Merge tag 'nfsd-4.10' of git://linux-nfs.org/~bfields/linux

Pull nfsd updates from Bruce Fields:
 "The one new feature is support for a new NFSv4.2 mode_umask attribute
  that makes ACL inheritance a little more useful in environments that
  default to restrictive umasks. Requires client-side support, also on
  its way for 4.10.

  Other than that, miscellaneous smaller fixes and cleanup, especially
  to the server rdma code"

[ The client side of the umask attribute was merged yesterday ]

* tag 'nfsd-4.10' of git://linux-nfs.org/~bfields/linux:
  nfsd: add support for the umask attribute
  sunrpc: use DEFINE_SPINLOCK()
  svcrdma: Further clean-up of svc_rdma_get_inv_rkey()
  svcrdma: Break up dprintk format in svc_rdma_accept()
  svcrdma: Remove unused variable in rdma_copy_tail()
  svcrdma: Remove unused variables in xprt_rdma_bc_allocate()
  svcrdma: Remove svc_rdma_op_ctxt::wc_status
  svcrdma: Remove DMA map accounting
  svcrdma: Remove BH-disabled spin locking in svc_rdma_send()
  svcrdma: Renovate sendto chunk list parsing
  svcauth_gss: Close connection when dropping an incoming message
  svcrdma: Clear xpt_bc_xps in xprt_setup_rdma_bc() error exit arm
  nfsd: constify reply_cache_stats_operations structure
  nfsd: update workqueue creation
  sunrpc: GFP_KERNEL should be GFP_NOFS in crypto code
  nfsd: catch errors in decode_fattr earlier
  nfsd: clean up supported attribute handling
  nfsd: fix error handling for clients that fail to return the layout
  nfsd: more robust allocation failure handling in nfsd_reply_cache_init

8 files changed, 103 insertions, 85 deletions

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 211dc2aed8e1..eb78109d666c 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -1061,7 +1061,7 @@ static const struct rpc_call_ops nfsd4_cb_ops = {
 
 int nfsd4_create_callback_queue(void)
 {
-	callback_wq = create_singlethread_workqueue("nfsd4_callbacks");
+	callback_wq = alloc_ordered_workqueue("nfsd4_callbacks", 0);
 	if (!callback_wq)
 		return -ENOMEM;
 	return 0;
diff --git a/fs/nfsd/nfs4layouts.c b/fs/nfsd/nfs4layouts.c
index 42aace4fc4c8..596205d939a1 100644
--- a/fs/nfsd/nfs4layouts.c
+++ b/fs/nfsd/nfs4layouts.c
@@ -686,10 +686,6 @@ nfsd4_cb_layout_done(struct nfsd4_callback *cb, struct rpc_task *task)
 			return 0;
 		}
 		/* Fallthrough */
-	case -NFS4ERR_NOMATCHING_LAYOUT:
-		trace_layout_recall_done(&ls->ls_stid.sc_stateid);
-		task->tk_status = 0;
-		return 1;
 	default:
 		/*
 		 * Unknown error or non-responding client, we'll need to fence.
@@ -702,6 +698,10 @@ nfsd4_cb_layout_done(struct nfsd4_callback *cb, struct rpc_task *task)
 		else
 			nfsd4_cb_layout_fail(ls);
 		return -1;
+	case -NFS4ERR_NOMATCHING_LAYOUT:
+		trace_layout_recall_done(&ls->ls_stid.sc_stateid);
+		task->tk_status = 0;
+		return 1;
 	}
 }
 
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index abb09b580389..74a6e573e061 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -96,33 +96,15 @@ check_attr_support(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 {
 	struct dentry *dentry = cstate->current_fh.fh_dentry;
 
-	/*
-	 * Check about attributes are supported by the NFSv4 server or not.
-	 * According to spec, unsupported attributes return ERR_ATTRNOTSUPP.
-	 */
-	if ((bmval[0] & ~nfsd_suppattrs0(cstate->minorversion)) ||
-	    (bmval[1] & ~nfsd_suppattrs1(cstate->minorversion)) ||
-	    (bmval[2] & ~nfsd_suppattrs2(cstate->minorversion)))
+	if (!nfsd_attrs_supported(cstate->minorversion, bmval))
 		return nfserr_attrnotsupp;
-
-	/*
-	 * Check FATTR4_WORD0_ACL can be supported
-	 * in current environment or not.
-	 */
-	if (bmval[0] & FATTR4_WORD0_ACL) {
-		if (!IS_POSIXACL(d_inode(dentry)))
-			return nfserr_attrnotsupp;
-	}
-
-	/*
-	 * According to spec, read-only attributes return ERR_INVAL.
-	 */
-	if (writable) {
-		if ((bmval[0] & ~writable[0]) || (bmval[1] & ~writable[1]) ||
-		    (bmval[2] & ~writable[2]))
-			return nfserr_inval;
-	}
-
+	if ((bmval[0] & FATTR4_WORD0_ACL) && !IS_POSIXACL(d_inode(dentry)))
+		return nfserr_attrnotsupp;
+	if (writable && !bmval_is_subset(bmval, writable))
+		return nfserr_inval;
+	if (writable && (bmval[2] & FATTR4_WORD2_MODE_UMASK) &&
+			(bmval[1] & FATTR4_WORD1_MODE))
+		return nfserr_inval;
 	return nfs_ok;
 }
 
@@ -695,9 +677,9 @@ nfsd4_getattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (getattr->ga_bmval[1] & NFSD_WRITEONLY_ATTRS_WORD1)
 		return nfserr_inval;
 
-	getattr->ga_bmval[0] &= nfsd_suppattrs0(cstate->minorversion);
-	getattr->ga_bmval[1] &= nfsd_suppattrs1(cstate->minorversion);
-	getattr->ga_bmval[2] &= nfsd_suppattrs2(cstate->minorversion);
+	getattr->ga_bmval[0] &= nfsd_suppattrs[cstate->minorversion][0];
+	getattr->ga_bmval[1] &= nfsd_suppattrs[cstate->minorversion][1];
+	getattr->ga_bmval[2] &= nfsd_suppattrs[cstate->minorversion][2];
 
 	getattr->ga_fhp = &cstate->current_fh;
 	return nfs_ok;
@@ -799,9 +781,9 @@ nfsd4_readdir(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (readdir->rd_bmval[1] & NFSD_WRITEONLY_ATTRS_WORD1)
 		return nfserr_inval;
 
-	readdir->rd_bmval[0] &= nfsd_suppattrs0(cstate->minorversion);
-	readdir->rd_bmval[1] &= nfsd_suppattrs1(cstate->minorversion);
-	readdir->rd_bmval[2] &= nfsd_suppattrs2(cstate->minorversion);
+	readdir->rd_bmval[0] &= nfsd_suppattrs[cstate->minorversion][0];
+	readdir->rd_bmval[1] &= nfsd_suppattrs[cstate->minorversion][1];
+	readdir->rd_bmval[2] &= nfsd_suppattrs[cstate->minorversion][2];
 
 	if ((cookie == 1) || (cookie == 2) ||
 	    (cookie == 0 && memcmp(readdir->rd_verf.data, zeroverf.data, NFS4_VERIFIER_SIZE)))
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index c2d2895a1ec1..79edde4577b2 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -33,6 +33,7 @@
  *  SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  */
 
+#include <linux/fs_struct.h>
 #include <linux/file.h>
 #include <linux/slab.h>
 #include <linux/namei.h>
@@ -57,6 +58,20 @@
 
 #define NFSDDBG_FACILITY		NFSDDBG_XDR
 
+u32 nfsd_suppattrs[3][3] = {
+	{NFSD4_SUPPORTED_ATTRS_WORD0,
+	 NFSD4_SUPPORTED_ATTRS_WORD1,
+	 NFSD4_SUPPORTED_ATTRS_WORD2},
+
+	{NFSD4_1_SUPPORTED_ATTRS_WORD0,
+	 NFSD4_1_SUPPORTED_ATTRS_WORD1,
+	 NFSD4_1_SUPPORTED_ATTRS_WORD2},
+
+	{NFSD4_1_SUPPORTED_ATTRS_WORD0,
+	 NFSD4_1_SUPPORTED_ATTRS_WORD1,
+	 NFSD4_2_SUPPORTED_ATTRS_WORD2},
+};
+
 /*
  * As per referral draft, the fsid for a referral MUST be different from the fsid of the containing
  * directory in order to indicate to the client that a filesystem boundary is present
@@ -285,7 +300,7 @@ nfsd4_decode_bitmap(struct nfsd4_compoundargs *argp, u32 *bmval)
 static __be32
 nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval,
 		   struct iattr *iattr, struct nfs4_acl **acl,
-		   struct xdr_netobj *label)
+		   struct xdr_netobj *label, int *umask)
 {
 	int expected_len, len = 0;
 	u32 dummy32;
@@ -296,6 +311,14 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval,
 	if ((status = nfsd4_decode_bitmap(argp, bmval)))
 		return status;
 
+	if (bmval[0] & ~NFSD_WRITEABLE_ATTRS_WORD0
+	    || bmval[1] & ~NFSD_WRITEABLE_ATTRS_WORD1
+	    || bmval[2] & ~NFSD_WRITEABLE_ATTRS_WORD2) {
+		if (nfsd_attrs_supported(argp->minorversion, bmval))
+			return nfserr_inval;
+		return nfserr_attrnotsupp;
+	}
+
 	READ_BUF(4);
 	expected_len = be32_to_cpup(p++);
 
@@ -435,12 +458,18 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval,
 			return nfserr_jukebox;
 	}
 #endif
-
-	if (bmval[0] & ~NFSD_WRITEABLE_ATTRS_WORD0
-	    || bmval[1] & ~NFSD_WRITEABLE_ATTRS_WORD1
-	    || bmval[2] & ~NFSD_WRITEABLE_ATTRS_WORD2)
-		READ_BUF(expected_len - len);
-	else if (len != expected_len)
+	if (bmval[2] & FATTR4_WORD2_MODE_UMASK) {
+		if (!umask)
+			goto xdr_error;
+		READ_BUF(8);
+		len += 8;
+		dummy32 = be32_to_cpup(p++);
+		iattr->ia_mode = dummy32 & (S_IFMT | S_IALLUGO);
+		dummy32 = be32_to_cpup(p++);
+		*umask = dummy32 & S_IRWXUGO;
+		iattr->ia_valid |= ATTR_MODE;
+	}
+	if (len != expected_len)
 		goto xdr_error;
 
 	DECODE_TAIL;
@@ -634,7 +663,8 @@ nfsd4_decode_create(struct nfsd4_compoundargs *argp, struct nfsd4_create *create
 		return status;
 
 	status = nfsd4_decode_fattr(argp, create->cr_bmval, &create->cr_iattr,
-				    &create->cr_acl, &create->cr_label);
+				    &create->cr_acl, &create->cr_label,
+				    &current->fs->umask);
 	if (status)
 		goto out;
 
@@ -879,13 +909,15 @@ nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open)
 	case NFS4_OPEN_NOCREATE:
 		break;
 	case NFS4_OPEN_CREATE:
+		current->fs->umask = 0;
 		READ_BUF(4);
 		open->op_createmode = be32_to_cpup(p++);
 		switch (open->op_createmode) {
 		case NFS4_CREATE_UNCHECKED:
 		case NFS4_CREATE_GUARDED:
 			status = nfsd4_decode_fattr(argp, open->op_bmval,
-				&open->op_iattr, &open->op_acl, &open->op_label);
+				&open->op_iattr, &open->op_acl, &open->op_label,
+				&current->fs->umask);
 			if (status)
 				goto out;
 			break;
@@ -899,7 +931,8 @@ nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open)
 			READ_BUF(NFS4_VERIFIER_SIZE);
 			COPYMEM(open->op_verf.data, NFS4_VERIFIER_SIZE);
 			status = nfsd4_decode_fattr(argp, open->op_bmval,
-				&open->op_iattr, &open->op_acl, &open->op_label);
+				&open->op_iattr, &open->op_acl, &open->op_label,
+				&current->fs->umask);
 			if (status)
 				goto out;
 			break;
@@ -1136,7 +1169,7 @@ nfsd4_decode_setattr(struct nfsd4_compoundargs *argp, struct nfsd4_setattr *seta
 	if (status)
 		return status;
 	return nfsd4_decode_fattr(argp, setattr->sa_bmval, &setattr->sa_iattr,
-				  &setattr->sa_acl, &setattr->sa_label);
+				  &setattr->sa_acl, &setattr->sa_label, NULL);
 }
 
 static __be32
@@ -2340,9 +2373,7 @@ nfsd4_encode_fattr(struct xdr_stream *xdr, struct svc_fh *fhp,
 	struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
 
 	BUG_ON(bmval1 & NFSD_WRITEONLY_ATTRS_WORD1);
-	BUG_ON(bmval0 & ~nfsd_suppattrs0(minorversion));
-	BUG_ON(bmval1 & ~nfsd_suppattrs1(minorversion));
-	BUG_ON(bmval2 & ~nfsd_suppattrs2(minorversion));
+	BUG_ON(!nfsd_attrs_supported(minorversion, bmval));
 
 	if (exp->ex_fslocs.migrated) {
 		status = fattr_handle_absent_fs(&bmval0, &bmval1, &bmval2, &rdattr_err);
@@ -2409,29 +2440,27 @@ nfsd4_encode_fattr(struct xdr_stream *xdr, struct svc_fh *fhp,
 	p++;                /* to be backfilled later */
 
 	if (bmval0 & FATTR4_WORD0_SUPPORTED_ATTRS) {
-		u32 word0 = nfsd_suppattrs0(minorversion);
-		u32 word1 = nfsd_suppattrs1(minorversion);
-		u32 word2 = nfsd_suppattrs2(minorversion);
+		u32 *supp = nfsd_suppattrs[minorversion];
 
 		if (!IS_POSIXACL(dentry->d_inode))
-			word0 &= ~FATTR4_WORD0_ACL;
+			supp[0] &= ~FATTR4_WORD0_ACL;
 		if (!contextsupport)
-			word2 &= ~FATTR4_WORD2_SECURITY_LABEL;
-		if (!word2) {
+			supp[2] &= ~FATTR4_WORD2_SECURITY_LABEL;
+		if (!supp[2]) {
 			p = xdr_reserve_space(xdr, 12);
 			if (!p)
 				goto out_resource;
 			*p++ = cpu_to_be32(2);
-			*p++ = cpu_to_be32(word0);
-			*p++ = cpu_to_be32(word1);
+			*p++ = cpu_to_be32(supp[0]);
+			*p++ = cpu_to_be32(supp[1]);
 		} else {
 			p = xdr_reserve_space(xdr, 16);
 			if (!p)
 				goto out_resource;
 			*p++ = cpu_to_be32(3);
-			*p++ = cpu_to_be32(word0);
-			*p++ = cpu_to_be32(word1);
-			*p++ = cpu_to_be32(word2);
+			*p++ = cpu_to_be32(supp[0]);
+			*p++ = cpu_to_be32(supp[1]);
+			*p++ = cpu_to_be32(supp[2]);
 		}
 	}
 	if (bmval0 & FATTR4_WORD0_TYPE) {
diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c
index 54cde9a5864e..d6b97b424ad1 100644
--- a/fs/nfsd/nfscache.c
+++ b/fs/nfsd/nfscache.c
@@ -9,6 +9,7 @@
  */
 
 #include <linux/slab.h>
+#include <linux/vmalloc.h>
 #include <linux/sunrpc/addr.h>
 #include <linux/highmem.h>
 #include <linux/log2.h>
@@ -174,8 +175,12 @@ int nfsd_reply_cache_init(void)
 		goto out_nomem;
 
 	drc_hashtbl = kcalloc(hashsize, sizeof(*drc_hashtbl), GFP_KERNEL);
-	if (!drc_hashtbl)
-		goto out_nomem;
+	if (!drc_hashtbl) {
+		drc_hashtbl = vzalloc(hashsize * sizeof(*drc_hashtbl));
+		if (!drc_hashtbl)
+			goto out_nomem;
+	}
+
 	for (i = 0; i < hashsize; i++) {
 		INIT_LIST_HEAD(&drc_hashtbl[i].lru_head);
 		spin_lock_init(&drc_hashtbl[i].cache_lock);
@@ -204,7 +209,7 @@ void nfsd_reply_cache_shutdown(void)
 		}
 	}
 
-	kfree (drc_hashtbl);
+	kvfree(drc_hashtbl);
 	drc_hashtbl = NULL;
 	drc_hashsize = 0;
 
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 2857e46d5cc5..f3b2f34b10a3 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -217,7 +217,7 @@ static const struct file_operations pool_stats_operations = {
 	.release	= nfsd_pool_stats_release,
 };
 
-static struct file_operations reply_cache_stats_operations = {
+static const struct file_operations reply_cache_stats_operations = {
 	.open		= nfsd_reply_cache_stats_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h
index 9446849888d5..d74c8c44dc35 100644
--- a/fs/nfsd/nfsd.h
+++ b/fs/nfsd/nfsd.h
@@ -359,44 +359,46 @@ void		nfsd_lockd_shutdown(void);
 
 #define NFSD4_2_SUPPORTED_ATTRS_WORD2 \
 	(NFSD4_1_SUPPORTED_ATTRS_WORD2 | \
+	FATTR4_WORD2_MODE_UMASK | \
 	NFSD4_2_SECURITY_ATTRS)
 
-static inline u32 nfsd_suppattrs0(u32 minorversion)
-{
-	return minorversion ? NFSD4_1_SUPPORTED_ATTRS_WORD0
-			    : NFSD4_SUPPORTED_ATTRS_WORD0;
-}
+extern u32 nfsd_suppattrs[3][3];
 
-static inline u32 nfsd_suppattrs1(u32 minorversion)
+static inline bool bmval_is_subset(u32 *bm1, u32 *bm2)
 {
-	return minorversion ? NFSD4_1_SUPPORTED_ATTRS_WORD1
-			    : NFSD4_SUPPORTED_ATTRS_WORD1;
+	return !((bm1[0] & ~bm2[0]) ||
+	         (bm1[1] & ~bm2[1]) ||
+		 (bm1[2] & ~bm2[2]));
 }
 
-static inline u32 nfsd_suppattrs2(u32 minorversion)
+static inline bool nfsd_attrs_supported(u32 minorversion, u32 *bmval)
 {
-	switch (minorversion) {
-	default: return NFSD4_2_SUPPORTED_ATTRS_WORD2;
-	case 1:  return NFSD4_1_SUPPORTED_ATTRS_WORD2;
-	case 0:  return NFSD4_SUPPORTED_ATTRS_WORD2;
-	}
+	return bmval_is_subset(bmval, nfsd_suppattrs[minorversion]);
 }
 
 /* These will return ERR_INVAL if specified in GETATTR or READDIR. */
 #define NFSD_WRITEONLY_ATTRS_WORD1 \
 	(FATTR4_WORD1_TIME_ACCESS_SET   | FATTR4_WORD1_TIME_MODIFY_SET)
 
-/* These are the only attrs allowed in CREATE/OPEN/SETATTR. */
+/*
+ * These are the only attrs allowed in CREATE/OPEN/SETATTR. Don't add
+ * a writeable attribute here without also adding code to parse it to
+ * nfsd4_decode_fattr().
+ */
 #define NFSD_WRITEABLE_ATTRS_WORD0 \
 	(FATTR4_WORD0_SIZE | FATTR4_WORD0_ACL)
 #define NFSD_WRITEABLE_ATTRS_WORD1 \
 	(FATTR4_WORD1_MODE | FATTR4_WORD1_OWNER | FATTR4_WORD1_OWNER_GROUP \
 	| FATTR4_WORD1_TIME_ACCESS_SET | FATTR4_WORD1_TIME_MODIFY_SET)
 #ifdef CONFIG_NFSD_V4_SECURITY_LABEL
-#define NFSD_WRITEABLE_ATTRS_WORD2 FATTR4_WORD2_SECURITY_LABEL
+#define MAYBE_FATTR4_WORD2_SECURITY_LABEL \
+	FATTR4_WORD2_SECURITY_LABEL
 #else
-#define NFSD_WRITEABLE_ATTRS_WORD2 0
+#define MAYBE_FATTR4_WORD2_SECURITY_LABEL 0
 #endif
+#define NFSD_WRITEABLE_ATTRS_WORD2 \
+	(FATTR4_WORD2_MODE_UMASK \
+	| MAYBE_FATTR4_WORD2_SECURITY_LABEL)
 
 #define NFSD_SUPPATTR_EXCLCREAT_WORD0 \
 	NFSD_WRITEABLE_ATTRS_WORD0
diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index a2b65fc56dd6..e6bfd96734c0 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -661,8 +661,8 @@ nfsd(void *vrqstp)
 	mutex_lock(&nfsd_mutex);
 
 	/* At this point, the thread shares current->fs
-	 * with the init process. We need to create files with a
-	 * umask of 0 instead of init's umask. */
+	 * with the init process. We need to create files with the
+	 * umask as defined by the client instead of init's umask. */
 	if (unshare_fs_struct() < 0) {
 		printk("Unable to start nfsd thread: out of memory\n");
 		goto out;