[PATCH 00/15] Various writeback related fixes

[PATCH 00/15] Various writeback related fixes

[Trond Myklebust]

A combination of writeback fixes for both regular NFS and pNFS.

Trond Myklebust (15):
  NFS: Revalidate the file size on a fatal write error
  NFS: Revalidate the file mapping on all fatal writeback errors
  SUNRPC: Remove broken gss_mech_list_pseudoflavors()
  NFS: Fix up fsync() when the server rebooted
  NFS/pnfs: Fix pnfs_generic_prepare_to_resend_writes()
  NFSv4: Improve read/write/commit tracing
  NFS: Fix fix of show_nfs_errors
  pNFS/flexfiles: Record resend attempts on I/O failure
  NFS: Clean up generic file read tracepoints
  NFS: Clean up generic writeback tracepoints
  NFS: Clean up generic file commit tracepoint
  pNFS/flexfiles: Add tracing for layout errors
  NFS: Improve tracing of permission calls
  NFS: When resending after a short write, reset the reply count to zero
  NFS: Fix nfs_direct_write_reschedule_io()

 fs/nfs/dir.c                           |   4 +-
 fs/nfs/direct.c                        |   7 +-
 fs/nfs/file.c                          |  37 ++--
 fs/nfs/flexfilelayout/flexfilelayout.c |  34 ++--
 fs/nfs/nfs3xdr.c                       |   5 +-
 fs/nfs/nfs4trace.c                     |   4 +
 fs/nfs/nfs4trace.h                     | 202 +++++++++++++++++----
 fs/nfs/nfs4xdr.c                       |   5 +-
 fs/nfs/nfstrace.h                      | 234 +++++++++++++++++--------
 fs/nfs/pnfs.h                          |   8 +-
 fs/nfs/pnfs_nfs.c                      |   7 +-
 fs/nfs/read.c                          |   7 +-
 fs/nfs/write.c                         |  29 ++-
 include/linux/sunrpc/auth.h            |   2 -
 include/linux/sunrpc/gss_api.h         |   3 -
 net/sunrpc/auth.c                      |  49 ------
 net/sunrpc/auth_gss/auth_gss.c         |   1 -
 net/sunrpc/auth_gss/gss_mech_switch.c  |  29 ---
 18 files changed, 415 insertions(+), 252 deletions(-)

-- 
2.24.1

[PATCH 01/15] NFS: Revalidate the file size on a fatal write error

[Trond Myklebust]

If we suffer a fatal error upon writing a file, which causes us to
need to revalidate the entire mapping, then we should also revalidate
the file size.

Fixes: d2ceb7e57086 ("NFS: Don't use page_file_mapping after removing the page")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/write.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 52cab65f91cf..f5170bc839aa 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -243,7 +243,15 @@ static void nfs_grow_file(struct page *page, unsigned int offset, unsigned int c
 /* A writeback failed: mark the page as bad, and invalidate the page cache */
 static void nfs_set_pageerror(struct address_space *mapping)
 {
+	struct inode *inode = mapping->host;
+
 	nfs_zap_mapping(mapping->host, mapping);
+	/* Force file size revalidation */
+	spin_lock(&inode->i_lock);
+	NFS_I(inode)->cache_validity |= NFS_INO_REVAL_FORCED |
+					NFS_INO_REVAL_PAGECACHE |
+					NFS_INO_INVALID_SIZE;
+	spin_unlock(&inode->i_lock);
 }
 
 static void nfs_mapping_set_error(struct page *page, int error)
-- 
2.24.1

[PATCH 02/15] NFS: Revalidate the file mapping on all fatal writeback errors

[Trond Myklebust]

If a write or commit failed, and the mapping sees a fatal error, we
need to revalidate the contents of that mapping.

Fixes: 06c9fdf3b9f1 ("NFS: On fatal writeback errors, we need to call nfs_inode_remove_request()")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/write.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index f5170bc839aa..83e6f691368c 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -256,8 +256,11 @@ static void nfs_set_pageerror(struct address_space *mapping)
 
 static void nfs_mapping_set_error(struct page *page, int error)
 {
+	struct address_space *mapping = page_file_mapping(page);
+
 	SetPageError(page);
-	mapping_set_error(page_file_mapping(page), error);
+	mapping_set_error(mapping, error);
+	nfs_set_pageerror(mapping);
 }
 
 /*
@@ -600,7 +603,6 @@ nfs_lock_and_join_requests(struct page *page)
 
 static void nfs_write_error(struct nfs_page *req, int error)
 {
-	nfs_set_pageerror(page_file_mapping(req->wb_page));
 	nfs_mapping_set_error(req->wb_page, error);
 	nfs_inode_remove_request(req);
 	nfs_end_page_writeback(req);
@@ -1006,7 +1008,6 @@ static void nfs_write_completion(struct nfs_pgio_header *hdr)
 		nfs_list_remove_request(req);
 		if (test_bit(NFS_IOHDR_ERROR, &hdr->flags) &&
 		    (hdr->good_bytes < bytes)) {
-			nfs_set_pageerror(page_file_mapping(req->wb_page));
 			nfs_mapping_set_error(req->wb_page, hdr->error);
 			goto remove_req;
 		}
-- 
2.24.1

[PATCH 03/15] SUNRPC: Remove broken gss_mech_list_pseudoflavors()

[Trond Myklebust]

Remove gss_mech_list_pseudoflavors() and its callers. This is part of
an unused API, and could leak an RCU reference if it were ever called.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 include/linux/sunrpc/auth.h           |  2 --
 include/linux/sunrpc/gss_api.h        |  3 --
 net/sunrpc/auth.c                     | 49 ---------------------------
 net/sunrpc/auth_gss/auth_gss.c        |  1 -
 net/sunrpc/auth_gss/gss_mech_switch.c | 29 ----------------
 5 files changed, 84 deletions(-)

diff --git a/include/linux/sunrpc/auth.h b/include/linux/sunrpc/auth.h
index e9ec742796e7..4f6b28487f28 100644
--- a/include/linux/sunrpc/auth.h
+++ b/include/linux/sunrpc/auth.h
@@ -113,7 +113,6 @@ struct rpc_authops {
 	int			(*hash_cred)(struct auth_cred *, unsigned int);
 	struct rpc_cred *	(*lookup_cred)(struct rpc_auth *, struct auth_cred *, int);
 	struct rpc_cred *	(*crcreate)(struct rpc_auth*, struct auth_cred *, int, gfp_t);
-	int			(*list_pseudoflavors)(rpc_authflavor_t *, int);
 	rpc_authflavor_t	(*info2flavor)(struct rpcsec_gss_info *);
 	int			(*flavor2info)(rpc_authflavor_t,
 						struct rpcsec_gss_info *);
@@ -158,7 +157,6 @@ rpc_authflavor_t	rpcauth_get_pseudoflavor(rpc_authflavor_t,
 				struct rpcsec_gss_info *);
 int			rpcauth_get_gssinfo(rpc_authflavor_t,
 				struct rpcsec_gss_info *);
-int			rpcauth_list_flavors(rpc_authflavor_t *, int);
 struct rpc_cred *	rpcauth_lookup_credcache(struct rpc_auth *, struct auth_cred *, int, gfp_t);
 void			rpcauth_init_cred(struct rpc_cred *, const struct auth_cred *, struct rpc_auth *, const struct rpc_credops *);
 struct rpc_cred *	rpcauth_lookupcred(struct rpc_auth *, int);
diff --git a/include/linux/sunrpc/gss_api.h b/include/linux/sunrpc/gss_api.h
index bd691e08be3b..4c4cb087f3e8 100644
--- a/include/linux/sunrpc/gss_api.h
+++ b/include/linux/sunrpc/gss_api.h
@@ -150,9 +150,6 @@ struct gss_api_mech *gss_mech_get_by_name(const char *);
 /* Similar, but get by pseudoflavor. */
 struct gss_api_mech *gss_mech_get_by_pseudoflavor(u32);
 
-/* Fill in an array with a list of supported pseudoflavors */
-int gss_mech_list_pseudoflavors(rpc_authflavor_t *, int);
-
 struct gss_api_mech * gss_mech_get(struct gss_api_mech *);
 
 /* For every successful gss_mech_get or gss_mech_get_by_* call there must be a
diff --git a/net/sunrpc/auth.c b/net/sunrpc/auth.c
index cdb05b48de44..5748ad0ba1bd 100644
--- a/net/sunrpc/auth.c
+++ b/net/sunrpc/auth.c
@@ -221,55 +221,6 @@ rpcauth_get_gssinfo(rpc_authflavor_t pseudoflavor, struct rpcsec_gss_info *info)
 }
 EXPORT_SYMBOL_GPL(rpcauth_get_gssinfo);
 
-/**
- * rpcauth_list_flavors - discover registered flavors and pseudoflavors
- * @array: array to fill in
- * @size: size of "array"
- *
- * Returns the number of array items filled in, or a negative errno.
- *
- * The returned array is not sorted by any policy.  Callers should not
- * rely on the order of the items in the returned array.
- */
-int
-rpcauth_list_flavors(rpc_authflavor_t *array, int size)
-{
-	const struct rpc_authops *ops;
-	rpc_authflavor_t flavor, pseudos[4];
-	int i, len, result = 0;
-
-	rcu_read_lock();
-	for (flavor = 0; flavor < RPC_AUTH_MAXFLAVOR; flavor++) {
-		ops = rcu_dereference(auth_flavors[flavor]);
-		if (result >= size) {
-			result = -ENOMEM;
-			break;
-		}
-
-		if (ops == NULL)
-			continue;
-		if (ops->list_pseudoflavors == NULL) {
-			array[result++] = ops->au_flavor;
-			continue;
-		}
-		len = ops->list_pseudoflavors(pseudos, ARRAY_SIZE(pseudos));
-		if (len < 0) {
-			result = len;
-			break;
-		}
-		for (i = 0; i < len; i++) {
-			if (result >= size) {
-				result = -ENOMEM;
-				break;
-			}
-			array[result++] = pseudos[i];
-		}
-	}
-	rcu_read_unlock();
-	return result;
-}
-EXPORT_SYMBOL_GPL(rpcauth_list_flavors);
-
 struct rpc_auth *
 rpcauth_create(const struct rpc_auth_create_args *args, struct rpc_clnt *clnt)
 {
diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index d75fddca44c9..24ca861815b1 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -2118,7 +2118,6 @@ static const struct rpc_authops authgss_ops = {
 	.hash_cred	= gss_hash_cred,
 	.lookup_cred	= gss_lookup_cred,
 	.crcreate	= gss_create_cred,
-	.list_pseudoflavors = gss_mech_list_pseudoflavors,
 	.info2flavor	= gss_mech_info2flavor,
 	.flavor2info	= gss_mech_flavor2info,
 };
diff --git a/net/sunrpc/auth_gss/gss_mech_switch.c b/net/sunrpc/auth_gss/gss_mech_switch.c
index 30b7de6f3d76..22d4be8e38c8 100644
--- a/net/sunrpc/auth_gss/gss_mech_switch.c
+++ b/net/sunrpc/auth_gss/gss_mech_switch.c
@@ -219,35 +219,6 @@ gss_mech_get_by_pseudoflavor(u32 pseudoflavor)
 	return gm;
 }
 
-/**
- * gss_mech_list_pseudoflavors - Discover registered GSS pseudoflavors
- * @array_ptr: array to fill in
- * @size: size of "array"
- *
- * Returns the number of array items filled in, or a negative errno.
- *
- * The returned array is not sorted by any policy.  Callers should not
- * rely on the order of the items in the returned array.
- */
-int gss_mech_list_pseudoflavors(rpc_authflavor_t *array_ptr, int size)
-{
-	struct gss_api_mech *pos = NULL;
-	int j, i = 0;
-
-	rcu_read_lock();
-	list_for_each_entry_rcu(pos, &registered_mechs, gm_list) {
-		for (j = 0; j < pos->gm_pf_num; j++) {
-			if (i >= size) {
-				spin_unlock(&registered_mechs_lock);
-				return -ENOMEM;
-			}
-			array_ptr[i++] = pos->gm_pfs[j].pseudoflavor;
-		}
-	}
-	rcu_read_unlock();
-	return i;
-}
-
 /**
  * gss_svc_to_pseudoflavor - map a GSS service number to a pseudoflavor
  * @gm: GSS mechanism handle
-- 
2.24.1

[PATCH 04/15] NFS: Fix up fsync() when the server rebooted

[Trond Myklebust]

Don't clear the NFS_CONTEXT_RESEND_WRITES flag until after calling
nfs_commit_inode(). Otherwise, if nfs_commit_inode() returns an
error, we end up with dirty pages in the page cache, but no tag
to tell us that those pages need resending.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/file.c | 37 ++++++++++++++++---------------------
 1 file changed, 16 insertions(+), 21 deletions(-)

diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 8eb731d9be3e..95a3445c8926 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -204,44 +204,39 @@ EXPORT_SYMBOL_GPL(nfs_file_mmap);
 static int
 nfs_file_fsync_commit(struct file *file, int datasync)
 {
-	struct nfs_open_context *ctx = nfs_file_open_context(file);
 	struct inode *inode = file_inode(file);
-	int do_resend, status;
-	int ret = 0;
+	int ret;
 
 	dprintk("NFS: fsync file(%pD2) datasync %d\n", file, datasync);
 
 	nfs_inc_stats(inode, NFSIOS_VFSFSYNC);
-	do_resend = test_and_clear_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags);
-	status = nfs_commit_inode(inode, FLUSH_SYNC);
-	if (status == 0)
-		status = file_check_and_advance_wb_err(file);
-	if (status < 0) {
-		ret = status;
-		goto out;
-	}
-	do_resend |= test_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags);
-	if (do_resend)
-		ret = -EAGAIN;
-out:
-	return ret;
+	ret = nfs_commit_inode(inode, FLUSH_SYNC);
+	if (ret < 0)
+		return ret;
+	return file_check_and_advance_wb_err(file);
 }
 
 int
 nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
-	int ret;
+	struct nfs_open_context *ctx = nfs_file_open_context(file);
 	struct inode *inode = file_inode(file);
+	int ret;
 
 	trace_nfs_fsync_enter(inode);
 
-	do {
+	for (;;) {
 		ret = file_write_and_wait_range(file, start, end);
 		if (ret != 0)
 			break;
 		ret = nfs_file_fsync_commit(file, datasync);
-		if (!ret)
-			ret = pnfs_sync_inode(inode, !!datasync);
+		if (ret != 0)
+			break;
+		ret = pnfs_sync_inode(inode, !!datasync);
+		if (ret != 0)
+			break;
+		if (!test_and_clear_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags))
+			break;
 		/*
 		 * If nfs_file_fsync_commit detected a server reboot, then
 		 * resend all dirty pages that might have been covered by
@@ -249,7 +244,7 @@ nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 		 */
 		start = 0;
 		end = LLONG_MAX;
-	} while (ret == -EAGAIN);
+	}
 
 	trace_nfs_fsync_exit(inode, ret);
 	return ret;
-- 
2.24.1

[PATCH 05/15] NFS/pnfs: Fix pnfs_generic_prepare_to_resend_writes()

[Trond Myklebust]

Instead of making assumptions about the commit verifier contents, change
the commit code to ensure we always check that the verifier was set
by the XDR code.

Fixes: f54bcf2ecee9 ("pnfs: Prepare for flexfiles by pulling out common code")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/direct.c   | 4 ++--
 fs/nfs/nfs3xdr.c  | 5 ++++-
 fs/nfs/nfs4xdr.c  | 5 ++++-
 fs/nfs/pnfs_nfs.c | 7 +++----
 fs/nfs/write.c    | 4 +++-
 5 files changed, 16 insertions(+), 9 deletions(-)

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 040a50fd9bf3..29f00da8a0b7 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -245,10 +245,10 @@ static int nfs_direct_cmp_commit_data_verf(struct nfs_direct_req *dreq,
 					 data->ds_commit_index);
 
 	/* verifier not set so always fail */
-	if (verfp->committed < 0)
+	if (verfp->committed < 0 || data->res.verf->committed <= NFS_UNSTABLE)
 		return 1;
 
-	return nfs_direct_cmp_verf(verfp, &data->verf);
+	return nfs_direct_cmp_verf(verfp, data->res.verf);
 }
 
 /**
diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c
index 927eb680f161..69971f6c840d 100644
--- a/fs/nfs/nfs3xdr.c
+++ b/fs/nfs/nfs3xdr.c
@@ -2334,6 +2334,7 @@ static int nfs3_xdr_dec_commit3res(struct rpc_rqst *req,
 				   void *data)
 {
 	struct nfs_commitres *result = data;
+	struct nfs_writeverf *verf = result->verf;
 	enum nfs_stat status;
 	int error;
 
@@ -2346,7 +2347,9 @@ static int nfs3_xdr_dec_commit3res(struct rpc_rqst *req,
 	result->op_status = status;
 	if (status != NFS3_OK)
 		goto out_status;
-	error = decode_writeverf3(xdr, &result->verf->verifier);
+	error = decode_writeverf3(xdr, &verf->verifier);
+	if (!error)
+		verf->committed = NFS_FILE_SYNC;
 out:
 	return error;
 out_status:
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 936c57779ff4..d0feef17db50 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -4313,11 +4313,14 @@ static int decode_write_verifier(struct xdr_stream *xdr, struct nfs_write_verifi
 
 static int decode_commit(struct xdr_stream *xdr, struct nfs_commitres *res)
 {
+	struct nfs_writeverf *verf = res->verf;
 	int status;
 
 	status = decode_op_hdr(xdr, OP_COMMIT);
 	if (!status)
-		status = decode_write_verifier(xdr, &res->verf->verifier);
+		status = decode_write_verifier(xdr, &verf->verifier);
+	if (!status)
+		verf->committed = NFS_FILE_SYNC;
 	return status;
 }
 
diff --git a/fs/nfs/pnfs_nfs.c b/fs/nfs/pnfs_nfs.c
index 82af4809b869..8b37e7f8e789 100644
--- a/fs/nfs/pnfs_nfs.c
+++ b/fs/nfs/pnfs_nfs.c
@@ -31,12 +31,11 @@ EXPORT_SYMBOL_GPL(pnfs_generic_rw_release);
 /* Fake up some data that will cause nfs_commit_release to retry the writes. */
 void pnfs_generic_prepare_to_resend_writes(struct nfs_commit_data *data)
 {
-	struct nfs_page *first = nfs_list_entry(data->pages.next);
+	struct nfs_writeverf *verf = data->res.verf;
 
 	data->task.tk_status = 0;
-	memcpy(&data->verf.verifier, &first->wb_verf,
-	       sizeof(data->verf.verifier));
-	data->verf.verifier.data[0]++; /* ensure verifier mismatch */
+	memset(&verf->verifier, 0, sizeof(verf->verifier));
+	verf->committed = NFS_UNSTABLE;
 }
 EXPORT_SYMBOL_GPL(pnfs_generic_prepare_to_resend_writes);
 
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 83e6f691368c..ab3d4611f0aa 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1838,6 +1838,7 @@ static void nfs_commit_done(struct rpc_task *task, void *calldata)
 
 static void nfs_commit_release_pages(struct nfs_commit_data *data)
 {
+	const struct nfs_writeverf *verf = data->res.verf;
 	struct nfs_page	*req;
 	int status = data->task.tk_status;
 	struct nfs_commit_info cinfo;
@@ -1865,7 +1866,8 @@ static void nfs_commit_release_pages(struct nfs_commit_data *data)
 
 		/* Okay, COMMIT succeeded, apparently. Check the verifier
 		 * returned by the server against all stored verfs. */
-		if (!nfs_write_verifier_cmp(&req->wb_verf, &data->verf.verifier)) {
+		if (verf->committed > NFS_UNSTABLE &&
+		    !nfs_write_verifier_cmp(&req->wb_verf, &verf->verifier)) {
 			/* We have a match */
 			if (req->wb_page)
 				nfs_inode_remove_request(req);
-- 
2.24.1

[PATCH 06/15] NFSv4: Improve read/write/commit tracing

[Trond Myklebust]

Ensure we always return the number of bytes read/written. Also display
the pnfs filehandle if it is in use.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/nfs4trace.h | 52 ++++++++++++++++++++++++++++++----------------
 1 file changed, 34 insertions(+), 18 deletions(-)

diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h
index e60b6fbd5ada..a3155b7f8063 100644
--- a/fs/nfs/nfs4trace.h
+++ b/fs/nfs/nfs4trace.h
@@ -1694,7 +1694,8 @@ DECLARE_EVENT_CLASS(nfs4_read_event,
 			__field(u32, fhandle)
 			__field(u64, fileid)
 			__field(loff_t, offset)
-			__field(size_t, count)
+			__field(u32, arg_count)
+			__field(u32, res_count)
 			__field(unsigned long, error)
 			__field(int, stateid_seq)
 			__field(u32, stateid_hash)
@@ -1702,13 +1703,18 @@ DECLARE_EVENT_CLASS(nfs4_read_event,
 
 		TP_fast_assign(
 			const struct inode *inode = hdr->inode;
+			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = hdr->args.fh ?
+						  hdr->args.fh : &nfsi->fh;
 			const struct nfs4_state *state =
 				hdr->args.context->state;
+
 			__entry->dev = inode->i_sb->s_dev;
-			__entry->fileid = NFS_FILEID(inode);
-			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(fh);
 			__entry->offset = hdr->args.offset;
-			__entry->count = hdr->args.count;
+			__entry->arg_count = hdr->args.count;
+			__entry->res_count = hdr->res.count;
 			__entry->error = error < 0 ? -error : 0;
 			__entry->stateid_seq =
 				be32_to_cpu(state->stateid.seqid);
@@ -1718,14 +1724,14 @@ DECLARE_EVENT_CLASS(nfs4_read_event,
 
 		TP_printk(
 			"error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld count=%zu stateid=%d:0x%08x",
+			"offset=%lld count=%u res=%u stateid=%d:0x%08x",
 			-__entry->error,
 			show_nfsv4_errors(__entry->error),
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
 			(long long)__entry->offset,
-			__entry->count,
+			__entry->arg_count, __entry->res_count,
 			__entry->stateid_seq, __entry->stateid_hash
 		)
 );
@@ -1754,7 +1760,8 @@ DECLARE_EVENT_CLASS(nfs4_write_event,
 			__field(u32, fhandle)
 			__field(u64, fileid)
 			__field(loff_t, offset)
-			__field(size_t, count)
+			__field(u32, arg_count)
+			__field(u32, res_count)
 			__field(unsigned long, error)
 			__field(int, stateid_seq)
 			__field(u32, stateid_hash)
@@ -1762,13 +1769,18 @@ DECLARE_EVENT_CLASS(nfs4_write_event,
 
 		TP_fast_assign(
 			const struct inode *inode = hdr->inode;
+			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = hdr->args.fh ?
+						  hdr->args.fh : &nfsi->fh;
 			const struct nfs4_state *state =
 				hdr->args.context->state;
+
 			__entry->dev = inode->i_sb->s_dev;
-			__entry->fileid = NFS_FILEID(inode);
-			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(fh);
 			__entry->offset = hdr->args.offset;
-			__entry->count = hdr->args.count;
+			__entry->arg_count = hdr->args.count;
+			__entry->res_count = hdr->res.count;
 			__entry->error = error < 0 ? -error : 0;
 			__entry->stateid_seq =
 				be32_to_cpu(state->stateid.seqid);
@@ -1778,14 +1790,14 @@ DECLARE_EVENT_CLASS(nfs4_write_event,
 
 		TP_printk(
 			"error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld count=%zu stateid=%d:0x%08x",
+			"offset=%lld count=%u res=%u stateid=%d:0x%08x",
 			-__entry->error,
 			show_nfsv4_errors(__entry->error),
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
 			(long long)__entry->offset,
-			__entry->count,
+			__entry->arg_count, __entry->res_count,
 			__entry->stateid_seq, __entry->stateid_hash
 		)
 );
@@ -1814,24 +1826,28 @@ DECLARE_EVENT_CLASS(nfs4_commit_event,
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
-			__field(loff_t, offset)
-			__field(size_t, count)
 			__field(unsigned long, error)
+			__field(loff_t, offset)
+			__field(u32, count)
 		),
 
 		TP_fast_assign(
 			const struct inode *inode = data->inode;
+			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = data->args.fh ?
+						  data->args.fh : &nfsi->fh;
+
 			__entry->dev = inode->i_sb->s_dev;
-			__entry->fileid = NFS_FILEID(inode);
-			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(fh);
 			__entry->offset = data->args.offset;
 			__entry->count = data->args.count;
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 		),
 
 		TP_printk(
 			"error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld count=%zu",
+			"offset=%lld count=%u",
 			-__entry->error,
 			show_nfsv4_errors(__entry->error),
 			MAJOR(__entry->dev), MINOR(__entry->dev),
-- 
2.24.1

[PATCH 07/15] NFS: Fix fix of show_nfs_errors

[Trond Myklebust]

Casting a negative value to an unsigned long is not the same as
converting it to its absolute value.

Fixes: 96650e2effa2 ("NFS: Fix show_nfs_errors macros again")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/nfs4trace.h | 33 +++++++++++++++++----------------
 1 file changed, 17 insertions(+), 16 deletions(-)

diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h
index a3155b7f8063..a291877c0c32 100644
--- a/fs/nfs/nfs4trace.h
+++ b/fs/nfs/nfs4trace.h
@@ -352,7 +352,7 @@ DECLARE_EVENT_CLASS(nfs4_clientid_event,
 		),
 
 		TP_fast_assign(
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__assign_str(dstaddr, clp->cl_hostname);
 		),
 
@@ -432,7 +432,8 @@ TRACE_EVENT(nfs4_sequence_done,
 			__entry->target_highest_slotid =
 					res->sr_target_highest_slotid;
 			__entry->status_flags = res->sr_status_flags;
-			__entry->error = res->sr_status;
+			__entry->error = res->sr_status < 0 ?
+					-res->sr_status : 0;
 		),
 		TP_printk(
 			"error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u "
@@ -640,7 +641,7 @@ TRACE_EVENT(nfs4_state_mgr_failed,
 		),
 
 		TP_fast_assign(
-			__entry->error = status;
+			__entry->error = status < 0 ? -status : 0;
 			__entry->state = clp->cl_state;
 			__assign_str(hostname, clp->cl_hostname);
 			__assign_str(section, section);
@@ -659,7 +660,7 @@ TRACE_EVENT(nfs4_xdr_status,
 		TP_PROTO(
 			const struct xdr_stream *xdr,
 			u32 op,
-			int error
+			u32 error
 		),
 
 		TP_ARGS(xdr, op, error),
@@ -849,7 +850,7 @@ TRACE_EVENT(nfs4_close,
 			__entry->fileid = NFS_FILEID(inode);
 			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
 			__entry->fmode = (__force unsigned int)state->state;
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->stateid_seq =
 				be32_to_cpu(args->stateid.seqid);
 			__entry->stateid_hash =
@@ -914,7 +915,7 @@ DECLARE_EVENT_CLASS(nfs4_lock_event,
 		TP_fast_assign(
 			const struct inode *inode = state->inode;
 
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->cmd = cmd;
 			__entry->type = request->fl_type;
 			__entry->start = request->fl_start;
@@ -986,7 +987,7 @@ TRACE_EVENT(nfs4_set_lock,
 		TP_fast_assign(
 			const struct inode *inode = state->inode;
 
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->cmd = cmd;
 			__entry->type = request->fl_type;
 			__entry->start = request->fl_start;
@@ -1164,7 +1165,7 @@ TRACE_EVENT(nfs4_delegreturn_exit,
 		TP_fast_assign(
 			__entry->dev = res->server->s_dev;
 			__entry->fhandle = nfs_fhandle_hash(args->fhandle);
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->stateid_seq =
 				be32_to_cpu(args->stateid->seqid);
 			__entry->stateid_hash =
@@ -1204,7 +1205,7 @@ DECLARE_EVENT_CLASS(nfs4_test_stateid_event,
 		TP_fast_assign(
 			const struct inode *inode = state->inode;
 
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = NFS_FILEID(inode);
 			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
@@ -1306,7 +1307,7 @@ TRACE_EVENT(nfs4_lookupp,
 		TP_fast_assign(
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->ino = NFS_FILEID(inode);
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 		),
 
 		TP_printk(
@@ -1342,7 +1343,7 @@ TRACE_EVENT(nfs4_rename,
 			__entry->dev = olddir->i_sb->s_dev;
 			__entry->olddir = NFS_FILEID(olddir);
 			__entry->newdir = NFS_FILEID(newdir);
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__assign_str(oldname, oldname->name);
 			__assign_str(newname, newname->name);
 		),
@@ -1433,7 +1434,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_stateid_event,
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = NFS_FILEID(inode);
 			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->stateid_seq =
 				be32_to_cpu(stateid->seqid);
 			__entry->stateid_hash =
@@ -1489,7 +1490,7 @@ DECLARE_EVENT_CLASS(nfs4_getattr_event,
 			__entry->valid = fattr->valid;
 			__entry->fhandle = nfs_fhandle_hash(fhandle);
 			__entry->fileid = (fattr->valid & NFS_ATTR_FATTR_FILEID) ? fattr->fileid : 0;
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 		),
 
 		TP_printk(
@@ -1536,7 +1537,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_callback_event,
 		),
 
 		TP_fast_assign(
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->fhandle = nfs_fhandle_hash(fhandle);
 			if (!IS_ERR_OR_NULL(inode)) {
 				__entry->fileid = NFS_FILEID(inode);
@@ -1593,7 +1594,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_stateid_callback_event,
 		),
 
 		TP_fast_assign(
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->fhandle = nfs_fhandle_hash(fhandle);
 			if (!IS_ERR_OR_NULL(inode)) {
 				__entry->fileid = NFS_FILEID(inode);
@@ -1912,7 +1913,7 @@ TRACE_EVENT(nfs4_layoutget,
 			__entry->iomode = args->iomode;
 			__entry->offset = args->offset;
 			__entry->count = args->length;
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->stateid_seq =
 				be32_to_cpu(state->stateid.seqid);
 			__entry->stateid_hash =
-- 
2.24.1

[PATCH 08/15] pNFS/flexfiles: Record resend attempts on I/O failure

[Trond Myklebust]

If the attempt to do pNFS fails, then record what action we
take to recover (resend, reset to pnfs or reset to mds).

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/flexfilelayout/flexfilelayout.c | 6 +++---
 fs/nfs/nfs4trace.h                     | 8 +++++++-
 fs/nfs/pnfs.h                          | 8 ++++----
 3 files changed, 14 insertions(+), 8 deletions(-)

diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c
index 5657b7f2611f..3163b78b1d2c 100644
--- a/fs/nfs/flexfilelayout/flexfilelayout.c
+++ b/fs/nfs/flexfilelayout/flexfilelayout.c
@@ -1321,7 +1321,6 @@ static int ff_layout_read_done_cb(struct rpc_task *task,
 	int new_idx = hdr->pgio_mirror_idx;
 	int err;
 
-	trace_nfs4_pnfs_read(hdr, task->tk_status);
 	if (task->tk_status < 0)
 		ff_layout_io_track_ds_error(hdr->lseg, hdr->pgio_mirror_idx,
 					    hdr->args.offset, hdr->args.count,
@@ -1331,6 +1330,7 @@ static int ff_layout_read_done_cb(struct rpc_task *task,
 					   hdr->ds_clp, hdr->lseg,
 					   hdr->pgio_mirror_idx);
 
+	trace_nfs4_pnfs_read(hdr, err);
 	clear_bit(NFS_IOHDR_RESEND_PNFS, &hdr->flags);
 	clear_bit(NFS_IOHDR_RESEND_MDS, &hdr->flags);
 	switch (err) {
@@ -1494,7 +1494,6 @@ static int ff_layout_write_done_cb(struct rpc_task *task,
 	loff_t end_offs = 0;
 	int err;
 
-	trace_nfs4_pnfs_write(hdr, task->tk_status);
 	if (task->tk_status < 0)
 		ff_layout_io_track_ds_error(hdr->lseg, hdr->pgio_mirror_idx,
 					    hdr->args.offset, hdr->args.count,
@@ -1504,6 +1503,7 @@ static int ff_layout_write_done_cb(struct rpc_task *task,
 					   hdr->ds_clp, hdr->lseg,
 					   hdr->pgio_mirror_idx);
 
+	trace_nfs4_pnfs_write(hdr, err);
 	clear_bit(NFS_IOHDR_RESEND_PNFS, &hdr->flags);
 	clear_bit(NFS_IOHDR_RESEND_MDS, &hdr->flags);
 	switch (err) {
@@ -1537,7 +1537,6 @@ static int ff_layout_commit_done_cb(struct rpc_task *task,
 {
 	int err;
 
-	trace_nfs4_pnfs_commit_ds(data, task->tk_status);
 	if (task->tk_status < 0)
 		ff_layout_io_track_ds_error(data->lseg, data->ds_commit_index,
 					    data->args.offset, data->args.count,
@@ -1546,6 +1545,7 @@ static int ff_layout_commit_done_cb(struct rpc_task *task,
 	err = ff_layout_async_handle_error(task, NULL, data->ds_clp,
 					   data->lseg, data->ds_commit_index);
 
+	trace_nfs4_pnfs_commit_ds(data, err);
 	switch (err) {
 	case -NFS4ERR_RESET_TO_PNFS:
 		pnfs_generic_prepare_to_resend_writes(data);
diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h
index a291877c0c32..a467f49989f2 100644
--- a/fs/nfs/nfs4trace.h
+++ b/fs/nfs/nfs4trace.h
@@ -155,6 +155,9 @@ TRACE_DEFINE_ENUM(NFS4ERR_WRONG_CRED);
 TRACE_DEFINE_ENUM(NFS4ERR_WRONG_TYPE);
 TRACE_DEFINE_ENUM(NFS4ERR_XDEV);
 
+TRACE_DEFINE_ENUM(NFS4ERR_RESET_TO_MDS);
+TRACE_DEFINE_ENUM(NFS4ERR_RESET_TO_PNFS);
+
 #define show_nfsv4_errors(error) \
 	__print_symbolic(error, \
 		{ NFS4_OK, "OK" }, \
@@ -305,7 +308,10 @@ TRACE_DEFINE_ENUM(NFS4ERR_XDEV);
 		{ NFS4ERR_WRONGSEC, "WRONGSEC" }, \
 		{ NFS4ERR_WRONG_CRED, "WRONG_CRED" }, \
 		{ NFS4ERR_WRONG_TYPE, "WRONG_TYPE" }, \
-		{ NFS4ERR_XDEV, "XDEV" })
+		{ NFS4ERR_XDEV, "XDEV" }, \
+		/* ***** Internal to Linux NFS client ***** */ \
+		{ NFS4ERR_RESET_TO_MDS, "RESET_TO_MDS" }, \
+		{ NFS4ERR_RESET_TO_PNFS, "RESET_TO_PNFS" })
 
 #define show_open_flags(flags) \
 	__print_flags(flags, "|", \
diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h
index f8a38065c7e4..0fafdadc9c8d 100644
--- a/fs/nfs/pnfs.h
+++ b/fs/nfs/pnfs.h
@@ -79,6 +79,10 @@ enum pnfs_try_status {
 	PNFS_TRY_AGAIN     = 2,
 };
 
+/* error codes for internal use */
+#define NFS4ERR_RESET_TO_MDS   12001
+#define NFS4ERR_RESET_TO_PNFS  12002
+
 #ifdef CONFIG_NFS_V4_1
 
 #define LAYOUT_NFSV4_1_MODULE_PREFIX "nfs-layouttype4"
@@ -91,10 +95,6 @@ enum pnfs_try_status {
 #define NFS4_DEF_DS_RETRANS 5
 #define PNFS_DEVICE_RETRY_TIMEOUT (120*HZ)
 
-/* error codes for internal use */
-#define NFS4ERR_RESET_TO_MDS   12001
-#define NFS4ERR_RESET_TO_PNFS  12002
-
 enum {
 	NFS_LAYOUT_RO_FAILED = 0,	/* get ro layout failed stop trying */
 	NFS_LAYOUT_RW_FAILED,		/* get rw layout failed stop trying */
-- 
2.24.1

[PATCH 09/15] NFS: Clean up generic file read tracepoints

[Trond Myklebust]

Clean up the generic file read tracepoints so they do pass the
full structures as arguments. Also ensure we report the number
of bytes actually read.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/nfstrace.h | 56 ++++++++++++++++++++++++++++-------------------
 fs/nfs/read.c     |  5 ++---
 2 files changed, 35 insertions(+), 26 deletions(-)

diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h
index f64a33d2a1d1..f0e869d21368 100644
--- a/fs/nfs/nfstrace.h
+++ b/fs/nfs/nfstrace.h
@@ -820,75 +820,85 @@ TRACE_EVENT(nfs_sillyrename_unlink,
 
 TRACE_EVENT(nfs_initiate_read,
 		TP_PROTO(
-			const struct inode *inode,
-			loff_t offset, unsigned long count
+			const struct nfs_pgio_header *hdr
 		),
 
-		TP_ARGS(inode, offset, count),
+		TP_ARGS(hdr),
 
 		TP_STRUCT__entry(
-			__field(loff_t, offset)
-			__field(unsigned long, count)
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(u32, count)
 		),
 
 		TP_fast_assign(
+			const struct inode *inode = hdr->inode;
 			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = hdr->args.fh ?
+						  hdr->args.fh : &nfsi->fh;
 
-			__entry->offset = offset;
-			__entry->count = count;
+			__entry->offset = hdr->args.offset;
+			__entry->count = hdr->args.count;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = nfsi->fileid;
-			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->fhandle = nfs_fhandle_hash(fh);
 		),
 
 		TP_printk(
 			"fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld count=%lu",
+			"offset=%lld count=%u",
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
-			__entry->offset, __entry->count
+			(long long)__entry->offset, __entry->count
 		)
 );
 
 TRACE_EVENT(nfs_readpage_done,
 		TP_PROTO(
-			const struct inode *inode,
-			int status, loff_t offset, bool eof
+			const struct rpc_task *task,
+			const struct nfs_pgio_header *hdr
 		),
 
-		TP_ARGS(inode, status, offset, eof),
+		TP_ARGS(task, hdr),
 
 		TP_STRUCT__entry(
-			__field(int, status)
-			__field(loff_t, offset)
-			__field(bool, eof)
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(u32, arg_count)
+			__field(u32, res_count)
+			__field(bool, eof)
+			__field(int, status)
 		),
 
 		TP_fast_assign(
+			const struct inode *inode = hdr->inode;
 			const struct nfs_inode *nfsi = NFS_I(inode);
-
-			__entry->status = status;
-			__entry->offset = offset;
-			__entry->eof = eof;
+			const struct nfs_fh *fh = hdr->args.fh ?
+						  hdr->args.fh : &nfsi->fh;
+
+			__entry->status = task->tk_status;
+			__entry->offset = hdr->args.offset;
+			__entry->arg_count = hdr->args.count;
+			__entry->res_count = hdr->res.count;
+			__entry->eof = hdr->res.eof;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = nfsi->fileid;
-			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->fhandle = nfs_fhandle_hash(fh);
 		),
 
 		TP_printk(
 			"fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld status=%d%s",
+			"offset=%lld count=%u res=%u status=%d%s",
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
-			__entry->offset, __entry->status,
+			(long long)__entry->offset, __entry->arg_count,
+			__entry->res_count, __entry->status,
 			__entry->eof ? " eof" : ""
 		)
 );
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index cfe0b586eadd..12deb3bdb2a0 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -214,7 +214,7 @@ static void nfs_initiate_read(struct nfs_pgio_header *hdr,
 
 	task_setup_data->flags |= swap_flags;
 	rpc_ops->read_setup(hdr, msg);
-	trace_nfs_initiate_read(inode, hdr->io_start, hdr->good_bytes);
+	trace_nfs_initiate_read(hdr);
 }
 
 static void
@@ -247,8 +247,7 @@ static int nfs_readpage_done(struct rpc_task *task,
 		return status;
 
 	nfs_add_stats(inode, NFSIOS_SERVERREADBYTES, hdr->res.count);
-	trace_nfs_readpage_done(inode, task->tk_status,
-				hdr->args.offset, hdr->res.eof);
+	trace_nfs_readpage_done(task, hdr);
 
 	if (task->tk_status == -ESTALE) {
 		set_bit(NFS_INO_STALE, &NFS_I(inode)->flags);
-- 
2.24.1

[PATCH 10/15] NFS: Clean up generic writeback tracepoints

[Trond Myklebust]

Clean up the generic writeback tracepoints so they do pass the
full structures as arguments. Also ensure we report the number
of bytes actually written.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/nfstrace.h | 73 ++++++++++++++++++++++++++---------------------
 fs/nfs/write.c    |  6 ++--
 2 files changed, 43 insertions(+), 36 deletions(-)

diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h
index f0e869d21368..3d8d36fe7b6f 100644
--- a/fs/nfs/nfstrace.h
+++ b/fs/nfs/nfstrace.h
@@ -915,87 +915,96 @@ TRACE_DEFINE_ENUM(NFS_FILE_SYNC);
 
 TRACE_EVENT(nfs_initiate_write,
 		TP_PROTO(
-			const struct inode *inode,
-			loff_t offset, unsigned long count,
-			enum nfs3_stable_how stable
+			const struct nfs_pgio_header *hdr
 		),
 
-		TP_ARGS(inode, offset, count, stable),
+		TP_ARGS(hdr),
 
 		TP_STRUCT__entry(
-			__field(loff_t, offset)
-			__field(unsigned long, count)
-			__field(enum nfs3_stable_how, stable)
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(u32, count)
+			__field(enum nfs3_stable_how, stable)
 		),
 
 		TP_fast_assign(
+			const struct inode *inode = hdr->inode;
 			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = hdr->args.fh ?
+						  hdr->args.fh : &nfsi->fh;
 
-			__entry->offset = offset;
-			__entry->count = count;
-			__entry->stable = stable;
+			__entry->offset = hdr->args.offset;
+			__entry->count = hdr->args.count;
+			__entry->stable = hdr->args.stable;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = nfsi->fileid;
-			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->fhandle = nfs_fhandle_hash(fh);
 		),
 
 		TP_printk(
 			"fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld count=%lu stable=%s",
+			"offset=%lld count=%u stable=%s",
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
-			__entry->offset, __entry->count,
+			(long long)__entry->offset, __entry->count,
 			nfs_show_stable(__entry->stable)
 		)
 );
 
 TRACE_EVENT(nfs_writeback_done,
 		TP_PROTO(
-			const struct inode *inode,
-			int status,
-			loff_t offset,
-			struct nfs_writeverf *writeverf
+			const struct rpc_task *task,
+			const struct nfs_pgio_header *hdr
 		),
 
-		TP_ARGS(inode, status, offset, writeverf),
+		TP_ARGS(task, hdr),
 
 		TP_STRUCT__entry(
-			__field(int, status)
-			__field(loff_t, offset)
-			__field(enum nfs3_stable_how, stable)
-			__field(unsigned long long, verifier)
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(u32, arg_count)
+			__field(u32, res_count)
+			__field(int, status)
+			__field(enum nfs3_stable_how, stable)
+			__array(char, verifier, NFS4_VERIFIER_SIZE)
 		),
 
 		TP_fast_assign(
+			const struct inode *inode = hdr->inode;
 			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = hdr->args.fh ?
+						  hdr->args.fh : &nfsi->fh;
+			const struct nfs_writeverf *verf = hdr->res.verf;
 
-			__entry->status = status;
-			__entry->offset = offset;
-			__entry->stable = writeverf->committed;
-			memcpy(&__entry->verifier, &writeverf->verifier,
-			       sizeof(__entry->verifier));
+			__entry->status = task->tk_status;
+			__entry->offset = hdr->args.offset;
+			__entry->arg_count = hdr->args.count;
+			__entry->res_count = hdr->res.count;
+			__entry->stable = verf->committed;
+			memcpy(__entry->verifier,
+				&verf->verifier,
+				NFS4_VERIFIER_SIZE);
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = nfsi->fileid;
-			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->fhandle = nfs_fhandle_hash(fh);
 		),
 
 		TP_printk(
 			"fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld status=%d stable=%s "
-			"verifier 0x%016llx",
+			"offset=%lld count=%u res=%u status=%d stable=%s "
+			"verifier=%s",
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
-			__entry->offset, __entry->status,
+			(long long)__entry->offset, __entry->arg_count,
+			__entry->res_count, __entry->status,
 			nfs_show_stable(__entry->stable),
-			__entry->verifier
+			__print_hex_str(__entry->verifier, NFS4_VERIFIER_SIZE)
 		)
 );
 
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index ab3d4611f0aa..d0f62aef3489 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1412,8 +1412,7 @@ static void nfs_initiate_write(struct nfs_pgio_header *hdr,
 
 	task_setup_data->priority = priority;
 	rpc_ops->write_setup(hdr, msg, &task_setup_data->rpc_client);
-	trace_nfs_initiate_write(hdr->inode, hdr->io_start, hdr->good_bytes,
-				 hdr->args.stable);
+	trace_nfs_initiate_write(hdr);
 }
 
 /* If a nfs_flush_* function fails, it should remove reqs from @head and
@@ -1577,8 +1576,7 @@ static int nfs_writeback_done(struct rpc_task *task,
 		return status;
 
 	nfs_add_stats(inode, NFSIOS_SERVERWRITTENBYTES, hdr->res.count);
-	trace_nfs_writeback_done(inode, task->tk_status,
-				 hdr->args.offset, hdr->res.verf);
+	trace_nfs_writeback_done(task, hdr);
 
 	if (hdr->res.verf->committed < hdr->args.stable &&
 	    task->tk_status >= 0) {
-- 
2.24.1

[PATCH 11/15] NFS: Clean up generic file commit tracepoint

[Trond Myklebust]

Clean up the generic file commit tracepoints to use a 64-bit value
for the verifier, and to display the pNFS filehandle, if it exists.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/nfstrace.h | 42 ++++++++++++++++++++++++++----------------
 fs/nfs/write.c    |  2 +-
 2 files changed, 27 insertions(+), 17 deletions(-)

diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h
index 3d8d36fe7b6f..51043f02e86f 100644
--- a/fs/nfs/nfstrace.h
+++ b/fs/nfs/nfstrace.h
@@ -1016,71 +1016,81 @@ TRACE_EVENT(nfs_initiate_commit,
 		TP_ARGS(data),
 
 		TP_STRUCT__entry(
-			__field(loff_t, offset)
-			__field(unsigned long, count)
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(u32, count)
 		),
 
 		TP_fast_assign(
 			const struct inode *inode = data->inode;
 			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = data->args.fh ?
+						  data->args.fh : &nfsi->fh;
 
 			__entry->offset = data->args.offset;
 			__entry->count = data->args.count;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = nfsi->fileid;
-			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->fhandle = nfs_fhandle_hash(fh);
 		),
 
 		TP_printk(
 			"fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld count=%lu",
+			"offset=%lld count=%u",
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
-			__entry->offset, __entry->count
+			(long long)__entry->offset, __entry->count
 		)
 );
 
 TRACE_EVENT(nfs_commit_done,
 		TP_PROTO(
+			const struct rpc_task *task,
 			const struct nfs_commit_data *data
 		),
 
-		TP_ARGS(data),
+		TP_ARGS(task, data),
 
 		TP_STRUCT__entry(
-			__field(int, status)
-			__field(loff_t, offset)
-			__field(unsigned long long, verifier)
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(int, status)
+			__field(enum nfs3_stable_how, stable)
+			__array(char, verifier, NFS4_VERIFIER_SIZE)
 		),
 
 		TP_fast_assign(
 			const struct inode *inode = data->inode;
 			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = data->args.fh ?
+						  data->args.fh : &nfsi->fh;
+			const struct nfs_writeverf *verf = data->res.verf;
 
-			__entry->status = data->res.op_status;
+			__entry->status = task->tk_status;
 			__entry->offset = data->args.offset;
-			memcpy(&__entry->verifier, &data->verf.verifier,
-			       sizeof(__entry->verifier));
+			__entry->stable = verf->committed;
+			memcpy(__entry->verifier,
+				&verf->verifier,
+				NFS4_VERIFIER_SIZE);
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = nfsi->fileid;
-			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->fhandle = nfs_fhandle_hash(fh);
 		),
 
 		TP_printk(
 			"fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld status=%d verifier 0x%016llx",
+			"offset=%lld status=%d stable=%s verifier=%s",
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
-			__entry->offset, __entry->status,
-			__entry->verifier
+			(long long)__entry->offset, __entry->status,
+			nfs_show_stable(__entry->stable),
+			__print_hex_str(__entry->verifier, NFS4_VERIFIER_SIZE)
 		)
 );
 
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index d0f62aef3489..985ddff46051 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1831,7 +1831,7 @@ static void nfs_commit_done(struct rpc_task *task, void *calldata)
 
 	/* Call the NFS version-specific code */
 	NFS_PROTO(data->inode)->commit_done(task, data);
-	trace_nfs_commit_done(data);
+	trace_nfs_commit_done(task, data);
 }
 
 static void nfs_commit_release_pages(struct nfs_commit_data *data)
-- 
2.24.1

[PATCH 12/15] pNFS/flexfiles: Add tracing for layout errors

[Trond Myklebust]

Trace layout errors for pNFS/flexfiles on read/write/commit operations.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/flexfilelayout/flexfilelayout.c |  28 +++++--
 fs/nfs/nfs4trace.c                     |   4 +
 fs/nfs/nfs4trace.h                     | 109 +++++++++++++++++++++++++
 3 files changed, 132 insertions(+), 9 deletions(-)

diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c
index 3163b78b1d2c..bb9148b83166 100644
--- a/fs/nfs/flexfilelayout/flexfilelayout.c
+++ b/fs/nfs/flexfilelayout/flexfilelayout.c
@@ -1266,9 +1266,10 @@ static int ff_layout_async_handle_error(struct rpc_task *task,
 
 static void ff_layout_io_track_ds_error(struct pnfs_layout_segment *lseg,
 					int idx, u64 offset, u64 length,
-					u32 status, int opnum, int error)
+					u32 *op_status, int opnum, int error)
 {
 	struct nfs4_ff_layout_mirror *mirror;
+	u32 status = *op_status;
 	int err;
 
 	if (status == 0) {
@@ -1286,10 +1287,10 @@ static void ff_layout_io_track_ds_error(struct pnfs_layout_segment *lseg,
 		case -ENOBUFS:
 		case -EPIPE:
 		case -EPERM:
-			status = NFS4ERR_NXIO;
+			*op_status = status = NFS4ERR_NXIO;
 			break;
 		case -EACCES:
-			status = NFS4ERR_ACCESS;
+			*op_status = status = NFS4ERR_ACCESS;
 			break;
 		default:
 			return;
@@ -1321,11 +1322,14 @@ static int ff_layout_read_done_cb(struct rpc_task *task,
 	int new_idx = hdr->pgio_mirror_idx;
 	int err;
 
-	if (task->tk_status < 0)
+	if (task->tk_status < 0) {
 		ff_layout_io_track_ds_error(hdr->lseg, hdr->pgio_mirror_idx,
 					    hdr->args.offset, hdr->args.count,
-					    hdr->res.op_status, OP_READ,
+					    &hdr->res.op_status, OP_READ,
 					    task->tk_status);
+		trace_ff_layout_read_error(hdr);
+	}
+
 	err = ff_layout_async_handle_error(task, hdr->args.context->state,
 					   hdr->ds_clp, hdr->lseg,
 					   hdr->pgio_mirror_idx);
@@ -1494,11 +1498,14 @@ static int ff_layout_write_done_cb(struct rpc_task *task,
 	loff_t end_offs = 0;
 	int err;
 
-	if (task->tk_status < 0)
+	if (task->tk_status < 0) {
 		ff_layout_io_track_ds_error(hdr->lseg, hdr->pgio_mirror_idx,
 					    hdr->args.offset, hdr->args.count,
-					    hdr->res.op_status, OP_WRITE,
+					    &hdr->res.op_status, OP_WRITE,
 					    task->tk_status);
+		trace_ff_layout_write_error(hdr);
+	}
+
 	err = ff_layout_async_handle_error(task, hdr->args.context->state,
 					   hdr->ds_clp, hdr->lseg,
 					   hdr->pgio_mirror_idx);
@@ -1537,11 +1544,14 @@ static int ff_layout_commit_done_cb(struct rpc_task *task,
 {
 	int err;
 
-	if (task->tk_status < 0)
+	if (task->tk_status < 0) {
 		ff_layout_io_track_ds_error(data->lseg, data->ds_commit_index,
 					    data->args.offset, data->args.count,
-					    data->res.op_status, OP_COMMIT,
+					    &data->res.op_status, OP_COMMIT,
 					    task->tk_status);
+		trace_ff_layout_commit_error(data);
+	}
+
 	err = ff_layout_async_handle_error(task, NULL, data->ds_clp,
 					   data->lseg, data->ds_commit_index);
 
diff --git a/fs/nfs/nfs4trace.c b/fs/nfs/nfs4trace.c
index 1a8f376b3f73..d9ac556bebcf 100644
--- a/fs/nfs/nfs4trace.c
+++ b/fs/nfs/nfs4trace.c
@@ -24,4 +24,8 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_read_done);
 EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_write_done);
 EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_read_pagelist);
 EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_write_pagelist);
+
+EXPORT_TRACEPOINT_SYMBOL_GPL(ff_layout_read_error);
+EXPORT_TRACEPOINT_SYMBOL_GPL(ff_layout_write_error);
+EXPORT_TRACEPOINT_SYMBOL_GPL(ff_layout_commit_error);
 #endif
diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h
index a467f49989f2..d39897daa284 100644
--- a/fs/nfs/nfs4trace.h
+++ b/fs/nfs/nfs4trace.h
@@ -2117,6 +2117,115 @@ DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_write_done);
 DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_read_pagelist);
 DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_write_pagelist);
 
+DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event,
+		TP_PROTO(
+			const struct nfs_pgio_header *hdr
+		),
+
+		TP_ARGS(hdr),
+
+		TP_STRUCT__entry(
+			__field(unsigned long, error)
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(u32, count)
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
+			__string(dstaddr, hdr->ds_clp ?
+				rpc_peeraddr2str(hdr->ds_clp->cl_rpcclient,
+					RPC_DISPLAY_ADDR) : "unknown")
+		),
+
+		TP_fast_assign(
+			const struct inode *inode = hdr->inode;
+
+			__entry->error = hdr->res.op_status;
+			__entry->fhandle = nfs_fhandle_hash(hdr->args.fh);
+			__entry->fileid = NFS_FILEID(inode);
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->offset = hdr->args.offset;
+			__entry->count = hdr->args.count;
+			__entry->stateid_seq =
+				be32_to_cpu(hdr->args.stateid.seqid);
+			__entry->stateid_hash =
+				nfs_stateid_hash(&hdr->args.stateid);
+			__assign_str(dstaddr, hdr->ds_clp ?
+				rpc_peeraddr2str(hdr->ds_clp->cl_rpcclient,
+					RPC_DISPLAY_ADDR) : "unknown");
+		),
+
+		TP_printk(
+			"error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"offset=%llu count=%u stateid=%d:0x%08x dstaddr=%s",
+			-__entry->error,
+			show_nfsv4_errors(__entry->error),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			__entry->offset, __entry->count,
+			__entry->stateid_seq, __entry->stateid_hash,
+			__get_str(dstaddr)
+		)
+);
+
+#define DEFINE_NFS4_FLEXFILES_IO_EVENT(name) \
+	DEFINE_EVENT(nfs4_flexfiles_io_event, name, \
+			TP_PROTO( \
+				const struct nfs_pgio_header *hdr \
+			), \
+			TP_ARGS(hdr))
+DEFINE_NFS4_FLEXFILES_IO_EVENT(ff_layout_read_error);
+DEFINE_NFS4_FLEXFILES_IO_EVENT(ff_layout_write_error);
+
+TRACE_EVENT(ff_layout_commit_error,
+		TP_PROTO(
+			const struct nfs_commit_data *data
+		),
+
+		TP_ARGS(data),
+
+		TP_STRUCT__entry(
+			__field(unsigned long, error)
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(u32, count)
+			__string(dstaddr, data->ds_clp ?
+				rpc_peeraddr2str(data->ds_clp->cl_rpcclient,
+					RPC_DISPLAY_ADDR) : "unknown")
+		),
+
+		TP_fast_assign(
+			const struct inode *inode = data->inode;
+
+			__entry->error = data->res.op_status;
+			__entry->fhandle = nfs_fhandle_hash(data->args.fh);
+			__entry->fileid = NFS_FILEID(inode);
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->offset = data->args.offset;
+			__entry->count = data->args.count;
+			__assign_str(dstaddr, data->ds_clp ?
+				rpc_peeraddr2str(data->ds_clp->cl_rpcclient,
+					RPC_DISPLAY_ADDR) : "unknown");
+		),
+
+		TP_printk(
+			"error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"offset=%llu count=%u dstaddr=%s",
+			-__entry->error,
+			show_nfsv4_errors(__entry->error),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			__entry->offset, __entry->count,
+			__get_str(dstaddr)
+		)
+);
+
+
 #endif /* CONFIG_NFS_V4_1 */
 
 #endif /* _TRACE_NFS4_H */
-- 
2.24.1

[PATCH 13/15] NFS: Improve tracing of permission calls

[Trond Myklebust]

On exit from nfs_do_access(), record the mask representing the requested
permissions, as well as the server-supplied set of access rights for
this user.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/dir.c      |  4 ++--
 fs/nfs/nfstrace.h | 61 ++++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 62 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index e180033e35cf..372c16b3042c 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -2476,7 +2476,7 @@ static int nfs_do_access(struct inode *inode, const struct cred *cred, int mask)
 {
 	struct nfs_access_entry cache;
 	bool may_block = (mask & MAY_NOT_BLOCK) == 0;
-	int cache_mask;
+	int cache_mask = -1;
 	int status;
 
 	trace_nfs_access_enter(inode);
@@ -2515,7 +2515,7 @@ static int nfs_do_access(struct inode *inode, const struct cred *cred, int mask)
 	if ((mask & ~cache_mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) != 0)
 		status = -EACCES;
 out:
-	trace_nfs_access_exit(inode, status);
+	trace_nfs_access_exit(inode, mask, cache_mask, status);
 	return status;
 }
 
diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h
index 51043f02e86f..b43f1b2501db 100644
--- a/fs/nfs/nfstrace.h
+++ b/fs/nfs/nfstrace.h
@@ -198,7 +198,66 @@ DEFINE_NFS_INODE_EVENT_DONE(nfs_writeback_inode_exit);
 DEFINE_NFS_INODE_EVENT(nfs_fsync_enter);
 DEFINE_NFS_INODE_EVENT_DONE(nfs_fsync_exit);
 DEFINE_NFS_INODE_EVENT(nfs_access_enter);
-DEFINE_NFS_INODE_EVENT_DONE(nfs_access_exit);
+
+TRACE_EVENT(nfs_access_exit,
+		TP_PROTO(
+			const struct inode *inode,
+			unsigned int mask,
+			unsigned int permitted,
+			int error
+		),
+
+		TP_ARGS(inode, mask, permitted, error),
+
+		TP_STRUCT__entry(
+			__field(unsigned long, error)
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(unsigned char, type)
+			__field(u64, fileid)
+			__field(u64, version)
+			__field(loff_t, size)
+			__field(unsigned long, nfsi_flags)
+			__field(unsigned long, cache_validity)
+			__field(unsigned int, mask)
+			__field(unsigned int, permitted)
+		),
+
+		TP_fast_assign(
+			const struct nfs_inode *nfsi = NFS_I(inode);
+			__entry->error = error < 0 ? -error : 0;
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->type = nfs_umode_to_dtype(inode->i_mode);
+			__entry->version = inode_peek_iversion_raw(inode);
+			__entry->size = i_size_read(inode);
+			__entry->nfsi_flags = nfsi->flags;
+			__entry->cache_validity = nfsi->cache_validity;
+			__entry->mask = mask;
+			__entry->permitted = permitted;
+		),
+
+		TP_printk(
+			"error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"type=%u (%s) version=%llu size=%lld "
+			"cache_validity=0x%lx (%s) nfs_flags=0x%lx (%s) "
+			"mask=0x%x permitted=0x%x",
+			-__entry->error, nfs_show_status(__entry->error),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			__entry->type,
+			nfs_show_file_type(__entry->type),
+			(unsigned long long)__entry->version,
+			(long long)__entry->size,
+			__entry->cache_validity,
+			nfs_show_cache_validity(__entry->cache_validity),
+			__entry->nfsi_flags,
+			nfs_show_nfsi_flags(__entry->nfsi_flags),
+			__entry->mask, __entry->permitted
+		)
+);
 
 TRACE_DEFINE_ENUM(LOOKUP_FOLLOW);
 TRACE_DEFINE_ENUM(LOOKUP_DIRECTORY);
-- 
2.24.1

[PATCH 14/15] NFS: When resending after a short write, reset the reply count to zero

[Trond Myklebust]

If we're resending a write due to a short read or write, ensure we
reset the reply count to zero.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/read.c  | 2 ++
 fs/nfs/write.c | 2 ++
 2 files changed, 4 insertions(+)

diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 12deb3bdb2a0..34bb9add2302 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -281,6 +281,8 @@ static void nfs_readpage_retry(struct rpc_task *task,
 	argp->offset += resp->count;
 	argp->pgbase += resp->count;
 	argp->count -= resp->count;
+	resp->count = 0;
+	resp->eof = 0;
 	rpc_restart_call_prepare(task);
 }
 
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 985ddff46051..2c2020546e24 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1656,6 +1656,8 @@ static void nfs_writeback_result(struct rpc_task *task,
 			 */
 			argp->stable = NFS_FILE_SYNC;
 		}
+		resp->count = 0;
+		resp->verf->committed = 0;
 		rpc_restart_call_prepare(task);
 	}
 }
-- 
2.24.1

[PATCH 15/15] NFS: Fix nfs_direct_write_reschedule_io()

[Trond Myklebust]

The 'hdr->good_bytes' is defined as the number of bytes we expect to
read or write starting at offset hdr->io_start. In the case of a partial
read/write we may end up adjusting hdr->args.offset and hdr->args.count
to skip I/O for data that was already read/written, and so we must ensure
the calculation takes that into account.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/direct.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 29f00da8a0b7..b768a0b42e82 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -824,7 +824,8 @@ static void nfs_direct_write_reschedule_io(struct nfs_pgio_header *hdr)
 		dreq->flags = NFS_ODIRECT_RESCHED_WRITES;
 		/* fake unstable write to let common nfs resend pages */
 		hdr->verf.committed = NFS_UNSTABLE;
-		hdr->good_bytes = hdr->args.count;
+		hdr->good_bytes = hdr->args.offset + hdr->args.count -
+			hdr->io_start;
 	}
 	spin_unlock(&dreq->lock);
 }
-- 
2.24.1