summaryrefslogtreecommitdiffstats
path: root/source4/ntvfs/cifs/vfs_cifs.c
diff options
context:
space:
mode:
Diffstat (limited to 'source4/ntvfs/cifs/vfs_cifs.c')
-rw-r--r--source4/ntvfs/cifs/vfs_cifs.c1263
1 files changed, 1263 insertions, 0 deletions
diff --git a/source4/ntvfs/cifs/vfs_cifs.c b/source4/ntvfs/cifs/vfs_cifs.c
new file mode 100644
index 0000000..121ff57
--- /dev/null
+++ b/source4/ntvfs/cifs/vfs_cifs.c
@@ -0,0 +1,1263 @@
+/*
+ Unix SMB/CIFS implementation.
+
+ CIFS-on-CIFS NTVFS filesystem backend
+
+ Copyright (C) Andrew Tridgell 2003
+ Copyright (C) James J Myers 2003 <myersjj@samba.org>
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 3 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program. If not, see <http://www.gnu.org/licenses/>.
+*/
+/*
+ this implements a CIFS->CIFS NTVFS filesystem backend.
+
+*/
+
+#include "includes.h"
+#include "libcli/raw/libcliraw.h"
+#include "libcli/raw/raw_proto.h"
+#include "libcli/smb_composite/smb_composite.h"
+#include "auth/auth.h"
+#include "auth/credentials/credentials.h"
+#include "ntvfs/ntvfs.h"
+#include "../lib/util/dlinklist.h"
+#include "param/param.h"
+#include "libcli/resolve/resolve.h"
+#include "../libcli/smb/smbXcli_base.h"
+
+struct cvfs_file {
+ struct cvfs_file *prev, *next;
+ uint16_t fnum;
+ struct ntvfs_handle *h;
+};
+
+/* this is stored in ntvfs_private */
+struct cvfs_private {
+ struct smbcli_tree *tree;
+ struct smbcli_transport *transport;
+ struct ntvfs_module_context *ntvfs;
+ struct async_info *pending;
+ struct cvfs_file *files;
+ bool map_generic;
+ bool map_trans2;
+};
+
+
+/* a structure used to pass information to an async handler */
+struct async_info {
+ struct async_info *next, *prev;
+ struct cvfs_private *cvfs;
+ struct ntvfs_request *req;
+ struct smbcli_request *c_req;
+ struct cvfs_file *f;
+ void *parms;
+};
+
+NTSTATUS ntvfs_cifs_init(TALLOC_CTX *);
+
+#define CHECK_UPSTREAM_OPEN do { \
+ if (!smbXcli_conn_is_connected(p->transport->conn)) { \
+ req->async_states->state|=NTVFS_ASYNC_STATE_CLOSE; \
+ return NT_STATUS_CONNECTION_DISCONNECTED; \
+ } \
+} while(0)
+
+#define SETUP_PID do { \
+ p->tree->session->pid = req->smbpid; \
+ CHECK_UPSTREAM_OPEN; \
+} while(0)
+
+#define SETUP_FILE_HERE(f) do { \
+ f = ntvfs_handle_get_backend_data(io->generic.in.file.ntvfs, ntvfs); \
+ if (!f) return NT_STATUS_INVALID_HANDLE; \
+ io->generic.in.file.fnum = f->fnum; \
+} while (0)
+
+#define SETUP_FILE do { \
+ struct cvfs_file *f; \
+ SETUP_FILE_HERE(f); \
+} while (0)
+
+#define SETUP_PID_AND_FILE do { \
+ SETUP_PID; \
+ SETUP_FILE; \
+} while (0)
+
+#define CIFS_SERVER "cifs:server"
+#define CIFS_USER "cifs:user"
+#define CIFS_PASSWORD "cifs:password"
+#define CIFS_DOMAIN "cifs:domain"
+#define CIFS_SHARE "cifs:share"
+#define CIFS_USE_MACHINE_ACCT "cifs:use-machine-account"
+#define CIFS_USE_S4U2PROXY "cifs:use-s4u2proxy"
+#define CIFS_MAP_GENERIC "cifs:map-generic"
+#define CIFS_MAP_TRANS2 "cifs:map-trans2"
+
+#define CIFS_USE_MACHINE_ACCT_DEFAULT false
+#define CIFS_USE_S4U2PROXY_DEFAULT false
+#define CIFS_MAP_GENERIC_DEFAULT false
+#define CIFS_MAP_TRANS2_DEFAULT true
+
+/*
+ a handler for oplock break events from the server - these need to be passed
+ along to the client
+ */
+static bool oplock_handler(struct smbcli_transport *transport, uint16_t tid, uint16_t fnum, uint8_t level, void *p_private)
+{
+ struct cvfs_private *p = p_private;
+ NTSTATUS status;
+ struct ntvfs_handle *h = NULL;
+ struct cvfs_file *f;
+
+ for (f=p->files; f; f=f->next) {
+ if (f->fnum != fnum) continue;
+ h = f->h;
+ break;
+ }
+
+ if (!h) {
+ DEBUG(5,("vfs_cifs: ignoring oplock break level %d for fnum %d\n", level, fnum));
+ return true;
+ }
+
+ DEBUG(5,("vfs_cifs: sending oplock break level %d for fnum %d\n", level, fnum));
+ status = ntvfs_send_oplock_break(p->ntvfs, h, level);
+ if (!NT_STATUS_IS_OK(status)) return false;
+ return true;
+}
+
+/*
+ connect to a share - used when a tree_connect operation comes in.
+*/
+static NTSTATUS cvfs_connect(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req,
+ union smb_tcon *tcon)
+{
+ NTSTATUS status;
+ struct cvfs_private *p;
+ const char *host, *user, *pass, *domain, *remote_share;
+ struct smb_composite_connect io;
+ struct composite_context *creq;
+ struct share_config *scfg = ntvfs->ctx->config;
+
+ struct cli_credentials *credentials;
+ bool machine_account;
+ bool s4u2proxy;
+ const char* sharename;
+ TALLOC_CTX *tmp_ctx;
+
+ tmp_ctx = talloc_new(req);
+ if (tmp_ctx == NULL) {
+ return NT_STATUS_NO_MEMORY;
+ }
+
+ switch (tcon->generic.level) {
+ case RAW_TCON_TCON:
+ sharename = tcon->tcon.in.service;
+ break;
+ case RAW_TCON_TCONX:
+ sharename = tcon->tconx.in.path;
+ break;
+ case RAW_TCON_SMB2:
+ sharename = tcon->smb2.in.path;
+ break;
+ default:
+ status = NT_STATUS_INVALID_LEVEL;
+ goto out;
+ }
+
+ if (strncmp(sharename, "\\\\", 2) == 0) {
+ char *str = strchr(sharename+2, '\\');
+ if (str) {
+ sharename = str + 1;
+ }
+ }
+
+ /* Here we need to determine which server to connect to.
+ * For now we use parametric options, type cifs.
+ */
+ host = share_string_option(tmp_ctx, scfg, CIFS_SERVER, NULL);
+ user = share_string_option(tmp_ctx, scfg, CIFS_USER, NULL);
+ pass = share_string_option(tmp_ctx, scfg, CIFS_PASSWORD, NULL);
+ domain = share_string_option(tmp_ctx, scfg, CIFS_DOMAIN, NULL);
+ remote_share = share_string_option(tmp_ctx, scfg, CIFS_SHARE, NULL);
+ if (!remote_share) {
+ remote_share = sharename;
+ }
+
+ machine_account = share_bool_option(scfg, CIFS_USE_MACHINE_ACCT, CIFS_USE_MACHINE_ACCT_DEFAULT);
+ s4u2proxy = share_bool_option(scfg, CIFS_USE_S4U2PROXY, CIFS_USE_S4U2PROXY_DEFAULT);
+
+ p = talloc_zero(ntvfs, struct cvfs_private);
+ if (!p) {
+ status = NT_STATUS_NO_MEMORY;
+ goto out;
+ }
+
+ ntvfs->private_data = p;
+
+ if (!host) {
+ DEBUG(1,("CIFS backend: You must supply server\n"));
+ status = NT_STATUS_INVALID_PARAMETER;
+ goto out;
+ }
+
+ if (user && pass) {
+ DEBUG(5, ("CIFS backend: Using specified password\n"));
+ credentials = cli_credentials_init(p);
+ if (!credentials) {
+ status = NT_STATUS_NO_MEMORY;
+ goto out;
+ }
+ cli_credentials_set_conf(credentials, ntvfs->ctx->lp_ctx);
+ cli_credentials_set_username(credentials, user, CRED_SPECIFIED);
+ if (domain) {
+ cli_credentials_set_domain(credentials, domain, CRED_SPECIFIED);
+ }
+ cli_credentials_set_password(credentials, pass, CRED_SPECIFIED);
+ } else if (machine_account) {
+ DEBUG(5, ("CIFS backend: Using machine account\n"));
+ credentials = cli_credentials_init_server(p,
+ ntvfs->ctx->lp_ctx);
+ if (credentials == NULL) {
+ status = NT_STATUS_NO_MEMORY;
+ goto out;
+ }
+ } else if (req->session_info->credentials) {
+ DEBUG(5, ("CIFS backend: Using delegated credentials\n"));
+ credentials = req->session_info->credentials;
+ } else if (s4u2proxy) {
+ struct ccache_container *ccc = NULL;
+ const char *err_str = NULL;
+ int ret;
+ char *impersonate_principal;
+ char *self_service;
+ char *target_service;
+
+ impersonate_principal = talloc_asprintf(req, "%s@%s",
+ req->session_info->info->account_name,
+ req->session_info->info->domain_name);
+
+ self_service = talloc_asprintf(req, "cifs/%s",
+ lpcfg_netbios_name(ntvfs->ctx->lp_ctx));
+
+ target_service = talloc_asprintf(req, "cifs/%s", host);
+
+ DEBUG(5, ("CIFS backend: Using S4U2Proxy credentials\n"));
+
+ credentials = cli_credentials_init_server(p,
+ ntvfs->ctx->lp_ctx);
+ if (credentials == NULL) {
+ status = NT_STATUS_NO_MEMORY;
+ goto out;
+ }
+ cli_credentials_invalidate_ccache(credentials, CRED_SPECIFIED);
+ cli_credentials_set_impersonate_principal(credentials,
+ impersonate_principal,
+ self_service);
+ cli_credentials_set_target_service(credentials, target_service);
+ ret = cli_credentials_get_ccache(credentials,
+ ntvfs->ctx->event_ctx,
+ ntvfs->ctx->lp_ctx,
+ &ccc,
+ &err_str);
+ if (ret != 0) {
+ status = NT_STATUS_CROSSREALM_DELEGATION_FAILURE;
+ DEBUG(1,("S4U2Proxy: cli_credentials_get_ccache() gave: ret[%d] str[%s] - %s\n",
+ ret, err_str, nt_errstr(status)));
+ goto out;
+ }
+
+ } else {
+ DEBUG(1,("CIFS backend: NO delegated credentials found: You must supply server, user and password or the client must supply delegated credentials\n"));
+ status = NT_STATUS_INTERNAL_ERROR;
+ goto out;
+ }
+
+ /* connect to the server, using the smbd event context */
+ io.in.dest_host = host;
+ io.in.dest_ports = lpcfg_smb_ports(ntvfs->ctx->lp_ctx);
+ io.in.socket_options = lpcfg_socket_options(ntvfs->ctx->lp_ctx);
+ io.in.called_name = host;
+ io.in.existing_conn = NULL;
+ io.in.credentials = credentials;
+ io.in.fallback_to_anonymous = false;
+ io.in.workgroup = lpcfg_workgroup(ntvfs->ctx->lp_ctx);
+ io.in.service = remote_share;
+ io.in.service_type = "?????";
+ io.in.gensec_settings = lpcfg_gensec_settings(p, ntvfs->ctx->lp_ctx);
+ lpcfg_smbcli_options(ntvfs->ctx->lp_ctx, &io.in.options);
+ lpcfg_smbcli_session_options(ntvfs->ctx->lp_ctx, &io.in.session_options);
+
+ if (!(ntvfs->ctx->client_caps & NTVFS_CLIENT_CAP_LEVEL_II_OPLOCKS)) {
+ io.in.options.use_level2_oplocks = false;
+ }
+
+ creq = smb_composite_connect_send(&io, p,
+ lpcfg_resolve_context(ntvfs->ctx->lp_ctx),
+ ntvfs->ctx->event_ctx);
+ status = smb_composite_connect_recv(creq, p);
+ if (!NT_STATUS_IS_OK(status)) {
+ goto out;
+ }
+
+ p->tree = io.out.tree;
+
+ p->transport = p->tree->session->transport;
+ SETUP_PID;
+ p->ntvfs = ntvfs;
+
+ ntvfs->ctx->fs_type = talloc_strdup(ntvfs->ctx, "NTFS");
+ if (ntvfs->ctx->fs_type == NULL) {
+ status = NT_STATUS_NO_MEMORY;
+ goto out;
+ }
+ ntvfs->ctx->dev_type = talloc_strdup(ntvfs->ctx, "A:");
+ if (ntvfs->ctx->dev_type == NULL) {
+ status = NT_STATUS_NO_MEMORY;
+ goto out;
+ }
+
+ if (tcon->generic.level == RAW_TCON_TCONX) {
+ tcon->tconx.out.fs_type = ntvfs->ctx->fs_type;
+ tcon->tconx.out.dev_type = ntvfs->ctx->dev_type;
+ }
+
+ /* we need to receive oplock break requests from the server */
+ smbcli_oplock_handler(p->transport, oplock_handler, p);
+
+ p->map_generic = share_bool_option(scfg, CIFS_MAP_GENERIC, CIFS_MAP_GENERIC_DEFAULT);
+
+ p->map_trans2 = share_bool_option(scfg, CIFS_MAP_TRANS2, CIFS_MAP_TRANS2_DEFAULT);
+
+ status = NT_STATUS_OK;
+
+out:
+ TALLOC_FREE(tmp_ctx);
+ return status;
+}
+
+/*
+ disconnect from a share
+*/
+static NTSTATUS cvfs_disconnect(struct ntvfs_module_context *ntvfs)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct async_info *a, *an;
+
+ /* first cleanup pending requests */
+ for (a=p->pending; a; a = an) {
+ an = a->next;
+ smbcli_request_destroy(a->c_req);
+ talloc_free(a);
+ }
+
+ talloc_free(p);
+ ntvfs->private_data = NULL;
+
+ return NT_STATUS_OK;
+}
+
+/*
+ destroy an async info structure
+*/
+static int async_info_destructor(struct async_info *async)
+{
+ DLIST_REMOVE(async->cvfs->pending, async);
+ return 0;
+}
+
+/*
+ a handler for simple async replies
+ this handler can only be used for functions that don't return any
+ parameters (those that just return a status code)
+ */
+static void async_simple(struct smbcli_request *c_req)
+{
+ struct async_info *async = c_req->async.private_data;
+ struct ntvfs_request *req = async->req;
+ req->async_states->status = smbcli_request_simple_recv(c_req);
+ talloc_free(async);
+ req->async_states->send_fn(req);
+}
+
+
+/* save some typing for the simple functions */
+#define ASYNC_RECV_TAIL_F(io, async_fn, file) do { \
+ if (!c_req) return NT_STATUS_UNSUCCESSFUL; \
+ { \
+ struct async_info *async; \
+ async = talloc(req, struct async_info); \
+ if (!async) return NT_STATUS_NO_MEMORY; \
+ async->parms = io; \
+ async->req = req; \
+ async->f = file; \
+ async->cvfs = p; \
+ async->c_req = c_req; \
+ DLIST_ADD(p->pending, async); \
+ c_req->async.private_data = async; \
+ talloc_set_destructor(async, async_info_destructor); \
+ } \
+ c_req->async.fn = async_fn; \
+ req->async_states->state |= NTVFS_ASYNC_STATE_ASYNC; \
+ return NT_STATUS_OK; \
+} while (0)
+
+#define ASYNC_RECV_TAIL(io, async_fn) ASYNC_RECV_TAIL_F(io, async_fn, NULL)
+
+#define SIMPLE_ASYNC_TAIL ASYNC_RECV_TAIL(NULL, async_simple)
+
+/*
+ delete a file - the dirtype specifies the file types to include in the search.
+ The name can contain CIFS wildcards, but rarely does (except with OS/2 clients)
+*/
+static NTSTATUS cvfs_unlink(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_unlink *unl)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID;
+
+ /* see if the front end will allow us to perform this
+ function asynchronously. */
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_unlink(p->tree, unl);
+ }
+
+ c_req = smb_raw_unlink_send(p->tree, unl);
+
+ SIMPLE_ASYNC_TAIL;
+}
+
+/*
+ a handler for async ioctl replies
+ */
+static void async_ioctl(struct smbcli_request *c_req)
+{
+ struct async_info *async = c_req->async.private_data;
+ struct ntvfs_request *req = async->req;
+ req->async_states->status = smb_raw_ioctl_recv(c_req, req, async->parms);
+ talloc_free(async);
+ req->async_states->send_fn(req);
+}
+
+/*
+ ioctl interface
+*/
+static NTSTATUS cvfs_ioctl(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_ioctl *io)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID_AND_FILE;
+
+ /* see if the front end will allow us to perform this
+ function asynchronously. */
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_ioctl(p->tree, req, io);
+ }
+
+ c_req = smb_raw_ioctl_send(p->tree, io);
+
+ ASYNC_RECV_TAIL(io, async_ioctl);
+}
+
+/*
+ check if a directory exists
+*/
+static NTSTATUS cvfs_chkpath(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_chkpath *cp)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_chkpath(p->tree, cp);
+ }
+
+ c_req = smb_raw_chkpath_send(p->tree, cp);
+
+ SIMPLE_ASYNC_TAIL;
+}
+
+/*
+ a handler for async qpathinfo replies
+ */
+static void async_qpathinfo(struct smbcli_request *c_req)
+{
+ struct async_info *async = c_req->async.private_data;
+ struct ntvfs_request *req = async->req;
+ req->async_states->status = smb_raw_pathinfo_recv(c_req, req, async->parms);
+ talloc_free(async);
+ req->async_states->send_fn(req);
+}
+
+/*
+ return info on a pathname
+*/
+static NTSTATUS cvfs_qpathinfo(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_fileinfo *info)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_pathinfo(p->tree, req, info);
+ }
+
+ c_req = smb_raw_pathinfo_send(p->tree, info);
+
+ ASYNC_RECV_TAIL(info, async_qpathinfo);
+}
+
+/*
+ a handler for async qfileinfo replies
+ */
+static void async_qfileinfo(struct smbcli_request *c_req)
+{
+ struct async_info *async = c_req->async.private_data;
+ struct ntvfs_request *req = async->req;
+ req->async_states->status = smb_raw_fileinfo_recv(c_req, req, async->parms);
+ talloc_free(async);
+ req->async_states->send_fn(req);
+}
+
+/*
+ query info on a open file
+*/
+static NTSTATUS cvfs_qfileinfo(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_fileinfo *io)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID_AND_FILE;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_fileinfo(p->tree, req, io);
+ }
+
+ c_req = smb_raw_fileinfo_send(p->tree, io);
+
+ ASYNC_RECV_TAIL(io, async_qfileinfo);
+}
+
+
+/*
+ set info on a pathname
+*/
+static NTSTATUS cvfs_setpathinfo(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_setfileinfo *st)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_setpathinfo(p->tree, st);
+ }
+
+ c_req = smb_raw_setpathinfo_send(p->tree, st);
+
+ SIMPLE_ASYNC_TAIL;
+}
+
+
+/*
+ a handler for async open replies
+ */
+static void async_open(struct smbcli_request *c_req)
+{
+ struct async_info *async = c_req->async.private_data;
+ struct cvfs_private *cvfs = async->cvfs;
+ struct ntvfs_request *req = async->req;
+ struct cvfs_file *f = async->f;
+ union smb_open *io = async->parms;
+ union smb_handle *file;
+ if (f == NULL) {
+ goto failed;
+ }
+ talloc_free(async);
+ req->async_states->status = smb_raw_open_recv(c_req, req, io);
+ SMB_OPEN_OUT_FILE(io, file);
+ if (file == NULL) {
+ goto failed;
+ }
+ f->fnum = file->fnum;
+ file->ntvfs = NULL;
+ if (!NT_STATUS_IS_OK(req->async_states->status)) goto failed;
+ req->async_states->status = ntvfs_handle_set_backend_data(f->h, cvfs->ntvfs, f);
+ if (!NT_STATUS_IS_OK(req->async_states->status)) goto failed;
+ file->ntvfs = f->h;
+ DLIST_ADD(cvfs->files, f);
+failed:
+ req->async_states->send_fn(req);
+}
+
+/*
+ open a file
+*/
+static NTSTATUS cvfs_open(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_open *io)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+ struct ntvfs_handle *h;
+ struct cvfs_file *f;
+ NTSTATUS status;
+
+ SETUP_PID;
+
+ if (io->generic.level != RAW_OPEN_GENERIC &&
+ p->map_generic) {
+ return ntvfs_map_open(ntvfs, req, io);
+ }
+
+ status = ntvfs_handle_new(ntvfs, req, &h);
+ NT_STATUS_NOT_OK_RETURN(status);
+
+ f = talloc_zero(h, struct cvfs_file);
+ NT_STATUS_HAVE_NO_MEMORY(f);
+ f->h = h;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ union smb_handle *file;
+
+ status = smb_raw_open(p->tree, req, io);
+ NT_STATUS_NOT_OK_RETURN(status);
+
+ SMB_OPEN_OUT_FILE(io, file);
+ if (file == NULL) {
+ return NT_STATUS_INVALID_PARAMETER;
+ }
+ f->fnum = file->fnum;
+ file->ntvfs = NULL;
+ status = ntvfs_handle_set_backend_data(f->h, p->ntvfs, f);
+ NT_STATUS_NOT_OK_RETURN(status);
+ file->ntvfs = f->h;
+ DLIST_ADD(p->files, f);
+
+ return NT_STATUS_OK;
+ }
+
+ c_req = smb_raw_open_send(p->tree, io);
+
+ ASYNC_RECV_TAIL_F(io, async_open, f);
+}
+
+/*
+ create a directory
+*/
+static NTSTATUS cvfs_mkdir(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_mkdir *md)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_mkdir(p->tree, md);
+ }
+
+ c_req = smb_raw_mkdir_send(p->tree, md);
+
+ SIMPLE_ASYNC_TAIL;
+}
+
+/*
+ remove a directory
+*/
+static NTSTATUS cvfs_rmdir(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, struct smb_rmdir *rd)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_rmdir(p->tree, rd);
+ }
+ c_req = smb_raw_rmdir_send(p->tree, rd);
+
+ SIMPLE_ASYNC_TAIL;
+}
+
+/*
+ rename a set of files
+*/
+static NTSTATUS cvfs_rename(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_rename *ren)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID;
+
+ if (ren->nttrans.level == RAW_RENAME_NTTRANS) {
+ struct cvfs_file *f;
+ f = ntvfs_handle_get_backend_data(ren->nttrans.in.file.ntvfs, ntvfs);
+ if (!f) return NT_STATUS_INVALID_HANDLE;
+ ren->nttrans.in.file.fnum = f->fnum;
+ }
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_rename(p->tree, ren);
+ }
+
+ c_req = smb_raw_rename_send(p->tree, ren);
+
+ SIMPLE_ASYNC_TAIL;
+}
+
+/*
+ copy a set of files
+*/
+static NTSTATUS cvfs_copy(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, struct smb_copy *cp)
+{
+ return NT_STATUS_NOT_SUPPORTED;
+}
+
+/*
+ a handler for async read replies
+ */
+static void async_read(struct smbcli_request *c_req)
+{
+ struct async_info *async = c_req->async.private_data;
+ struct ntvfs_request *req = async->req;
+ req->async_states->status = smb_raw_read_recv(c_req, async->parms);
+ talloc_free(async);
+ req->async_states->send_fn(req);
+}
+
+/*
+ read from a file
+*/
+static NTSTATUS cvfs_read(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_read *io)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID;
+
+ if (io->generic.level != RAW_READ_GENERIC &&
+ p->map_generic) {
+ return ntvfs_map_read(ntvfs, req, io);
+ }
+
+ SETUP_FILE;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_read(p->tree, io);
+ }
+
+ c_req = smb_raw_read_send(p->tree, io);
+
+ ASYNC_RECV_TAIL(io, async_read);
+}
+
+/*
+ a handler for async write replies
+ */
+static void async_write(struct smbcli_request *c_req)
+{
+ struct async_info *async = c_req->async.private_data;
+ struct ntvfs_request *req = async->req;
+ req->async_states->status = smb_raw_write_recv(c_req, async->parms);
+ talloc_free(async);
+ req->async_states->send_fn(req);
+}
+
+/*
+ write to a file
+*/
+static NTSTATUS cvfs_write(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_write *io)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID;
+
+ if (io->generic.level != RAW_WRITE_GENERIC &&
+ p->map_generic) {
+ return ntvfs_map_write(ntvfs, req, io);
+ }
+ SETUP_FILE;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_write(p->tree, io);
+ }
+
+ c_req = smb_raw_write_send(p->tree, io);
+
+ ASYNC_RECV_TAIL(io, async_write);
+}
+
+/*
+ a handler for async seek replies
+ */
+static void async_seek(struct smbcli_request *c_req)
+{
+ struct async_info *async = c_req->async.private_data;
+ struct ntvfs_request *req = async->req;
+ req->async_states->status = smb_raw_seek_recv(c_req, async->parms);
+ talloc_free(async);
+ req->async_states->send_fn(req);
+}
+
+/*
+ seek in a file
+*/
+static NTSTATUS cvfs_seek(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req,
+ union smb_seek *io)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID_AND_FILE;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_seek(p->tree, io);
+ }
+
+ c_req = smb_raw_seek_send(p->tree, io);
+
+ ASYNC_RECV_TAIL(io, async_seek);
+}
+
+/*
+ flush a file
+*/
+static NTSTATUS cvfs_flush(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req,
+ union smb_flush *io)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID;
+ switch (io->generic.level) {
+ case RAW_FLUSH_FLUSH:
+ SETUP_FILE;
+ break;
+ case RAW_FLUSH_ALL:
+ io->generic.in.file.fnum = 0xFFFF;
+ break;
+ case RAW_FLUSH_SMB2:
+ return NT_STATUS_INVALID_LEVEL;
+ }
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_flush(p->tree, io);
+ }
+
+ c_req = smb_raw_flush_send(p->tree, io);
+
+ SIMPLE_ASYNC_TAIL;
+}
+
+/*
+ close a file
+*/
+static NTSTATUS cvfs_close(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_close *io)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+ struct cvfs_file *f;
+ union smb_close io2;
+
+ SETUP_PID;
+
+ if (io->generic.level != RAW_CLOSE_GENERIC &&
+ p->map_generic) {
+ return ntvfs_map_close(ntvfs, req, io);
+ }
+
+ if (io->generic.level == RAW_CLOSE_GENERIC) {
+ ZERO_STRUCT(io2);
+ io2.close.level = RAW_CLOSE_CLOSE;
+ io2.close.in.file = io->generic.in.file;
+ io2.close.in.write_time = io->generic.in.write_time;
+ io = &io2;
+ }
+
+ SETUP_FILE_HERE(f);
+ /* Note, we aren't free-ing f, or it's h here. Should we?
+ even if file-close fails, we'll remove it from the list,
+ what else would we do? Maybe we should not remove until
+ after the proxied call completes? */
+ DLIST_REMOVE(p->files, f);
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_close(p->tree, io);
+ }
+
+ c_req = smb_raw_close_send(p->tree, io);
+
+ SIMPLE_ASYNC_TAIL;
+}
+
+/*
+ exit - closing files open by the pid
+*/
+static NTSTATUS cvfs_exit(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_exit(p->tree->session);
+ }
+
+ c_req = smb_raw_exit_send(p->tree->session);
+
+ SIMPLE_ASYNC_TAIL;
+}
+
+/*
+ logoff - closing files open by the user
+*/
+static NTSTATUS cvfs_logoff(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req)
+{
+ /* we can't do this right in the cifs backend .... */
+ return NT_STATUS_OK;
+}
+
+/*
+ setup for an async call - nothing to do yet
+*/
+static NTSTATUS cvfs_async_setup(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req,
+ void *private_data)
+{
+ return NT_STATUS_OK;
+}
+
+/*
+ cancel an async call
+*/
+static NTSTATUS cvfs_cancel(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct async_info *a;
+
+ /* find the matching request */
+ for (a=p->pending;a;a=a->next) {
+ if (a->req == req) {
+ break;
+ }
+ }
+
+ if (a == NULL) {
+ return NT_STATUS_INVALID_PARAMETER;
+ }
+
+ return smb_raw_ntcancel(a->c_req);
+}
+
+/*
+ lock a byte range
+*/
+static NTSTATUS cvfs_lock(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_lock *io)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID;
+
+ if (io->generic.level != RAW_LOCK_GENERIC &&
+ p->map_generic) {
+ return ntvfs_map_lock(ntvfs, req, io);
+ }
+ SETUP_FILE;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_lock(p->tree, io);
+ }
+
+ c_req = smb_raw_lock_send(p->tree, io);
+ SIMPLE_ASYNC_TAIL;
+}
+
+/*
+ set info on a open file
+*/
+static NTSTATUS cvfs_setfileinfo(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req,
+ union smb_setfileinfo *io)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID_AND_FILE;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_setfileinfo(p->tree, io);
+ }
+ c_req = smb_raw_setfileinfo_send(p->tree, io);
+
+ SIMPLE_ASYNC_TAIL;
+}
+
+
+/*
+ a handler for async fsinfo replies
+ */
+static void async_fsinfo(struct smbcli_request *c_req)
+{
+ struct async_info *async = c_req->async.private_data;
+ struct ntvfs_request *req = async->req;
+ req->async_states->status = smb_raw_fsinfo_recv(c_req, req, async->parms);
+ talloc_free(async);
+ req->async_states->send_fn(req);
+}
+
+/*
+ return filesystem space info
+*/
+static NTSTATUS cvfs_fsinfo(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_fsinfo *fs)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ SETUP_PID;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_fsinfo(p->tree, req, fs);
+ }
+
+ c_req = smb_raw_fsinfo_send(p->tree, req, fs);
+
+ ASYNC_RECV_TAIL(fs, async_fsinfo);
+}
+
+/*
+ return print queue info
+*/
+static NTSTATUS cvfs_lpq(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_lpq *lpq)
+{
+ return NT_STATUS_NOT_SUPPORTED;
+}
+
+/*
+ list files in a directory matching a wildcard pattern
+*/
+static NTSTATUS cvfs_search_first(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_search_first *io,
+ void *search_private,
+ bool (*callback)(void *, const union smb_search_data *))
+{
+ struct cvfs_private *p = ntvfs->private_data;
+
+ SETUP_PID;
+
+ return smb_raw_search_first(p->tree, req, io, search_private, callback);
+}
+
+/* continue a search */
+static NTSTATUS cvfs_search_next(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_search_next *io,
+ void *search_private,
+ bool (*callback)(void *, const union smb_search_data *))
+{
+ struct cvfs_private *p = ntvfs->private_data;
+
+ SETUP_PID;
+
+ return smb_raw_search_next(p->tree, req, io, search_private, callback);
+}
+
+/* close a search */
+static NTSTATUS cvfs_search_close(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req, union smb_search_close *io)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+
+ SETUP_PID;
+
+ return smb_raw_search_close(p->tree, io);
+}
+
+/*
+ a handler for async trans2 replies
+ */
+static void async_trans2(struct smbcli_request *c_req)
+{
+ struct async_info *async = c_req->async.private_data;
+ struct ntvfs_request *req = async->req;
+ req->async_states->status = smb_raw_trans2_recv(c_req, req, async->parms);
+ talloc_free(async);
+ req->async_states->send_fn(req);
+}
+
+/* raw trans2 */
+static NTSTATUS cvfs_trans2(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req,
+ struct smb_trans2 *trans2)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+
+ if (p->map_trans2) {
+ return NT_STATUS_NOT_IMPLEMENTED;
+ }
+
+ SETUP_PID;
+
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return smb_raw_trans2(p->tree, req, trans2);
+ }
+
+ c_req = smb_raw_trans2_send(p->tree, trans2);
+
+ ASYNC_RECV_TAIL(trans2, async_trans2);
+}
+
+
+/* SMBtrans - not used on file shares */
+static NTSTATUS cvfs_trans(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req,
+ struct smb_trans2 *trans2)
+{
+ return NT_STATUS_ACCESS_DENIED;
+}
+
+/*
+ a handler for async change notify replies
+ */
+static void async_changenotify(struct smbcli_request *c_req)
+{
+ struct async_info *async = c_req->async.private_data;
+ struct ntvfs_request *req = async->req;
+ req->async_states->status = smb_raw_changenotify_recv(c_req, req, async->parms);
+ talloc_free(async);
+ req->async_states->send_fn(req);
+}
+
+/* change notify request - always async */
+static NTSTATUS cvfs_notify(struct ntvfs_module_context *ntvfs,
+ struct ntvfs_request *req,
+ union smb_notify *io)
+{
+ struct cvfs_private *p = ntvfs->private_data;
+ struct smbcli_request *c_req;
+ int saved_timeout = p->transport->options.request_timeout;
+ struct cvfs_file *f;
+
+ if (io->nttrans.level != RAW_NOTIFY_NTTRANS) {
+ return NT_STATUS_NOT_IMPLEMENTED;
+ }
+
+ SETUP_PID;
+
+ f = ntvfs_handle_get_backend_data(io->nttrans.in.file.ntvfs, ntvfs);
+ if (!f) return NT_STATUS_INVALID_HANDLE;
+ io->nttrans.in.file.fnum = f->fnum;
+
+ /* this request doesn't make sense unless its async */
+ if (!(req->async_states->state & NTVFS_ASYNC_STATE_MAY_ASYNC)) {
+ return NT_STATUS_INVALID_PARAMETER;
+ }
+
+ /* we must not timeout on notify requests - they wait
+ forever */
+ p->transport->options.request_timeout = 0;
+
+ c_req = smb_raw_changenotify_send(p->tree, io);
+
+ p->transport->options.request_timeout = saved_timeout;
+
+ ASYNC_RECV_TAIL(io, async_changenotify);
+}
+
+/*
+ initialise the CIFS->CIFS backend, registering ourselves with the ntvfs subsystem
+ */
+NTSTATUS ntvfs_cifs_init(TALLOC_CTX *ctx)
+{
+ NTSTATUS ret;
+ struct ntvfs_ops ops;
+ NTVFS_CURRENT_CRITICAL_SIZES(vers);
+
+ ZERO_STRUCT(ops);
+
+ /* fill in the name and type */
+ ops.name = "cifs";
+ ops.type = NTVFS_DISK;
+
+ /* fill in all the operations */
+ ops.connect_fn = cvfs_connect;
+ ops.disconnect_fn = cvfs_disconnect;
+ ops.unlink_fn = cvfs_unlink;
+ ops.chkpath_fn = cvfs_chkpath;
+ ops.qpathinfo_fn = cvfs_qpathinfo;
+ ops.setpathinfo_fn = cvfs_setpathinfo;
+ ops.open_fn = cvfs_open;
+ ops.mkdir_fn = cvfs_mkdir;
+ ops.rmdir_fn = cvfs_rmdir;
+ ops.rename_fn = cvfs_rename;
+ ops.copy_fn = cvfs_copy;
+ ops.ioctl_fn = cvfs_ioctl;
+ ops.read_fn = cvfs_read;
+ ops.write_fn = cvfs_write;
+ ops.seek_fn = cvfs_seek;
+ ops.flush_fn = cvfs_flush;
+ ops.close_fn = cvfs_close;
+ ops.exit_fn = cvfs_exit;
+ ops.lock_fn = cvfs_lock;
+ ops.setfileinfo_fn = cvfs_setfileinfo;
+ ops.qfileinfo_fn = cvfs_qfileinfo;
+ ops.fsinfo_fn = cvfs_fsinfo;
+ ops.lpq_fn = cvfs_lpq;
+ ops.search_first_fn = cvfs_search_first;
+ ops.search_next_fn = cvfs_search_next;
+ ops.search_close_fn = cvfs_search_close;
+ ops.trans_fn = cvfs_trans;
+ ops.logoff_fn = cvfs_logoff;
+ ops.async_setup_fn = cvfs_async_setup;
+ ops.cancel_fn = cvfs_cancel;
+ ops.notify_fn = cvfs_notify;
+ ops.trans2_fn = cvfs_trans2;
+
+ /* register ourselves with the NTVFS subsystem. We register
+ under the name 'cifs'. */
+ ret = ntvfs_register(&ops, &vers);
+
+ if (!NT_STATUS_IS_OK(ret)) {
+ DEBUG(0,("Failed to register CIFS backend!\n"));
+ }
+
+ return ret;
+}