Merge branch 'for-2.6.40' of git://linux-nfs.org/~bfields/linux

* 'for-2.6.40' of git://linux-nfs.org/~bfields/linux: (22 commits)
  nfsd: make local functions static
  NFSD: Remove unused variable from nfsd4_decode_bind_conn_to_session()
  NFSD: Check status from nfsd4_map_bcts_dir()
  NFSD: Remove setting unused variable in nfsd_vfs_read()
  nfsd41: error out on repeated RECLAIM_COMPLETE
  nfsd41: compare request's opcnt with session's maxops at nfsd4_sequence
  nfsd v4.1 lOCKT clientid field must be ignored
  nfsd41: add flag checking for create_session
  nfsd41: make sure nfs server process OPEN with EXCLUSIVE4_1 correctly
  nfsd4: fix wrongsec handling for PUTFH + op cases
  nfsd4: make fh_verify responsibility of nfsd_lookup_dentry caller
  nfsd4: introduce OPDESC helper
  nfsd4: allow fh_verify caller to skip pseudoflavor checks
  nfsd: distinguish functions of NFSD_MAY_* flags
  svcrpc: complete svsk processing on cb receive failure
  svcrpc: take advantage of tcp autotuning
  SUNRPC: Don't wait for full record to receive tcp data
  svcrpc: copy cb reply instead of pages
  svcrpc: close connection if client sends short packet
  svcrpc: note network-order types in svc_process_calldir
  ...

fs/nfsd/export.c

@@ -1354,12 +1354,6 @@ exp_pseudoroot(struct svc_rqst *rqstp, struct svc_fh *fhp)
 	if (IS_ERR(exp))
 		return nfserrno(PTR_ERR(exp));
 	rv = fh_compose(fhp, exp, exp->ex_path.dentry, NULL);
-	if (rv)
-		goto out;
-	rv = check_nfsd_access(exp, rqstp);
-	if (rv)
-		fh_put(fhp);
-out:
 	exp_put(exp);
 	return rv;
 }

fs/nfsd/nfs3proc.c

@@ -245,7 +245,7 @@ nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp,
 	}
 
 	/* Now create the file and set attributes */
-	nfserr = nfsd_create_v3(rqstp, dirfhp, argp->name, argp->len,
+	nfserr = do_nfsd_create(rqstp, dirfhp, argp->name, argp->len,
 				attr, newfhp,
 				argp->createmode, argp->verf, NULL, NULL);
 

fs/nfsd/nfs3xdr.c

@@ -842,7 +842,7 @@ compose_entry_fh(struct nfsd3_readdirres *cd, struct svc_fh *fhp,
 	return rv;
 }
 
-__be32 *encode_entryplus_baggage(struct nfsd3_readdirres *cd, __be32 *p, const char *name, int namlen)
+static __be32 *encode_entryplus_baggage(struct nfsd3_readdirres *cd, __be32 *p, const char *name, int namlen)
 {
 	struct svc_fh	fh;
 	int err;

fs/nfsd/nfs4proc.c

@@ -196,9 +196,9 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o
 
 		/*
 		 * Note: create modes (UNCHECKED,GUARDED...) are the same
-		 * in NFSv4 as in v3.
+		 * in NFSv4 as in v3 except EXCLUSIVE4_1.
 		 */
-		status = nfsd_create_v3(rqstp, current_fh, open->op_fname.data,
+		status = do_nfsd_create(rqstp, current_fh, open->op_fname.data,
 					open->op_fname.len, &open->op_iattr,
 					&resfh, open->op_createmode,
 					(u32 *)open->op_verf.data,
@@ -403,7 +403,7 @@ nfsd4_putfh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	cstate->current_fh.fh_handle.fh_size = putfh->pf_fhlen;
 	memcpy(&cstate->current_fh.fh_handle.fh_base, putfh->pf_fhval,
 	       putfh->pf_fhlen);
-	return fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP);
+	return fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_BYPASS_GSS);
 }
 
 static __be32
@@ -762,6 +762,9 @@ nfsd4_secinfo(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	__be32 err;
 
 	fh_init(&resfh, NFS4_FHSIZE);
+	err = fh_verify(rqstp, &cstate->current_fh, S_IFDIR, NFSD_MAY_EXEC);
+	if (err)
+		return err;
 	err = nfsd_lookup_dentry(rqstp, &cstate->current_fh,
 				    secinfo->si_name, secinfo->si_namelen,
 				    &exp, &dentry);
@@ -986,6 +989,9 @@ enum nfsd4_op_flags {
 	ALLOWED_WITHOUT_FH = 1 << 0,	/* No current filehandle required */
 	ALLOWED_ON_ABSENT_FS = 1 << 1,	/* ops processed on absent fs */
 	ALLOWED_AS_FIRST_OP = 1 << 2,	/* ops reqired first in compound */
+	/* For rfc 5661 section 2.6.3.1.1: */
+	OP_HANDLES_WRONGSEC = 1 << 3,
+	OP_IS_PUTFH_LIKE = 1 << 4,
 };
 
 struct nfsd4_operation {
@@ -1031,6 +1037,44 @@ static __be32 nfs41_check_op_ordering(struct nfsd4_compoundargs *args)
 	return nfs_ok;
 }
 
+static inline struct nfsd4_operation *OPDESC(struct nfsd4_op *op)
+{
+	return &nfsd4_ops[op->opnum];
+}
+
+static bool need_wrongsec_check(struct svc_rqst *rqstp)
+{
+	struct nfsd4_compoundres *resp = rqstp->rq_resp;
+	struct nfsd4_compoundargs *argp = rqstp->rq_argp;
+	struct nfsd4_op *this = &argp->ops[resp->opcnt - 1];
+	struct nfsd4_op *next = &argp->ops[resp->opcnt];
+	struct nfsd4_operation *thisd;
+	struct nfsd4_operation *nextd;
+
+	thisd = OPDESC(this);
+	/*
+	 * Most ops check wronsec on our own; only the putfh-like ops
+	 * have special rules.
+	 */
+	if (!(thisd->op_flags & OP_IS_PUTFH_LIKE))
+		return false;
+	/*
+	 * rfc 5661 2.6.3.1.1.6: don't bother erroring out a
+	 * put-filehandle operation if we're not going to use the
+	 * result:
+	 */
+	if (argp->opcnt == resp->opcnt)
+		return false;
+
+	nextd = OPDESC(next);
+	/*
+	 * Rest of 2.6.3.1.1: certain operations will return WRONGSEC
+	 * errors themselves as necessary; others should check for them
+	 * now:
+	 */
+	return !(nextd->op_flags & OP_HANDLES_WRONGSEC);
+}
+
 /*
  * COMPOUND call.
  */
@@ -1108,7 +1152,7 @@ nfsd4_proc_compound(struct svc_rqst *rqstp,
 			goto encode_op;
 		}
 
-		opdesc = &nfsd4_ops[op->opnum];
+		opdesc = OPDESC(op);
 
 		if (!cstate->current_fh.fh_dentry) {
 			if (!(opdesc->op_flags & ALLOWED_WITHOUT_FH)) {
@@ -1126,6 +1170,9 @@ nfsd4_proc_compound(struct svc_rqst *rqstp,
 		else
 			BUG_ON(op->status == nfs_ok);
 
+		if (!op->status && need_wrongsec_check(rqstp))
+			op->status = check_nfsd_access(cstate->current_fh.fh_export, rqstp);
+
 encode_op:
 		/* Only from SEQUENCE */
 		if (resp->cstate.status == nfserr_replay_cache) {
@@ -1217,10 +1264,12 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	},
 	[OP_LOOKUP] = {
 		.op_func = (nfsd4op_func)nfsd4_lookup,
+		.op_flags = OP_HANDLES_WRONGSEC,
 		.op_name = "OP_LOOKUP",
 	},
 	[OP_LOOKUPP] = {
 		.op_func = (nfsd4op_func)nfsd4_lookupp,
+		.op_flags = OP_HANDLES_WRONGSEC,
 		.op_name = "OP_LOOKUPP",
 	},
 	[OP_NVERIFY] = {
@@ -1229,6 +1278,7 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	},
 	[OP_OPEN] = {
 		.op_func = (nfsd4op_func)nfsd4_open,
+		.op_flags = OP_HANDLES_WRONGSEC,
 		.op_name = "OP_OPEN",
 	},
 	[OP_OPEN_CONFIRM] = {
@@ -1241,17 +1291,20 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	},
 	[OP_PUTFH] = {
 		.op_func = (nfsd4op_func)nfsd4_putfh,
-		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS,
+		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
+				| OP_IS_PUTFH_LIKE,
 		.op_name = "OP_PUTFH",
 	},
 	[OP_PUTPUBFH] = {
 		.op_func = (nfsd4op_func)nfsd4_putrootfh,
-		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS,
+		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
+				| OP_IS_PUTFH_LIKE,
 		.op_name = "OP_PUTPUBFH",
 	},
 	[OP_PUTROOTFH] = {
 		.op_func = (nfsd4op_func)nfsd4_putrootfh,
-		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS,
+		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
+				| OP_IS_PUTFH_LIKE,
 		.op_name = "OP_PUTROOTFH",
 	},
 	[OP_READ] = {
@@ -1281,15 +1334,18 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	},
 	[OP_RESTOREFH] = {
 		.op_func = (nfsd4op_func)nfsd4_restorefh,
-		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS,
+		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
+				| OP_IS_PUTFH_LIKE,
 		.op_name = "OP_RESTOREFH",
 	},
 	[OP_SAVEFH] = {
 		.op_func = (nfsd4op_func)nfsd4_savefh,
+		.op_flags = OP_HANDLES_WRONGSEC,
 		.op_name = "OP_SAVEFH",
 	},
 	[OP_SECINFO] = {
 		.op_func = (nfsd4op_func)nfsd4_secinfo,
+		.op_flags = OP_HANDLES_WRONGSEC,
 		.op_name = "OP_SECINFO",
 	},
 	[OP_SETATTR] = {
@@ -1353,6 +1409,7 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	},
 	[OP_SECINFO_NO_NAME] = {
 		.op_func = (nfsd4op_func)nfsd4_secinfo_no_name,
+		.op_flags = OP_HANDLES_WRONGSEC,
 		.op_name = "OP_SECINFO_NO_NAME",
 	},
 };

fs/nfsd/nfs4state.c

@@ -1519,6 +1519,9 @@ nfsd4_create_session(struct svc_rqst *rqstp,
 	bool confirm_me = false;
 	int status = 0;
 
+	if (cr_ses->flags & ~SESSION4_FLAG_MASK_A)
+		return nfserr_inval;
+
 	nfs4_lock_state();
 	unconf = find_unconfirmed_client(&cr_ses->clientid);
 	conf = find_confirmed_client(&cr_ses->clientid);
@@ -1637,8 +1640,9 @@ __be32 nfsd4_bind_conn_to_session(struct svc_rqst *rqstp,
 		return nfserr_badsession;
 
 	status = nfsd4_map_bcts_dir(&bcts->dir);
+	if (!status)
 		nfsd4_new_conn(rqstp, cstate->session, bcts->dir);
-	return nfs_ok;
+	return status;
 }
 
 static bool nfsd4_compound_in_session(struct nfsd4_session *session, struct nfs4_sessionid *sid)
@@ -1725,6 +1729,13 @@ static void nfsd4_sequence_check_conn(struct nfsd4_conn *new, struct nfsd4_sessi
 	return;
 }
 
+static bool nfsd4_session_too_many_ops(struct svc_rqst *rqstp, struct nfsd4_session *session)
+{
+	struct nfsd4_compoundargs *args = rqstp->rq_argp;
+
+	return args->opcnt > session->se_fchannel.maxops;
+}
+
 __be32
 nfsd4_sequence(struct svc_rqst *rqstp,
 	       struct nfsd4_compound_state *cstate,
@@ -1753,6 +1764,10 @@ nfsd4_sequence(struct svc_rqst *rqstp,
 	if (!session)
 		goto out;
 
+	status = nfserr_too_many_ops;
+	if (nfsd4_session_too_many_ops(rqstp, session))
+		goto out;
+
 	status = nfserr_badslot;
 	if (seq->slotid >= session->se_fchannel.maxreqs)
 		goto out;
@@ -1808,6 +1823,8 @@ nfsd4_sequence(struct svc_rqst *rqstp,
 __be32
 nfsd4_reclaim_complete(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_reclaim_complete *rc)
 {
+	int status = 0;
+
 	if (rc->rca_one_fs) {
 		if (!cstate->current_fh.fh_dentry)
 			return nfserr_nofilehandle;
@@ -1817,9 +1834,14 @@ nfsd4_reclaim_complete(struct svc_rqst *rqstp, struct nfsd4_compound_state *csta
 		 */
 		 return nfs_ok;
 	}
+
 	nfs4_lock_state();
-	if (is_client_expired(cstate->session->se_client)) {
-		nfs4_unlock_state();
+	status = nfserr_complete_already;
+	if (cstate->session->se_client->cl_firststate)
+		goto out;
+
+	status = nfserr_stale_clientid;
+	if (is_client_expired(cstate->session->se_client))
 		/*
 		 * The following error isn't really legal.
 		 * But we only get here if the client just explicitly
@@ -1827,11 +1849,13 @@ nfsd4_reclaim_complete(struct svc_rqst *rqstp, struct nfsd4_compound_state *csta
 		 * error it gets back on an operation for the dead
 		 * client.
 		 */
-		return nfserr_stale_clientid;
-	}
+		goto out;
+
+	status = nfs_ok;
 	nfsd4_create_clid_dir(cstate->session->se_client);
+out:
 	nfs4_unlock_state();
-	return nfs_ok;
+	return status;
 }
 
 __be32
@@ -2462,7 +2486,7 @@ find_delegation_file(struct nfs4_file *fp, stateid_t *stid)
 	return NULL;
 }
 
-int share_access_to_flags(u32 share_access)
+static int share_access_to_flags(u32 share_access)
 {
 	share_access &= ~NFS4_SHARE_WANT_MASK;
 
@@ -2882,7 +2906,7 @@ nfsd4_renew(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	return status;
 }
 
-struct lock_manager nfsd4_manager = {
+static struct lock_manager nfsd4_manager = {
 };
 
 static void

fs/nfsd/nfs4xdr.c

@@ -424,15 +424,12 @@ nfsd4_decode_access(struct nfsd4_compoundargs *argp, struct nfsd4_access *access
 static __be32 nfsd4_decode_bind_conn_to_session(struct nfsd4_compoundargs *argp, struct nfsd4_bind_conn_to_session *bcts)
 {
 	DECODE_HEAD;
-	u32 dummy;
 
 	READ_BUF(NFS4_MAX_SESSIONID_LEN + 8);
 	COPYMEM(bcts->sessionid.data, NFS4_MAX_SESSIONID_LEN);
 	READ32(bcts->dir);
-	/* XXX: Perhaps Tom Tucker could help us figure out how we
-	 * should be using ctsa_use_conn_in_rdma_mode: */
-	READ32(dummy);
-
+	/* XXX: skipping ctsa_use_conn_in_rdma_mode.  Perhaps Tom Tucker
+	 * could help us figure out we should be using it. */
 	DECODE_TAIL;
 }
 
@@ -588,8 +585,6 @@ nfsd4_decode_lockt(struct nfsd4_compoundargs *argp, struct nfsd4_lockt *lockt)
 	READ_BUF(lockt->lt_owner.len);
 	READMEM(lockt->lt_owner.data, lockt->lt_owner.len);
 
-	if (argp->minorversion && !zero_clientid(&lockt->lt_clientid))
-		return nfserr_inval;
 	DECODE_TAIL;
 }
 
@@ -3120,7 +3115,7 @@ nfsd4_encode_destroy_session(struct nfsd4_compoundres *resp, int nfserr,
 	return nfserr;
 }
 
-__be32
+static __be32
 nfsd4_encode_sequence(struct nfsd4_compoundres *resp, int nfserr,
 		      struct nfsd4_sequence *seq)
 {

fs/nfsd/nfsfh.c

@@ -344,7 +344,7 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access)
 	 * which clients virtually always use auth_sys for,
 	 * even while using RPCSEC_GSS for NFS.
 	 */
-	if (access & NFSD_MAY_LOCK)
+	if (access & NFSD_MAY_LOCK || access & NFSD_MAY_BYPASS_GSS)
 		goto skip_pseudoflavor_check;
 	/*
 	 * Clients may expect to be able to use auth_sys during mount,

fs/nfsd/vfs.c

@@ -181,16 +181,10 @@ nfsd_lookup_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	struct svc_export	*exp;
 	struct dentry		*dparent;
 	struct dentry		*dentry;
-	__be32			err;
 	int			host_err;
 
 	dprintk("nfsd: nfsd_lookup(fh %s, %.*s)\n", SVCFH_fmt(fhp), len,name);
 
-	/* Obtain dentry and export. */
-	err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_EXEC);
-	if (err)
-		return err;
-
 	dparent = fhp->fh_dentry;
 	exp  = fhp->fh_export;
 	exp_get(exp);
@@ -254,6 +248,9 @@ nfsd_lookup(struct svc_rqst *rqstp, struct svc_fh *fhp, const char *name,
 	struct dentry		*dentry;
 	__be32 err;
 
+	err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_EXEC);
+	if (err)
+		return err;
 	err = nfsd_lookup_dentry(rqstp, fhp, name, len, &exp, &dentry);
 	if (err)
 		return err;
@@ -877,13 +874,11 @@ static __be32
 nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
               loff_t offset, struct kvec *vec, int vlen, unsigned long *count)
 {
-	struct inode *inode;
 	mm_segment_t	oldfs;
 	__be32		err;
 	int		host_err;
 
 	err = nfserr_perm;
-	inode = file->f_path.dentry->d_inode;
 
 	if (file->f_op->splice_read && rqstp->rq_splice_ok) {
 		struct splice_desc sd = {
@@ -1340,11 +1335,18 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 }
 
 #ifdef CONFIG_NFSD_V3
+
+static inline int nfsd_create_is_exclusive(int createmode)
+{
+	return createmode == NFS3_CREATE_EXCLUSIVE
+	       || createmode == NFS4_CREATE_EXCLUSIVE4_1;
+}
+
 /*
- * NFSv3 version of nfsd_create
+ * NFSv3 and NFSv4 version of nfsd_create
  */
 __be32
-nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp,
+do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 		char *fname, int flen, struct iattr *iap,
 		struct svc_fh *resfhp, int createmode, u32 *verifier,
 	        int *truncp, int *created)
@@ -1396,7 +1398,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	if (err)
 		goto out;
 
-	if (createmode == NFS3_CREATE_EXCLUSIVE) {
+	if (nfsd_create_is_exclusive(createmode)) {
 		/* solaris7 gets confused (bugid 4218508) if these have
 		 * the high bit set, so just clear the high bits. If this is
 		 * ever changed to use different attrs for storing the
@@ -1437,6 +1439,11 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp,
 			    && dchild->d_inode->i_atime.tv_sec == v_atime
 			    && dchild->d_inode->i_size  == 0 )
 				break;
+		case NFS4_CREATE_EXCLUSIVE4_1:
+			if (   dchild->d_inode->i_mtime.tv_sec == v_mtime
+			    && dchild->d_inode->i_atime.tv_sec == v_atime
+			    && dchild->d_inode->i_size  == 0 )
+				goto set_attr;
 			 /* fallthru */
 		case NFS3_CREATE_GUARDED:
 			err = nfserr_exist;
@@ -1455,7 +1462,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp,
 
 	nfsd_check_ignore_resizing(iap);
 
-	if (createmode == NFS3_CREATE_EXCLUSIVE) {
+	if (nfsd_create_is_exclusive(createmode)) {
 		/* Cram the verifier into atime/mtime */
 		iap->ia_valid = ATTR_MTIME|ATTR_ATIME
 			| ATTR_MTIME_SET|ATTR_ATIME_SET;
@@ -2034,7 +2041,7 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp,
 	struct inode	*inode = dentry->d_inode;
 	int		err;
 
-	if (acc == NFSD_MAY_NOP)
+	if ((acc & NFSD_MAY_MASK) == NFSD_MAY_NOP)
 		return 0;
 #if 0
 	dprintk("nfsd: permission 0x%x%s%s%s%s%s%s%s mode 0%o%s%s%s\n",

fs/nfsd/vfs.h

@@ -17,10 +17,14 @@
 #define NFSD_MAY_SATTR		8
 #define NFSD_MAY_TRUNC		16
 #define NFSD_MAY_LOCK		32
+#define NFSD_MAY_MASK		63
+
+/* extra hints to permission and open routines: */
 #define NFSD_MAY_OWNER_OVERRIDE	64
 #define NFSD_MAY_LOCAL_ACCESS	128 /* IRIX doing local access check on device special file*/
 #define NFSD_MAY_BYPASS_GSS_ON_ROOT 256
 #define NFSD_MAY_NOT_BREAK_LEASE 512
+#define NFSD_MAY_BYPASS_GSS	1024
 
 #define NFSD_MAY_CREATE		(NFSD_MAY_EXEC|NFSD_MAY_WRITE)
 #define NFSD_MAY_REMOVE		(NFSD_MAY_EXEC|NFSD_MAY_WRITE|NFSD_MAY_TRUNC)
@@ -54,7 +58,7 @@ __be32		nfsd_create(struct svc_rqst *, struct svc_fh *,
 				int type, dev_t rdev, struct svc_fh *res);
 #ifdef CONFIG_NFSD_V3
 __be32		nfsd_access(struct svc_rqst *, struct svc_fh *, u32 *, u32 *);
-__be32		nfsd_create_v3(struct svc_rqst *, struct svc_fh *,
+__be32		do_nfsd_create(struct svc_rqst *, struct svc_fh *,
 				char *name, int len, struct iattr *attrs,
 				struct svc_fh *res, int createmode,
 				u32 *verifier, int *truncp, int *created);

include/linux/nfs4.h

@@ -574,6 +574,8 @@ struct nfs4_sessionid {
 #define SESSION4_BACK_CHAN	0x002
 #define SESSION4_RDMA		0x004
 
+#define SESSION4_FLAG_MASK_A	0x007
+
 enum state_protect_how4 {
 	SP4_NONE	= 0,
 	SP4_MACH_CRED	= 1,

include/linux/sunrpc/svcsock.h

@@ -28,6 +28,7 @@ struct svc_sock {
 	/* private TCP part */
 	u32			sk_reclen;	/* length of record */
 	u32			sk_tcplen;	/* current read length */
+	struct page *		sk_pages[RPCSVC_MAXPAGES];	/* received data */
 };
 
 /*

net/sunrpc/svcsock.c

@@ -387,6 +387,33 @@ static int svc_recvfrom(struct svc_rqst *rqstp, struct kvec *iov, int nr,
 	return len;
 }
 
+static int svc_partial_recvfrom(struct svc_rqst *rqstp,
+				struct kvec *iov, int nr,
+				int buflen, unsigned int base)
+{
+	size_t save_iovlen;
+	void __user *save_iovbase;
+	unsigned int i;
+	int ret;
+
+	if (base == 0)
+		return svc_recvfrom(rqstp, iov, nr, buflen);
+
+	for (i = 0; i < nr; i++) {
+		if (iov[i].iov_len > base)
+			break;
+		base -= iov[i].iov_len;
+	}
+	save_iovlen = iov[i].iov_len;
+	save_iovbase = iov[i].iov_base;
+	iov[i].iov_len -= base;
+	iov[i].iov_base += base;
+	ret = svc_recvfrom(rqstp, &iov[i], nr - i, buflen);
+	iov[i].iov_len = save_iovlen;
+	iov[i].iov_base = save_iovbase;
+	return ret;
+}
+
 /*
  * Set socket snd and rcv buffer lengths
  */
@@ -409,7 +436,6 @@ static void svc_sock_setbufsize(struct socket *sock, unsigned int snd,
 	lock_sock(sock->sk);
 	sock->sk->sk_sndbuf = snd * 2;
 	sock->sk->sk_rcvbuf = rcv * 2;
-	sock->sk->sk_userlocks |= SOCK_SNDBUF_LOCK|SOCK_RCVBUF_LOCK;
 	sock->sk->sk_write_space(sock->sk);
 	release_sock(sock->sk);
 #endif
@@ -884,6 +910,56 @@ static struct svc_xprt *svc_tcp_accept(struct svc_xprt *xprt)
 	return NULL;
 }
 
+static unsigned int svc_tcp_restore_pages(struct svc_sock *svsk, struct svc_rqst *rqstp)
+{
+	unsigned int i, len, npages;
+
+	if (svsk->sk_tcplen <= sizeof(rpc_fraghdr))
+		return 0;
+	len = svsk->sk_tcplen - sizeof(rpc_fraghdr);
+	npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	for (i = 0; i < npages; i++) {
+		if (rqstp->rq_pages[i] != NULL)
+			put_page(rqstp->rq_pages[i]);
+		BUG_ON(svsk->sk_pages[i] == NULL);
+		rqstp->rq_pages[i] = svsk->sk_pages[i];
+		svsk->sk_pages[i] = NULL;
+	}
+	rqstp->rq_arg.head[0].iov_base = page_address(rqstp->rq_pages[0]);
+	return len;
+}
+
+static void svc_tcp_save_pages(struct svc_sock *svsk, struct svc_rqst *rqstp)
+{
+	unsigned int i, len, npages;
+
+	if (svsk->sk_tcplen <= sizeof(rpc_fraghdr))
+		return;
+	len = svsk->sk_tcplen - sizeof(rpc_fraghdr);
+	npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	for (i = 0; i < npages; i++) {
+		svsk->sk_pages[i] = rqstp->rq_pages[i];
+		rqstp->rq_pages[i] = NULL;
+	}
+}
+
+static void svc_tcp_clear_pages(struct svc_sock *svsk)
+{
+	unsigned int i, len, npages;
+
+	if (svsk->sk_tcplen <= sizeof(rpc_fraghdr))
+		goto out;
+	len = svsk->sk_tcplen - sizeof(rpc_fraghdr);
+	npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	for (i = 0; i < npages; i++) {
+		BUG_ON(svsk->sk_pages[i] == NULL);
+		put_page(svsk->sk_pages[i]);
+		svsk->sk_pages[i] = NULL;
+	}
+out:
+	svsk->sk_tcplen = 0;
+}
+
 /*
  * Receive data.
  * If we haven't gotten the record length yet, get the next four bytes.
@@ -893,31 +969,15 @@ static struct svc_xprt *svc_tcp_accept(struct svc_xprt *xprt)
 static int svc_tcp_recv_record(struct svc_sock *svsk, struct svc_rqst *rqstp)
 {
 	struct svc_serv	*serv = svsk->sk_xprt.xpt_server;
+	unsigned int want;
 	int len;
 
-	if (test_and_clear_bit(XPT_CHNGBUF, &svsk->sk_xprt.xpt_flags))
-		/* sndbuf needs to have room for one request
-		 * per thread, otherwise we can stall even when the
-		 * network isn't a bottleneck.
-		 *
-		 * We count all threads rather than threads in a
-		 * particular pool, which provides an upper bound
-		 * on the number of threads which will access the socket.
-		 *
-		 * rcvbuf just needs to be able to hold a few requests.
-		 * Normally they will be removed from the queue
-		 * as soon a a complete request arrives.
-		 */
-		svc_sock_setbufsize(svsk->sk_sock,
-				    (serv->sv_nrthreads+3) * serv->sv_max_mesg,
-				    3 * serv->sv_max_mesg);
-
 	clear_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 
 	if (svsk->sk_tcplen < sizeof(rpc_fraghdr)) {
-		int		want = sizeof(rpc_fraghdr) - svsk->sk_tcplen;
 		struct kvec	iov;
 
+		want = sizeof(rpc_fraghdr) - svsk->sk_tcplen;
 		iov.iov_base = ((char *) &svsk->sk_reclen) + svsk->sk_tcplen;
 		iov.iov_len  = want;
 		if ((len = svc_recvfrom(rqstp, &iov, 1, want)) < 0)
@@ -927,7 +987,7 @@ static int svc_tcp_recv_record(struct svc_sock *svsk, struct svc_rqst *rqstp)
 		if (len < want) {
 			dprintk("svc: short recvfrom while reading record "
 				"length (%d of %d)\n", len, want);
-			goto err_again; /* record header not complete */
+			return -EAGAIN;
 		}
 
 		svsk->sk_reclen = ntohl(svsk->sk_reclen);
@@ -954,54 +1014,32 @@ static int svc_tcp_recv_record(struct svc_sock *svsk, struct svc_rqst *rqstp)
 		}
 	}
 
-	/* Check whether enough data is available */
-	len = svc_recv_available(svsk);
-	if (len < 0)
-		goto error;
+	if (svsk->sk_reclen < 8)
+		goto err_delete; /* client is nuts. */
 
-	if (len < svsk->sk_reclen) {
-		dprintk("svc: incomplete TCP record (%d of %d)\n",
-			len, svsk->sk_reclen);
-		goto err_again;	/* record not complete */
-	}
 	len = svsk->sk_reclen;
-	set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 
 	return len;
- error:
-	if (len == -EAGAIN)
-		dprintk("RPC: TCP recv_record got EAGAIN\n");
+error:
+	dprintk("RPC: TCP recv_record got %d\n", len);
 	return len;
- err_delete:
+err_delete:
 	set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags);
- err_again:
 	return -EAGAIN;
 }
 
-static int svc_process_calldir(struct svc_sock *svsk, struct svc_rqst *rqstp,
-			       struct rpc_rqst **reqpp, struct kvec *vec)
+static int receive_cb_reply(struct svc_sock *svsk, struct svc_rqst *rqstp)
 {
+	struct rpc_xprt *bc_xprt = svsk->sk_xprt.xpt_bc_xprt;
 	struct rpc_rqst *req = NULL;
-	u32 *p;
-	u32 xid;
-	u32 calldir;
-	int len;
+	struct kvec *src, *dst;
+	__be32 *p = (__be32 *)rqstp->rq_arg.head[0].iov_base;
+	__be32 xid;
+	__be32 calldir;
 
-	len = svc_recvfrom(rqstp, vec, 1, 8);
-	if (len < 0)
-		goto error;
-
-	p = (u32 *)rqstp->rq_arg.head[0].iov_base;
 	xid = *p++;
 	calldir = *p;
 
-	if (calldir == 0) {
-		/* REQUEST is the most common case */
-		vec[0] = rqstp->rq_arg.head[0];
-	} else {
-		/* REPLY */
-		struct rpc_xprt *bc_xprt = svsk->sk_xprt.xpt_bc_xprt;
-
 	if (bc_xprt)
 		req = xprt_lookup_rqst(bc_xprt, xid);
 
@@ -1011,26 +1049,40 @@ static int svc_process_calldir(struct svc_sock *svsk, struct svc_rqst *rqstp,
 			"calldir 0x%x xpt_bc_xprt %p xid %08x\n",
 			__func__, ntohl(calldir),
 			bc_xprt, xid);
-			vec[0] = rqstp->rq_arg.head[0];
-			goto out;
+		return -EAGAIN;
 	}
 
-		memcpy(&req->rq_private_buf, &req->rq_rcv_buf,
-		       sizeof(struct xdr_buf));
-		/* copy the xid and call direction */
-		memcpy(req->rq_private_buf.head[0].iov_base,
-		       rqstp->rq_arg.head[0].iov_base, 8);
-		vec[0] = req->rq_private_buf.head[0];
+	memcpy(&req->rq_private_buf, &req->rq_rcv_buf, sizeof(struct xdr_buf));
+	/*
+	 * XXX!: cheating for now!  Only copying HEAD.
+	 * But we know this is good enough for now (in fact, for any
+	 * callback reply in the forseeable future).
+	 */
+	dst = &req->rq_private_buf.head[0];
+	src = &rqstp->rq_arg.head[0];
+	if (dst->iov_len < src->iov_len)
+		return -EAGAIN; /* whatever; just giving up. */
+	memcpy(dst->iov_base, src->iov_base, src->iov_len);
+	xprt_complete_rqst(req->rq_task, svsk->sk_reclen);
+	rqstp->rq_arg.len = 0;
+	return 0;
+}
+
+static int copy_pages_to_kvecs(struct kvec *vec, struct page **pages, int len)
+{
+	int i = 0;
+	int t = 0;
+
+	while (t < len) {
+		vec[i].iov_base = page_address(pages[i]);
+		vec[i].iov_len = PAGE_SIZE;
+		i++;
+		t += PAGE_SIZE;
 	}
- out:
-	vec[0].iov_base += 8;
-	vec[0].iov_len -= 8;
-	len = svsk->sk_reclen - 8;
- error:
-	*reqpp = req;
-	return len;
+	return i;
 }
 
+
 /*
  * Receive data from a TCP socket.
  */
@@ -1041,8 +1093,10 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp)
 	struct svc_serv	*serv = svsk->sk_xprt.xpt_server;
 	int		len;
 	struct kvec *vec;
-	int pnum, vlen;
-	struct rpc_rqst *req = NULL;
+	unsigned int want, base;
+	__be32 *p;
+	__be32 calldir;
+	int pnum;
 
 	dprintk("svc: tcp_recv %p data %d conn %d close %d\n",
 		svsk, test_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags),
@@ -1053,87 +1107,73 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp)
 	if (len < 0)
 		goto error;
 
+	base = svc_tcp_restore_pages(svsk, rqstp);
+	want = svsk->sk_reclen - base;
+
 	vec = rqstp->rq_vec;
-	vec[0] = rqstp->rq_arg.head[0];
-	vlen = PAGE_SIZE;
 
-	/*
-	 * We have enough data for the whole tcp record. Let's try and read the
-	 * first 8 bytes to get the xid and the call direction. We can use this
-	 * to figure out if this is a call or a reply to a callback. If
-	 * sk_reclen is < 8 (xid and calldir), then this is a malformed packet.
-	 * In that case, don't bother with the calldir and just read the data.
-	 * It will be rejected in svc_process.
-	 */
-	if (len >= 8) {
-		len = svc_process_calldir(svsk, rqstp, &req, vec);
-		if (len < 0)
-			goto err_again;
-		vlen -= 8;
-	}
+	pnum = copy_pages_to_kvecs(&vec[0], &rqstp->rq_pages[0],
+						svsk->sk_reclen);
 
-	pnum = 1;
-	while (vlen < len) {
-		vec[pnum].iov_base = (req) ?
-			page_address(req->rq_private_buf.pages[pnum - 1]) :
-			page_address(rqstp->rq_pages[pnum]);
-		vec[pnum].iov_len = PAGE_SIZE;
-		pnum++;
-		vlen += PAGE_SIZE;
-	}
 	rqstp->rq_respages = &rqstp->rq_pages[pnum];
 
 	/* Now receive data */
-	len = svc_recvfrom(rqstp, vec, pnum, len);
-	if (len < 0)
-		goto err_again;
-
-	/*
-	 * Account for the 8 bytes we read earlier
-	 */
-	len += 8;
-
-	if (req) {
-		xprt_complete_rqst(req->rq_task, len);
-		len = 0;
-		goto out;
+	len = svc_partial_recvfrom(rqstp, vec, pnum, want, base);
+	if (len >= 0)
+		svsk->sk_tcplen += len;
+	if (len != want) {
+		if (len < 0 && len != -EAGAIN)
+			goto err_other;
+		svc_tcp_save_pages(svsk, rqstp);
+		dprintk("svc: incomplete TCP record (%d of %d)\n",
+			svsk->sk_tcplen, svsk->sk_reclen);
+		goto err_noclose;
 	}
-	dprintk("svc: TCP complete record (%d bytes)\n", len);
-	rqstp->rq_arg.len = len;
+
+	rqstp->rq_arg.len = svsk->sk_reclen;
 	rqstp->rq_arg.page_base = 0;
-	if (len <= rqstp->rq_arg.head[0].iov_len) {
-		rqstp->rq_arg.head[0].iov_len = len;
+	if (rqstp->rq_arg.len <= rqstp->rq_arg.head[0].iov_len) {
+		rqstp->rq_arg.head[0].iov_len = rqstp->rq_arg.len;
 		rqstp->rq_arg.page_len = 0;
-	} else {
-		rqstp->rq_arg.page_len = len - rqstp->rq_arg.head[0].iov_len;
-	}
+	} else
+		rqstp->rq_arg.page_len = rqstp->rq_arg.len - rqstp->rq_arg.head[0].iov_len;
 
 	rqstp->rq_xprt_ctxt   = NULL;
 	rqstp->rq_prot	      = IPPROTO_TCP;
 
-out:
+	p = (__be32 *)rqstp->rq_arg.head[0].iov_base;
+	calldir = p[1];
+	if (calldir)
+		len = receive_cb_reply(svsk, rqstp);
+
 	/* Reset TCP read info */
 	svsk->sk_reclen = 0;
 	svsk->sk_tcplen = 0;
+	/* If we have more data, signal svc_xprt_enqueue() to try again */
+	if (svc_recv_available(svsk) > sizeof(rpc_fraghdr))
+		set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
+
+	if (len < 0)
+		goto error;
 
 	svc_xprt_copy_addrs(rqstp, &svsk->sk_xprt);
 	if (serv->sv_stats)
 		serv->sv_stats->nettcpcnt++;
 
-	return len;
+	dprintk("svc: TCP complete record (%d bytes)\n", rqstp->rq_arg.len);
+	return rqstp->rq_arg.len;
 
-err_again:
-	if (len == -EAGAIN) {
-		dprintk("RPC: TCP recvfrom got EAGAIN\n");
-		return len;
-	}
 error:
-	if (len != -EAGAIN) {
+	if (len != -EAGAIN)
+		goto err_other;
+	dprintk("RPC: TCP recvfrom got EAGAIN\n");
+	return -EAGAIN;
+err_other:
 	printk(KERN_NOTICE "%s: recvfrom returned errno %d\n",
 	       svsk->sk_xprt.xpt_server->sv_name, -len);
 	set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags);
-	}
-	return -EAGAIN;
+err_noclose:
+	return -EAGAIN;	/* record not complete */
 }
 
 /*
@@ -1304,18 +1344,10 @@ static void svc_tcp_init(struct svc_sock *svsk, struct svc_serv *serv)
 
 		svsk->sk_reclen = 0;
 		svsk->sk_tcplen = 0;
+		memset(&svsk->sk_pages[0], 0, sizeof(svsk->sk_pages));
 
 		tcp_sk(sk)->nonagle |= TCP_NAGLE_OFF;
 
-		/* initialise setting must have enough space to
-		 * receive and respond to one request.
-		 * svc_tcp_recvfrom will re-adjust if necessary
-		 */
-		svc_sock_setbufsize(svsk->sk_sock,
-				    3 * svsk->sk_xprt.xpt_server->sv_max_mesg,
-				    3 * svsk->sk_xprt.xpt_server->sv_max_mesg);
-
-		set_bit(XPT_CHNGBUF, &svsk->sk_xprt.xpt_flags);
 		set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 		if (sk->sk_state != TCP_ESTABLISHED)
 			set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags);
@@ -1379,8 +1411,14 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv,
 	/* Initialize the socket */
 	if (sock->type == SOCK_DGRAM)
 		svc_udp_init(svsk, serv);
-	else
+	else {
+		/* initialise setting must have enough space to
+		 * receive and respond to one request.
+		 */
+		svc_sock_setbufsize(svsk->sk_sock, 4 * serv->sv_max_mesg,
+					4 * serv->sv_max_mesg);
 		svc_tcp_init(svsk, serv);
+	}
 
 	dprintk("svc: svc_setup_socket created %p (inet %p)\n",
 				svsk, svsk->sk_sk);
@@ -1562,8 +1600,10 @@ static void svc_tcp_sock_detach(struct svc_xprt *xprt)
 
 	svc_sock_detach(xprt);
 
-	if (!test_bit(XPT_LISTENER, &xprt->xpt_flags))
+	if (!test_bit(XPT_LISTENER, &xprt->xpt_flags)) {
+		svc_tcp_clear_pages(svsk);
 		kernel_sock_shutdown(svsk->sk_sock, SHUT_RDWR);
+	}
 }
 
 /*