[PATCH kNFSd 0 of 9] Introduction

[PATCH kNFSd 0 of 9] Introduction

[NeilBrown]

9 patches for knfsd in 2.6.9-rc1-mm2

The first three affect all versions.  The remaining are v4 specific.

1 - Add O_NONBLOCK to calls to break_lease so the nfs server doesn't block
    pointlessly.  Return nfserr_jukebox for v3, or just drop the reques for
    v2 if we get a EWOULDBLOCK.

2 - Change a return of "ESTALE" or "EACCES" in a situation where ACCES is 
    argueable the more correct error, and real people have had real problems
    that were really fixed.

3 - Fix some indentation.

4 - Check is filesystem supports acls before returning "acl supported" attribute

5-9 assorted v4 locking changes.

NeilBrown


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

[PATCH kNFSd 1 of 9] Calls to break_lease in nfsd should be O_NONBLOCKing.

[NeilBrown]

If we would block, we return "err-jukebox" for nfsv3,
or just drop the request for v2.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

### Diffstat output
 ./fs/nfsd/nfsproc.c |    1 +
 ./fs/nfsd/nfssvc.c  |    2 ++
 ./fs/nfsd/vfs.c     |    8 ++++----
 3 files changed, 7 insertions(+), 4 deletions(-)

diff ./fs/nfsd/nfsproc.c~current~ ./fs/nfsd/nfsproc.c
--- ./fs/nfsd/nfsproc.c~current~	2004-09-03 11:34:26.000000000 +1000
+++ ./fs/nfsd/nfsproc.c	2004-09-03 11:34:29.000000000 +1000
@@ -586,6 +586,7 @@ nfserrno (int errno)
 		{ nfserr_dquot, -EDQUOT },
 #endif
 		{ nfserr_stale, -ESTALE },
+		{ nfserr_jukebox, -EWOULDBLOCK },
 		{ nfserr_jukebox, -ETIMEDOUT },
 		{ nfserr_dropit, -EAGAIN },
 		{ nfserr_dropit, -ENOMEM },

diff ./fs/nfsd/nfssvc.c~current~ ./fs/nfsd/nfssvc.c
--- ./fs/nfsd/nfssvc.c~current~	2004-09-03 11:34:26.000000000 +1000
+++ ./fs/nfsd/nfssvc.c	2004-09-03 11:34:29.000000000 +1000
@@ -328,6 +328,8 @@ nfsd_dispatch(struct svc_rqst *rqstp, u3
 
 	/* Now call the procedure handler, and encode NFS status. */
 	nfserr = proc->pc_func(rqstp, rqstp->rq_argp, rqstp->rq_resp);
+	if (nfserr == nfserr_jukebox && rqstp->rq_vers == 2)
+		nfserr = nfserr_dropit;
 	if (nfserr == nfserr_dropit) {
 		dprintk("nfsd: Dropping request due to malloc failure!\n");
 		nfsd_cache_update(rqstp, RC_NOCACHE, NULL);

diff ./fs/nfsd/vfs.c~current~ ./fs/nfsd/vfs.c
--- ./fs/nfsd/vfs.c~current~	2004-09-03 11:34:26.000000000 +1000
+++ ./fs/nfsd/vfs.c	2004-09-03 11:34:29.000000000 +1000
@@ -303,8 +303,8 @@ nfsd_setattr(struct svc_rqst *rqstp, str
 		 * If we are changing the size of the file, then
 		 * we need to break all leases.
 		 */
-		err = break_lease(inode, FMODE_WRITE);
-		if (err)
+		err = break_lease(inode, FMODE_WRITE | O_NONBLOCK);
+		if (err) /* ENOMEM or EWOULDBLOCK */
 			goto out_nfserr;
 
 		err = get_write_access(inode);
@@ -669,8 +669,8 @@ nfsd_open(struct svc_rqst *rqstp, struct
 	 * Check to see if there are any leases on this file.
 	 * This may block while leases are broken.
 	 */
-	err = break_lease(inode, (access & MAY_WRITE) ? FMODE_WRITE : 0);
-	if (err)
+	err = break_lease(inode, O_NONBLOCK | ((access & MAY_WRITE) ? FMODE_WRITE : 0));
+	if (err) /* NOMEM or WOULDBLOCK */
 		goto out_nfserr;
 
 	if (access & MAY_WRITE) {


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

[PATCH kNFSd 6 of 9] nfsd4 could leak a stateid in an error path

[NeilBrown]

nfsd4 could leak a stateid in a case of kmalloc failure; fix.

Signed-off-by: Andy Adamson <andros@umich.edu>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

### Diffstat output
 ./fs/nfsd/nfs4state.c |    4 +++-
 1 files changed, 3 insertions(+), 1 deletion(-)

diff ./fs/nfsd/nfs4state.c~current~ ./fs/nfsd/nfs4state.c
--- ./fs/nfsd/nfs4state.c~current~	2004-09-03 11:57:20.000000000 +1000
+++ ./fs/nfsd/nfs4state.c	2004-09-03 11:57:34.000000000 +1000
@@ -2153,8 +2153,10 @@ nfsd4_lock(struct svc_rqst *rqstp, struc
 		if (!(lock->lk_stateowner = alloc_init_lock_stateowner(strhashval, open_sop->so_client, open_stp, lock)))
 			goto out;
 		if ((lock_stp = alloc_init_lock_stateid(lock->lk_stateowner, 
-						fp, open_stp)) == NULL)
+						fp, open_stp)) == NULL) {
+			release_stateowner(lock->lk_stateowner);
 			goto out;
+		}
 		/* bump the open seqid used to create the lock */
 		open_sop->so_seqid++;
 	} else {


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

[PATCH kNFSd 8 of 9] nfsd4: store current->tgid instead of lockowner hash in fl_pid

[NeilBrown]

The file_lock.fl_pid is no longer used in posix_same_owner() tests.  So just
set it to current->tgid for informational purposes.

Signed-off-by: Andy Adamson <andros@umich.edu>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

### Diffstat output
 ./fs/nfsd/nfs4state.c |    9 +++++----
 1 files changed, 5 insertions(+), 4 deletions(-)

diff ./fs/nfsd/nfs4state.c~current~ ./fs/nfsd/nfs4state.c
--- ./fs/nfsd/nfs4state.c~current~	2004-09-03 11:57:34.000000000 +1000
+++ ./fs/nfsd/nfs4state.c	2004-09-03 11:57:34.000000000 +1000
@@ -1950,9 +1950,10 @@ static inline void
 nfs4_set_lock_denied(struct file_lock *fl, struct nfsd4_lock_denied *deny)
 {
 	struct nfs4_stateowner *sop = (struct nfs4_stateowner *) fl->fl_owner;
+	unsigned int hval = lockownerid_hashval(sop->so_id);
 
 	deny->ld_sop = NULL;
-	if (nfs4_verify_lock_stateowner(sop, fl->fl_pid))
+	if (nfs4_verify_lock_stateowner(sop, hval))
 		deny->ld_sop = sop;
 	deny->ld_start = fl->fl_start;
 	deny->ld_length = ~(u64)0;
@@ -2196,7 +2197,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struc
 		goto out;
 	}
 	file_lock.fl_owner = (fl_owner_t) lock->lk_stateowner;
-	file_lock.fl_pid = lockownerid_hashval(lock->lk_stateowner->so_id);
+	file_lock.fl_pid = current->tgid;
 	file_lock.fl_file = filp;
 	file_lock.fl_flags = FL_POSIX;
 
@@ -2322,7 +2323,7 @@ nfsd4_lockt(struct svc_rqst *rqstp, stru
 	sop = lockt->lt_stateowner;
 	if (sop) {
 		file_lock.fl_owner = (fl_owner_t) sop;
-		file_lock.fl_pid = lockownerid_hashval(sop->so_id);
+		file_lock.fl_pid = current->tgid;
 	} else {
 		file_lock.fl_owner = NULL;
 		file_lock.fl_pid = 0;
@@ -2386,7 +2387,7 @@ nfsd4_locku(struct svc_rqst *rqstp, stru
 	locks_init_lock(&file_lock);
 	file_lock.fl_type = F_UNLCK;
 	file_lock.fl_owner = (fl_owner_t) locku->lu_stateowner;
-	file_lock.fl_pid = lockownerid_hashval(locku->lu_stateowner->so_id);
+	file_lock.fl_pid = current->tgid;
 	file_lock.fl_file = filp;
 	file_lock.fl_flags = FL_POSIX; 
 	file_lock.fl_start = locku->lu_offset;


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

[PATCH kNFSd 3 of 9] fix incorrect indentation in fh_verify

[NeilBrown]

fix incorrect indentation in fh_verify

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

### Diffstat output
 ./fs/nfsd/nfsfh.c |    8 ++++----
 1 files changed, 4 insertions(+), 4 deletions(-)

diff ./fs/nfsd/nfsfh.c~current~ ./fs/nfsd/nfsfh.c
--- ./fs/nfsd/nfsfh.c~current~	2004-09-03 11:56:47.000000000 +1000
+++ ./fs/nfsd/nfsfh.c	2004-09-03 11:56:56.000000000 +1000
@@ -190,10 +190,10 @@ fh_verify(struct svc_rqst *rqstp, struct
 			dentry = dget(exp->ex_dentry);
 		else {
 			struct export_operations *nop = exp->ex_mnt->mnt_sb->s_export_op;
-				dentry = CALL(nop,decode_fh)(exp->ex_mnt->mnt_sb,
-							     datap, data_left,
-							     fileid_type,
-							     nfsd_acceptable, exp);
+			dentry = CALL(nop,decode_fh)(exp->ex_mnt->mnt_sb,
+						     datap, data_left,
+						     fileid_type,
+						     nfsd_acceptable, exp);
 		}
 		if (dentry == NULL)
 			goto out;


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

[PATCH kNFSd 4 of 9] nfsd4: Support acl_support attribute

[NeilBrown]

The nfs4 attributes supported_attrs and aclsupport should not be
static; they need to depend on the exported filesystem's acl support.  Test the
latter by attempting to get an acl, and adjust the returned attributes
apropriately.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

### Diffstat output
 ./fs/nfsd/nfs4xdr.c |   25 +++++++++++++++----------
 1 files changed, 15 insertions(+), 10 deletions(-)

diff ./fs/nfsd/nfs4xdr.c~current~ ./fs/nfsd/nfs4xdr.c
--- ./fs/nfsd/nfs4xdr.c~current~	2004-09-03 11:56:48.000000000 +1000
+++ ./fs/nfsd/nfs4xdr.c	2004-09-03 11:57:04.000000000 +1000
@@ -1414,6 +1414,7 @@ nfsd4_encode_fattr(struct svc_fh *fhp, s
 	u64 dummy64;
 	u32 *p = buffer;
 	int status;
+	int aclsupport = 0;
 	struct nfs4_acl *acl = NULL;
 
 	BUG_ON(bmval1 & NFSD_WRITEONLY_ATTRS_WORD1);
@@ -1437,12 +1438,16 @@ nfsd4_encode_fattr(struct svc_fh *fhp, s
 			goto out;
 		fhp = &tempfh;
 	}
-	if (bmval0 & FATTR4_WORD0_ACL) {
+	if (bmval0 & (FATTR4_WORD0_ACL | FATTR4_WORD0_ACLSUPPORT
+			| FATTR4_WORD0_SUPPORTED_ATTRS)) {
 		status = nfsd4_get_nfs4_acl(rqstp, dentry, &acl);
-		if (status == -EOPNOTSUPP)
-			bmval0 &= ~FATTR4_WORD0_ACL;
-		else if (status < 0)
-			goto out_nfserr;
+		aclsupport = (status == 0);
+		if (bmval0 & FATTR4_WORD0_ACL) {
+			if (status == -EOPNOTSUPP)
+				bmval0 &= ~FATTR4_WORD0_ACL;
+			else if (status != 0)
+				goto out_nfserr;
+		}
 	}
 	if ((buflen -= 16) < 0)
 		goto out_resource;
@@ -1456,9 +1461,9 @@ nfsd4_encode_fattr(struct svc_fh *fhp, s
 		if ((buflen -= 12) < 0)
 			goto out_resource;
 		WRITE32(2);
-		/* XXX Should depend on exported filesystem (e.g.
-		 * for acl support) */
-		WRITE32(NFSD_SUPPORTED_ATTRS_WORD0);
+		WRITE32(aclsupport ?
+			NFSD_SUPPORTED_ATTRS_WORD0 :
+			NFSD_SUPPORTED_ATTRS_WORD0 & ~FATTR4_WORD0_ACL);
 		WRITE32(NFSD_SUPPORTED_ATTRS_WORD1);
 	}
 	if (bmval0 & FATTR4_WORD0_TYPE) {
@@ -1565,8 +1570,8 @@ out_acl:
 	if (bmval0 & FATTR4_WORD0_ACLSUPPORT) {
 		if ((buflen -= 4) < 0)
 			goto out_resource;
-		/* XXX: should depend on exported filesystem: */
-		WRITE32(ACL4_SUPPORT_ALLOW_ACL|ACL4_SUPPORT_DENY_ACL);
+		WRITE32(aclsupport ?
+			ACL4_SUPPORT_ALLOW_ACL|ACL4_SUPPORT_DENY_ACL : 0);
 	}
 	if (bmval0 & FATTR4_WORD0_CANSETTIME) {
 		if ((buflen -= 4) < 0)


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

[PATCH kNFSd 9 of 9] Remove redundant initialization in nfsd4_lockt

[NeilBrown]

No need to set fl_owner and fl_pid to 0, since that's already been done by
locks_init_lock.

Signed-off-by: Andy Adamson <andros@umich.edu>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

### Diffstat output
 ./fs/nfsd/nfs4state.c |   12 +++---------
 1 files changed, 3 insertions(+), 9 deletions(-)

diff ./fs/nfsd/nfs4state.c~current~ ./fs/nfsd/nfs4state.c
--- ./fs/nfsd/nfs4state.c~current~	2004-09-03 11:57:34.000000000 +1000
+++ ./fs/nfsd/nfs4state.c	2004-09-03 11:57:34.000000000 +1000
@@ -2268,7 +2268,6 @@ int
 nfsd4_lockt(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lockt *lockt)
 {
 	struct inode *inode;
-	struct nfs4_stateowner *sop;
 	struct file file;
 	struct file_lock file_lock;
 	struct file_lock *conflicting_lock;
@@ -2320,14 +2319,9 @@ nfsd4_lockt(struct svc_rqst *rqstp, stru
 	find_lockstateowner_str(strhashval, &lockt->lt_owner,
 					&lockt->lt_clientid, 
 					&lockt->lt_stateowner);
-	sop = lockt->lt_stateowner;
-	if (sop) {
-		file_lock.fl_owner = (fl_owner_t) sop;
-		file_lock.fl_pid = current->tgid;
-	} else {
-		file_lock.fl_owner = NULL;
-		file_lock.fl_pid = 0;
-	}
+	if (lockt->lt_stateowner)
+		file_lock.fl_owner = (fl_owner_t)lockt->lt_stateowner;
+	file_lock.fl_pid = current->tgid;
 	file_lock.fl_flags = FL_POSIX;
 
 	file_lock.fl_start = lockt->lt_offset;


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

[PATCH kNFSd 5 of 9] trivial cleanup of nfs4state.c

[NeilBrown]

Whitespace cleanup, fix one dprintk, remove superfluous casts of NULL.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

### Diffstat output
 ./fs/nfsd/nfs4state.c |   72 +++++++++++++++++++++++++-------------------------
 1 files changed, 36 insertions(+), 36 deletions(-)

diff ./fs/nfsd/nfs4state.c~current~ ./fs/nfsd/nfs4state.c
--- ./fs/nfsd/nfs4state.c~current~	2004-09-03 11:56:48.000000000 +1000
+++ ./fs/nfsd/nfs4state.c	2004-09-03 11:57:20.000000000 +1000
@@ -236,7 +236,7 @@ static struct nfs4_client *
 create_client(struct xdr_netobj name) {
 	struct nfs4_client *clp;
 
-	if(!(clp = alloc_client(name)))
+	if (!(clp = alloc_client(name)))
 		goto out;
 	INIT_LIST_HEAD(&clp->cl_idhash);
 	INIT_LIST_HEAD(&clp->cl_strhash);
@@ -268,7 +268,7 @@ copy_cred(struct svc_cred *target, struc
 
 static int
 cmp_name(struct xdr_netobj *n1, struct xdr_netobj *n2) {
-	if(!n1 || !n2)
+	if (!n1 || !n2)
 		return 0;
 	return((n1->len == n2->len) && !memcmp(n1->data, n2->data, n2->len));
 }
@@ -399,7 +399,7 @@ parse_ipv4(unsigned int addr_len, char *
 			return 0;
 		}
 		cbaddr |= (temp << shift);
-		if(shift > 0)
+		if (shift > 0)
 		shift -= 8;
 	}
 	*cbaddrp = cbaddr;
@@ -411,7 +411,7 @@ parse_ipv4(unsigned int addr_len, char *
 			return 0;
 		}
 		cbport |= (temp << shift);
-		if(shift > 0)
+		if (shift > 0)
 			shift -= 8;
 	}
 	*cbportp = cbport;
@@ -423,7 +423,7 @@ gen_callback(struct nfs4_client *clp, st
 {
 	struct nfs4_callback *cb = &clp->cl_callback;
 
-	if( !(parse_ipv4(se->se_callback_addr_len, se->se_callback_addr_val,
+	if ( !(parse_ipv4(se->se_callback_addr_len, se->se_callback_addr_val,
 		         &cb->cb_addr, &cb->cb_port))) {
 		printk(KERN_INFO "NFSD: BAD callback address. client will not receive delegations\n");
 		cb->cb_parsed = 0;
@@ -431,9 +431,9 @@ gen_callback(struct nfs4_client *clp, st
 	}
 	cb->cb_netid.len = se->se_callback_netid_len;
 	cb->cb_netid.data = se->se_callback_netid_val;
-        cb->cb_prog = se->se_callback_prog;
-        cb->cb_ident = se->se_callback_ident;
-        cb->cb_parsed = 1;
+	cb->cb_prog = se->se_callback_prog;
+	cb->cb_ident = se->se_callback_ident;
+	cb->cb_parsed = 1;
 }
 
 /*
@@ -812,7 +812,7 @@ alloc_init_file(unsigned int hashval, st
 		alloc_file++;
 		return fp;
 	}
-	return (struct nfs4_file *)NULL;
+	return NULL;
 }
 
 static void
@@ -825,7 +825,7 @@ release_all_files(void)
 		while (!list_empty(&file_hashtbl[i])) {
 			fp = list_entry(file_hashtbl[i].next, struct nfs4_file, fi_hash);
 			/* this should never be more than once... */
-			if(!list_empty(&fp->fi_perfile)) {
+			if (!list_empty(&fp->fi_perfile)) {
 				printk("ERROR: release_all_files: file %p is open, creating dangling state !!!\n",fp);
 			}
 			release_file(fp);
@@ -839,20 +839,20 @@ alloc_stateowner(struct xdr_netobj *owne
 	struct nfs4_stateowner *sop;
 
 	if ((sop = kmalloc(sizeof(struct nfs4_stateowner),GFP_KERNEL))) {
-		if((sop->so_owner.data = kmalloc(owner->len, GFP_KERNEL))) {
+		if ((sop->so_owner.data = kmalloc(owner->len, GFP_KERNEL))) {
 			memcpy(sop->so_owner.data, owner->data, owner->len);
 			sop->so_owner.len = owner->len;
 			return sop;
 		} 
 		kfree(sop);
 	}
-	return (struct nfs4_stateowner *)NULL;
+	return NULL;
 }
 
 /* should use a slab cache */
 static void
 free_stateowner(struct nfs4_stateowner *sop) {
-	if(sop) {
+	if (sop) {
 		kfree(sop->so_owner.data);
 		kfree(sop);
 		sop = NULL;
@@ -867,7 +867,7 @@ alloc_init_open_stateowner(unsigned int 
 	unsigned int idhashval;
 
 	if (!(sop = alloc_stateowner(&open->op_owner)))
-		return (struct nfs4_stateowner *)NULL;
+		return NULL;
 	idhashval = ownerid_hashval(current_ownerid);
 	INIT_LIST_HEAD(&sop->so_idhash);
 	INIT_LIST_HEAD(&sop->so_strhash);
@@ -921,7 +921,7 @@ release_stateowner(struct nfs4_stateowne
 	while (!list_empty(&sop->so_perfilestate)) {
 		stp = list_entry(sop->so_perfilestate.next, 
 			struct nfs4_stateid, st_perfilestate);
-		if(sop->so_is_open_owner)
+		if (sop->so_is_open_owner)
 			release_stateid(stp, OPEN_STATE);
 		else
 			release_stateid(stp, LOCK_STATE);
@@ -960,7 +960,7 @@ release_stateid(struct nfs4_stateid *stp
 	list_del_perfile++;
 	list_del(&stp->st_perfile);
 	list_del(&stp->st_perfilestate);
-	if((stp->st_vfs_set) && (flags & OPEN_STATE)) {
+	if ((stp->st_vfs_set) && (flags & OPEN_STATE)) {
 		release_stateid_lockowner(stp);
 		nfsd_close(stp->st_vfs_file);
 		vfsclose++;
@@ -991,8 +991,8 @@ move_to_close_lru(struct nfs4_stateowner
 	list_del_init(&sop->so_strhash);
 	list_del_init(&sop->so_perlockowner);
 
-        list_add_tail(&sop->so_close_lru, &close_lru);
-        sop->so_time = get_seconds();
+	list_add_tail(&sop->so_close_lru, &close_lru);
+	sop->so_time = get_seconds();
 }
 
 void
@@ -1030,7 +1030,7 @@ find_openstateowner_str(unsigned int has
 	struct nfs4_stateowner *local = NULL;
 
 	list_for_each_entry(local, &ownerstr_hashtbl[hashval], so_strhash) {
-		if(!cmp_owner_str(local, &open->op_owner, &open->op_clientid)) 
+		if (!cmp_owner_str(local, &open->op_owner, &open->op_clientid))
 			continue;
 		*op = local;
 		return(1);
@@ -1078,7 +1078,7 @@ set_access(unsigned int *access, unsigne
 
 	*access = 0;
 	for (i = 1; i < 4; i++) {
-		if(test_bit(i, &bmap))
+		if (test_bit(i, &bmap))
 			*access |= i;
 	}
 }
@@ -1089,7 +1089,7 @@ set_deny(unsigned int *deny, unsigned lo
 
 	*deny = 0;
 	for (i = 0; i < 4; i++) {
-		if(test_bit(i, &bmap))
+		if (test_bit(i, &bmap))
 			*deny |= i ;
 	}
 }
@@ -1278,7 +1278,7 @@ nfsd4_process_open2(struct svc_rqst *rqs
 		/* Search for conflicting share reservations */
 		status = nfserr_share_denied;
 		list_for_each_entry(stq, &fp->fi_perfile, st_perfile) {
-			if(stq->st_stateowner == sop) {
+			if (stq->st_stateowner == sop) {
 				stp = stq;
 				continue;
 			}
@@ -1484,7 +1484,7 @@ find_openstateowner_id(u32 st_id, int fl
 	dprintk("NFSD: find_openstateowner_id %d\n", st_id);
 	if (flags & CLOSE_STATE) {
 		list_for_each_entry(local, &close_lru, so_close_lru) {
-			if(local->so_id == st_id)
+			if (local->so_id == st_id)
 				return local;
 		}
 	}
@@ -1888,16 +1888,16 @@ find_stateid(stateid_t *stid, int flags)
 	if ((flags & LOCK_STATE) || (flags & RDWR_STATE)) {
 		hashval = stateid_hashval(st_id, f_id);
 		list_for_each_entry(local, &lockstateid_hashtbl[hashval], st_hash) {
-			if((local->st_stateid.si_stateownerid == st_id) &&
-			   (local->st_stateid.si_fileid == f_id))
+			if ((local->st_stateid.si_stateownerid == st_id) &&
+			    (local->st_stateid.si_fileid == f_id))
 				return local;
 		}
 	} 
 	if ((flags & OPEN_STATE) || (flags & RDWR_STATE)) {
 		hashval = stateid_hashval(st_id, f_id);
 		list_for_each_entry(local, &stateid_hashtbl[hashval], st_hash) {
-			if((local->st_stateid.si_stateownerid == st_id) &&
-			   (local->st_stateid.si_fileid == f_id))
+			if ((local->st_stateid.si_stateownerid == st_id) &&
+			    (local->st_stateid.si_fileid == f_id))
 				return local;
 		}
 	} else
@@ -1967,7 +1967,7 @@ find_lockstateowner(struct xdr_netobj *o
 
 	for (i = 0; i < LOCK_HASH_SIZE; i++) {
 		list_for_each_entry(local, &lock_ownerid_hashtbl[i], so_idhash) {
-			if(!cmp_owner_str(local, owner, clid))
+			if (!cmp_owner_str(local, owner, clid))
 				continue;
 			return local;
 		}
@@ -1980,7 +1980,7 @@ find_lockstateowner_str(unsigned int has
 	struct nfs4_stateowner *local = NULL;
 
 	list_for_each_entry(local, &lock_ownerstr_hashtbl[hashval], so_strhash) {
-		if(!cmp_owner_str(local, owner, clid)) 
+		if (!cmp_owner_str(local, owner, clid))
 			continue;
 		*op = local;
 		return(1);
@@ -2005,7 +2005,7 @@ alloc_init_lock_stateowner(unsigned int 
 	unsigned int idhashval;
 
 	if (!(sop = alloc_stateowner(&lock->lk_new_owner)))
-		return (struct nfs4_stateowner *)NULL;
+		return NULL;
 	idhashval = lockownerid_hashval(current_ownerid);
 	INIT_LIST_HEAD(&sop->so_idhash);
 	INIT_LIST_HEAD(&sop->so_strhash);
@@ -2068,7 +2068,7 @@ int
 check_lock_length(u64 offset, u64 length)
 {
 	return ((length == 0)  || ((length != ~(u64)0) &&
- 	     LOFF_OVERFLOW(offset, length)));
+	     LOFF_OVERFLOW(offset, length)));
 }
 
 /*
@@ -2210,7 +2210,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struc
 	status = posix_lock_file(filp, &file_lock);
 	if (file_lock.fl_ops && file_lock.fl_ops->fl_release_private)
 		file_lock.fl_ops->fl_release_private(&file_lock);
-	dprintk("NFSD: nfsd4_lock: posix_test_lock passed. posix_lock_file status %d\n",status);
+	dprintk("NFSD: nfsd4_lock: posix_lock_file status %d\n",status);
 	switch (-status) {
 	case 0: /* success! */
 		update_stateid(&lock_stp->st_stateid);
@@ -2459,7 +2459,7 @@ nfsd4_release_lockowner(struct svc_rqst 
 
 	nfs4_lock_state();
 
-        status = nfs_ok;
+	status = nfs_ok;
 	local = find_lockstateowner(owner, clid);
 	if (local) {
 		struct nfs4_stateid *stp;
@@ -2469,7 +2469,7 @@ nfsd4_release_lockowner(struct svc_rqst 
 		status = nfserr_locks_held;
 		list_for_each_entry(stp, &local->so_perfilestate,
 				st_perfilestate) {
-			if(stp->st_vfs_set) {
+			if (stp->st_vfs_set) {
 				if (check_for_locks(stp->st_vfs_file, local))
 					goto out;
 			}
@@ -2496,7 +2496,7 @@ alloc_reclaim(int namelen)
 		kfree(crp);
 		return NULL;
 	}
-       	return crp;
+	return crp;
 }
 
 /*
@@ -2566,7 +2566,7 @@ nfs4_find_reclaim_client(clientid_t *cli
 	strhashval = clientstr_hashval(client->cl_name.data,
 	                              client->cl_name.len);
 	list_for_each_entry(crp, &reclaim_str_hashtbl[strhashval], cr_strhash) {
-		if(cmp_name(&crp->cr_name, &client->cl_name)) {
+		if (cmp_name(&crp->cr_name, &client->cl_name)) {
 			return crp;
 		}
 	}


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

[PATCH kNFSd 2 of 9] Return EACCES instead of ESTALE for certain filehandle lookup failures

[NeilBrown]

As 'nfs_acceptable' uses the authnetication user in it's checks,
the filehandle might not be truely "stale" - EACCES is safer.

Thanks Olaf Kirch <okir@suse.de> and others.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

### Diffstat output
 ./fs/exportfs/expfs.c |    7 ++++++-
 1 files changed, 6 insertions(+), 1 deletion(-)

diff ./fs/exportfs/expfs.c~current~ ./fs/exportfs/expfs.c
--- ./fs/exportfs/expfs.c~current~	2004-09-03 11:34:26.000000000 +1000
+++ ./fs/exportfs/expfs.c	2004-09-03 11:37:53.000000000 +1000
@@ -283,7 +283,12 @@ find_exported_dentry(struct super_block 
 
 	/* drat - I just cannot find anything acceptable */
 	dput(result);
-	return ERR_PTR(-ESTALE);
+	/* It might be justifiable to return ESTALE here,
+	 * but the filehandle at-least looks reasonable good
+	 * and it just be a permission problem, so returning
+	 * -EACCESS is safer
+	 */
+	return ERR_PTR(-EACCES);
 
  err_target:
 	dput(target_dir);


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

[PATCH kNFSd 7 of 9] nfsd4: postpone release of stateowner on CLOSE

[NeilBrown]

Postpone the release of a stateowner on CLOSE for lease time to enable the
CLOSE replay cache. Place stateowner on the close_lru list to be reaped by the
laundromat service.

Signed-off-by: Andy Adamson <andros@umich.edu>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

### Diffstat output
 ./fs/nfsd/nfs4state.c |   18 +++++++++++-------
 1 files changed, 11 insertions(+), 7 deletions(-)

diff ./fs/nfsd/nfs4state.c~current~ ./fs/nfsd/nfs4state.c
--- ./fs/nfsd/nfs4state.c~current~	2004-09-03 11:57:34.000000000 +1000
+++ ./fs/nfsd/nfs4state.c	2004-09-03 11:57:34.000000000 +1000
@@ -908,7 +908,7 @@ release_stateid_lockowner(struct nfs4_st
 }
 
 static void
-release_stateowner(struct nfs4_stateowner *sop)
+unhash_stateowner(struct nfs4_stateowner *sop)
 {
 	struct nfs4_stateid *stp;
 
@@ -916,7 +916,6 @@ release_stateowner(struct nfs4_stateowne
 	list_del(&sop->so_strhash);
 	list_del(&sop->so_perclient);
 	list_del(&sop->so_perlockowner);
-	list_del(&sop->so_close_lru);
 	del_perclient++;
 	while (!list_empty(&sop->so_perfilestate)) {
 		stp = list_entry(sop->so_perfilestate.next, 
@@ -926,6 +925,13 @@ release_stateowner(struct nfs4_stateowne
 		else
 			release_stateid(stp, LOCK_STATE);
 	}
+}
+
+static void
+release_stateowner(struct nfs4_stateowner *sop)
+{
+	unhash_stateowner(sop);
+	list_del(&sop->so_close_lru);
 	free_stateowner(sop);
 }
 
@@ -986,11 +992,8 @@ void
 move_to_close_lru(struct nfs4_stateowner *sop)
 {
 	dprintk("NFSD: move_to_close_lru nfs4_stateowner %p\n", sop);
-	/* remove stateowner from all other hash lists except perclient */
-	list_del_init(&sop->so_idhash);
-	list_del_init(&sop->so_strhash);
-	list_del_init(&sop->so_perlockowner);
 
+	unhash_stateowner(sop);
 	list_add_tail(&sop->so_close_lru, &close_lru);
 	sop->so_time = get_seconds();
 }
@@ -1456,7 +1459,8 @@ nfs4_laundromat(void)
 		}
 		dprintk("NFSD: purging unused open stateowner (so_id %d)\n",
 			sop->so_id);
-		release_stateowner(sop);
+		list_del(&sop->so_close_lru);
+		free_stateowner(sop);
 	}
 	if (clientid_val < NFSD_LAUNDROMAT_MINTIMEOUT)
 		clientid_val = NFSD_LAUNDROMAT_MINTIMEOUT;


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

Re: [PATCH kNFSd 1 of 9] Calls to break_lease in nfsd should be O_NONBLOCKing.

[J. Bruce Fields]

On Fri, Sep 03, 2004 at 12:34:25PM +1000, NeilBrown wrote:
> 
> If we would block, we return "err-jukebox" for nfsv3,
> or just drop the request for v2.

Sorry, only just noticed there's a problem here:

> diff ./fs/nfsd/nfsproc.c~current~ ./fs/nfsd/nfsproc.c
> --- ./fs/nfsd/nfsproc.c~current~	2004-09-03 11:34:26.000000000 +1000
> +++ ./fs/nfsd/nfsproc.c	2004-09-03 11:34:29.000000000 +1000
> @@ -586,6 +586,7 @@ nfserrno (int errno)
>  		{ nfserr_dquot, -EDQUOT },
>  #endif
>  		{ nfserr_stale, -ESTALE },
> +		{ nfserr_jukebox, -EWOULDBLOCK },
>  		{ nfserr_jukebox, -ETIMEDOUT },
>  		{ nfserr_dropit, -EAGAIN },
>  		{ nfserr_dropit, -ENOMEM },

Note that EWOULDBLOCK is defined to be EAGAIN in
include/asm-generic/errno.h, so the effect of this is to map EAGAIN to
nfserr_jukebox instead of nfserr_dropit.

The cache upcall deferral code depends on EAGAIN being mapped to
nfserr_dropit.  So now whenever cache upcall deferral happens, the
client gets a nfserr_jukebox response, which isn't what we want.

--b.


-------------------------------------------------------
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us what you think of them. Give us
Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
http://productguide.itmanagersjournal.com/guidepromo.tmpl
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs