Import kernel part of SMB/CIFS requester.

Add smbfs(CIFS) filesystem.

Userland part will be in the ports tree for a while.

Obtained from:	smbfs-1.3.7-dev package.

1 files changed, 874 insertions, 0 deletions

diff --git a/sys/netsmb/smb_conn.c b/sys/netsmb/smb_conn.c
new file mode 100644
index 0000000..54f5c0b
--- /dev/null
+++ b/sys/netsmb/smb_conn.c
@@ -0,0 +1,874 @@
+/*
+ * Copyright (c) 2000-2001 Boris Popov
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. All advertising materials mentioning features or use of this software
+ *    must display the following acknowledgement:
+ *    This product includes software developed by Boris Popov.
+ * 4. Neither the name of the author nor the names of any co-contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * $FreeBSD$
+ */
+
+/*
+ * Connection engine.
+ */
+
+#include <sys/param.h>
+#include <sys/systm.h>
+#include <sys/kernel.h>
+#include <sys/malloc.h>
+#include <sys/proc.h>
+#include <sys/lock.h>
+#include <sys/sysctl.h>
+#include <sys/socketvar.h>
+
+#include <sys/iconv.h>
+
+#include <netsmb/smb.h>
+#include <netsmb/smb_subr.h>
+#include <netsmb/smb_conn.h>
+#include <netsmb/smb_tran.h>
+#include <netsmb/smb_trantcp.h>
+
+static struct smb_connobj smb_vclist;
+static int smb_vcnext = 1;	/* next unique id for VC */
+
+extern struct linker_set sysctl_net_smb;
+
+SYSCTL_NODE(_net, OID_AUTO, smb, CTLFLAG_RW, NULL, "SMB protocol");
+
+MALLOC_DEFINE(M_SMBCONN, "SMB conn", "SMB connection");
+
+static void smb_co_init(struct smb_connobj *cp, int level, char *objname,
+	struct proc *p);
+static void smb_co_done(struct smb_connobj *cp);
+static int  smb_co_lockstatus(struct smb_connobj *cp, struct proc *p);
+
+static int  smb_vc_disconnect(struct smb_vc *vcp);
+static void smb_vc_free(struct smb_connobj *cp);
+static void smb_vc_gone(struct smb_connobj *cp, struct smb_cred *scred);
+static smb_co_free_t smb_share_free;
+static smb_co_gone_t smb_share_gone;
+
+static int  smb_sysctl_treedump(SYSCTL_HANDLER_ARGS);
+
+SYSCTL_PROC(_net_smb, OID_AUTO, treedump, CTLFLAG_RD | CTLTYPE_OPAQUE,
+	    NULL, 0, smb_sysctl_treedump, "S,treedump", "Requester tree");
+
+int
+smb_sm_init(void)
+{
+
+	smb_co_init(&smb_vclist, SMBL_SM, "smbsm", curproc);
+	smb_co_unlock(&smb_vclist, 0, curproc);
+	return 0;
+}
+
+int
+smb_sm_done(void)
+{
+
+	/* XXX: hold the mutex */
+	if (smb_vclist.co_usecount > 1) {
+		SMBERROR("%d connections still active\n", smb_vclist.co_usecount - 1);
+		return EBUSY;
+	}
+	smb_co_done(&smb_vclist);
+	return 0;
+}
+
+static int
+smb_sm_lockvclist(int flags, struct proc *p)
+{
+
+	return smb_co_lock(&smb_vclist, flags | LK_CANRECURSE, p);
+}
+
+static void
+smb_sm_unlockvclist(struct proc *p)
+{
+
+	smb_co_unlock(&smb_vclist, LK_RELEASE, p);
+}
+
+static int
+smb_sm_lookupint(struct smb_vcspec *vcspec, struct smb_sharespec *shspec,
+	struct smb_cred *scred,	struct smb_vc **vcpp)
+{
+	struct proc *p = scred->scr_p;
+	struct smb_vc *vcp;
+	int exact = 1;
+	int error;
+
+	vcspec->shspec = shspec;
+	error = ENOENT;
+	SMBCO_FOREACH((struct smb_connobj*)vcp, &smb_vclist) {
+		error = smb_vc_lock(vcp, LK_EXCLUSIVE, p);
+		if (error)
+			continue;
+		itry {
+			if ((vcp->obj.co_flags & SMBV_PRIVATE) ||
+			    !CONNADDREQ(vcp->vc_paddr, vcspec->sap) ||
+			    strcmp(vcp->vc_username, vcspec->username) != 0)
+				ithrow(1);
+			if (vcspec->owner != SMBM_ANY_OWNER) {
+				if (vcp->vc_uid != vcspec->owner)
+					ithrow(1);
+			} else
+				exact = 0;
+			if (vcspec->group != SMBM_ANY_GROUP) {
+				if (vcp->vc_grp != vcspec->group)
+					ithrow(1);
+			} else
+				exact = 0;
+
+			if (vcspec->mode & SMBM_EXACT) {
+				if (!exact ||
+				    (vcspec->mode & SMBM_MASK) != vcp->vc_mode)
+					ithrow(1);
+			}
+			if (smb_vc_access(vcp, scred, vcspec->mode) != 0)
+				ithrow(1);
+			vcspec->ssp = NULL;
+			if (shspec)
+				ithrow(smb_vc_lookupshare(vcp, shspec, scred, &vcspec->ssp));
+			error = 0;
+			break;
+		} icatch(error) {
+			smb_vc_unlock(vcp, 0, p);
+		} ifinally {
+		} iendtry;
+		if (error == 0)
+			break;
+	}
+	if (vcp) {
+		smb_vc_ref(vcp, p);
+		*vcpp = vcp;
+	}
+	return error;
+}
+
+int
+smb_sm_lookup(struct smb_vcspec *vcspec, struct smb_sharespec *shspec,
+	struct smb_cred *scred,	struct smb_vc **vcpp)
+{
+	struct proc *p = scred->scr_p;
+	struct smb_vc *vcp;
+	struct smb_share *ssp = NULL;
+	int error;
+
+	*vcpp = vcp = NULL;
+
+	error = smb_sm_lockvclist(LK_EXCLUSIVE, p);
+	if (error)
+		return error;
+	error = smb_sm_lookupint(vcspec, shspec, scred, vcpp);
+	if (error == 0 || (vcspec->flags & SMBV_CREATE) == 0) {
+		smb_sm_unlockvclist(p);
+		return error;
+	}
+	error = smb_sm_lookupint(vcspec, NULL, scred, &vcp);
+	if (error) {
+		error = smb_vc_create(vcspec, scred, &vcp);
+		if (error)
+			goto out;
+		error = smb_vc_connect(vcp, scred);
+		if (error)
+			goto out;
+	}
+	if (shspec == NULL)
+		goto out;
+	error = smb_share_create(vcp, shspec, scred, &ssp);
+	if (error)
+		goto out;
+	error = smb_smb_treeconnect(ssp, scred);
+	if (error == 0)
+		vcspec->ssp = ssp;
+	else
+		smb_share_put(ssp, scred);
+out:
+	smb_sm_unlockvclist(p);
+	if (error == 0)
+		*vcpp = vcp;
+	else if (vcp)
+		smb_vc_put(vcp, scred);
+	return error;
+}
+
+/*
+ * Common code for connection object
+ */
+static void
+smb_co_init(struct smb_connobj *cp, int level, char *objname, struct proc *p)
+{
+	SLIST_INIT(&cp->co_children);
+	smb_sl_init(&cp->co_interlock, objname);
+	lockinit(&cp->co_lock, PZERO, objname, 0, 0);
+	cp->co_level = level;
+	cp->co_usecount = 1;
+	KASSERT(smb_co_lock(cp, LK_EXCLUSIVE, p) == 0, ("smb_co_init: lock failed"));
+}
+
+static void
+smb_co_done(struct smb_connobj *cp)
+{
+	smb_sl_destroy(&cp->co_interlock);
+	lockdestroy(&cp->co_lock);
+}
+
+static void
+smb_co_gone(struct smb_connobj *cp, struct smb_cred *scred)
+{
+	struct smb_connobj *parent;
+
+	if (cp->co_gone)
+		cp->co_gone(cp, scred);
+	parent = cp->co_parent;
+	if (parent) {
+		smb_co_lock(parent, LK_EXCLUSIVE, scred->scr_p);
+		SLIST_REMOVE(&parent->co_children, cp, smb_connobj, co_next);
+		smb_co_put(parent, scred);
+	}
+	if (cp->co_free)
+		cp->co_free(cp);
+}
+
+void
+smb_co_ref(struct smb_connobj *cp, struct proc *p)
+{
+
+	SMB_CO_LOCK(cp);
+	cp->co_usecount++;
+	SMB_CO_UNLOCK(cp);
+}
+
+void
+smb_co_rele(struct smb_connobj *cp, struct smb_cred *scred)
+{
+	struct proc *p = scred->scr_p;
+
+	SMB_CO_LOCK(cp);
+	if (cp->co_usecount > 1) {
+		cp->co_usecount--;
+		SMB_CO_UNLOCK(cp);
+		return;
+	}
+	if (cp->co_usecount == 0) {
+		SMBERROR("negative use_count for object %d", cp->co_level);
+		SMB_CO_UNLOCK(cp);
+		return;
+	}
+	cp->co_usecount--;
+	cp->co_flags |= SMBO_GONE;
+
+	lockmgr(&cp->co_lock, LK_DRAIN | LK_INTERLOCK, &cp->co_interlock, p);
+	smb_co_gone(cp, scred);
+}
+
+int
+smb_co_get(struct smb_connobj *cp, int flags, struct smb_cred *scred)
+{
+	int error;
+
+	if ((flags & LK_INTERLOCK) == 0)
+		SMB_CO_LOCK(cp);
+	cp->co_usecount++;
+	error = smb_co_lock(cp, flags | LK_INTERLOCK, scred->scr_p);
+	if (error) {
+		SMB_CO_LOCK(cp);
+		cp->co_usecount--;
+		SMB_CO_UNLOCK(cp);
+		return error;
+	}
+	return 0;
+}
+
+void
+smb_co_put(struct smb_connobj *cp, struct smb_cred *scred)
+{
+	struct proc *p = scred->scr_p;
+	int flags;
+
+	flags = LK_RELEASE;
+	SMB_CO_LOCK(cp);
+	if (cp->co_usecount > 1) {
+		cp->co_usecount--;
+	} else if (cp->co_usecount == 1) {
+		cp->co_usecount--;
+		cp->co_flags |= SMBO_GONE;
+		flags = LK_DRAIN;
+	} else {
+		SMBERROR("negative usecount");
+	}
+	lockmgr(&cp->co_lock, LK_RELEASE | LK_INTERLOCK, &cp->co_interlock, p);
+	if ((cp->co_flags & SMBO_GONE) == 0)
+		return;
+	lockmgr(&cp->co_lock, LK_DRAIN, NULL, p);
+	smb_co_gone(cp, scred);
+}
+
+int
+smb_co_lockstatus(struct smb_connobj *cp, struct proc *p)
+{
+	return lockstatus(&cp->co_lock, p);
+}
+
+int
+smb_co_lock(struct smb_connobj *cp, int flags, struct proc *p)
+{
+
+	if (cp->co_flags & SMBO_GONE)
+		return EINVAL;
+	if ((flags & LK_TYPE_MASK) == 0)
+		flags |= LK_EXCLUSIVE;
+	if (smb_co_lockstatus(cp, p) == LK_EXCLUSIVE && 
+	    (flags & LK_CANRECURSE) == 0) {
+		SMBERROR("recursive lock for object %d\n", cp->co_level);
+		return 0;
+	}
+	return lockmgr(&cp->co_lock, flags, &cp->co_interlock, p);
+}
+
+void
+smb_co_unlock(struct smb_connobj *cp, int flags, struct proc *p)
+{
+	(void)lockmgr(&cp->co_lock, flags | LK_RELEASE, &cp->co_interlock, p);
+}
+
+static void
+smb_co_addchild(struct smb_connobj *parent, struct smb_connobj *child)
+{
+	KASSERT(smb_co_lockstatus(parent, curproc) == LK_EXCLUSIVE, ("smb_co_addchild: parent not locked"));
+	KASSERT(smb_co_lockstatus(child, curproc) == LK_EXCLUSIVE, ("smb_co_addchild: child not locked"));
+
+	smb_co_ref(parent, curproc);
+	SLIST_INSERT_HEAD(&parent->co_children, child, co_next);
+	child->co_parent = parent;
+}
+
+/*
+ * Session implementation
+ */
+
+int
+smb_vc_create(struct smb_vcspec *vcspec,
+	struct smb_cred *scred, struct smb_vc **vcpp)
+{
+	struct smb_vc *vcp;
+	struct proc *p = scred->scr_p;
+	struct ucred *cred = scred->scr_cred;
+	uid_t uid = vcspec->owner;
+	gid_t gid = vcspec->group;
+	uid_t realuid = cred->cr_uid;
+	char *domain = vcspec->domain;
+	int error, isroot;
+
+	isroot = smb_suser(cred) == 0;
+	/*
+	 * Only superuser can create VCs with different uid and gid
+	 */
+	if (uid != SMBM_ANY_OWNER && uid != realuid && !isroot)
+		return EPERM;
+	if (gid != SMBM_ANY_GROUP && !groupmember(gid, cred) && !isroot)
+		return EPERM;
+
+	vcp = smb_zmalloc(sizeof(*vcp), M_SMBCONN, M_WAITOK);
+	smb_co_init(VCTOCP(vcp), SMBL_VC, "smb_vc", p);
+	vcp->obj.co_free = smb_vc_free;
+	vcp->obj.co_gone = smb_vc_gone;
+	vcp->vc_number = smb_vcnext++;
+	vcp->vc_timo = SMB_DEFRQTIMO;
+	vcp->vc_smbuid = SMB_UID_UNKNOWN;
+	vcp->vc_mode = vcspec->rights & SMBM_MASK;
+	vcp->obj.co_flags = vcspec->flags & (SMBV_PRIVATE | SMBV_SINGLESHARE);
+	vcp->vc_tdesc = &smb_tran_nbtcp_desc;
+
+	if (uid == SMBM_ANY_OWNER)
+		uid = realuid;
+	if (gid == SMBM_ANY_GROUP)
+		gid = cred->cr_groups[0];
+	vcp->vc_uid = uid;
+	vcp->vc_grp = gid;
+
+	smb_sl_init(&vcp->vc_stlock, "vcstlock");
+	error = 0;
+	itry {
+		vcp->vc_paddr = dup_sockaddr(vcspec->sap, 1);
+		ierror(vcp->vc_paddr == NULL, ENOMEM);
+
+		vcp->vc_laddr = dup_sockaddr(vcspec->lap, 1);
+		ierror(vcp->vc_laddr == NULL, ENOMEM);
+
+		ierror((vcp->vc_pass = smb_strdup(vcspec->pass)) == NULL, ENOMEM);
+
+		vcp->vc_domain = smb_strdup((domain && domain[0]) ? domain : "NODOMAIN");
+		ierror(vcp->vc_domain == NULL, ENOMEM);
+
+		ierror((vcp->vc_srvname = smb_strdup(vcspec->srvname)) == NULL, ENOMEM);
+		ierror((vcp->vc_username = smb_strdup(vcspec->username)) == NULL, ENOMEM);
+
+		ithrow(iconv_open("tolower", vcspec->localcs, &vcp->vc_tolower));
+		ithrow(iconv_open("toupper", vcspec->localcs, &vcp->vc_toupper));
+		if (vcspec->servercs[0]) {
+			ithrow(iconv_open(vcspec->servercs, vcspec->localcs,
+			    &vcp->vc_toserver));
+			ithrow(iconv_open(vcspec->localcs, vcspec->servercs,
+			    &vcp->vc_tolocal));
+		}
+
+		ithrow(smb_iod_create(vcp));
+		*vcpp = vcp;
+		smb_co_addchild(&smb_vclist, VCTOCP(vcp));
+	} icatch(error) {
+		smb_vc_put(vcp, scred);
+	} ifinally {
+	} iendtry;
+	return error;
+}
+
+static void
+smb_vc_free(struct smb_connobj *cp)
+{
+	struct smb_vc *vcp = CPTOVC(cp);
+
+	if (vcp->vc_iod)
+		smb_iod_destroy(vcp->vc_iod);
+	SMB_STRFREE(vcp->vc_username);
+	SMB_STRFREE(vcp->vc_srvname);
+	SMB_STRFREE(vcp->vc_pass);
+	SMB_STRFREE(vcp->vc_domain);
+	if (vcp->vc_paddr)
+		free(vcp->vc_paddr, M_SONAME);
+	if (vcp->vc_laddr)
+		free(vcp->vc_laddr, M_SONAME);
+	if (vcp->vc_tolower)
+		iconv_close(vcp->vc_tolower);
+	if (vcp->vc_toupper)
+		iconv_close(vcp->vc_toupper);
+	if (vcp->vc_tolocal)
+		iconv_close(vcp->vc_tolocal);
+	if (vcp->vc_toserver)
+		iconv_close(vcp->vc_toserver);
+	smb_co_done(VCTOCP(vcp));
+	smb_sl_destroy(&vcp->vc_stlock);
+	free(vcp, M_SMBCONN);
+}
+
+/*
+ * Called when use count of VC dropped to zero.
+ * VC should be locked on enter with LK_DRAIN.
+ */
+static void
+smb_vc_gone(struct smb_connobj *cp, struct smb_cred *scred)
+{
+	struct smb_vc *vcp = CPTOVC(cp);
+
+	smb_vc_disconnect(vcp);
+}
+
+void
+smb_vc_ref(struct smb_vc *vcp, struct proc *p)
+{
+	smb_co_ref(VCTOCP(vcp), p);
+}
+
+void
+smb_vc_rele(struct smb_vc *vcp, struct smb_cred *scred)
+{
+	smb_co_rele(VCTOCP(vcp), scred);
+}
+
+int
+smb_vc_get(struct smb_vc *vcp, int flags, struct smb_cred *scred)
+{
+	return smb_co_get(VCTOCP(vcp), flags, scred);
+}
+
+void
+smb_vc_put(struct smb_vc *vcp, struct smb_cred *scred)
+{
+	smb_co_put(VCTOCP(vcp), scred);
+}
+
+int
+smb_vc_lock(struct smb_vc *vcp, int flags, struct proc *p)
+{
+	return smb_co_lock(VCTOCP(vcp), flags, p);
+}
+
+void
+smb_vc_unlock(struct smb_vc *vcp, int flags, struct proc *p)
+{
+	smb_co_unlock(VCTOCP(vcp), flags, p);
+}
+
+int
+smb_vc_access(struct smb_vc *vcp, struct smb_cred *scred, mode_t mode)
+{
+	struct ucred *cred = scred->scr_cred;
+
+	if (smb_suser(cred) == 0 || cred->cr_uid == vcp->vc_uid)
+		return 0;
+	mode >>= 3;
+	if (!groupmember(vcp->vc_grp, cred))
+		mode >>= 3;
+	return (vcp->vc_mode & mode) == mode ? 0 : EACCES;
+}
+
+static int
+smb_vc_cmpshare(struct smb_share *ssp, struct smb_sharespec *dp)
+{
+	int exact = 1;
+
+	if (strcmp(ssp->ss_name, dp->name) != 0)
+		return 1;
+	if (dp->owner != SMBM_ANY_OWNER) {
+		if (ssp->ss_uid != dp->owner)
+			return 1;
+	} else
+		exact = 0;
+	if (dp->group != SMBM_ANY_GROUP) {
+		if (ssp->ss_grp != dp->group)
+			return 1;
+	} else
+		exact = 0;
+
+	if (dp->mode & SMBM_EXACT) {
+		if (!exact)
+			return 1;
+		return (dp->mode & SMBM_MASK) == ssp->ss_mode ? 0 : 1;
+	}
+	if (smb_share_access(ssp, dp->scred, dp->mode) != 0)
+		return 1;
+	return 0;
+}
+
+/*
+ * Lookup share in the given VC. Share referenced and locked on return.
+ * VC expected to be locked on entry and will be left locked on exit.
+ */
+int
+smb_vc_lookupshare(struct smb_vc *vcp, struct smb_sharespec *dp,
+	struct smb_cred *scred,	struct smb_share **sspp)
+{
+	struct proc *p = scred->scr_p;
+	struct smb_share *ssp = NULL;
+	int error;
+
+	*sspp = NULL;
+	dp->scred = scred;
+	SMBCO_FOREACH((struct smb_connobj*)ssp, VCTOCP(vcp)) {
+		error = smb_share_lock(ssp, LK_EXCLUSIVE, p);
+		if (error)
+			continue;
+		if (smb_vc_cmpshare(ssp, dp) == 0)
+			break;
+		smb_share_unlock(ssp, 0, p);
+	}
+	if (ssp) {
+		smb_share_ref(ssp, p);
+		*sspp = ssp;
+		error = 0;
+	} else
+		error = ENOENT;
+	return error;
+}
+
+int
+smb_vc_connect(struct smb_vc *vcp, struct smb_cred *scred)
+{
+
+	return smb_iod_request(vcp->vc_iod, SMBIOD_EV_CONNECT | SMBIOD_EV_SYNC, NULL);
+}
+
+/*
+ * Destroy VC to server, invalidate shares linked with it.
+ * Transport should be locked on entry.
+ */
+int
+smb_vc_disconnect(struct smb_vc *vcp)
+{
+
+	smb_iod_request(vcp->vc_iod, SMBIOD_EV_DISCONNECT | SMBIOD_EV_SYNC, NULL);
+	return 0;
+}
+
+static char smb_emptypass[] = "";
+
+const char *
+smb_vc_getpass(struct smb_vc *vcp)
+{
+	if (vcp->vc_pass)
+		return vcp->vc_pass;
+	return smb_emptypass;
+}
+
+static int
+smb_vc_getinfo(struct smb_vc *vcp, struct smb_vc_info *vip)
+{
+	bzero(vip, sizeof(struct smb_vc_info));
+	vip->itype = SMB_INFO_VC;
+	vip->usecount = vcp->obj.co_usecount;
+	vip->uid = vcp->vc_uid;
+	vip->gid = vcp->vc_grp;
+	vip->mode = vcp->vc_mode;
+	vip->flags = vcp->obj.co_flags;
+	vip->sopt = vcp->vc_sopt;
+	vip->iodstate = vcp->vc_iod->iod_state;
+	bzero(&vip->sopt.sv_skey, sizeof(vip->sopt.sv_skey));
+	snprintf(vip->srvname, sizeof(vip->srvname), "%s", vcp->vc_srvname);
+	snprintf(vip->vcname, sizeof(vip->vcname), "%s", vcp->vc_username);
+	return 0;
+}
+
+u_short
+smb_vc_nextmid(struct smb_vc *vcp)
+{
+	u_short r;
+
+	SMB_CO_LOCK(&vcp->obj);
+	r = vcp->vc_mid++;
+	SMB_CO_UNLOCK(&vcp->obj);
+	return r;
+}
+
+/*
+ * Share implementation
+ */
+/*
+ * Allocate share structure and attach it to the given VC
+ * Connection expected to be locked on entry. Share will be returned
+ * in locked state.
+ */
+int
+smb_share_create(struct smb_vc *vcp, struct smb_sharespec *shspec,
+	struct smb_cred *scred, struct smb_share **sspp)
+{
+	struct smb_share *ssp;
+	struct proc *p = scred->scr_p;
+	struct ucred *cred = scred->scr_cred;
+	uid_t realuid = cred->cr_uid;
+	uid_t uid = shspec->owner;
+	gid_t gid = shspec->group;
+	int error, isroot;
+
+	isroot = smb_suser(cred) == 0;
+	/*
+	 * Only superuser can create shares with different uid and gid
+	 */
+	if (uid != SMBM_ANY_OWNER && uid != realuid && !isroot)
+		return EPERM;
+	if (gid != SMBM_ANY_GROUP && !groupmember(gid, cred) && !isroot)
+		return EPERM;
+	error = smb_vc_lookupshare(vcp, shspec, scred, &ssp);
+	if (!error) {
+		smb_share_put(ssp, scred);
+		return EEXIST;
+	}
+	if (uid == SMBM_ANY_OWNER)
+		uid = realuid;
+	if (gid == SMBM_ANY_GROUP)
+		gid = cred->cr_groups[0];
+	ssp = smb_zmalloc(sizeof(*ssp), M_SMBCONN, M_WAITOK);
+	smb_co_init(SSTOCP(ssp), SMBL_SHARE, "smbss", p);
+	ssp->obj.co_free = smb_share_free;
+	ssp->obj.co_gone = smb_share_gone;
+	smb_sl_init(&ssp->ss_stlock, "ssstlock");
+	ssp->ss_name = smb_strdup(shspec->name);
+	if (shspec->pass && shspec->pass[0])
+		ssp->ss_pass = smb_strdup(shspec->pass);
+	ssp->ss_type = shspec->stype;
+	ssp->ss_tid = SMB_TID_UNKNOWN;
+	ssp->ss_uid = uid;
+	ssp->ss_grp = gid;
+	ssp->ss_mode = shspec->rights & SMBM_MASK;
+	smb_co_addchild(VCTOCP(vcp), SSTOCP(ssp));
+	*sspp = ssp;
+	return 0;
+}
+
+static void
+smb_share_free(struct smb_connobj *cp)
+{
+	struct smb_share *ssp = CPTOSS(cp);
+
+	SMB_STRFREE(ssp->ss_name);
+	SMB_STRFREE(ssp->ss_pass);
+	smb_sl_destroy(&ssp->ss_stlock);
+	smb_co_done(SSTOCP(ssp));
+	free(ssp, M_SMBCONN);
+}
+
+static void
+smb_share_gone(struct smb_connobj *cp, struct smb_cred *scred)
+{
+	struct smb_share *ssp = CPTOSS(cp);
+
+	smb_smb_treedisconnect(ssp, scred);
+}
+
+void
+smb_share_ref(struct smb_share *ssp, struct proc *p)
+{
+	smb_co_ref(SSTOCP(ssp), p);
+}
+
+void
+smb_share_rele(struct smb_share *ssp, struct smb_cred *scred)
+{
+	smb_co_rele(SSTOCP(ssp), scred);
+}
+
+int
+smb_share_get(struct smb_share *ssp, int flags, struct smb_cred *scred)
+{
+	return smb_co_get(SSTOCP(ssp), flags, scred);
+}
+
+void
+smb_share_put(struct smb_share *ssp, struct smb_cred *scred)
+{
+	smb_co_put(SSTOCP(ssp), scred);
+}
+
+int
+smb_share_lock(struct smb_share *ssp, int flags, struct proc *p)
+{
+	return smb_co_lock(SSTOCP(ssp), flags, p);
+}
+
+void
+smb_share_unlock(struct smb_share *ssp, int flags, struct proc *p)
+{
+	smb_co_unlock(SSTOCP(ssp), flags, p);
+}
+
+int
+smb_share_access(struct smb_share *ssp, struct smb_cred *scred, mode_t mode)
+{
+	struct ucred *cred = scred->scr_cred;
+
+	if (smb_suser(cred) == 0 || cred->cr_uid == ssp->ss_uid)
+		return 0;
+	mode >>= 3;
+	if (!groupmember(ssp->ss_grp, cred))
+		mode >>= 3;
+	return (ssp->ss_mode & mode) == mode ? 0 : EACCES;
+}
+
+void
+smb_share_invalidate(struct smb_share *ssp)
+{
+	ssp->ss_tid = SMB_TID_UNKNOWN;
+}
+
+int
+smb_share_valid(struct smb_share *ssp)
+{
+	return ssp->ss_tid != SMB_TID_UNKNOWN &&
+	    ssp->ss_vcgenid == SSTOVC(ssp)->vc_genid;
+}
+
+const char*
+smb_share_getpass(struct smb_share *ssp)
+{
+	struct smb_vc *vcp;
+
+	if (ssp->ss_pass)
+		return ssp->ss_pass;
+	vcp = SSTOVC(ssp);
+	if (vcp->vc_pass)
+		return vcp->vc_pass;
+	return smb_emptypass;
+}
+
+static int
+smb_share_getinfo(struct smb_share *ssp, struct smb_share_info *sip)
+{
+	bzero(sip, sizeof(struct smb_share_info));
+	sip->itype = SMB_INFO_SHARE;
+	sip->usecount = ssp->obj.co_usecount;
+	sip->tid  = ssp->ss_tid;
+	sip->type= ssp->ss_type;
+	sip->uid = ssp->ss_uid;
+	sip->gid = ssp->ss_grp;
+	sip->mode= ssp->ss_mode;
+	sip->flags = ssp->obj.co_flags;
+	snprintf(sip->sname, sizeof(sip->sname), "%s", ssp->ss_name);
+	return 0;
+}
+
+/*
+ * Dump an entire tree into sysctl call
+ */
+static int
+smb_sysctl_treedump(SYSCTL_HANDLER_ARGS)
+{
+	struct proc *p = req->p;
+	struct smb_cred scred;
+	struct smb_vc *vcp;
+	struct smb_share *ssp;
+	struct smb_vc_info vci;
+	struct smb_share_info ssi;
+	int error, itype;
+
+	smb_makescred(&scred, p, p->p_ucred);
+	error = smb_sm_lockvclist(LK_SHARED, p);
+	if (error)
+		return error;
+	SMBCO_FOREACH((struct smb_connobj*)vcp, &smb_vclist) {
+		error = smb_vc_lock(vcp, LK_SHARED, p);
+		if (error)
+			continue;
+		smb_vc_getinfo(vcp, &vci);
+		error = SYSCTL_OUT(req, &vci, sizeof(struct smb_vc_info));
+		if (error) {
+			smb_vc_unlock(vcp, 0, p);
+			break;
+		}
+		SMBCO_FOREACH((struct smb_connobj*)ssp, VCTOCP(vcp)) {
+			error = smb_share_lock(ssp, LK_SHARED, p);
+			if (error) {
+				error = 0;
+				continue;
+			}
+			smb_share_getinfo(ssp, &ssi);
+			smb_share_unlock(ssp, 0, p);
+			error = SYSCTL_OUT(req, &ssi, sizeof(struct smb_share_info));
+			if (error)
+				break;
+		}
+		smb_vc_unlock(vcp, 0, p);
+		if (error)
+			break;
+	}
+	if (!error) {
+		itype = SMB_INFO_NONE;
+		error = SYSCTL_OUT(req, &itype, sizeof(itype));
+	}
+	smb_sm_unlockvclist(p);
+	return error;
+}