block_proc_layoutget(struct inode *inode, const struct svc_fh *fhp,
 		struct nfsd4_layoutget *args)
@@ -141,20 +113,13 @@ out_layoutunavailable:
 }
 
 static __be32
-nfsd4_block_proc_layoutcommit(struct inode *inode,
-		struct nfsd4_layoutcommit *lcp)
+nfsd4_block_commit_blocks(struct inode *inode, struct nfsd4_layoutcommit *lcp,
+		struct iomap *iomaps, int nr_iomaps)
 {
 	loff_t new_size = lcp->lc_last_wr + 1;
 	struct iattr iattr = { .ia_valid = 0 };
-	struct iomap *iomaps;
-	int nr_iomaps;
 	int error;
 
-	nr_iomaps = nfsd4_block_decode_layoutupdate(lcp->lc_up_layout,
-			lcp->lc_up_len, &iomaps, 1 << inode->i_blkbits);
-	if (nr_iomaps < 0)
-		return nfserrno(nr_iomaps);
-
 	if (lcp->lc_mtime.tv_nsec == UTIME_NOW ||
 	    timespec_compare(&lcp->lc_mtime, &inode->i_mtime) < 0)
 		lcp->lc_mtime = current_fs_time(inode->i_sb);
@@ -172,6 +137,54 @@ nfsd4_block_proc_layoutcommit(struct inode *inode,
 	return nfserrno(error);
 }
 
+#ifdef CONFIG_NFSD_BLOCKLAYOUT
+static int
+nfsd4_block_get_device_info_simple(struct super_block *sb,
+		struct nfsd4_getdeviceinfo *gdp)
+{
+	struct pnfs_block_deviceaddr *dev;
+	struct pnfs_block_volume *b;
+
+	dev = kzalloc(sizeof(struct pnfs_block_deviceaddr) +
+		      sizeof(struct pnfs_block_volume), GFP_KERNEL);
+	if (!dev)
+		return -ENOMEM;
+	gdp->gd_device = dev;
+
+	dev->nr_volumes = 1;
+	b = &dev->volumes[0];
+
+	b->type = PNFS_BLOCK_VOLUME_SIMPLE;
+	b->simple.sig_len = PNFS_BLOCK_UUID_LEN;
+	return sb->s_export_op->get_uuid(sb, b->simple.sig, &b->simple.sig_len,
+			&b->simple.offset);
+}
+
+static __be32
+nfsd4_block_proc_getdeviceinfo(struct super_block *sb,
+		struct nfs4_client *clp,
+		struct nfsd4_getdeviceinfo *gdp)
+{
+	if (sb->s_bdev != sb->s_bdev->bd_contains)
+		return nfserr_inval;
+	return nfserrno(nfsd4_block_get_device_info_simple(sb, gdp));
+}
+
+static __be32
+nfsd4_block_proc_layoutcommit(struct inode *inode,
+		struct nfsd4_layoutcommit *lcp)
+{
+	struct iomap *iomaps;
+	int nr_iomaps;
+
+	nr_iomaps = nfsd4_block_decode_layoutupdate(lcp->lc_up_layout,
+			lcp->lc_up_len, &iomaps, 1 << inode->i_blkbits);
+	if (nr_iomaps < 0)
+		return nfserrno(nr_iomaps);
+
+	return nfsd4_block_commit_blocks(inode, lcp, iomaps, nr_iomaps);
+}
+
 const struct nfsd4_layout_ops bl_layout_ops = {
 	/*
 	 * Pretend that we send notification to the client.  This is a blatant
@@ -190,3 +203,206 @@ const struct nfsd4_layout_ops bl_layout_ops = {
 	.encode_layoutget	= nfsd4_block_encode_layoutget,
 	.proc_layoutcommit	= nfsd4_block_proc_layoutcommit,
 };
+#endif /* CONFIG_NFSD_BLOCKLAYOUT */
+
+#ifdef CONFIG_NFSD_SCSILAYOUT
+static int nfsd4_scsi_identify_device(struct block_device *bdev,
+		struct pnfs_block_volume *b)
+{
+	struct request_queue *q = bdev->bd_disk->queue;
+	struct request *rq;
+	size_t bufflen = 252, len, id_len;
+	u8 *buf, *d, type, assoc;
+	int error;
+
+	buf = kzalloc(bufflen, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	rq = blk_get_request(q, READ, GFP_KERNEL);
+	if (IS_ERR(rq)) {
+		error = -ENOMEM;
+		goto out_free_buf;
+	}
+	blk_rq_set_block_pc(rq);
+
+	error = blk_rq_map_kern(q, rq, buf, bufflen, GFP_KERNEL);
+	if (error)
+		goto out_put_request;
+
+	rq->cmd[0] = INQUIRY;
+	rq->cmd[1] = 1;
+	rq->cmd[2] = 0x83;
+	rq->cmd[3] = bufflen >> 8;
+	rq->cmd[4] = bufflen & 0xff;
+	rq->cmd_len = COMMAND_SIZE(INQUIRY);
+
+	error = blk_execute_rq(rq->q, NULL, rq, 1);
+	if (error) {
+		pr_err("pNFS: INQUIRY 0x83 failed with: %x\n",
+			rq->errors);
+		goto out_put_request;
+	}
+
+	len = (buf[2] << 8) + buf[3] + 4;
+	if (len > bufflen) {
+		pr_err("pNFS: INQUIRY 0x83 response invalid (len = %zd)\n",
+			len);
+		goto out_put_request;
+	}
+
+	d = buf + 4;
+	for (d = buf + 4; d < buf + len; d += id_len + 4) {
+		id_len = d[3];
+		type = d[1] & 0xf;
+		assoc = (d[1] >> 4) & 0x3;
+
+		/*
+		 * We only care about a EUI-64 and NAA designator types
+		 * with LU association.
+		 */
+		if (assoc != 0x00)
+			continue;
+		if (type != 0x02 && type != 0x03)
+			continue;
+		if (id_len != 8 && id_len != 12 && id_len != 16)
+			continue;
+
+		b->scsi.code_set = PS_CODE_SET_BINARY;
+		b->scsi.designator_type = type == 0x02 ?
+			PS_DESIGNATOR_EUI64 : PS_DESIGNATOR_NAA;
+		b->scsi.designator_len = id_len;
+		memcpy(b->scsi.designator, d + 4, id_len);
+
+		/*
+		 * If we found a 8 or 12 byte descriptor continue on to
+		 * see if a 16 byte one is available.  If we find a
+		 * 16 byte descriptor we're done.
+		 */
+		if (id_len == 16)
+			break;
+	}
+
+out_put_request:
+	blk_put_request(rq);
+out_free_buf:
+	kfree(buf);
+	return error;
+}
+
+#define NFSD_MDS_PR_KEY		0x0100000000000000
+
+/*
+ * We use the client ID as a unique key for the reservations.
+ * This allows us to easily fence a client when recalls fail.
+ */
+static u64 nfsd4_scsi_pr_key(struct nfs4_client *clp)
+{
+	return ((u64)clp->cl_clientid.cl_boot << 32) | clp->cl_clientid.cl_id;
+}
+
+static int
+nfsd4_block_get_device_info_scsi(struct super_block *sb,
+		struct nfs4_client *clp,
+		struct nfsd4_getdeviceinfo *gdp)
+{
+	struct pnfs_block_deviceaddr *dev;
+	struct pnfs_block_volume *b;
+	const struct pr_ops *ops;
+	int error;
+
+	dev = kzalloc(sizeof(struct pnfs_block_deviceaddr) +
+		      sizeof(struct pnfs_block_volume), GFP_KERNEL);
+	if (!dev)
+		return -ENOMEM;
+	gdp->gd_device = dev;
+
+	dev->nr_volumes = 1;
+	b = &dev->volumes[0];
+
+	b->type = PNFS_BLOCK_VOLUME_SCSI;
+	b->scsi.pr_key = nfsd4_scsi_pr_key(clp);
+
+	error = nfsd4_scsi_identify_device(sb->s_bdev, b);
+	if (error)
+		return error;
+
+	ops = sb->s_bdev->bd_disk->fops->pr_ops;
+	if (!ops) {
+		pr_err("pNFS: device %s does not support PRs.\n",
+			sb->s_id);
+		return -EINVAL;
+	}
+
+	error = ops->pr_register(sb->s_bdev, 0, NFSD_MDS_PR_KEY, true);
+	if (error) {
+		pr_err("pNFS: failed to register key for device %s.\n",
+			sb->s_id);
+		return -EINVAL;
+	}
+
+	error = ops->pr_reserve(sb->s_bdev, NFSD_MDS_PR_KEY,
+			PR_EXCLUSIVE_ACCESS_REG_ONLY, 0);
+	if (error) {
+		pr_err("pNFS: failed to reserve device %s.\n",
+			sb->s_id);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static __be32
+nfsd4_scsi_proc_getdeviceinfo(struct super_block *sb,
+		struct nfs4_client *clp,
+		struct nfsd4_getdeviceinfo *gdp)
+{
+	if (sb->s_bdev != sb->s_bdev->bd_contains)
+		return nfserr_inval;
+	return nfserrno(nfsd4_block_get_device_info_scsi(sb, clp, gdp));
+}
+static __be32
+nfsd4_scsi_proc_layoutcommit(struct inode *inode,
+		struct nfsd4_layoutcommit *lcp)
+{
+	struct iomap *iomaps;
+	int nr_iomaps;
+
+	nr_iomaps = nfsd4_scsi_decode_layoutupdate(lcp->lc_up_layout,
+			lcp->lc_up_len, &iomaps, 1 << inode->i_blkbits);
+	if (nr_iomaps < 0)
+		return nfserrno(nr_iomaps);
+
+	return nfsd4_block_commit_blocks(inode, lcp, iomaps, nr_iomaps);
+}
+
+static void
+nfsd4_scsi_fence_client(struct nfs4_layout_stateid *ls)
+{
+	struct nfs4_client *clp = ls->ls_stid.sc_client;
+	struct block_device *bdev = ls->ls_file->f_path.mnt->mnt_sb->s_bdev;
+
+	bdev->bd_disk->fops->pr_ops->pr_preempt(bdev, NFSD_MDS_PR_KEY,
+			nfsd4_scsi_pr_key(clp), 0, true);
+}
+
+const struct nfsd4_layout_ops scsi_layout_ops = {
+	/*
+	 * Pretend that we send notification to the client.  This is a blatant
+	 * lie to force recent Linux clients to cache our device IDs.
+	 * We rarely ever change the device ID, so the harm of leaking deviceids
+	 * for a while isn't too bad.  Unfortunately RFC5661 is a complete mess
+	 * in this regard, but I filed errata 4119 for this a while ago, and
+	 * hopefully the Linux client will eventually start caching deviceids
+	 * without this again.
+	 */
+	.notify_types		=
+			NOTIFY_DEVICEID4_DELETE | NOTIFY_DEVICEID4_CHANGE,
+	.proc_getdeviceinfo	= nfsd4_scsi_proc_getdeviceinfo,
+	.encode_getdeviceinfo	= nfsd4_block_encode_getdeviceinfo,
+	.proc_layoutget		= nfsd4_block_proc_layoutget,
+	.encode_layoutget	= nfsd4_block_encode_layoutget,
+	.proc_layoutcommit	= nfsd4_scsi_proc_layoutcommit,
+	.fence_client		= nfsd4_scsi_fence_client,
+};
+#endif /* CONFIG_NFSD_SCSILAYOUT */
diff --git a/fs/nfsd/blocklayoutxdr.c b/fs/nfsd/blocklayoutxdr.c
index 6d834dc9b..6c3b316f9 100644
--- a/fs/nfsd/blocklayoutxdr.c
+++ b/fs/nfsd/blocklayoutxdr.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2014 Christoph Hellwig.
+ * Copyright (c) 2014-2016 Christoph Hellwig.
  */
 #include <linux/sunrpc/svc.h>
 #include <linux/exportfs.h>
@@ -53,6 +53,18 @@ nfsd4_block_encode_volume(struct xdr_stream *xdr, struct pnfs_block_volume *b)
 		p = xdr_encode_hyper(p, b->simple.offset);
 		p = xdr_encode_opaque(p, b->simple.sig, b->simple.sig_len);
 		break;
+	case PNFS_BLOCK_VOLUME_SCSI:
+		len = 4 + 4 + 4 + 4 + b->scsi.designator_len + 8;
+		p = xdr_reserve_space(xdr, len);
+		if (!p)
+			return -ETOOSMALL;
+
+		*p++ = cpu_to_be32(b->type);
+		*p++ = cpu_to_be32(b->scsi.code_set);
+		*p++ = cpu_to_be32(b->scsi.designator_type);
+		p = xdr_encode_opaque(p, b->scsi.designator, b->scsi.designator_len);
+		p = xdr_encode_hyper(p, b->scsi.pr_key);
+		break;
 	default:
 		return -ENOTSUPP;
 	}
@@ -93,18 +105,22 @@ nfsd4_block_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
 		u32 block_size)
 {
 	struct iomap *iomaps;
-	u32 nr_iomaps, expected, i;
+	u32 nr_iomaps, i;
 
 	if (len < sizeof(u32)) {
 		dprintk("%s: extent array too small: %u\n", __func__, len);
 		return -EINVAL;
 	}
+	len -= sizeof(u32);
+	if (len % PNFS_BLOCK_EXTENT_SIZE) {
+		dprintk("%s: extent array invalid: %u\n", __func__, len);
+		return -EINVAL;
+	}
 
 	nr_iomaps = be32_to_cpup(p++);
-	expected = sizeof(__be32) + nr_iomaps * PNFS_BLOCK_EXTENT_SIZE;
-	if (len != expected) {
+	if (nr_iomaps != len / PNFS_BLOCK_EXTENT_SIZE) {
 		dprintk("%s: extent array size mismatch: %u/%u\n",
-			__func__, len, expected);
+			__func__, len, nr_iomaps);
 		return -EINVAL;
 	}
 
@@ -155,3 +171,54 @@ fail:
 	kfree(iomaps);
 	return -EINVAL;
 }
+
+int
+nfsd4_scsi_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
+		u32 block_size)
+{
+	struct iomap *iomaps;
+	u32 nr_iomaps, expected, i;
+
+	if (len < sizeof(u32)) {
+		dprintk("%s: extent array too small: %u\n", __func__, len);
+		return -EINVAL;
+	}
+
+	nr_iomaps = be32_to_cpup(p++);
+	expected = sizeof(__be32) + nr_iomaps * PNFS_SCSI_RANGE_SIZE;
+	if (len != expected) {
+		dprintk("%s: extent array size mismatch: %u/%u\n",
+			__func__, len, expected);
+		return -EINVAL;
+	}
+
+	iomaps = kcalloc(nr_iomaps, sizeof(*iomaps), GFP_KERNEL);
+	if (!iomaps) {
+		dprintk("%s: failed to allocate extent array\n", __func__);
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < nr_iomaps; i++) {
+		u64 val;
+
+		p = xdr_decode_hyper(p, &val);
+		if (val & (block_size - 1)) {
+			dprintk("%s: unaligned offset 0x%llx\n", __func__, val);
+			goto fail;
+		}
+		iomaps[i].offset = val;
+
+		p = xdr_decode_hyper(p, &val);
+		if (val & (block_size - 1)) {
+			dprintk("%s: unaligned length 0x%llx\n", __func__, val);
+			goto fail;
+		}
+		iomaps[i].length = val;
+	}
+
+	*iomapp = iomaps;
+	return nr_iomaps;
+fail:
+	kfree(iomaps);
+	return -EINVAL;
+}
diff --git a/fs/nfsd/blocklayoutxdr.h b/fs/nfsd/blocklayoutxdr.h
index 6de925fe8..397bc7563 100644
--- a/fs/nfsd/blocklayoutxdr.h
+++ b/fs/nfsd/blocklayoutxdr.h
@@ -15,6 +15,11 @@ struct pnfs_block_extent {
 	enum pnfs_block_extent_state	es;
 };
 
+struct pnfs_block_range {
+	u64				foff;
+	u64				len;
+};
+
 /*
  * Random upper cap for the uuid length to avoid unbounded allocation.
  * Not actually limited by the protocol.
@@ -29,6 +34,13 @@ struct pnfs_block_volume {
 			u32		sig_len;
 			u8		sig[PNFS_BLOCK_UUID_LEN];
 		} simple;
+		struct {
+			enum scsi_code_set		code_set;
+			enum scsi_designator_type	designator_type;
+			int				designator_len;
+			u8				designator[256];
+			u64				pr_key;
+		} scsi;
 	};
 };
 
@@ -43,5 +55,7 @@ __be32 nfsd4_block_encode_layoutget(struct xdr_stream *xdr,
 		struct nfsd4_layoutget *lgp);
 int nfsd4_block_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
 		u32 block_size);
+int nfsd4_scsi_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
+		u32 block_size);
 
 #endif /* _NFSD_BLOCKLAYOUTXDR_H */
diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c
index 7b755b7f7..51c3b06e8 100644
--- a/fs/nfsd/nfs3proc.c
+++ b/fs/nfsd/nfs3proc.c
@@ -147,6 +147,7 @@ nfsd3_proc_read(struct svc_rqst *rqstp, struct nfsd3_readargs *argp,
 {
 	__be32	nfserr;
 	u32	max_blocksize = svc_max_payload(rqstp);
+	unsigned long cnt = min(argp->count, max_blocksize);
 
 	dprintk("nfsd: READ(3) %s %lu bytes at %Lu\n",
 				SVCFH_fmt(&argp->fh),
@@ -157,7 +158,7 @@ nfsd3_proc_read(struct svc_rqst *rqstp, struct nfsd3_readargs *argp,
 	 * 1 (status) + 22 (post_op_attr) + 1 (count) + 1 (eof)
 	 * + 1 (xdr opaque byte count) = 26
 	 */
-	resp->count = min(argp->count, max_blocksize);
+	resp->count = cnt;
 	svc_reserve_auth(rqstp, ((1 + NFS3_POST_OP_ATTR_WORDS + 3)<<2) + resp->count +4);
 
 	fh_copy(&resp->fh, &argp->fh);
@@ -167,8 +168,8 @@ nfsd3_proc_read(struct svc_rqst *rqstp, struct nfsd3_readargs *argp,
 				  &resp->count);
 	if (nfserr == 0) {
 		struct inode	*inode = d_inode(resp->fh.fh_dentry);
-
-		resp->eof = (argp->offset + resp->count) >= inode->i_size;
+		resp->eof = nfsd_eof_on_read(cnt, resp->count, argp->offset,
+							inode->i_size);
 	}
 
 	RETURN_STATUS(nfserr);
diff --git a/fs/nfsd/nfs4layouts.c b/fs/nfsd/nfs4layouts.c
index ce2d010d3..825c7bc8d 100644
--- a/fs/nfsd/nfs4layouts.c
+++ b/fs/nfsd/nfs4layouts.c
@@ -1,6 +1,7 @@
 /*
  * Copyright (c) 2014 Christoph Hellwig.
  */
+#include <linux/blkdev.h>
 #include <linux/kmod.h>
 #include <linux/file.h>
 #include <linux/jhash.h>
@@ -26,7 +27,12 @@ static const struct nfsd4_callback_ops nfsd4_cb_layout_ops;
 static const struct lock_manager_operations nfsd4_layouts_lm_ops;
 
 const struct nfsd4_layout_ops *nfsd4_layout_ops[LAYOUT_TYPE_MAX] =  {
+#ifdef CONFIG_NFSD_BLOCKLAYOUT
 	[LAYOUT_BLOCK_VOLUME]	= &bl_layout_ops,
+#endif
+#ifdef CONFIG_NFSD_SCSILAYOUT
+	[LAYOUT_SCSI]		= &scsi_layout_ops,
+#endif
 };
 
 /* pNFS device ID to export fsid mapping */
@@ -121,10 +127,24 @@ void nfsd4_setup_layout_type(struct svc_export *exp)
 	if (!(exp->ex_flags & NFSEXP_PNFS))
 		return;
 
+	/*
+	 * Check if the file system supports exporting a block-like layout.
+	 * If the block device supports reservations prefer the SCSI layout,
+	 * otherwise advertise the block layout.
+	 */
+#ifdef CONFIG_NFSD_BLOCKLAYOUT
 	if (sb->s_export_op->get_uuid &&
 	    sb->s_export_op->map_blocks &&
 	    sb->s_export_op->commit_blocks)
 		exp->ex_layout_type = LAYOUT_BLOCK_VOLUME;
+#endif
+#ifdef CONFIG_NFSD_SCSILAYOUT
+	/* overwrite block layout selection if needed */
+	if (sb->s_export_op->map_blocks &&
+	    sb->s_export_op->commit_blocks &&
+	    sb->s_bdev && sb->s_bdev->bd_disk->fops->pr_ops)
+		exp->ex_layout_type = LAYOUT_SCSI;
+#endif
 }
 
 static void
@@ -590,8 +610,6 @@ nfsd4_cb_layout_fail(struct nfs4_layout_stateid *ls)
 
 	rpc_ntop((struct sockaddr *)&clp->cl_addr, addr_str, sizeof(addr_str));
 
-	trace_layout_recall_fail(&ls->ls_stid.sc_stateid);
-
 	printk(KERN_WARNING
 		"nfsd: client %s failed to respond to layout recall. "
 		"  Fencing..\n", addr_str);
@@ -626,6 +644,7 @@ nfsd4_cb_layout_done(struct nfsd4_callback *cb, struct rpc_task *task)
 		container_of(cb, struct nfs4_layout_stateid, ls_recall);
 	struct nfsd_net *nn;
 	ktime_t now, cutoff;
+	const struct nfsd4_layout_ops *ops;
 	LIST_HEAD(reaplist);
 
 
@@ -661,7 +680,13 @@ nfsd4_cb_layout_done(struct nfsd4_callback *cb, struct rpc_task *task)
 		/*
 		 * Unknown error or non-responding client, we'll need to fence.
 		 */
-		nfsd4_cb_layout_fail(ls);
+		trace_layout_recall_fail(&ls->ls_stid.sc_stateid);
+
+		ops = nfsd4_layout_ops[ls->ls_layout_type];
+		if (ops->fence_client)
+			ops->fence_client(ls);
+		else
+			nfsd4_cb_layout_fail(ls);
 		return -1;
 	}
 }
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index f8082c7cd..de1ff1d98 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -864,12 +864,10 @@ static __be32
 nfsd4_secinfo(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	      struct nfsd4_secinfo *secinfo)
 {
-	struct svc_fh resfh;
 	struct svc_export *exp;
 	struct dentry *dentry;
 	__be32 err;
 
-	fh_init(&resfh, NFS4_FHSIZE);
 	err = fh_verify(rqstp, &cstate->current_fh, S_IFDIR, NFSD_MAY_EXEC);
 	if (err)
 		return err;
@@ -1270,8 +1268,10 @@ nfsd4_getdeviceinfo(struct svc_rqst *rqstp,
 		goto out;
 
 	nfserr = nfs_ok;
-	if (gdp->gd_maxcount != 0)
-		nfserr = ops->proc_getdeviceinfo(exp->ex_path.mnt->mnt_sb, gdp);
+	if (gdp->gd_maxcount != 0) {
+		nfserr = ops->proc_getdeviceinfo(exp->ex_path.mnt->mnt_sb,
+					cstate->session->se_client, gdp);
+	}
 
 	gdp->gd_notify_types &= ops->notify_types;
 out:
diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c
index dc8ebecf5..66eaeb1e8 100644
--- a/fs/nfsd/nfs4recover.c
+++ b/fs/nfsd/nfs4recover.c
@@ -32,10 +32,10 @@
 *
 */
 
+#include <crypto/hash.h>
 #include <linux/file.h>
 #include <linux/slab.h>
 #include <linux/namei.h>
-#include <linux/crypto.h>
 #include <linux/sched.h>
 #include <linux/fs.h>
 #include <linux/module.h>
@@ -104,29 +104,35 @@ static int
 nfs4_make_rec_clidname(char *dname, const struct xdr_netobj *clname)
 {
 	struct xdr_netobj cksum;
-	struct hash_desc desc;
-	struct scatterlist sg;
+	struct crypto_shash *tfm;
 	int status;
 
 	dprintk("NFSD: nfs4_make_rec_clidname for %.*s\n",
 			clname->len, clname->data);
-	desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP;
-	desc.tfm = crypto_alloc_hash("md5", 0, CRYPTO_ALG_ASYNC);
-	if (IS_ERR(desc.tfm)) {
-		status = PTR_ERR(desc.tfm);
+	tfm = crypto_alloc_shash("md5", 0, 0);
+	if (IS_ERR(tfm)) {
+		status = PTR_ERR(tfm);
 		goto out_no_tfm;
 	}
 
-	cksum.len = crypto_hash_digestsize(desc.tfm);
+	cksum.len = crypto_shash_digestsize(tfm);
 	cksum.data = kmalloc(cksum.len, GFP_KERNEL);
 	if (cksum.data == NULL) {
 		status = -ENOMEM;
  		goto out;
 	}
 
-	sg_init_one(&sg, clname->data, clname->len);
+	{
+		SHASH_DESC_ON_STACK(desc, tfm);
+
+		desc->tfm = tfm;
+		desc->flags = CRYPTO_TFM_REQ_MAY_SLEEP;
+
+		status = crypto_shash_digest(desc, clname->data, clname->len,
+					     cksum.data);
+		shash_desc_zero(desc);
+	}
 
-	status = crypto_hash_digest(&desc, &sg, sg.length, cksum.data);
 	if (status)
 		goto out;
 
@@ -135,7 +141,7 @@ nfs4_make_rec_clidname(char *dname, const struct xdr_netobj *clname)
 	status = 0;
 out:
 	kfree(cksum.data);
-	crypto_free_hash(desc.tfm);
+	crypto_free_shash(tfm);
 out_no_tfm:
 	return status;
 }
@@ -1260,6 +1266,7 @@ nfsd4_umh_cltrack_init(struct net *net)
 	/* XXX: The usermode helper s not working in container yet. */
 	if (net != &init_net) {
 		pr_warn("NFSD: attempt to initialize umh client tracking in a container ignored.\n");
+		kfree(grace_start);
 		return -EINVAL;
 	}
 
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index c484a2b6c..0462eeddf 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -2408,7 +2408,8 @@ nfsd4_exchange_id(struct svc_rqst *rqstp,
 	default:				/* checked by xdr code */
 		WARN_ON_ONCE(1);
 	case SP4_SSV:
-		return nfserr_encr_alg_unsupp;
+		status = nfserr_encr_alg_unsupp;
+		goto out_nolock;
 	}
 
 	/* Cases below refer to rfc 5661 section 18.35.4: */
@@ -2586,21 +2587,26 @@ static __be32 check_forechannel_attrs(struct nfsd4_channel_attrs *ca, struct nfs
 	return nfs_ok;
 }
 
+/*
+ * Server's NFSv4.1 backchannel support is AUTH_SYS-only for now.
+ * These are based on similar macros in linux/sunrpc/msg_prot.h .
+ */
+#define RPC_MAX_HEADER_WITH_AUTH_SYS \
+	(RPC_CALLHDRSIZE + 2 * (2 + UNX_CALLSLACK))
+
+#define RPC_MAX_REPHEADER_WITH_AUTH_SYS \
+	(RPC_REPHDRSIZE + (2 + NUL_REPLYSLACK))
+
 #define NFSD_CB_MAX_REQ_SZ	((NFS4_enc_cb_recall_sz + \
-				 RPC_MAX_HEADER_WITH_AUTH) * sizeof(__be32))
+				 RPC_MAX_HEADER_WITH_AUTH_SYS) * sizeof(__be32))
 #define NFSD_CB_MAX_RESP_SZ	((NFS4_dec_cb_recall_sz + \
-				 RPC_MAX_REPHEADER_WITH_AUTH) * sizeof(__be32))
+				 RPC_MAX_REPHEADER_WITH_AUTH_SYS) * \
+				 sizeof(__be32))
 
 static __be32 check_backchannel_attrs(struct nfsd4_channel_attrs *ca)
 {
 	ca->headerpadsz = 0;
 
-	/*
-	 * These RPC_MAX_HEADER macros are overkill, especially since we
-	 * don't even do gss on the backchannel yet.  But this is still
-	 * less than 1k.  Tighten up this estimate in the unlikely event
-	 * it turns out to be a problem for some client:
-	 */
 	if (ca->maxreq_sz < NFSD_CB_MAX_REQ_SZ)
 		return nfserr_toosmall;
 	if (ca->maxresp_sz < NFSD_CB_MAX_RESP_SZ)
@@ -2710,10 +2716,9 @@ nfsd4_create_session(struct svc_rqst *rqstp,
 		goto out_free_conn;
 	}
 	status = nfs_ok;
-	/*
-	 * We do not support RDMA or persistent sessions
-	 */
+	/* Persistent sessions are not supported */
 	cr_ses->flags &= ~SESSION4_PERSIST;
+	/* Upshifting from TCP to RDMA is not supported */
 	cr_ses->flags &= ~SESSION4_RDMA;
 
 	init_session(rqstp, new, conf, cr_ses);
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index 1600ec470..9df898ba6 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -3063,7 +3063,7 @@ static __be32 nfsd4_encode_bind_conn_to_session(struct nfsd4_compoundres *resp,
 		p = xdr_encode_opaque_fixed(p, bcts->sessionid.data,
 						NFS4_MAX_SESSIONID_LEN);
 		*p++ = cpu_to_be32(bcts->dir);
-		/* Sorry, we do not yet support RDMA over 4.1: */
+		/* Upshifting from TCP to RDMA is not supported */
 		*p++ = cpu_to_be32(0);
 	}
 	return nfserr;
@@ -3365,6 +3365,7 @@ static __be32 nfsd4_encode_splice_read(
 	struct xdr_stream *xdr = &resp->xdr;
 	struct xdr_buf *buf = xdr->buf;
 	u32 eof;
+	long len;
 	int space_left;
 	__be32 nfserr;
 	__be32 *p = xdr->p - 2;
@@ -3373,6 +3374,7 @@ static __be32 nfsd4_encode_splice_read(
 	if (xdr->end - xdr->p < 1)
 		return nfserr_resource;
 
+	len = maxcount;
 	nfserr = nfsd_splice_read(read->rd_rqstp, file,
 				  read->rd_offset, &maxcount);
 	if (nfserr) {
@@ -3385,8 +3387,8 @@ static __be32 nfsd4_encode_splice_read(
 		return nfserr;
 	}
 
-	eof = (read->rd_offset + maxcount >=
-	       d_inode(read->rd_fhp->fh_dentry)->i_size);
+	eof = nfsd_eof_on_read(len, maxcount, read->rd_offset,
+				d_inode(read->rd_fhp->fh_dentry)->i_size);
 
 	*(p++) = htonl(eof);
 	*(p++) = htonl(maxcount);
@@ -3456,14 +3458,15 @@ static __be32 nfsd4_encode_readv(struct nfsd4_compoundres *resp,
 	}
 	read->rd_vlen = v;
 
+	len = maxcount;
 	nfserr = nfsd_readv(file, read->rd_offset, resp->rqstp->rq_vec,
 			read->rd_vlen, &maxcount);
 	if (nfserr)
 		return nfserr;
 	xdr_truncate_encode(xdr, starting_len + 8 + ((maxcount+3)&~3));
 
-	eof = (read->rd_offset + maxcount >=
-	       d_inode(read->rd_fhp->fh_dentry)->i_size);
+	eof = nfsd_eof_on_read(len, maxcount, read->rd_offset,
+				d_inode(read->rd_fhp->fh_dentry)->i_size);
 
 	tmp = htonl(eof);
 	write_bytes_to_xdr_buf(xdr->buf, starting_len    , &tmp, 4);
diff --git a/fs/nfsd/pnfs.h b/fs/nfsd/pnfs.h
index d4c445367..7d073b9b1 100644
--- a/fs/nfsd/pnfs.h
+++ b/fs/nfsd/pnfs.h
@@ -21,6 +21,7 @@ struct nfsd4_layout_ops {
 	u32		notify_types;
 
 	__be32 (*proc_getdeviceinfo)(struct super_block *sb,
+			struct nfs4_client *clp,
 			struct nfsd4_getdeviceinfo *gdevp);
 	__be32 (*encode_getdeviceinfo)(struct xdr_stream *xdr,
 			struct nfsd4_getdeviceinfo *gdevp);
@@ -32,10 +33,17 @@ struct nfsd4_layout_ops {
 
 	__be32 (*proc_layoutcommit)(struct inode *inode,
 			struct nfsd4_layoutcommit *lcp);
+
+	void (*fence_client)(struct nfs4_layout_stateid *ls);
 };
 
 extern const struct nfsd4_layout_ops *nfsd4_layout_ops[];
+#ifdef CONFIG_NFSD_BLOCKLAYOUT
 extern const struct nfsd4_layout_ops bl_layout_ops;
+#endif
+#ifdef CONFIG_NFSD_SCSILAYOUT
+extern const struct nfsd4_layout_ops scsi_layout_ops;
+#endif
 
 __be32 nfsd4_preprocess_layout_stateid(struct svc_rqst *rqstp,
 		struct nfsd4_compound_state *cstate, stateid_t *stateid,
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 5d2a57e4c..d40010e4f 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -870,7 +870,7 @@ __be32 nfsd_readv(struct file *file, loff_t offset, struct kvec *vec, int vlen,
 
 	oldfs = get_fs();
 	set_fs(KERNEL_DS);
-	host_err = vfs_readv(file, (struct iovec __user *)vec, vlen, &offset);
+	host_err = vfs_readv(file, (struct iovec __user *)vec, vlen, &offset, 0);
 	set_fs(oldfs);
 	return nfsd_finish_read(file, count, host_err);
 }
@@ -957,7 +957,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 
 	/* Write the data. */
 	oldfs = get_fs(); set_fs(KERNEL_DS);
-	host_err = vfs_writev(file, (struct iovec __user *)vec, vlen, &pos);
+	host_err = vfs_writev(file, (struct iovec __user *)vec, vlen, &pos, 0);
 	set_fs(oldfs);
 	if (host_err < 0)
 		goto out_nfserr;
diff --git a/fs/nfsd/vfs.h b/fs/nfsd/vfs.h
index c11ba316f..2d573ec05 100644
--- a/fs/nfsd/vfs.h
+++ b/fs/nfsd/vfs.h
@@ -139,4 +139,23 @@ static inline int nfsd_create_is_exclusive(int createmode)
 	       || createmode == NFS4_CREATE_EXCLUSIVE4_1;
 }
 
+static inline bool nfsd_eof_on_read(long requested, long read,
+				loff_t offset, loff_t size)
+{
+	/* We assume a short read means eof: */
+	if (requested > read)
+		return true;
+	/*
+	 * A non-short read might also reach end of file.  The spec
+	 * still requires us to set eof in that case.
+	 *
+	 * Further operations may have modified the file size since
+	 * the read, so the following check is not atomic with the read.
+	 * We've only seen that cause a problem for a client in the case
+	 * where the read returned a count of 0 without setting eof.
+	 * That case was fixed by the addition of the above check.
+	 */
+	return (offset + read >= size);
+}
+
 #endif /* LINUX_NFSD_VFS_H */