qemu/ui/vnc.c

2741 lines
79 KiB
C
Raw Normal View History

/*
* QEMU VNC display driver
*
* Copyright (C) 2006 Anthony Liguori <anthony@codemonkey.ws>
* Copyright (C) 2006 Fabrice Bellard
* Copyright (C) 2009 Red Hat, Inc
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to deal
* in the Software without restriction, including without limitation the rights
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
* copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
* THE SOFTWARE.
*/
#include "vnc.h"
#include "vnc-jobs.h"
#include "sysemu.h"
#include "qemu_socket.h"
#include "qemu-timer.h"
Support ACLs for controlling VNC access ("Daniel P. Berrange") This patch introduces a generic internal API for access control lists to be used by network servers in QEMU. It adds support for checking these ACL in the VNC server, in two places. The first ACL is for the SASL authentication mechanism, checking the SASL username. This ACL is called 'vnc.username'. The second is for the TLS authentication mechanism, when x509 client certificates are turned on, checking against the Distinguished Name of the client. This ACL is called 'vnc.x509dname' The internal API provides for an ACL with the following characteristics - A unique name, eg vnc.username, and vnc.x509dname. - A default policy, allow or deny - An ordered series of match rules, with allow or deny policy If none of the match rules apply, then the default policy is used. There is a monitor API to manipulate the ACLs, which I'll describe via examples (qemu) acl show vnc.username policy: allow (qemu) acl policy vnc.username denya acl: policy set to 'deny' (qemu) acl allow vnc.username fred acl: added rule at position 1 (qemu) acl allow vnc.username bob acl: added rule at position 2 (qemu) acl allow vnc.username joe 1 acl: added rule at position 1 (qemu) acl show vnc.username policy: deny 0: allow fred 1: allow joe 2: allow bob (qemu) acl show vnc.x509dname policy: allow (qemu) acl policy vnc.x509dname deny acl: policy set to 'deny' (qemu) acl allow vnc.x509dname C=GB,O=ACME,L=London,CN=* acl: added rule at position 1 (qemu) acl allow vnc.x509dname C=GB,O=ACME,L=Boston,CN=bob acl: added rule at position 2 (qemu) acl show vnc.x509dname policy: deny 0: allow C=GB,O=ACME,L=London,CN=* 1: allow C=GB,O=ACME,L=Boston,CN=bob By default the VNC server will not use any ACLs, allowing access to the server if the user successfully authenticates. To enable use of ACLs to restrict user access, the ',acl' flag should be given when starting QEMU. The initial ACL activated will be a 'deny all' policy and should be customized using monitor commands. eg enable SASL auth and ACLs qemu .... -vnc localhost:1,sasl,acl The next patch will provide a way to load a pre-defined ACL when starting up Makefile | 6 + b/acl.c | 185 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ b/acl.h | 74 ++++++++++++++++++++++ configure | 18 +++++ monitor.c | 95 ++++++++++++++++++++++++++++ qemu-doc.texi | 49 ++++++++++++++ vnc-auth-sasl.c | 16 +++- vnc-auth-sasl.h | 7 ++ vnc-tls.c | 19 +++++ vnc-tls.h | 3 vnc.c | 21 ++++++ vnc.h | 3 12 files changed, 491 insertions(+), 5 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6726 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:37 +03:00
#include "acl.h"
#include "qemu-objects.h"
#define VNC_REFRESH_INTERVAL_BASE 30
#define VNC_REFRESH_INTERVAL_INC 50
#define VNC_REFRESH_INTERVAL_MAX 2000
#include "vnc_keysym.h"
#include "d3des.h"
#define count_bits(c, v) { \
for (c = 0; v; v >>= 1) \
{ \
c += v & 1; \
} \
}
static VncDisplay *vnc_display; /* needed for info vnc */
DisplayState interface change (Stefano Stabellini) This patch changes the DisplayState interface adding support for multiple frontends at the same time (sdl and vnc) and implements most of the benefit of the shared_buf patch without the added complexity. Currently DisplayState is managed by sdl (or vnc) and sdl (or vnc) is also responsible for allocating the data and setting the depth. Vga.c (or another backend) will do any necessary conversion. The idea is to change it so that is vga.c (or another backend) together with console.c that fully manage the DisplayState interface allocating data and setting the depth (either 16 or 32 bit, if the guest uses a different resolution or is in text mode, vga.c (or another backend) is in charge of doing the conversion seamlessly). The other idea is that DisplayState supports *multiple* frontends like sdl and vnc; each of them can register some callbacks to be called when a display event occurs. The interesting changes are: - the new structures and related functions in console.h and console.c in particular the following functions are very helpful to manage a DisplaySurface: qemu_create_displaysurface qemu_resize_displaysurface qemu_create_displaysurface_from qemu_free_displaysurface - console_select and qemu_console_resize in console.c this two functions manage multiple consoles on a single host display - moving code around in hw/vga.c as for the shared_buf patch this is necessary to be able to handle a dynamic DisplaySurface bpp - changes to vga_draw_graphic in hw/vga.c this is the place where the DisplaySurface buffer is shared with the videoram, when possible; Compared to the last version the only changes are: - do not remove support to dpy_copy in cirrus_vga - change the name of the displaysurface handling functions Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6336 c046a42c-6fe2-441c-8c8c-71466251a162
2009-01-16 01:14:11 +03:00
static DisplayChangeListener *dcl;
static int vnc_cursor_define(VncState *vs);
static char *addr_to_string(const char *format,
struct sockaddr_storage *sa,
socklen_t salen) {
char *addr;
char host[NI_MAXHOST];
char serv[NI_MAXSERV];
int err;
size_t addrlen;
if ((err = getnameinfo((struct sockaddr *)sa, salen,
host, sizeof(host),
serv, sizeof(serv),
NI_NUMERICHOST | NI_NUMERICSERV)) != 0) {
VNC_DEBUG("Cannot resolve address %d: %s\n",
err, gai_strerror(err));
return NULL;
}
/* Enough for the existing format + the 2 vars we're
* substituting in. */
addrlen = strlen(format) + strlen(host) + strlen(serv);
addr = qemu_malloc(addrlen + 1);
snprintf(addr, addrlen, format, host, serv);
addr[addrlen] = '\0';
return addr;
}
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
char *vnc_socket_local_addr(const char *format, int fd) {
struct sockaddr_storage sa;
socklen_t salen;
salen = sizeof(sa);
if (getsockname(fd, (struct sockaddr*)&sa, &salen) < 0)
return NULL;
return addr_to_string(format, &sa, salen);
}
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
char *vnc_socket_remote_addr(const char *format, int fd) {
struct sockaddr_storage sa;
socklen_t salen;
salen = sizeof(sa);
if (getpeername(fd, (struct sockaddr*)&sa, &salen) < 0)
return NULL;
return addr_to_string(format, &sa, salen);
}
static int put_addr_qdict(QDict *qdict, struct sockaddr_storage *sa,
socklen_t salen)
{
char host[NI_MAXHOST];
char serv[NI_MAXSERV];
int err;
if ((err = getnameinfo((struct sockaddr *)sa, salen,
host, sizeof(host),
serv, sizeof(serv),
NI_NUMERICHOST | NI_NUMERICSERV)) != 0) {
VNC_DEBUG("Cannot resolve address %d: %s\n",
err, gai_strerror(err));
return -1;
}
qdict_put(qdict, "host", qstring_from_str(host));
qdict_put(qdict, "service", qstring_from_str(serv));
qdict_put(qdict, "family",qstring_from_str(inet_strfamily(sa->ss_family)));
return 0;
}
static int vnc_server_addr_put(QDict *qdict, int fd)
{
struct sockaddr_storage sa;
socklen_t salen;
salen = sizeof(sa);
if (getsockname(fd, (struct sockaddr*)&sa, &salen) < 0) {
return -1;
}
return put_addr_qdict(qdict, &sa, salen);
}
static int vnc_qdict_remote_addr(QDict *qdict, int fd)
{
struct sockaddr_storage sa;
socklen_t salen;
salen = sizeof(sa);
if (getpeername(fd, (struct sockaddr*)&sa, &salen) < 0) {
return -1;
}
return put_addr_qdict(qdict, &sa, salen);
}
static const char *vnc_auth_name(VncDisplay *vd) {
switch (vd->auth) {
case VNC_AUTH_INVALID:
return "invalid";
case VNC_AUTH_NONE:
return "none";
case VNC_AUTH_VNC:
return "vnc";
case VNC_AUTH_RA2:
return "ra2";
case VNC_AUTH_RA2NE:
return "ra2ne";
case VNC_AUTH_TIGHT:
return "tight";
case VNC_AUTH_ULTRA:
return "ultra";
case VNC_AUTH_TLS:
return "tls";
case VNC_AUTH_VENCRYPT:
#ifdef CONFIG_VNC_TLS
switch (vd->subauth) {
case VNC_AUTH_VENCRYPT_PLAIN:
return "vencrypt+plain";
case VNC_AUTH_VENCRYPT_TLSNONE:
return "vencrypt+tls+none";
case VNC_AUTH_VENCRYPT_TLSVNC:
return "vencrypt+tls+vnc";
case VNC_AUTH_VENCRYPT_TLSPLAIN:
return "vencrypt+tls+plain";
case VNC_AUTH_VENCRYPT_X509NONE:
return "vencrypt+x509+none";
case VNC_AUTH_VENCRYPT_X509VNC:
return "vencrypt+x509+vnc";
case VNC_AUTH_VENCRYPT_X509PLAIN:
return "vencrypt+x509+plain";
case VNC_AUTH_VENCRYPT_TLSSASL:
return "vencrypt+tls+sasl";
case VNC_AUTH_VENCRYPT_X509SASL:
return "vencrypt+x509+sasl";
default:
return "vencrypt";
}
#else
return "vencrypt";
#endif
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
case VNC_AUTH_SASL:
return "sasl";
}
return "unknown";
}
static int vnc_server_info_put(QDict *qdict)
{
if (vnc_server_addr_put(qdict, vnc_display->lsock) < 0) {
return -1;
}
qdict_put(qdict, "auth", qstring_from_str(vnc_auth_name(vnc_display)));
return 0;
}
static void vnc_client_cache_auth(VncState *client)
{
#if defined(CONFIG_VNC_TLS) || defined(CONFIG_VNC_SASL)
QDict *qdict;
#endif
if (!client->info) {
return;
}
#if defined(CONFIG_VNC_TLS) || defined(CONFIG_VNC_SASL)
qdict = qobject_to_qdict(client->info);
#endif
#ifdef CONFIG_VNC_TLS
if (client->tls.session &&
client->tls.dname) {
qdict_put(qdict, "x509_dname", qstring_from_str(client->tls.dname));
}
#endif
#ifdef CONFIG_VNC_SASL
if (client->sasl.conn &&
client->sasl.username) {
qdict_put(qdict, "sasl_username",
qstring_from_str(client->sasl.username));
}
#endif
}
static void vnc_client_cache_addr(VncState *client)
{
QDict *qdict;
qdict = qdict_new();
if (vnc_qdict_remote_addr(qdict, client->csock) < 0) {
QDECREF(qdict);
/* XXX: how to report the error? */
return;
}
client->info = QOBJECT(qdict);
}
static void vnc_qmp_event(VncState *vs, MonitorEvent event)
{
QDict *server;
QObject *data;
if (!vs->info) {
return;
}
server = qdict_new();
if (vnc_server_info_put(server) < 0) {
QDECREF(server);
return;
}
data = qobject_from_jsonf("{ 'client': %p, 'server': %p }",
vs->info, QOBJECT(server));
monitor_protocol_event(event, data);
qobject_incref(vs->info);
qobject_decref(data);
}
static void info_vnc_iter(QObject *obj, void *opaque)
{
QDict *client;
Monitor *mon = opaque;
client = qobject_to_qdict(obj);
monitor_printf(mon, "Client:\n");
monitor_printf(mon, " address: %s:%s\n",
qdict_get_str(client, "host"),
qdict_get_str(client, "service"));
#ifdef CONFIG_VNC_TLS
monitor_printf(mon, " x509_dname: %s\n",
qdict_haskey(client, "x509_dname") ?
qdict_get_str(client, "x509_dname") : "none");
#endif
#ifdef CONFIG_VNC_SASL
monitor_printf(mon, " username: %s\n",
qdict_haskey(client, "sasl_username") ?
qdict_get_str(client, "sasl_username") : "none");
#endif
}
void do_info_vnc_print(Monitor *mon, const QObject *data)
{
QDict *server;
QList *clients;
server = qobject_to_qdict(data);
if (qdict_get_bool(server, "enabled") == 0) {
monitor_printf(mon, "Server: disabled\n");
return;
}
monitor_printf(mon, "Server:\n");
monitor_printf(mon, " address: %s:%s\n",
qdict_get_str(server, "host"),
qdict_get_str(server, "service"));
monitor_printf(mon, " auth: %s\n", qdict_get_str(server, "auth"));
clients = qdict_get_qlist(server, "clients");
if (qlist_empty(clients)) {
monitor_printf(mon, "Client: none\n");
} else {
qlist_iter(clients, info_vnc_iter, mon);
}
}
void do_info_vnc(Monitor *mon, QObject **ret_data)
{
if (vnc_display == NULL || vnc_display->display == NULL) {
*ret_data = qobject_from_jsonf("{ 'enabled': false }");
} else {
QList *clist;
VncState *client;
clist = qlist_new();
QTAILQ_FOREACH(client, &vnc_display->clients, next) {
if (client->info) {
/* incref so that it's not freed by upper layers */
qobject_incref(client->info);
qlist_append_obj(clist, client->info);
}
}
*ret_data = qobject_from_jsonf("{ 'enabled': true, 'clients': %p }",
QOBJECT(clist));
assert(*ret_data != NULL);
if (vnc_server_info_put(qobject_to_qdict(*ret_data)) < 0) {
qobject_decref(*ret_data);
*ret_data = NULL;
}
}
}
/* TODO
1) Get the queue working for IO.
2) there is some weirdness when using the -S option (the screen is grey
and not totally invalidated
3) resolutions > 1024
*/
static int vnc_update_client(VncState *vs, int has_dirty);
static int vnc_update_client_sync(VncState *vs, int has_dirty);
static void vnc_disconnect_start(VncState *vs);
static void vnc_disconnect_finish(VncState *vs);
static void vnc_init_timer(VncDisplay *vd);
static void vnc_remove_timer(VncDisplay *vd);
static void vnc_colordepth(VncState *vs);
static void framebuffer_update_request(VncState *vs, int incremental,
int x_position, int y_position,
int w, int h);
static void vnc_refresh(void *opaque);
static int vnc_refresh_server_surface(VncDisplay *vd);
static inline void vnc_set_bit(uint32_t *d, int k)
{
d[k >> 5] |= 1 << (k & 0x1f);
}
static inline void vnc_clear_bit(uint32_t *d, int k)
{
d[k >> 5] &= ~(1 << (k & 0x1f));
}
static inline void vnc_set_bits(uint32_t *d, int n, int nb_words)
{
int j;
j = 0;
while (n >= 32) {
d[j++] = -1;
n -= 32;
}
if (n > 0)
d[j++] = (1 << n) - 1;
while (j < nb_words)
d[j++] = 0;
}
static inline int vnc_get_bit(const uint32_t *d, int k)
{
return (d[k >> 5] >> (k & 0x1f)) & 1;
}
static inline int vnc_and_bits(const uint32_t *d1, const uint32_t *d2,
int nb_words)
{
int i;
for(i = 0; i < nb_words; i++) {
if ((d1[i] & d2[i]) != 0)
return 1;
}
return 0;
}
static void vnc_dpy_update(DisplayState *ds, int x, int y, int w, int h)
{
int i;
VncDisplay *vd = ds->opaque;
struct VncSurface *s = &vd->guest;
h += y;
/* round x down to ensure the loop only spans one 16-pixel block per,
iteration. otherwise, if (x % 16) != 0, the last iteration may span
two 16-pixel blocks but we only mark the first as dirty
*/
w += (x % 16);
x -= (x % 16);
vnc: cleanup surface handling, fix screen corruption bug. (Gerd Hoffmann) This patch killes the old_data hack in the qemu server and replaces it with a clean separation of the guest-visible display surface and the vnc server display surface. Both guest and server surface have their own dirty bitmap for tracking screen updates. Workflow is this: (1) The guest writes to the guest surface. With shared buffers being active the guest writes are directly visible to the vnc server code. Note that this may happen in parallel to the vnc server code running (today only in xenfb, once we have vcpu threads in qemu also for other display adapters). (2) vnc_update() callback tags the specified area in the guest dirty map. (3) vnc_update_client() will first walk through the guest dirty map. It will compare guest and server surface for all regions tagged dirty and in case the screen content really did change the server surface and dirty map are updated. Note: old code used old_data in a simliar way, so this does *not* introduce an extra memcpy. (4) Then vnc_update_cient() will send the updates to the vnc client using the server surface and dirty map. Note: old code used the guest-visible surface instead, causing screen corruption in case of guest screen updates running in parallel. The separate dirty bitmap also has the nice effect that forced screen updates can be done cleanly by simply tagging the area in both guest and server dirty map. The old, hackish way was memset(old_data, 42, size) to trick the code checking for screen changes. Signed-off-by: Gerd Hoffmann <kraxel@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6860 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-20 18:59:14 +03:00
x = MIN(x, s->ds->width);
y = MIN(y, s->ds->height);
w = MIN(x + w, s->ds->width) - x;
h = MIN(h, s->ds->height);
for (; y < h; y++)
for (i = 0; i < w; i += 16)
vnc: cleanup surface handling, fix screen corruption bug. (Gerd Hoffmann) This patch killes the old_data hack in the qemu server and replaces it with a clean separation of the guest-visible display surface and the vnc server display surface. Both guest and server surface have their own dirty bitmap for tracking screen updates. Workflow is this: (1) The guest writes to the guest surface. With shared buffers being active the guest writes are directly visible to the vnc server code. Note that this may happen in parallel to the vnc server code running (today only in xenfb, once we have vcpu threads in qemu also for other display adapters). (2) vnc_update() callback tags the specified area in the guest dirty map. (3) vnc_update_client() will first walk through the guest dirty map. It will compare guest and server surface for all regions tagged dirty and in case the screen content really did change the server surface and dirty map are updated. Note: old code used old_data in a simliar way, so this does *not* introduce an extra memcpy. (4) Then vnc_update_cient() will send the updates to the vnc client using the server surface and dirty map. Note: old code used the guest-visible surface instead, causing screen corruption in case of guest screen updates running in parallel. The separate dirty bitmap also has the nice effect that forced screen updates can be done cleanly by simply tagging the area in both guest and server dirty map. The old, hackish way was memset(old_data, 42, size) to trick the code checking for screen changes. Signed-off-by: Gerd Hoffmann <kraxel@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6860 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-20 18:59:14 +03:00
vnc_set_bit(s->dirty[y], (x + i) / 16);
}
void vnc_framebuffer_update(VncState *vs, int x, int y, int w, int h,
int32_t encoding)
{
vnc_write_u16(vs, x);
vnc_write_u16(vs, y);
vnc_write_u16(vs, w);
vnc_write_u16(vs, h);
vnc_write_s32(vs, encoding);
}
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
void buffer_reserve(Buffer *buffer, size_t len)
{
if ((buffer->capacity - buffer->offset) < len) {
buffer->capacity += (len + 1024);
buffer->buffer = qemu_realloc(buffer->buffer, buffer->capacity);
if (buffer->buffer == NULL) {
fprintf(stderr, "vnc: out of memory\n");
exit(1);
}
}
}
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
int buffer_empty(Buffer *buffer)
{
return buffer->offset == 0;
}
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
uint8_t *buffer_end(Buffer *buffer)
{
return buffer->buffer + buffer->offset;
}
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
void buffer_reset(Buffer *buffer)
{
buffer->offset = 0;
}
void buffer_free(Buffer *buffer)
{
qemu_free(buffer->buffer);
buffer->offset = 0;
buffer->capacity = 0;
buffer->buffer = NULL;
}
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
void buffer_append(Buffer *buffer, const void *data, size_t len)
{
memcpy(buffer->buffer + buffer->offset, data, len);
buffer->offset += len;
}
static void vnc_desktop_resize(VncState *vs)
{
DisplayState *ds = vs->ds;
if (vs->csock == -1 || !vnc_has_feature(vs, VNC_FEATURE_RESIZE)) {
return;
}
if (vs->client_width == ds_get_width(ds) &&
vs->client_height == ds_get_height(ds)) {
return;
}
vs->client_width = ds_get_width(ds);
vs->client_height = ds_get_height(ds);
vnc_lock_output(vs);
vnc_write_u8(vs, VNC_MSG_SERVER_FRAMEBUFFER_UPDATE);
vnc_write_u8(vs, 0);
vnc_write_u16(vs, 1); /* number of rects */
vnc_framebuffer_update(vs, 0, 0, vs->client_width, vs->client_height,
VNC_ENCODING_DESKTOPRESIZE);
vnc_unlock_output(vs);
vnc_flush(vs);
}
#ifdef CONFIG_VNC_THREAD
static void vnc_abort_display_jobs(VncDisplay *vd)
{
VncState *vs;
QTAILQ_FOREACH(vs, &vd->clients, next) {
vnc_lock_output(vs);
vs->abort = true;
vnc_unlock_output(vs);
}
QTAILQ_FOREACH(vs, &vd->clients, next) {
vnc_jobs_join(vs);
}
QTAILQ_FOREACH(vs, &vd->clients, next) {
vnc_lock_output(vs);
vs->abort = false;
vnc_unlock_output(vs);
}
}
#else
static void vnc_abort_display_jobs(VncDisplay *vd)
{
}
#endif
static void vnc_dpy_resize(DisplayState *ds)
{
VncDisplay *vd = ds->opaque;
VncState *vs;
vnc_abort_display_jobs(vd);
/* server surface */
if (!vd->server)
vd->server = qemu_mallocz(sizeof(*vd->server));
if (vd->server->data)
qemu_free(vd->server->data);
*(vd->server) = *(ds->surface);
vd->server->data = qemu_mallocz(vd->server->linesize *
vd->server->height);
vnc: cleanup surface handling, fix screen corruption bug. (Gerd Hoffmann) This patch killes the old_data hack in the qemu server and replaces it with a clean separation of the guest-visible display surface and the vnc server display surface. Both guest and server surface have their own dirty bitmap for tracking screen updates. Workflow is this: (1) The guest writes to the guest surface. With shared buffers being active the guest writes are directly visible to the vnc server code. Note that this may happen in parallel to the vnc server code running (today only in xenfb, once we have vcpu threads in qemu also for other display adapters). (2) vnc_update() callback tags the specified area in the guest dirty map. (3) vnc_update_client() will first walk through the guest dirty map. It will compare guest and server surface for all regions tagged dirty and in case the screen content really did change the server surface and dirty map are updated. Note: old code used old_data in a simliar way, so this does *not* introduce an extra memcpy. (4) Then vnc_update_cient() will send the updates to the vnc client using the server surface and dirty map. Note: old code used the guest-visible surface instead, causing screen corruption in case of guest screen updates running in parallel. The separate dirty bitmap also has the nice effect that forced screen updates can be done cleanly by simply tagging the area in both guest and server dirty map. The old, hackish way was memset(old_data, 42, size) to trick the code checking for screen changes. Signed-off-by: Gerd Hoffmann <kraxel@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6860 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-20 18:59:14 +03:00
/* guest surface */
if (!vd->guest.ds)
vd->guest.ds = qemu_mallocz(sizeof(*vd->guest.ds));
if (ds_get_bytes_per_pixel(ds) != vd->guest.ds->pf.bytes_per_pixel)
console_color_init(ds);
*(vd->guest.ds) = *(ds->surface);
memset(vd->guest.dirty, 0xFF, sizeof(vd->guest.dirty));
QTAILQ_FOREACH(vs, &vd->clients, next) {
vnc_colordepth(vs);
vnc_desktop_resize(vs);
if (vs->vd->cursor) {
vnc_cursor_define(vs);
}
memset(vs->dirty, 0xFF, sizeof(vs->dirty));
}
}
/* fastest code */
static void vnc_write_pixels_copy(VncState *vs, struct PixelFormat *pf,
void *pixels, int size)
{
vnc_write(vs, pixels, size);
}
/* slowest but generic code. */
void vnc_convert_pixel(VncState *vs, uint8_t *buf, uint32_t v)
{
uint8_t r, g, b;
VncDisplay *vd = vs->vd;
r = ((((v & vd->server->pf.rmask) >> vd->server->pf.rshift) << vs->clientds.pf.rbits) >>
vd->server->pf.rbits);
g = ((((v & vd->server->pf.gmask) >> vd->server->pf.gshift) << vs->clientds.pf.gbits) >>
vd->server->pf.gbits);
b = ((((v & vd->server->pf.bmask) >> vd->server->pf.bshift) << vs->clientds.pf.bbits) >>
vd->server->pf.bbits);
v = (r << vs->clientds.pf.rshift) |
(g << vs->clientds.pf.gshift) |
(b << vs->clientds.pf.bshift);
switch(vs->clientds.pf.bytes_per_pixel) {
case 1:
buf[0] = v;
break;
case 2:
if (vs->clientds.flags & QEMU_BIG_ENDIAN_FLAG) {
buf[0] = v >> 8;
buf[1] = v;
} else {
buf[1] = v >> 8;
buf[0] = v;
}
break;
default:
case 4:
if (vs->clientds.flags & QEMU_BIG_ENDIAN_FLAG) {
buf[0] = v >> 24;
buf[1] = v >> 16;
buf[2] = v >> 8;
buf[3] = v;
} else {
buf[3] = v >> 24;
buf[2] = v >> 16;
buf[1] = v >> 8;
buf[0] = v;
}
break;
}
}
static void vnc_write_pixels_generic(VncState *vs, struct PixelFormat *pf,
void *pixels1, int size)
{
uint8_t buf[4];
if (pf->bytes_per_pixel == 4) {
uint32_t *pixels = pixels1;
int n, i;
n = size >> 2;
for(i = 0; i < n; i++) {
vnc_convert_pixel(vs, buf, pixels[i]);
vnc_write(vs, buf, vs->clientds.pf.bytes_per_pixel);
}
} else if (pf->bytes_per_pixel == 2) {
uint16_t *pixels = pixels1;
int n, i;
n = size >> 1;
for(i = 0; i < n; i++) {
vnc_convert_pixel(vs, buf, pixels[i]);
vnc_write(vs, buf, vs->clientds.pf.bytes_per_pixel);
}
} else if (pf->bytes_per_pixel == 1) {
uint8_t *pixels = pixels1;
int n, i;
n = size;
for(i = 0; i < n; i++) {
vnc_convert_pixel(vs, buf, pixels[i]);
vnc_write(vs, buf, vs->clientds.pf.bytes_per_pixel);
}
} else {
fprintf(stderr, "vnc_write_pixels_generic: VncState color depth not supported\n");
}
}
int vnc_raw_send_framebuffer_update(VncState *vs, int x, int y, int w, int h)
{
int i;
uint8_t *row;
VncDisplay *vd = vs->vd;
row = vd->server->data + y * ds_get_linesize(vs->ds) + x * ds_get_bytes_per_pixel(vs->ds);
for (i = 0; i < h; i++) {
vs->write_pixels(vs, &vd->server->pf, row, w * ds_get_bytes_per_pixel(vs->ds));
row += ds_get_linesize(vs->ds);
}
return 1;
}
int vnc_send_framebuffer_update(VncState *vs, int x, int y, int w, int h)
{
int n = 0;
switch(vs->vnc_encoding) {
case VNC_ENCODING_ZLIB:
n = vnc_zlib_send_framebuffer_update(vs, x, y, w, h);
break;
case VNC_ENCODING_HEXTILE:
vnc_framebuffer_update(vs, x, y, w, h, VNC_ENCODING_HEXTILE);
n = vnc_hextile_send_framebuffer_update(vs, x, y, w, h);
break;
case VNC_ENCODING_TIGHT:
n = vnc_tight_send_framebuffer_update(vs, x, y, w, h);
break;
case VNC_ENCODING_TIGHT_PNG:
n = vnc_tight_png_send_framebuffer_update(vs, x, y, w, h);
break;
default:
vnc_framebuffer_update(vs, x, y, w, h, VNC_ENCODING_RAW);
n = vnc_raw_send_framebuffer_update(vs, x, y, w, h);
break;
}
return n;
}
static void vnc_copy(VncState *vs, int src_x, int src_y, int dst_x, int dst_y, int w, int h)
{
/* send bitblit op to the vnc client */
vnc_lock_output(vs);
vnc_write_u8(vs, VNC_MSG_SERVER_FRAMEBUFFER_UPDATE);
vnc_write_u8(vs, 0);
vnc_write_u16(vs, 1); /* number of rects */
vnc_framebuffer_update(vs, dst_x, dst_y, w, h, VNC_ENCODING_COPYRECT);
vnc_write_u16(vs, src_x);
vnc_write_u16(vs, src_y);
vnc_unlock_output(vs);
vnc_flush(vs);
}
static void vnc_dpy_copy(DisplayState *ds, int src_x, int src_y, int dst_x, int dst_y, int w, int h)
{
VncDisplay *vd = ds->opaque;
VncState *vs, *vn;
uint8_t *src_row;
uint8_t *dst_row;
int i,x,y,pitch,depth,inc,w_lim,s;
int cmp_bytes;
vnc_refresh_server_surface(vd);
QTAILQ_FOREACH_SAFE(vs, &vd->clients, next, vn) {
if (vnc_has_feature(vs, VNC_FEATURE_COPYRECT)) {
vs->force_update = 1;
vnc_update_client_sync(vs, 1);
/* vs might be free()ed here */
}
}
/* do bitblit op on the local surface too */
pitch = ds_get_linesize(vd->ds);
depth = ds_get_bytes_per_pixel(vd->ds);
src_row = vd->server->data + pitch * src_y + depth * src_x;
dst_row = vd->server->data + pitch * dst_y + depth * dst_x;
y = dst_y;
inc = 1;
if (dst_y > src_y) {
/* copy backwards */
src_row += pitch * (h-1);
dst_row += pitch * (h-1);
pitch = -pitch;
y = dst_y + h - 1;
inc = -1;
}
w_lim = w - (16 - (dst_x % 16));
if (w_lim < 0)
w_lim = w;
else
w_lim = w - (w_lim % 16);
for (i = 0; i < h; i++) {
for (x = 0; x <= w_lim;
x += s, src_row += cmp_bytes, dst_row += cmp_bytes) {
if (x == w_lim) {
if ((s = w - w_lim) == 0)
break;
} else if (!x) {
s = (16 - (dst_x % 16));
s = MIN(s, w_lim);
} else {
s = 16;
}
cmp_bytes = s * depth;
if (memcmp(src_row, dst_row, cmp_bytes) == 0)
continue;
memmove(dst_row, src_row, cmp_bytes);
QTAILQ_FOREACH(vs, &vd->clients, next) {
if (!vnc_has_feature(vs, VNC_FEATURE_COPYRECT)) {
vnc_set_bit(vs->dirty[y], ((x + dst_x) / 16));
}
}
}
src_row += pitch - w * depth;
dst_row += pitch - w * depth;
y += inc;
}
QTAILQ_FOREACH(vs, &vd->clients, next) {
if (vnc_has_feature(vs, VNC_FEATURE_COPYRECT)) {
vnc_copy(vs, src_x, src_y, dst_x, dst_y, w, h);
}
}
}
static void vnc_mouse_set(int x, int y, int visible)
{
/* can we ask the client(s) to move the pointer ??? */
}
static int vnc_cursor_define(VncState *vs)
{
QEMUCursor *c = vs->vd->cursor;
PixelFormat pf = qemu_default_pixelformat(32);
int isize;
if (vnc_has_feature(vs, VNC_FEATURE_RICH_CURSOR)) {
vnc_lock_output(vs);
vnc_write_u8(vs, VNC_MSG_SERVER_FRAMEBUFFER_UPDATE);
vnc_write_u8(vs, 0); /* padding */
vnc_write_u16(vs, 1); /* # of rects */
vnc_framebuffer_update(vs, c->hot_x, c->hot_y, c->width, c->height,
VNC_ENCODING_RICH_CURSOR);
isize = c->width * c->height * vs->clientds.pf.bytes_per_pixel;
vnc_write_pixels_generic(vs, &pf, c->data, isize);
vnc_write(vs, vs->vd->cursor_mask, vs->vd->cursor_msize);
vnc_unlock_output(vs);
return 0;
}
return -1;
}
static void vnc_dpy_cursor_define(QEMUCursor *c)
{
VncDisplay *vd = vnc_display;
VncState *vs;
cursor_put(vd->cursor);
qemu_free(vd->cursor_mask);
vd->cursor = c;
cursor_get(vd->cursor);
vd->cursor_msize = cursor_get_mono_bpl(c) * c->height;
vd->cursor_mask = qemu_mallocz(vd->cursor_msize);
cursor_get_mono_mask(c, 0, vd->cursor_mask);
QTAILQ_FOREACH(vs, &vd->clients, next) {
vnc_cursor_define(vs);
}
}
static int find_and_clear_dirty_height(struct VncState *vs,
vnc: cleanup surface handling, fix screen corruption bug. (Gerd Hoffmann) This patch killes the old_data hack in the qemu server and replaces it with a clean separation of the guest-visible display surface and the vnc server display surface. Both guest and server surface have their own dirty bitmap for tracking screen updates. Workflow is this: (1) The guest writes to the guest surface. With shared buffers being active the guest writes are directly visible to the vnc server code. Note that this may happen in parallel to the vnc server code running (today only in xenfb, once we have vcpu threads in qemu also for other display adapters). (2) vnc_update() callback tags the specified area in the guest dirty map. (3) vnc_update_client() will first walk through the guest dirty map. It will compare guest and server surface for all regions tagged dirty and in case the screen content really did change the server surface and dirty map are updated. Note: old code used old_data in a simliar way, so this does *not* introduce an extra memcpy. (4) Then vnc_update_cient() will send the updates to the vnc client using the server surface and dirty map. Note: old code used the guest-visible surface instead, causing screen corruption in case of guest screen updates running in parallel. The separate dirty bitmap also has the nice effect that forced screen updates can be done cleanly by simply tagging the area in both guest and server dirty map. The old, hackish way was memset(old_data, 42, size) to trick the code checking for screen changes. Signed-off-by: Gerd Hoffmann <kraxel@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6860 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-20 18:59:14 +03:00
int y, int last_x, int x)
{
int h;
VncDisplay *vd = vs->vd;
for (h = 1; h < (vd->server->height - y); h++) {
int tmp_x;
if (!vnc_get_bit(vs->dirty[y + h], last_x))
break;
for (tmp_x = last_x; tmp_x < x; tmp_x++)
vnc_clear_bit(vs->dirty[y + h], tmp_x);
}
return h;
}
#ifdef CONFIG_VNC_THREAD
static int vnc_update_client_sync(VncState *vs, int has_dirty)
{
int ret = vnc_update_client(vs, has_dirty);
vnc_jobs_join(vs);
return ret;
}
#else
static int vnc_update_client_sync(VncState *vs, int has_dirty)
{
return vnc_update_client(vs, has_dirty);
}
#endif
static int vnc_update_client(VncState *vs, int has_dirty)
{
if (vs->need_update && vs->csock != -1) {
VncDisplay *vd = vs->vd;
VncJob *job;
int y;
int width, height;
int n = 0;
if (vs->output.offset && !vs->audio_cap && !vs->force_update)
/* kernel send buffers are full -> drop frames to throttle */
return 0;
if (!has_dirty && !vs->audio_cap && !vs->force_update)
return 0;
vnc: cleanup surface handling, fix screen corruption bug. (Gerd Hoffmann) This patch killes the old_data hack in the qemu server and replaces it with a clean separation of the guest-visible display surface and the vnc server display surface. Both guest and server surface have their own dirty bitmap for tracking screen updates. Workflow is this: (1) The guest writes to the guest surface. With shared buffers being active the guest writes are directly visible to the vnc server code. Note that this may happen in parallel to the vnc server code running (today only in xenfb, once we have vcpu threads in qemu also for other display adapters). (2) vnc_update() callback tags the specified area in the guest dirty map. (3) vnc_update_client() will first walk through the guest dirty map. It will compare guest and server surface for all regions tagged dirty and in case the screen content really did change the server surface and dirty map are updated. Note: old code used old_data in a simliar way, so this does *not* introduce an extra memcpy. (4) Then vnc_update_cient() will send the updates to the vnc client using the server surface and dirty map. Note: old code used the guest-visible surface instead, causing screen corruption in case of guest screen updates running in parallel. The separate dirty bitmap also has the nice effect that forced screen updates can be done cleanly by simply tagging the area in both guest and server dirty map. The old, hackish way was memset(old_data, 42, size) to trick the code checking for screen changes. Signed-off-by: Gerd Hoffmann <kraxel@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6860 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-20 18:59:14 +03:00
/*
* Send screen updates to the vnc client using the server
* surface and server dirty map. guest surface updates
* happening in parallel don't disturb us, the next pass will
* send them to the client.
*/
job = vnc_job_new(vs);
width = MIN(vd->server->width, vs->client_width);
height = MIN(vd->server->height, vs->client_height);
for (y = 0; y < height; y++) {
int x;
int last_x = -1;
for (x = 0; x < width / 16; x++) {
if (vnc_get_bit(vs->dirty[y], x)) {
if (last_x == -1) {
last_x = x;
}
vnc_clear_bit(vs->dirty[y], x);
} else {
if (last_x != -1) {
int h = find_and_clear_dirty_height(vs, y, last_x, x);
n += vnc_job_add_rect(job, last_x * 16, y,
(x - last_x) * 16, h);
}
last_x = -1;
}
}
if (last_x != -1) {
int h = find_and_clear_dirty_height(vs, y, last_x, x);
n += vnc_job_add_rect(job, last_x * 16, y,
(x - last_x) * 16, h);
}
}
vnc_job_push(job);
vs->force_update = 0;
return n;
}
if (vs->csock == -1)
vnc_disconnect_finish(vs);
return 0;
}
/* audio */
static void audio_capture_notify(void *opaque, audcnotification_e cmd)
{
VncState *vs = opaque;
switch (cmd) {
case AUD_CNOTIFY_DISABLE:
vnc_lock_output(vs);
vnc_write_u8(vs, VNC_MSG_SERVER_QEMU);
vnc_write_u8(vs, VNC_MSG_SERVER_QEMU_AUDIO);
vnc_write_u16(vs, VNC_MSG_SERVER_QEMU_AUDIO_END);
vnc_unlock_output(vs);
vnc_flush(vs);
break;
case AUD_CNOTIFY_ENABLE:
vnc_lock_output(vs);
vnc_write_u8(vs, VNC_MSG_SERVER_QEMU);
vnc_write_u8(vs, VNC_MSG_SERVER_QEMU_AUDIO);
vnc_write_u16(vs, VNC_MSG_SERVER_QEMU_AUDIO_BEGIN);
vnc_unlock_output(vs);
vnc_flush(vs);
break;
}
}
static void audio_capture_destroy(void *opaque)
{
}
static void audio_capture(void *opaque, void *buf, int size)
{
VncState *vs = opaque;
vnc_lock_output(vs);
vnc_write_u8(vs, VNC_MSG_SERVER_QEMU);
vnc_write_u8(vs, VNC_MSG_SERVER_QEMU_AUDIO);
vnc_write_u16(vs, VNC_MSG_SERVER_QEMU_AUDIO_DATA);
vnc_write_u32(vs, size);
vnc_write(vs, buf, size);
vnc_unlock_output(vs);
vnc_flush(vs);
}
static void audio_add(VncState *vs)
{
struct audio_capture_ops ops;
if (vs->audio_cap) {
monitor: Separate "default monitor" and "current monitor" cleanly Commits 376253ec..731b0364 introduced global variable cur_mon, which points to the "default monitor" (if any), except during execution of monitor_read() or monitor_control_read() it points to the monitor from which we're reading instead (the "current monitor"). Monitor command handlers run within monitor_read() or monitor_control_read(). Default monitor and current monitor are really separate things, and squashing them together is confusing and error-prone. For instance, usb_host_scan() can run both in "info usbhost" and periodically via usb_host_auto_check(). It prints to cur_mon, which is what we want in the former case: the monitor executing "info usbhost". But since that's the default monitor in the latter case, it periodically spams the default monitor there. A few places use cur_mon to log stuff to the default monitor. If we ever log something while cur_mon points to current monitor instead of default monitor, the log temporarily "jumps" to another monitor. Whether that can or cannot happen isn't always obvious. Maybe logging to the default monitor (which may not even exist) is a bad idea, and we should log to stderr or a logfile instead. But that's outside the scope of this commit. Change cur_mon to point to the current monitor. Create new default_mon to point to the default monitor. Update users of cur_mon accordingly. This fixes the periodical spamming of the default monitor by usb_host_scan(). It also stops "log jumping", should that problem exist.
2010-02-18 13:41:55 +03:00
monitor_printf(default_mon, "audio already running\n");
return;
}
ops.notify = audio_capture_notify;
ops.destroy = audio_capture_destroy;
ops.capture = audio_capture;
vs->audio_cap = AUD_add_capture(&vs->as, &ops, vs);
if (!vs->audio_cap) {
monitor: Separate "default monitor" and "current monitor" cleanly Commits 376253ec..731b0364 introduced global variable cur_mon, which points to the "default monitor" (if any), except during execution of monitor_read() or monitor_control_read() it points to the monitor from which we're reading instead (the "current monitor"). Monitor command handlers run within monitor_read() or monitor_control_read(). Default monitor and current monitor are really separate things, and squashing them together is confusing and error-prone. For instance, usb_host_scan() can run both in "info usbhost" and periodically via usb_host_auto_check(). It prints to cur_mon, which is what we want in the former case: the monitor executing "info usbhost". But since that's the default monitor in the latter case, it periodically spams the default monitor there. A few places use cur_mon to log stuff to the default monitor. If we ever log something while cur_mon points to current monitor instead of default monitor, the log temporarily "jumps" to another monitor. Whether that can or cannot happen isn't always obvious. Maybe logging to the default monitor (which may not even exist) is a bad idea, and we should log to stderr or a logfile instead. But that's outside the scope of this commit. Change cur_mon to point to the current monitor. Create new default_mon to point to the default monitor. Update users of cur_mon accordingly. This fixes the periodical spamming of the default monitor by usb_host_scan(). It also stops "log jumping", should that problem exist.
2010-02-18 13:41:55 +03:00
monitor_printf(default_mon, "Failed to add audio capture\n");
}
}
static void audio_del(VncState *vs)
{
if (vs->audio_cap) {
AUD_del_capture(vs->audio_cap, vs);
vs->audio_cap = NULL;
}
}
static void vnc_disconnect_start(VncState *vs)
{
if (vs->csock == -1)
return;
qemu_set_fd_handler2(vs->csock, NULL, NULL, NULL, NULL);
closesocket(vs->csock);
vs->csock = -1;
}
static void vnc_disconnect_finish(VncState *vs)
{
vnc_jobs_join(vs); /* Wait encoding jobs */
vnc_lock_output(vs);
vnc_qmp_event(vs, QEVENT_VNC_DISCONNECTED);
buffer_free(&vs->input);
buffer_free(&vs->output);
qobject_decref(vs->info);
vnc_zlib_clear(vs);
vnc_tight_clear(vs);
#ifdef CONFIG_VNC_TLS
vnc_tls_client_cleanup(vs);
#endif /* CONFIG_VNC_TLS */
#ifdef CONFIG_VNC_SASL
vnc_sasl_client_cleanup(vs);
#endif /* CONFIG_VNC_SASL */
audio_del(vs);
QTAILQ_REMOVE(&vs->vd->clients, vs, next);
if (QTAILQ_EMPTY(&vs->vd->clients)) {
dcl->idle = 1;
}
qemu_remove_mouse_mode_change_notifier(&vs->mouse_mode_notifier);
vnc_remove_timer(vs->vd);
if (vs->vd->lock_key_sync)
qemu_remove_led_event_handler(vs->led);
vnc_unlock_output(vs);
#ifdef CONFIG_VNC_THREAD
qemu_mutex_destroy(&vs->output_mutex);
#endif
qemu_free(vs);
}
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
int vnc_client_io_error(VncState *vs, int ret, int last_errno)
{
if (ret == 0 || ret == -1) {
if (ret == -1) {
switch (last_errno) {
case EINTR:
case EAGAIN:
#ifdef _WIN32
case WSAEWOULDBLOCK:
#endif
return 0;
default:
break;
}
}
VNC_DEBUG("Closing down client sock: ret %d, errno %d\n",
ret, ret < 0 ? last_errno : 0);
vnc_disconnect_start(vs);
vnc: cleanup surface handling, fix screen corruption bug. (Gerd Hoffmann) This patch killes the old_data hack in the qemu server and replaces it with a clean separation of the guest-visible display surface and the vnc server display surface. Both guest and server surface have their own dirty bitmap for tracking screen updates. Workflow is this: (1) The guest writes to the guest surface. With shared buffers being active the guest writes are directly visible to the vnc server code. Note that this may happen in parallel to the vnc server code running (today only in xenfb, once we have vcpu threads in qemu also for other display adapters). (2) vnc_update() callback tags the specified area in the guest dirty map. (3) vnc_update_client() will first walk through the guest dirty map. It will compare guest and server surface for all regions tagged dirty and in case the screen content really did change the server surface and dirty map are updated. Note: old code used old_data in a simliar way, so this does *not* introduce an extra memcpy. (4) Then vnc_update_cient() will send the updates to the vnc client using the server surface and dirty map. Note: old code used the guest-visible surface instead, causing screen corruption in case of guest screen updates running in parallel. The separate dirty bitmap also has the nice effect that forced screen updates can be done cleanly by simply tagging the area in both guest and server dirty map. The old, hackish way was memset(old_data, 42, size) to trick the code checking for screen changes. Signed-off-by: Gerd Hoffmann <kraxel@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6860 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-20 18:59:14 +03:00
return 0;
}
return ret;
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
void vnc_client_error(VncState *vs)
{
VNC_DEBUG("Closing down client sock: protocol error\n");
vnc_disconnect_start(vs);
}
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
/*
* Called to write a chunk of data to the client socket. The data may
* be the raw data, or may have already been encoded by SASL.
* The data will be written either straight onto the socket, or
* written via the GNUTLS wrappers, if TLS/SSL encryption is enabled
*
* NB, it is theoretically possible to have 2 layers of encryption,
* both SASL, and this TLS layer. It is highly unlikely in practice
* though, since SASL encryption will typically be a no-op if TLS
* is active
*
* Returns the number of bytes written, which may be less than
* the requested 'datalen' if the socket would block. Returns
* -1 on error, and disconnects the client socket.
*/
long vnc_client_write_buf(VncState *vs, const uint8_t *data, size_t datalen)
{
long ret;
#ifdef CONFIG_VNC_TLS
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
if (vs->tls.session) {
ret = gnutls_write(vs->tls.session, data, datalen);
if (ret < 0) {
if (ret == GNUTLS_E_AGAIN)
errno = EAGAIN;
else
errno = EIO;
ret = -1;
}
} else
#endif /* CONFIG_VNC_TLS */
ret = send(vs->csock, (const void *)data, datalen, 0);
VNC_DEBUG("Wrote wire %p %zd -> %ld\n", data, datalen, ret);
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
return vnc_client_io_error(vs, ret, socket_error());
}
/*
* Called to write buffered data to the client socket, when not
* using any SASL SSF encryption layers. Will write as much data
* as possible without blocking. If all buffered data is written,
* will switch the FD poll() handler back to read monitoring.
*
* Returns the number of bytes written, which may be less than
* the buffered output data if the socket would block. Returns
* -1 on error, and disconnects the client socket.
*/
static long vnc_client_write_plain(VncState *vs)
{
long ret;
#ifdef CONFIG_VNC_SASL
VNC_DEBUG("Write Plain: Pending output %p size %zd offset %zd. Wait SSF %d\n",
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
vs->output.buffer, vs->output.capacity, vs->output.offset,
vs->sasl.waitWriteSSF);
if (vs->sasl.conn &&
vs->sasl.runSSF &&
vs->sasl.waitWriteSSF) {
ret = vnc_client_write_buf(vs, vs->output.buffer, vs->sasl.waitWriteSSF);
if (ret)
vs->sasl.waitWriteSSF -= ret;
} else
#endif /* CONFIG_VNC_SASL */
ret = vnc_client_write_buf(vs, vs->output.buffer, vs->output.offset);
if (!ret)
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
return 0;
memmove(vs->output.buffer, vs->output.buffer + ret, (vs->output.offset - ret));
vs->output.offset -= ret;
if (vs->output.offset == 0) {
qemu_set_fd_handler2(vs->csock, NULL, vnc_client_read, NULL, vs);
}
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
return ret;
}
/*
* First function called whenever there is data to be written to
* the client socket. Will delegate actual work according to whether
* SASL SSF layers are enabled (thus requiring encryption calls)
*/
static void vnc_client_write_locked(void *opaque)
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
{
VncState *vs = opaque;
#ifdef CONFIG_VNC_SASL
if (vs->sasl.conn &&
vs->sasl.runSSF &&
!vs->sasl.waitWriteSSF) {
vnc_client_write_sasl(vs);
} else
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
#endif /* CONFIG_VNC_SASL */
vnc_client_write_plain(vs);
}
void vnc_client_write(void *opaque)
{
VncState *vs = opaque;
vnc_lock_output(vs);
if (vs->output.offset) {
vnc_client_write_locked(opaque);
} else if (vs->csock != -1) {
qemu_set_fd_handler2(vs->csock, NULL, vnc_client_read, NULL, vs);
}
vnc_unlock_output(vs);
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
void vnc_read_when(VncState *vs, VncReadEvent *func, size_t expecting)
{
vs->read_handler = func;
vs->read_handler_expect = expecting;
}
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
/*
* Called to read a chunk of data from the client socket. The data may
* be the raw data, or may need to be further decoded by SASL.
* The data will be read either straight from to the socket, or
* read via the GNUTLS wrappers, if TLS/SSL encryption is enabled
*
* NB, it is theoretically possible to have 2 layers of encryption,
* both SASL, and this TLS layer. It is highly unlikely in practice
* though, since SASL encryption will typically be a no-op if TLS
* is active
*
* Returns the number of bytes read, which may be less than
* the requested 'datalen' if the socket would block. Returns
* -1 on error, and disconnects the client socket.
*/
long vnc_client_read_buf(VncState *vs, uint8_t *data, size_t datalen)
{
long ret;
#ifdef CONFIG_VNC_TLS
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
if (vs->tls.session) {
ret = gnutls_read(vs->tls.session, data, datalen);
if (ret < 0) {
if (ret == GNUTLS_E_AGAIN)
errno = EAGAIN;
else
errno = EIO;
ret = -1;
}
} else
#endif /* CONFIG_VNC_TLS */
ret = recv(vs->csock, (void *)data, datalen, 0);
VNC_DEBUG("Read wire %p %zd -> %ld\n", data, datalen, ret);
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
return vnc_client_io_error(vs, ret, socket_error());
}
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
/*
* Called to read data from the client socket to the input buffer,
* when not using any SASL SSF encryption layers. Will read as much
* data as possible without blocking.
*
* Returns the number of bytes read. Returns -1 on error, and
* disconnects the client socket.
*/
static long vnc_client_read_plain(VncState *vs)
{
int ret;
VNC_DEBUG("Read plain %p size %zd offset %zd\n",
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
vs->input.buffer, vs->input.capacity, vs->input.offset);
buffer_reserve(&vs->input, 4096);
ret = vnc_client_read_buf(vs, buffer_end(&vs->input), 4096);
if (!ret)
return 0;
vs->input.offset += ret;
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
return ret;
}
/*
* First function called whenever there is more data to be read from
* the client socket. Will delegate actual work according to whether
* SASL SSF layers are enabled (thus requiring decryption calls)
*/
void vnc_client_read(void *opaque)
{
VncState *vs = opaque;
long ret;
#ifdef CONFIG_VNC_SASL
if (vs->sasl.conn && vs->sasl.runSSF)
ret = vnc_client_read_sasl(vs);
else
#endif /* CONFIG_VNC_SASL */
ret = vnc_client_read_plain(vs);
if (!ret) {
if (vs->csock == -1)
vnc_disconnect_finish(vs);
return;
}
while (vs->read_handler && vs->input.offset >= vs->read_handler_expect) {
size_t len = vs->read_handler_expect;
int ret;
ret = vs->read_handler(vs, vs->input.buffer, len);
if (vs->csock == -1) {
vnc_disconnect_finish(vs);
return;
}
if (!ret) {
memmove(vs->input.buffer, vs->input.buffer + len, (vs->input.offset - len));
vs->input.offset -= len;
} else {
vs->read_handler_expect = ret;
}
}
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
void vnc_write(VncState *vs, const void *data, size_t len)
{
buffer_reserve(&vs->output, len);
if (vs->csock != -1 && buffer_empty(&vs->output)) {
qemu_set_fd_handler2(vs->csock, NULL, vnc_client_read, vnc_client_write, vs);
}
buffer_append(&vs->output, data, len);
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
void vnc_write_s32(VncState *vs, int32_t value)
{
vnc_write_u32(vs, *(uint32_t *)&value);
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
void vnc_write_u32(VncState *vs, uint32_t value)
{
uint8_t buf[4];
buf[0] = (value >> 24) & 0xFF;
buf[1] = (value >> 16) & 0xFF;
buf[2] = (value >> 8) & 0xFF;
buf[3] = value & 0xFF;
vnc_write(vs, buf, 4);
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
void vnc_write_u16(VncState *vs, uint16_t value)
{
uint8_t buf[2];
buf[0] = (value >> 8) & 0xFF;
buf[1] = value & 0xFF;
vnc_write(vs, buf, 2);
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
void vnc_write_u8(VncState *vs, uint8_t value)
{
vnc_write(vs, (char *)&value, 1);
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
void vnc_flush(VncState *vs)
{
vnc_lock_output(vs);
if (vs->csock != -1 && vs->output.offset) {
vnc_client_write_locked(vs);
}
vnc_unlock_output(vs);
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
uint8_t read_u8(uint8_t *data, size_t offset)
{
return data[offset];
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
uint16_t read_u16(uint8_t *data, size_t offset)
{
return ((data[offset] & 0xFF) << 8) | (data[offset + 1] & 0xFF);
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
int32_t read_s32(uint8_t *data, size_t offset)
{
return (int32_t)((data[offset] << 24) | (data[offset + 1] << 16) |
(data[offset + 2] << 8) | data[offset + 3]);
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
uint32_t read_u32(uint8_t *data, size_t offset)
{
return ((data[offset] << 24) | (data[offset + 1] << 16) |
(data[offset + 2] << 8) | data[offset + 3]);
}
static void client_cut_text(VncState *vs, size_t len, uint8_t *text)
{
}
static void check_pointer_type_change(Notifier *notifier)
{
VncState *vs = container_of(notifier, VncState, mouse_mode_notifier);
int absolute = kbd_mouse_is_absolute();
if (vnc_has_feature(vs, VNC_FEATURE_POINTER_TYPE_CHANGE) && vs->absolute != absolute) {
vnc_lock_output(vs);
vnc_write_u8(vs, VNC_MSG_SERVER_FRAMEBUFFER_UPDATE);
vnc_write_u8(vs, 0);
vnc_write_u16(vs, 1);
vnc_framebuffer_update(vs, absolute, 0,
ds_get_width(vs->ds), ds_get_height(vs->ds),
VNC_ENCODING_POINTER_TYPE_CHANGE);
vnc_unlock_output(vs);
vnc_flush(vs);
}
vs->absolute = absolute;
}
static void pointer_event(VncState *vs, int button_mask, int x, int y)
{
int buttons = 0;
int dz = 0;
if (button_mask & 0x01)
buttons |= MOUSE_EVENT_LBUTTON;
if (button_mask & 0x02)
buttons |= MOUSE_EVENT_MBUTTON;
if (button_mask & 0x04)
buttons |= MOUSE_EVENT_RBUTTON;
if (button_mask & 0x08)
dz = -1;
if (button_mask & 0x10)
dz = 1;
if (vs->absolute) {
kbd_mouse_event(ds_get_width(vs->ds) > 1 ?
x * 0x7FFF / (ds_get_width(vs->ds) - 1) : 0x4000,
ds_get_height(vs->ds) > 1 ?
y * 0x7FFF / (ds_get_height(vs->ds) - 1) : 0x4000,
dz, buttons);
} else if (vnc_has_feature(vs, VNC_FEATURE_POINTER_TYPE_CHANGE)) {
x -= 0x7FFF;
y -= 0x7FFF;
kbd_mouse_event(x, y, dz, buttons);
} else {
if (vs->last_x != -1)
kbd_mouse_event(x - vs->last_x,
y - vs->last_y,
dz, buttons);
vs->last_x = x;
vs->last_y = y;
}
}
static void reset_keys(VncState *vs)
{
int i;
for(i = 0; i < 256; i++) {
if (vs->modifiers_state[i]) {
if (i & SCANCODE_GREY)
kbd_put_keycode(SCANCODE_EMUL0);
kbd_put_keycode(i | SCANCODE_UP);
vs->modifiers_state[i] = 0;
}
}
}
static void press_key(VncState *vs, int keysym)
{
int keycode = keysym2scancode(vs->vd->kbd_layout, keysym) & SCANCODE_KEYMASK;
if (keycode & SCANCODE_GREY)
kbd_put_keycode(SCANCODE_EMUL0);
kbd_put_keycode(keycode & SCANCODE_KEYCODEMASK);
if (keycode & SCANCODE_GREY)
kbd_put_keycode(SCANCODE_EMUL0);
kbd_put_keycode(keycode | SCANCODE_UP);
}
static void kbd_leds(void *opaque, int ledstate)
{
VncState *vs = opaque;
int caps, num;
caps = ledstate & QEMU_CAPS_LOCK_LED ? 1 : 0;
num = ledstate & QEMU_NUM_LOCK_LED ? 1 : 0;
if (vs->modifiers_state[0x3a] != caps) {
vs->modifiers_state[0x3a] = caps;
}
if (vs->modifiers_state[0x45] != num) {
vs->modifiers_state[0x45] = num;
}
}
static void do_key_event(VncState *vs, int down, int keycode, int sym)
{
/* QEMU console switch */
switch(keycode) {
case 0x2a: /* Left Shift */
case 0x36: /* Right Shift */
case 0x1d: /* Left CTRL */
case 0x9d: /* Right CTRL */
case 0x38: /* Left ALT */
case 0xb8: /* Right ALT */
if (down)
vs->modifiers_state[keycode] = 1;
else
vs->modifiers_state[keycode] = 0;
break;
case 0x02 ... 0x0a: /* '1' to '9' keys */
if (down && vs->modifiers_state[0x1d] && vs->modifiers_state[0x38]) {
/* Reset the modifiers sent to the current console */
reset_keys(vs);
console_select(keycode - 0x02);
return;
}
break;
case 0x3a: /* CapsLock */
case 0x45: /* NumLock */
if (down)
vs->modifiers_state[keycode] ^= 1;
break;
}
if (vs->vd->lock_key_sync &&
keycode_is_keypad(vs->vd->kbd_layout, keycode)) {
/* If the numlock state needs to change then simulate an additional
keypress before sending this one. This will happen if the user
toggles numlock away from the VNC window.
*/
if (keysym_is_numlock(vs->vd->kbd_layout, sym & 0xFFFF)) {
if (!vs->modifiers_state[0x45]) {
vs->modifiers_state[0x45] = 1;
press_key(vs, 0xff7f);
}
} else {
if (vs->modifiers_state[0x45]) {
vs->modifiers_state[0x45] = 0;
press_key(vs, 0xff7f);
}
}
}
if (vs->vd->lock_key_sync &&
((sym >= 'A' && sym <= 'Z') || (sym >= 'a' && sym <= 'z'))) {
/* If the capslock state needs to change then simulate an additional
keypress before sending this one. This will happen if the user
toggles capslock away from the VNC window.
*/
int uppercase = !!(sym >= 'A' && sym <= 'Z');
int shift = !!(vs->modifiers_state[0x2a] | vs->modifiers_state[0x36]);
int capslock = !!(vs->modifiers_state[0x3a]);
if (capslock) {
if (uppercase == shift) {
vs->modifiers_state[0x3a] = 0;
press_key(vs, 0xffe5);
}
} else {
if (uppercase != shift) {
vs->modifiers_state[0x3a] = 1;
press_key(vs, 0xffe5);
}
}
}
if (is_graphic_console()) {
if (keycode & SCANCODE_GREY)
kbd_put_keycode(SCANCODE_EMUL0);
if (down)
kbd_put_keycode(keycode & SCANCODE_KEYCODEMASK);
else
kbd_put_keycode(keycode | SCANCODE_UP);
} else {
/* QEMU console emulation */
if (down) {
int numlock = vs->modifiers_state[0x45];
switch (keycode) {
case 0x2a: /* Left Shift */
case 0x36: /* Right Shift */
case 0x1d: /* Left CTRL */
case 0x9d: /* Right CTRL */
case 0x38: /* Left ALT */
case 0xb8: /* Right ALT */
break;
case 0xc8:
kbd_put_keysym(QEMU_KEY_UP);
break;
case 0xd0:
kbd_put_keysym(QEMU_KEY_DOWN);
break;
case 0xcb:
kbd_put_keysym(QEMU_KEY_LEFT);
break;
case 0xcd:
kbd_put_keysym(QEMU_KEY_RIGHT);
break;
case 0xd3:
kbd_put_keysym(QEMU_KEY_DELETE);
break;
case 0xc7:
kbd_put_keysym(QEMU_KEY_HOME);
break;
case 0xcf:
kbd_put_keysym(QEMU_KEY_END);
break;
case 0xc9:
kbd_put_keysym(QEMU_KEY_PAGEUP);
break;
case 0xd1:
kbd_put_keysym(QEMU_KEY_PAGEDOWN);
break;
case 0x47:
kbd_put_keysym(numlock ? '7' : QEMU_KEY_HOME);
break;
case 0x48:
kbd_put_keysym(numlock ? '8' : QEMU_KEY_UP);
break;
case 0x49:
kbd_put_keysym(numlock ? '9' : QEMU_KEY_PAGEUP);
break;
case 0x4b:
kbd_put_keysym(numlock ? '4' : QEMU_KEY_LEFT);
break;
case 0x4c:
kbd_put_keysym('5');
break;
case 0x4d:
kbd_put_keysym(numlock ? '6' : QEMU_KEY_RIGHT);
break;
case 0x4f:
kbd_put_keysym(numlock ? '1' : QEMU_KEY_END);
break;
case 0x50:
kbd_put_keysym(numlock ? '2' : QEMU_KEY_DOWN);
break;
case 0x51:
kbd_put_keysym(numlock ? '3' : QEMU_KEY_PAGEDOWN);
break;
case 0x52:
kbd_put_keysym('0');
break;
case 0x53:
kbd_put_keysym(numlock ? '.' : QEMU_KEY_DELETE);
break;
case 0xb5:
kbd_put_keysym('/');
break;
case 0x37:
kbd_put_keysym('*');
break;
case 0x4a:
kbd_put_keysym('-');
break;
case 0x4e:
kbd_put_keysym('+');
break;
case 0x9c:
kbd_put_keysym('\n');
break;
default:
kbd_put_keysym(sym);
break;
}
}
}
}
static void key_event(VncState *vs, int down, uint32_t sym)
{
int keycode;
int lsym = sym;
if (lsym >= 'A' && lsym <= 'Z' && is_graphic_console()) {
lsym = lsym - 'A' + 'a';
}
keycode = keysym2scancode(vs->vd->kbd_layout, lsym & 0xFFFF) & SCANCODE_KEYMASK;
do_key_event(vs, down, keycode, sym);
}
static void ext_key_event(VncState *vs, int down,
uint32_t sym, uint16_t keycode)
{
/* if the user specifies a keyboard layout, always use it */
if (keyboard_layout)
key_event(vs, down, sym);
else
do_key_event(vs, down, keycode, sym);
}
static void framebuffer_update_request(VncState *vs, int incremental,
int x_position, int y_position,
int w, int h)
{
if (y_position > ds_get_height(vs->ds))
y_position = ds_get_height(vs->ds);
if (y_position + h >= ds_get_height(vs->ds))
h = ds_get_height(vs->ds) - y_position;
int i;
vs->need_update = 1;
if (!incremental) {
vs->force_update = 1;
for (i = 0; i < h; i++) {
vnc_set_bits(vs->dirty[y_position + i],
(ds_get_width(vs->ds) / 16), VNC_DIRTY_WORDS);
}
}
}
static void send_ext_key_event_ack(VncState *vs)
{
vnc_lock_output(vs);
vnc_write_u8(vs, VNC_MSG_SERVER_FRAMEBUFFER_UPDATE);
vnc_write_u8(vs, 0);
vnc_write_u16(vs, 1);
vnc_framebuffer_update(vs, 0, 0, ds_get_width(vs->ds), ds_get_height(vs->ds),
VNC_ENCODING_EXT_KEY_EVENT);
vnc_unlock_output(vs);
vnc_flush(vs);
}
static void send_ext_audio_ack(VncState *vs)
{
vnc_lock_output(vs);
vnc_write_u8(vs, VNC_MSG_SERVER_FRAMEBUFFER_UPDATE);
vnc_write_u8(vs, 0);
vnc_write_u16(vs, 1);
vnc_framebuffer_update(vs, 0, 0, ds_get_width(vs->ds), ds_get_height(vs->ds),
VNC_ENCODING_AUDIO);
vnc_unlock_output(vs);
vnc_flush(vs);
}
static void set_encodings(VncState *vs, int32_t *encodings, size_t n_encodings)
{
int i;
unsigned int enc = 0;
vs->features = 0;
vs->vnc_encoding = 0;
vs->tight.compression = 9;
vs->tight.quality = -1; /* Lossless by default */
vs->absolute = -1;
/*
* Start from the end because the encodings are sent in order of preference.
* This way the prefered encoding (first encoding defined in the array)
* will be set at the end of the loop.
*/
for (i = n_encodings - 1; i >= 0; i--) {
enc = encodings[i];
switch (enc) {
case VNC_ENCODING_RAW:
vs->vnc_encoding = enc;
break;
case VNC_ENCODING_COPYRECT:
vs->features |= VNC_FEATURE_COPYRECT_MASK;
break;
case VNC_ENCODING_HEXTILE:
vs->features |= VNC_FEATURE_HEXTILE_MASK;
vs->vnc_encoding = enc;
break;
case VNC_ENCODING_TIGHT:
vs->features |= VNC_FEATURE_TIGHT_MASK;
vs->vnc_encoding = enc;
break;
case VNC_ENCODING_TIGHT_PNG:
vs->features |= VNC_FEATURE_TIGHT_PNG_MASK;
vs->vnc_encoding = enc;
break;
case VNC_ENCODING_ZLIB:
vs->features |= VNC_FEATURE_ZLIB_MASK;
vs->vnc_encoding = enc;
break;
case VNC_ENCODING_DESKTOPRESIZE:
vs->features |= VNC_FEATURE_RESIZE_MASK;
break;
case VNC_ENCODING_POINTER_TYPE_CHANGE:
vs->features |= VNC_FEATURE_POINTER_TYPE_CHANGE_MASK;
break;
case VNC_ENCODING_RICH_CURSOR:
vs->features |= VNC_FEATURE_RICH_CURSOR_MASK;
break;
case VNC_ENCODING_EXT_KEY_EVENT:
send_ext_key_event_ack(vs);
break;
case VNC_ENCODING_AUDIO:
send_ext_audio_ack(vs);
break;
case VNC_ENCODING_WMVi:
vs->features |= VNC_FEATURE_WMVI_MASK;
break;
case VNC_ENCODING_COMPRESSLEVEL0 ... VNC_ENCODING_COMPRESSLEVEL0 + 9:
vs->tight.compression = (enc & 0x0F);
break;
case VNC_ENCODING_QUALITYLEVEL0 ... VNC_ENCODING_QUALITYLEVEL0 + 9:
vs->tight.quality = (enc & 0x0F);
break;
default:
VNC_DEBUG("Unknown encoding: %d (0x%.8x): %d\n", i, enc, enc);
break;
}
}
vnc_desktop_resize(vs);
check_pointer_type_change(&vs->mouse_mode_notifier);
}
static void set_pixel_conversion(VncState *vs)
{
if ((vs->clientds.flags & QEMU_BIG_ENDIAN_FLAG) ==
(vs->ds->surface->flags & QEMU_BIG_ENDIAN_FLAG) &&
!memcmp(&(vs->clientds.pf), &(vs->ds->surface->pf), sizeof(PixelFormat))) {
vs->write_pixels = vnc_write_pixels_copy;
vnc_hextile_set_pixel_conversion(vs, 0);
} else {
vs->write_pixels = vnc_write_pixels_generic;
vnc_hextile_set_pixel_conversion(vs, 1);
}
}
static void set_pixel_format(VncState *vs,
int bits_per_pixel, int depth,
int big_endian_flag, int true_color_flag,
int red_max, int green_max, int blue_max,
int red_shift, int green_shift, int blue_shift)
{
if (!true_color_flag) {
vnc_client_error(vs);
return;
}
vs->clientds = *(vs->vd->guest.ds);
vs->clientds.pf.rmax = red_max;
count_bits(vs->clientds.pf.rbits, red_max);
vs->clientds.pf.rshift = red_shift;
vs->clientds.pf.rmask = red_max << red_shift;
vs->clientds.pf.gmax = green_max;
count_bits(vs->clientds.pf.gbits, green_max);
vs->clientds.pf.gshift = green_shift;
vs->clientds.pf.gmask = green_max << green_shift;
vs->clientds.pf.bmax = blue_max;
count_bits(vs->clientds.pf.bbits, blue_max);
vs->clientds.pf.bshift = blue_shift;
vs->clientds.pf.bmask = blue_max << blue_shift;
vs->clientds.pf.bits_per_pixel = bits_per_pixel;
vs->clientds.pf.bytes_per_pixel = bits_per_pixel / 8;
vs->clientds.pf.depth = bits_per_pixel == 32 ? 24 : bits_per_pixel;
vs->clientds.flags = big_endian_flag ? QEMU_BIG_ENDIAN_FLAG : 0x00;
set_pixel_conversion(vs);
vga_hw_invalidate();
vga_hw_update();
}
static void pixel_format_message (VncState *vs) {
char pad[3] = { 0, 0, 0 };
vnc_write_u8(vs, vs->ds->surface->pf.bits_per_pixel); /* bits-per-pixel */
vnc_write_u8(vs, vs->ds->surface->pf.depth); /* depth */
#ifdef HOST_WORDS_BIGENDIAN
vnc_write_u8(vs, 1); /* big-endian-flag */
#else
vnc_write_u8(vs, 0); /* big-endian-flag */
#endif
vnc_write_u8(vs, 1); /* true-color-flag */
vnc_write_u16(vs, vs->ds->surface->pf.rmax); /* red-max */
vnc_write_u16(vs, vs->ds->surface->pf.gmax); /* green-max */
vnc_write_u16(vs, vs->ds->surface->pf.bmax); /* blue-max */
vnc_write_u8(vs, vs->ds->surface->pf.rshift); /* red-shift */
vnc_write_u8(vs, vs->ds->surface->pf.gshift); /* green-shift */
vnc_write_u8(vs, vs->ds->surface->pf.bshift); /* blue-shift */
vnc_hextile_set_pixel_conversion(vs, 0);
vs->clientds = *(vs->ds->surface);
vs->clientds.flags &= ~QEMU_ALLOCATED_FLAG;
vs->write_pixels = vnc_write_pixels_copy;
vnc_write(vs, pad, 3); /* padding */
}
DisplayState interface change (Stefano Stabellini) This patch changes the DisplayState interface adding support for multiple frontends at the same time (sdl and vnc) and implements most of the benefit of the shared_buf patch without the added complexity. Currently DisplayState is managed by sdl (or vnc) and sdl (or vnc) is also responsible for allocating the data and setting the depth. Vga.c (or another backend) will do any necessary conversion. The idea is to change it so that is vga.c (or another backend) together with console.c that fully manage the DisplayState interface allocating data and setting the depth (either 16 or 32 bit, if the guest uses a different resolution or is in text mode, vga.c (or another backend) is in charge of doing the conversion seamlessly). The other idea is that DisplayState supports *multiple* frontends like sdl and vnc; each of them can register some callbacks to be called when a display event occurs. The interesting changes are: - the new structures and related functions in console.h and console.c in particular the following functions are very helpful to manage a DisplaySurface: qemu_create_displaysurface qemu_resize_displaysurface qemu_create_displaysurface_from qemu_free_displaysurface - console_select and qemu_console_resize in console.c this two functions manage multiple consoles on a single host display - moving code around in hw/vga.c as for the shared_buf patch this is necessary to be able to handle a dynamic DisplaySurface bpp - changes to vga_draw_graphic in hw/vga.c this is the place where the DisplaySurface buffer is shared with the videoram, when possible; Compared to the last version the only changes are: - do not remove support to dpy_copy in cirrus_vga - change the name of the displaysurface handling functions Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6336 c046a42c-6fe2-441c-8c8c-71466251a162
2009-01-16 01:14:11 +03:00
static void vnc_dpy_setdata(DisplayState *ds)
{
/* We don't have to do anything */
}
static void vnc_colordepth(VncState *vs)
{
if (vnc_has_feature(vs, VNC_FEATURE_WMVI)) {
/* Sending a WMVi message to notify the client*/
vnc_lock_output(vs);
vnc_write_u8(vs, VNC_MSG_SERVER_FRAMEBUFFER_UPDATE);
vnc_write_u8(vs, 0);
vnc_write_u16(vs, 1); /* number of rects */
vnc_framebuffer_update(vs, 0, 0, ds_get_width(vs->ds),
ds_get_height(vs->ds), VNC_ENCODING_WMVi);
pixel_format_message(vs);
vnc_unlock_output(vs);
vnc_flush(vs);
} else {
set_pixel_conversion(vs);
}
}
static int protocol_client_msg(VncState *vs, uint8_t *data, size_t len)
{
int i;
uint16_t limit;
VncDisplay *vd = vs->vd;
if (data[0] > 3) {
vd->timer_interval = VNC_REFRESH_INTERVAL_BASE;
if (!qemu_timer_expired(vd->timer, qemu_get_clock(rt_clock) + vd->timer_interval))
qemu_mod_timer(vd->timer, qemu_get_clock(rt_clock) + vd->timer_interval);
}
switch (data[0]) {
case VNC_MSG_CLIENT_SET_PIXEL_FORMAT:
if (len == 1)
return 20;
set_pixel_format(vs, read_u8(data, 4), read_u8(data, 5),
read_u8(data, 6), read_u8(data, 7),
read_u16(data, 8), read_u16(data, 10),
read_u16(data, 12), read_u8(data, 14),
read_u8(data, 15), read_u8(data, 16));
break;
case VNC_MSG_CLIENT_SET_ENCODINGS:
if (len == 1)
return 4;
if (len == 4) {
limit = read_u16(data, 2);
if (limit > 0)
return 4 + (limit * 4);
} else
limit = read_u16(data, 2);
for (i = 0; i < limit; i++) {
int32_t val = read_s32(data, 4 + (i * 4));
memcpy(data + 4 + (i * 4), &val, sizeof(val));
}
set_encodings(vs, (int32_t *)(data + 4), limit);
break;
case VNC_MSG_CLIENT_FRAMEBUFFER_UPDATE_REQUEST:
if (len == 1)
return 10;
framebuffer_update_request(vs,
read_u8(data, 1), read_u16(data, 2), read_u16(data, 4),
read_u16(data, 6), read_u16(data, 8));
break;
case VNC_MSG_CLIENT_KEY_EVENT:
if (len == 1)
return 8;
key_event(vs, read_u8(data, 1), read_u32(data, 4));
break;
case VNC_MSG_CLIENT_POINTER_EVENT:
if (len == 1)
return 6;
pointer_event(vs, read_u8(data, 1), read_u16(data, 2), read_u16(data, 4));
break;
case VNC_MSG_CLIENT_CUT_TEXT:
if (len == 1)
return 8;
if (len == 8) {
uint32_t dlen = read_u32(data, 4);
if (dlen > 0)
return 8 + dlen;
}
client_cut_text(vs, read_u32(data, 4), data + 8);
break;
case VNC_MSG_CLIENT_QEMU:
if (len == 1)
return 2;
switch (read_u8(data, 1)) {
case VNC_MSG_CLIENT_QEMU_EXT_KEY_EVENT:
if (len == 2)
return 12;
ext_key_event(vs, read_u16(data, 2),
read_u32(data, 4), read_u32(data, 8));
break;
case VNC_MSG_CLIENT_QEMU_AUDIO:
if (len == 2)
return 4;
switch (read_u16 (data, 2)) {
case VNC_MSG_CLIENT_QEMU_AUDIO_ENABLE:
audio_add(vs);
break;
case VNC_MSG_CLIENT_QEMU_AUDIO_DISABLE:
audio_del(vs);
break;
case VNC_MSG_CLIENT_QEMU_AUDIO_SET_FORMAT:
if (len == 4)
return 10;
switch (read_u8(data, 4)) {
case 0: vs->as.fmt = AUD_FMT_U8; break;
case 1: vs->as.fmt = AUD_FMT_S8; break;
case 2: vs->as.fmt = AUD_FMT_U16; break;
case 3: vs->as.fmt = AUD_FMT_S16; break;
case 4: vs->as.fmt = AUD_FMT_U32; break;
case 5: vs->as.fmt = AUD_FMT_S32; break;
default:
printf("Invalid audio format %d\n", read_u8(data, 4));
vnc_client_error(vs);
break;
}
vs->as.nchannels = read_u8(data, 5);
if (vs->as.nchannels != 1 && vs->as.nchannels != 2) {
printf("Invalid audio channel coount %d\n",
read_u8(data, 5));
vnc_client_error(vs);
break;
}
vs->as.freq = read_u32(data, 6);
break;
default:
printf ("Invalid audio message %d\n", read_u8(data, 4));
vnc_client_error(vs);
break;
}
break;
default:
printf("Msg: %d\n", read_u16(data, 0));
vnc_client_error(vs);
break;
}
break;
default:
printf("Msg: %d\n", data[0]);
vnc_client_error(vs);
break;
}
vnc_read_when(vs, protocol_client_msg, 1);
return 0;
}
static int protocol_client_init(VncState *vs, uint8_t *data, size_t len)
{
char buf[1024];
int size;
vs->client_width = ds_get_width(vs->ds);
vs->client_height = ds_get_height(vs->ds);
vnc_write_u16(vs, vs->client_width);
vnc_write_u16(vs, vs->client_height);
pixel_format_message(vs);
if (qemu_name)
size = snprintf(buf, sizeof(buf), "QEMU (%s)", qemu_name);
else
size = snprintf(buf, sizeof(buf), "QEMU");
vnc_write_u32(vs, size);
vnc_write(vs, buf, size);
vnc_flush(vs);
vnc_client_cache_auth(vs);
vnc_qmp_event(vs, QEVENT_VNC_INITIALIZED);
vnc_read_when(vs, protocol_client_msg, 1);
return 0;
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
void start_client_init(VncState *vs)
{
vnc_read_when(vs, protocol_client_init, 1);
}
static void make_challenge(VncState *vs)
{
int i;
srand(time(NULL)+getpid()+getpid()*987654+rand());
for (i = 0 ; i < sizeof(vs->challenge) ; i++)
vs->challenge[i] = (int) (256.0*rand()/(RAND_MAX+1.0));
}
static int protocol_client_auth_vnc(VncState *vs, uint8_t *data, size_t len)
{
unsigned char response[VNC_AUTH_CHALLENGE_SIZE];
int i, j, pwlen;
unsigned char key[8];
if (!vs->vd->password || !vs->vd->password[0]) {
VNC_DEBUG("No password configured on server");
vnc_write_u32(vs, 1); /* Reject auth */
if (vs->minor >= 8) {
static const char err[] = "Authentication failed";
vnc_write_u32(vs, sizeof(err));
vnc_write(vs, err, sizeof(err));
}
vnc_flush(vs);
vnc_client_error(vs);
return 0;
}
memcpy(response, vs->challenge, VNC_AUTH_CHALLENGE_SIZE);
/* Calculate the expected challenge response */
pwlen = strlen(vs->vd->password);
for (i=0; i<sizeof(key); i++)
key[i] = i<pwlen ? vs->vd->password[i] : 0;
deskey(key, EN0);
for (j = 0; j < VNC_AUTH_CHALLENGE_SIZE; j += 8)
des(response+j, response+j);
/* Compare expected vs actual challenge response */
if (memcmp(response, data, VNC_AUTH_CHALLENGE_SIZE) != 0) {
VNC_DEBUG("Client challenge reponse did not match\n");
vnc_write_u32(vs, 1); /* Reject auth */
if (vs->minor >= 8) {
static const char err[] = "Authentication failed";
vnc_write_u32(vs, sizeof(err));
vnc_write(vs, err, sizeof(err));
}
vnc_flush(vs);
vnc_client_error(vs);
} else {
VNC_DEBUG("Accepting VNC challenge response\n");
vnc_write_u32(vs, 0); /* Accept auth */
vnc_flush(vs);
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
start_client_init(vs);
}
return 0;
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
void start_auth_vnc(VncState *vs)
{
make_challenge(vs);
/* Send client a 'random' challenge */
vnc_write(vs, vs->challenge, sizeof(vs->challenge));
vnc_flush(vs);
vnc_read_when(vs, protocol_client_auth_vnc, sizeof(vs->challenge));
}
static int protocol_client_auth(VncState *vs, uint8_t *data, size_t len)
{
/* We only advertise 1 auth scheme at a time, so client
* must pick the one we sent. Verify this */
if (data[0] != vs->vd->auth) { /* Reject auth */
VNC_DEBUG("Reject auth %d because it didn't match advertized\n", (int)data[0]);
vnc_write_u32(vs, 1);
if (vs->minor >= 8) {
static const char err[] = "Authentication failed";
vnc_write_u32(vs, sizeof(err));
vnc_write(vs, err, sizeof(err));
}
vnc_client_error(vs);
} else { /* Accept requested auth */
VNC_DEBUG("Client requested auth %d\n", (int)data[0]);
switch (vs->vd->auth) {
case VNC_AUTH_NONE:
VNC_DEBUG("Accept auth none\n");
if (vs->minor >= 8) {
vnc_write_u32(vs, 0); /* Accept auth completion */
vnc_flush(vs);
}
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
start_client_init(vs);
break;
case VNC_AUTH_VNC:
VNC_DEBUG("Start VNC auth\n");
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
start_auth_vnc(vs);
break;
#ifdef CONFIG_VNC_TLS
case VNC_AUTH_VENCRYPT:
VNC_DEBUG("Accept VeNCrypt auth\n");;
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
start_auth_vencrypt(vs);
break;
#endif /* CONFIG_VNC_TLS */
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
#ifdef CONFIG_VNC_SASL
case VNC_AUTH_SASL:
VNC_DEBUG("Accept SASL auth\n");
start_auth_sasl(vs);
break;
#endif /* CONFIG_VNC_SASL */
default: /* Should not be possible, but just in case */
VNC_DEBUG("Reject auth %d server code bug\n", vs->vd->auth);
vnc_write_u8(vs, 1);
if (vs->minor >= 8) {
static const char err[] = "Authentication failed";
vnc_write_u32(vs, sizeof(err));
vnc_write(vs, err, sizeof(err));
}
vnc_client_error(vs);
}
}
return 0;
}
static int protocol_version(VncState *vs, uint8_t *version, size_t len)
{
char local[13];
memcpy(local, version, 12);
local[12] = 0;
if (sscanf(local, "RFB %03d.%03d\n", &vs->major, &vs->minor) != 2) {
VNC_DEBUG("Malformed protocol version %s\n", local);
vnc_client_error(vs);
return 0;
}
VNC_DEBUG("Client request protocol version %d.%d\n", vs->major, vs->minor);
if (vs->major != 3 ||
(vs->minor != 3 &&
vs->minor != 4 &&
vs->minor != 5 &&
vs->minor != 7 &&
vs->minor != 8)) {
VNC_DEBUG("Unsupported client version\n");
vnc_write_u32(vs, VNC_AUTH_INVALID);
vnc_flush(vs);
vnc_client_error(vs);
return 0;
}
/* Some broken clients report v3.4 or v3.5, which spec requires to be treated
* as equivalent to v3.3 by servers
*/
if (vs->minor == 4 || vs->minor == 5)
vs->minor = 3;
if (vs->minor == 3) {
if (vs->vd->auth == VNC_AUTH_NONE) {
VNC_DEBUG("Tell client auth none\n");
vnc_write_u32(vs, vs->vd->auth);
vnc_flush(vs);
start_client_init(vs);
} else if (vs->vd->auth == VNC_AUTH_VNC) {
VNC_DEBUG("Tell client VNC auth\n");
vnc_write_u32(vs, vs->vd->auth);
vnc_flush(vs);
start_auth_vnc(vs);
} else {
VNC_DEBUG("Unsupported auth %d for protocol 3.3\n", vs->vd->auth);
vnc_write_u32(vs, VNC_AUTH_INVALID);
vnc_flush(vs);
vnc_client_error(vs);
}
} else {
VNC_DEBUG("Telling client we support auth %d\n", vs->vd->auth);
vnc_write_u8(vs, 1); /* num auth */
vnc_write_u8(vs, vs->vd->auth);
vnc_read_when(vs, protocol_client_auth, 1);
vnc_flush(vs);
}
return 0;
}
static int vnc_refresh_server_surface(VncDisplay *vd)
{
int y;
uint8_t *guest_row;
uint8_t *server_row;
int cmp_bytes;
uint32_t width_mask[VNC_DIRTY_WORDS];
VncState *vs;
int has_dirty = 0;
/*
* Walk through the guest dirty map.
* Check and copy modified bits from guest to server surface.
* Update server dirty map.
*/
vnc_set_bits(width_mask, (ds_get_width(vd->ds) / 16), VNC_DIRTY_WORDS);
cmp_bytes = 16 * ds_get_bytes_per_pixel(vd->ds);
guest_row = vd->guest.ds->data;
server_row = vd->server->data;
for (y = 0; y < vd->guest.ds->height; y++) {
if (vnc_and_bits(vd->guest.dirty[y], width_mask, VNC_DIRTY_WORDS)) {
int x;
uint8_t *guest_ptr;
uint8_t *server_ptr;
guest_ptr = guest_row;
server_ptr = server_row;
for (x = 0; x < vd->guest.ds->width;
x += 16, guest_ptr += cmp_bytes, server_ptr += cmp_bytes) {
if (!vnc_get_bit(vd->guest.dirty[y], (x / 16)))
continue;
vnc_clear_bit(vd->guest.dirty[y], (x / 16));
if (memcmp(server_ptr, guest_ptr, cmp_bytes) == 0)
continue;
memcpy(server_ptr, guest_ptr, cmp_bytes);
QTAILQ_FOREACH(vs, &vd->clients, next) {
vnc_set_bit(vs->dirty[y], (x / 16));
}
has_dirty++;
}
}
guest_row += ds_get_linesize(vd->ds);
server_row += ds_get_linesize(vd->ds);
}
return has_dirty;
}
static void vnc_refresh(void *opaque)
{
VncDisplay *vd = opaque;
VncState *vs, *vn;
int has_dirty, rects = 0;
vga_hw_update();
if (vnc_trylock_display(vd)) {
vd->timer_interval = VNC_REFRESH_INTERVAL_BASE;
qemu_mod_timer(vd->timer, qemu_get_clock(rt_clock) +
vd->timer_interval);
return;
}
has_dirty = vnc_refresh_server_surface(vd);
vnc_unlock_display(vd);
QTAILQ_FOREACH_SAFE(vs, &vd->clients, next, vn) {
rects += vnc_update_client(vs, has_dirty);
/* vs might be free()ed here */
}
/* vd->timer could be NULL now if the last client disconnected,
* in this case don't update the timer */
if (vd->timer == NULL)
return;
if (has_dirty && rects) {
vd->timer_interval /= 2;
if (vd->timer_interval < VNC_REFRESH_INTERVAL_BASE)
vd->timer_interval = VNC_REFRESH_INTERVAL_BASE;
} else {
vd->timer_interval += VNC_REFRESH_INTERVAL_INC;
if (vd->timer_interval > VNC_REFRESH_INTERVAL_MAX)
vd->timer_interval = VNC_REFRESH_INTERVAL_MAX;
}
qemu_mod_timer(vd->timer, qemu_get_clock(rt_clock) + vd->timer_interval);
}
static void vnc_init_timer(VncDisplay *vd)
{
vd->timer_interval = VNC_REFRESH_INTERVAL_BASE;
if (vd->timer == NULL && !QTAILQ_EMPTY(&vd->clients)) {
vd->timer = qemu_new_timer(rt_clock, vnc_refresh, vd);
vnc_refresh(vd);
}
}
static void vnc_remove_timer(VncDisplay *vd)
{
if (vd->timer != NULL && QTAILQ_EMPTY(&vd->clients)) {
qemu_del_timer(vd->timer);
qemu_free_timer(vd->timer);
vd->timer = NULL;
}
}
static void vnc_connect(VncDisplay *vd, int csock)
{
VncState *vs = qemu_mallocz(sizeof(VncState));
vs->csock = csock;
VNC_DEBUG("New client on socket %d\n", csock);
DisplayState interface change (Stefano Stabellini) This patch changes the DisplayState interface adding support for multiple frontends at the same time (sdl and vnc) and implements most of the benefit of the shared_buf patch without the added complexity. Currently DisplayState is managed by sdl (or vnc) and sdl (or vnc) is also responsible for allocating the data and setting the depth. Vga.c (or another backend) will do any necessary conversion. The idea is to change it so that is vga.c (or another backend) together with console.c that fully manage the DisplayState interface allocating data and setting the depth (either 16 or 32 bit, if the guest uses a different resolution or is in text mode, vga.c (or another backend) is in charge of doing the conversion seamlessly). The other idea is that DisplayState supports *multiple* frontends like sdl and vnc; each of them can register some callbacks to be called when a display event occurs. The interesting changes are: - the new structures and related functions in console.h and console.c in particular the following functions are very helpful to manage a DisplaySurface: qemu_create_displaysurface qemu_resize_displaysurface qemu_create_displaysurface_from qemu_free_displaysurface - console_select and qemu_console_resize in console.c this two functions manage multiple consoles on a single host display - moving code around in hw/vga.c as for the shared_buf patch this is necessary to be able to handle a dynamic DisplaySurface bpp - changes to vga_draw_graphic in hw/vga.c this is the place where the DisplaySurface buffer is shared with the videoram, when possible; Compared to the last version the only changes are: - do not remove support to dpy_copy in cirrus_vga - change the name of the displaysurface handling functions Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6336 c046a42c-6fe2-441c-8c8c-71466251a162
2009-01-16 01:14:11 +03:00
dcl->idle = 0;
socket_set_nonblock(vs->csock);
qemu_set_fd_handler2(vs->csock, NULL, vnc_client_read, NULL, vs);
vnc_client_cache_addr(vs);
vnc_qmp_event(vs, QEVENT_VNC_CONNECTED);
vs->vd = vd;
vs->ds = vd->ds;
vs->last_x = -1;
vs->last_y = -1;
vs->as.freq = 44100;
vs->as.nchannels = 2;
vs->as.fmt = AUD_FMT_S16;
vs->as.endianness = 0;
#ifdef CONFIG_VNC_THREAD
qemu_mutex_init(&vs->output_mutex);
#endif
QTAILQ_INSERT_HEAD(&vd->clients, vs, next);
vga_hw_update();
vnc_write(vs, "RFB 003.008\n", 12);
vnc_flush(vs);
vnc_read_when(vs, protocol_version, 12);
reset_keys(vs);
if (vs->vd->lock_key_sync)
vs->led = qemu_add_led_event_handler(kbd_leds, vs);
vs->mouse_mode_notifier.notify = check_pointer_type_change;
qemu_add_mouse_mode_change_notifier(&vs->mouse_mode_notifier);
vnc_init_timer(vd);
/* vs might be free()ed here */
}
static void vnc_listen_read(void *opaque)
{
VncDisplay *vs = opaque;
struct sockaddr_in addr;
socklen_t addrlen = sizeof(addr);
/* Catch-up */
vga_hw_update();
int csock = qemu_accept(vs->lsock, (struct sockaddr *)&addr, &addrlen);
if (csock != -1) {
vnc_connect(vs, csock);
}
}
void vnc_display_init(DisplayState *ds)
{
VncDisplay *vs = qemu_mallocz(sizeof(*vs));
DisplayState interface change (Stefano Stabellini) This patch changes the DisplayState interface adding support for multiple frontends at the same time (sdl and vnc) and implements most of the benefit of the shared_buf patch without the added complexity. Currently DisplayState is managed by sdl (or vnc) and sdl (or vnc) is also responsible for allocating the data and setting the depth. Vga.c (or another backend) will do any necessary conversion. The idea is to change it so that is vga.c (or another backend) together with console.c that fully manage the DisplayState interface allocating data and setting the depth (either 16 or 32 bit, if the guest uses a different resolution or is in text mode, vga.c (or another backend) is in charge of doing the conversion seamlessly). The other idea is that DisplayState supports *multiple* frontends like sdl and vnc; each of them can register some callbacks to be called when a display event occurs. The interesting changes are: - the new structures and related functions in console.h and console.c in particular the following functions are very helpful to manage a DisplaySurface: qemu_create_displaysurface qemu_resize_displaysurface qemu_create_displaysurface_from qemu_free_displaysurface - console_select and qemu_console_resize in console.c this two functions manage multiple consoles on a single host display - moving code around in hw/vga.c as for the shared_buf patch this is necessary to be able to handle a dynamic DisplaySurface bpp - changes to vga_draw_graphic in hw/vga.c this is the place where the DisplaySurface buffer is shared with the videoram, when possible; Compared to the last version the only changes are: - do not remove support to dpy_copy in cirrus_vga - change the name of the displaysurface handling functions Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6336 c046a42c-6fe2-441c-8c8c-71466251a162
2009-01-16 01:14:11 +03:00
dcl = qemu_mallocz(sizeof(DisplayChangeListener));
ds->opaque = vs;
DisplayState interface change (Stefano Stabellini) This patch changes the DisplayState interface adding support for multiple frontends at the same time (sdl and vnc) and implements most of the benefit of the shared_buf patch without the added complexity. Currently DisplayState is managed by sdl (or vnc) and sdl (or vnc) is also responsible for allocating the data and setting the depth. Vga.c (or another backend) will do any necessary conversion. The idea is to change it so that is vga.c (or another backend) together with console.c that fully manage the DisplayState interface allocating data and setting the depth (either 16 or 32 bit, if the guest uses a different resolution or is in text mode, vga.c (or another backend) is in charge of doing the conversion seamlessly). The other idea is that DisplayState supports *multiple* frontends like sdl and vnc; each of them can register some callbacks to be called when a display event occurs. The interesting changes are: - the new structures and related functions in console.h and console.c in particular the following functions are very helpful to manage a DisplaySurface: qemu_create_displaysurface qemu_resize_displaysurface qemu_create_displaysurface_from qemu_free_displaysurface - console_select and qemu_console_resize in console.c this two functions manage multiple consoles on a single host display - moving code around in hw/vga.c as for the shared_buf patch this is necessary to be able to handle a dynamic DisplaySurface bpp - changes to vga_draw_graphic in hw/vga.c this is the place where the DisplaySurface buffer is shared with the videoram, when possible; Compared to the last version the only changes are: - do not remove support to dpy_copy in cirrus_vga - change the name of the displaysurface handling functions Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6336 c046a42c-6fe2-441c-8c8c-71466251a162
2009-01-16 01:14:11 +03:00
dcl->idle = 1;
vnc_display = vs;
vs->lsock = -1;
vs->ds = ds;
QTAILQ_INIT(&vs->clients);
if (keyboard_layout)
vs->kbd_layout = init_keyboard_layout(name2keysym, keyboard_layout);
else
vs->kbd_layout = init_keyboard_layout(name2keysym, "en-us");
if (!vs->kbd_layout)
exit(1);
#ifdef CONFIG_VNC_THREAD
qemu_mutex_init(&vs->mutex);
vnc_start_worker_thread();
#endif
dcl->dpy_copy = vnc_dpy_copy;
DisplayState interface change (Stefano Stabellini) This patch changes the DisplayState interface adding support for multiple frontends at the same time (sdl and vnc) and implements most of the benefit of the shared_buf patch without the added complexity. Currently DisplayState is managed by sdl (or vnc) and sdl (or vnc) is also responsible for allocating the data and setting the depth. Vga.c (or another backend) will do any necessary conversion. The idea is to change it so that is vga.c (or another backend) together with console.c that fully manage the DisplayState interface allocating data and setting the depth (either 16 or 32 bit, if the guest uses a different resolution or is in text mode, vga.c (or another backend) is in charge of doing the conversion seamlessly). The other idea is that DisplayState supports *multiple* frontends like sdl and vnc; each of them can register some callbacks to be called when a display event occurs. The interesting changes are: - the new structures and related functions in console.h and console.c in particular the following functions are very helpful to manage a DisplaySurface: qemu_create_displaysurface qemu_resize_displaysurface qemu_create_displaysurface_from qemu_free_displaysurface - console_select and qemu_console_resize in console.c this two functions manage multiple consoles on a single host display - moving code around in hw/vga.c as for the shared_buf patch this is necessary to be able to handle a dynamic DisplaySurface bpp - changes to vga_draw_graphic in hw/vga.c this is the place where the DisplaySurface buffer is shared with the videoram, when possible; Compared to the last version the only changes are: - do not remove support to dpy_copy in cirrus_vga - change the name of the displaysurface handling functions Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6336 c046a42c-6fe2-441c-8c8c-71466251a162
2009-01-16 01:14:11 +03:00
dcl->dpy_update = vnc_dpy_update;
dcl->dpy_resize = vnc_dpy_resize;
dcl->dpy_setdata = vnc_dpy_setdata;
register_displaychangelistener(ds, dcl);
ds->mouse_set = vnc_mouse_set;
ds->cursor_define = vnc_dpy_cursor_define;
}
void vnc_display_close(DisplayState *ds)
{
VncDisplay *vs = ds ? (VncDisplay *)ds->opaque : vnc_display;
if (!vs)
return;
if (vs->display) {
qemu_free(vs->display);
vs->display = NULL;
}
if (vs->lsock != -1) {
qemu_set_fd_handler2(vs->lsock, NULL, NULL, NULL, NULL);
close(vs->lsock);
vs->lsock = -1;
}
vs->auth = VNC_AUTH_INVALID;
#ifdef CONFIG_VNC_TLS
vs->subauth = VNC_AUTH_INVALID;
Move TLS auth into separate file ("Daniel P. Berrange") This patch refactors the existing TLS code to make the main VNC code more managable. The code moves to two new files - vnc-tls.c: generic helpers for TLS handshake & credential setup - vnc-auth-vencrypt.c: the actual VNC TLS authentication mechanism. The reason for this split is that there are other TLS based auth mechanisms which we may like to use in the future. These can all share the same vnc-tls.c routines. In addition this will facilitate anyone who may want to port the vnc-tls.c file to allow for choice of GNUTLS & NSS for impl. The TLS state is moved out of the VncState struct, and into a separate VncStateTLS struct, defined in vnc-tls.h. This is then referenced from the main VncState. End size of the struct is the same, but it keeps things a little more managable. The vnc.h file gains a bunch more function prototypes, for functions in vnc.c that were previously static, but now need to be accessed from the separate auth code files. The only TLS related code still in the main vl.c is the command line argument handling / setup, and the low level I/O routines calling gnutls_send/recv. Makefile | 11 b/vnc-auth-vencrypt.c | 167 ++++++++++++++ b/vnc-auth-vencrypt.h | 33 ++ b/vnc-tls.c | 414 +++++++++++++++++++++++++++++++++++ b/vnc-tls.h | 70 ++++++ vnc.c | 581 +++----------------------------------------------- vnc.h | 76 ++++-- 7 files changed, 780 insertions(+), 572 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6723 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:23 +03:00
vs->tls.x509verify = 0;
#endif
}
int vnc_display_password(DisplayState *ds, const char *password)
{
VncDisplay *vs = ds ? (VncDisplay *)ds->opaque : vnc_display;
if (!vs) {
return -1;
}
if (vs->password) {
qemu_free(vs->password);
vs->password = NULL;
}
if (password && password[0]) {
if (!(vs->password = qemu_strdup(password)))
return -1;
if (vs->auth == VNC_AUTH_NONE) {
vs->auth = VNC_AUTH_VNC;
}
} else {
vs->auth = VNC_AUTH_NONE;
}
return 0;
}
char *vnc_display_local_addr(DisplayState *ds)
{
VncDisplay *vs = ds ? (VncDisplay *)ds->opaque : vnc_display;
return vnc_socket_local_addr("%s:%s", vs->lsock);
}
int vnc_display_open(DisplayState *ds, const char *display)
{
VncDisplay *vs = ds ? (VncDisplay *)ds->opaque : vnc_display;
const char *options;
int password = 0;
int reverse = 0;
#ifdef CONFIG_VNC_TLS
int tls = 0, x509 = 0;
#endif
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
#ifdef CONFIG_VNC_SASL
int sasl = 0;
int saslErr;
#endif
#if defined(CONFIG_VNC_TLS) || defined(CONFIG_VNC_SASL)
Support ACLs for controlling VNC access ("Daniel P. Berrange") This patch introduces a generic internal API for access control lists to be used by network servers in QEMU. It adds support for checking these ACL in the VNC server, in two places. The first ACL is for the SASL authentication mechanism, checking the SASL username. This ACL is called 'vnc.username'. The second is for the TLS authentication mechanism, when x509 client certificates are turned on, checking against the Distinguished Name of the client. This ACL is called 'vnc.x509dname' The internal API provides for an ACL with the following characteristics - A unique name, eg vnc.username, and vnc.x509dname. - A default policy, allow or deny - An ordered series of match rules, with allow or deny policy If none of the match rules apply, then the default policy is used. There is a monitor API to manipulate the ACLs, which I'll describe via examples (qemu) acl show vnc.username policy: allow (qemu) acl policy vnc.username denya acl: policy set to 'deny' (qemu) acl allow vnc.username fred acl: added rule at position 1 (qemu) acl allow vnc.username bob acl: added rule at position 2 (qemu) acl allow vnc.username joe 1 acl: added rule at position 1 (qemu) acl show vnc.username policy: deny 0: allow fred 1: allow joe 2: allow bob (qemu) acl show vnc.x509dname policy: allow (qemu) acl policy vnc.x509dname deny acl: policy set to 'deny' (qemu) acl allow vnc.x509dname C=GB,O=ACME,L=London,CN=* acl: added rule at position 1 (qemu) acl allow vnc.x509dname C=GB,O=ACME,L=Boston,CN=bob acl: added rule at position 2 (qemu) acl show vnc.x509dname policy: deny 0: allow C=GB,O=ACME,L=London,CN=* 1: allow C=GB,O=ACME,L=Boston,CN=bob By default the VNC server will not use any ACLs, allowing access to the server if the user successfully authenticates. To enable use of ACLs to restrict user access, the ',acl' flag should be given when starting QEMU. The initial ACL activated will be a 'deny all' policy and should be customized using monitor commands. eg enable SASL auth and ACLs qemu .... -vnc localhost:1,sasl,acl The next patch will provide a way to load a pre-defined ACL when starting up Makefile | 6 + b/acl.c | 185 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ b/acl.h | 74 ++++++++++++++++++++++ configure | 18 +++++ monitor.c | 95 ++++++++++++++++++++++++++++ qemu-doc.texi | 49 ++++++++++++++ vnc-auth-sasl.c | 16 +++- vnc-auth-sasl.h | 7 ++ vnc-tls.c | 19 +++++ vnc-tls.h | 3 vnc.c | 21 ++++++ vnc.h | 3 12 files changed, 491 insertions(+), 5 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6726 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:37 +03:00
int acl = 0;
#endif
int lock_key_sync = 1;
if (!vnc_display)
return -1;
vnc_display_close(ds);
if (strcmp(display, "none") == 0)
return 0;
if (!(vs->display = strdup(display)))
return -1;
options = display;
while ((options = strchr(options, ','))) {
options++;
if (strncmp(options, "password", 8) == 0) {
password = 1; /* Require password auth */
} else if (strncmp(options, "reverse", 7) == 0) {
reverse = 1;
} else if (strncmp(options, "no-lock-key-sync", 9) == 0) {
lock_key_sync = 0;
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
#ifdef CONFIG_VNC_SASL
} else if (strncmp(options, "sasl", 4) == 0) {
sasl = 1; /* Require SASL auth */
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
#endif
#ifdef CONFIG_VNC_TLS
} else if (strncmp(options, "tls", 3) == 0) {
tls = 1; /* Require TLS */
} else if (strncmp(options, "x509", 4) == 0) {
char *start, *end;
x509 = 1; /* Require x509 certificates */
if (strncmp(options, "x509verify", 10) == 0)
vs->tls.x509verify = 1; /* ...and verify client certs */
/* Now check for 'x509=/some/path' postfix
* and use that to setup x509 certificate/key paths */
start = strchr(options, '=');
end = strchr(options, ',');
if (start && (!end || (start < end))) {
int len = end ? end-(start+1) : strlen(start+1);
char *path = qemu_strndup(start + 1, len);
VNC_DEBUG("Trying certificate path '%s'\n", path);
if (vnc_tls_set_x509_creds_dir(vs, path) < 0) {
fprintf(stderr, "Failed to find x509 certificates/keys in %s\n", path);
qemu_free(path);
qemu_free(vs->display);
vs->display = NULL;
return -1;
}
qemu_free(path);
} else {
fprintf(stderr, "No certificate path provided\n");
qemu_free(vs->display);
vs->display = NULL;
return -1;
}
#endif
#if defined(CONFIG_VNC_TLS) || defined(CONFIG_VNC_SASL)
} else if (strncmp(options, "acl", 3) == 0) {
acl = 1;
#endif
} else if (strncmp(options, "lossy", 5) == 0) {
vs->lossy = true;
}
}
Support ACLs for controlling VNC access ("Daniel P. Berrange") This patch introduces a generic internal API for access control lists to be used by network servers in QEMU. It adds support for checking these ACL in the VNC server, in two places. The first ACL is for the SASL authentication mechanism, checking the SASL username. This ACL is called 'vnc.username'. The second is for the TLS authentication mechanism, when x509 client certificates are turned on, checking against the Distinguished Name of the client. This ACL is called 'vnc.x509dname' The internal API provides for an ACL with the following characteristics - A unique name, eg vnc.username, and vnc.x509dname. - A default policy, allow or deny - An ordered series of match rules, with allow or deny policy If none of the match rules apply, then the default policy is used. There is a monitor API to manipulate the ACLs, which I'll describe via examples (qemu) acl show vnc.username policy: allow (qemu) acl policy vnc.username denya acl: policy set to 'deny' (qemu) acl allow vnc.username fred acl: added rule at position 1 (qemu) acl allow vnc.username bob acl: added rule at position 2 (qemu) acl allow vnc.username joe 1 acl: added rule at position 1 (qemu) acl show vnc.username policy: deny 0: allow fred 1: allow joe 2: allow bob (qemu) acl show vnc.x509dname policy: allow (qemu) acl policy vnc.x509dname deny acl: policy set to 'deny' (qemu) acl allow vnc.x509dname C=GB,O=ACME,L=London,CN=* acl: added rule at position 1 (qemu) acl allow vnc.x509dname C=GB,O=ACME,L=Boston,CN=bob acl: added rule at position 2 (qemu) acl show vnc.x509dname policy: deny 0: allow C=GB,O=ACME,L=London,CN=* 1: allow C=GB,O=ACME,L=Boston,CN=bob By default the VNC server will not use any ACLs, allowing access to the server if the user successfully authenticates. To enable use of ACLs to restrict user access, the ',acl' flag should be given when starting QEMU. The initial ACL activated will be a 'deny all' policy and should be customized using monitor commands. eg enable SASL auth and ACLs qemu .... -vnc localhost:1,sasl,acl The next patch will provide a way to load a pre-defined ACL when starting up Makefile | 6 + b/acl.c | 185 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ b/acl.h | 74 ++++++++++++++++++++++ configure | 18 +++++ monitor.c | 95 ++++++++++++++++++++++++++++ qemu-doc.texi | 49 ++++++++++++++ vnc-auth-sasl.c | 16 +++- vnc-auth-sasl.h | 7 ++ vnc-tls.c | 19 +++++ vnc-tls.h | 3 vnc.c | 21 ++++++ vnc.h | 3 12 files changed, 491 insertions(+), 5 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6726 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:37 +03:00
#ifdef CONFIG_VNC_TLS
if (acl && x509 && vs->tls.x509verify) {
if (!(vs->tls.acl = qemu_acl_init("vnc.x509dname"))) {
fprintf(stderr, "Failed to create x509 dname ACL\n");
exit(1);
}
Support ACLs for controlling VNC access ("Daniel P. Berrange") This patch introduces a generic internal API for access control lists to be used by network servers in QEMU. It adds support for checking these ACL in the VNC server, in two places. The first ACL is for the SASL authentication mechanism, checking the SASL username. This ACL is called 'vnc.username'. The second is for the TLS authentication mechanism, when x509 client certificates are turned on, checking against the Distinguished Name of the client. This ACL is called 'vnc.x509dname' The internal API provides for an ACL with the following characteristics - A unique name, eg vnc.username, and vnc.x509dname. - A default policy, allow or deny - An ordered series of match rules, with allow or deny policy If none of the match rules apply, then the default policy is used. There is a monitor API to manipulate the ACLs, which I'll describe via examples (qemu) acl show vnc.username policy: allow (qemu) acl policy vnc.username denya acl: policy set to 'deny' (qemu) acl allow vnc.username fred acl: added rule at position 1 (qemu) acl allow vnc.username bob acl: added rule at position 2 (qemu) acl allow vnc.username joe 1 acl: added rule at position 1 (qemu) acl show vnc.username policy: deny 0: allow fred 1: allow joe 2: allow bob (qemu) acl show vnc.x509dname policy: allow (qemu) acl policy vnc.x509dname deny acl: policy set to 'deny' (qemu) acl allow vnc.x509dname C=GB,O=ACME,L=London,CN=* acl: added rule at position 1 (qemu) acl allow vnc.x509dname C=GB,O=ACME,L=Boston,CN=bob acl: added rule at position 2 (qemu) acl show vnc.x509dname policy: deny 0: allow C=GB,O=ACME,L=London,CN=* 1: allow C=GB,O=ACME,L=Boston,CN=bob By default the VNC server will not use any ACLs, allowing access to the server if the user successfully authenticates. To enable use of ACLs to restrict user access, the ',acl' flag should be given when starting QEMU. The initial ACL activated will be a 'deny all' policy and should be customized using monitor commands. eg enable SASL auth and ACLs qemu .... -vnc localhost:1,sasl,acl The next patch will provide a way to load a pre-defined ACL when starting up Makefile | 6 + b/acl.c | 185 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ b/acl.h | 74 ++++++++++++++++++++++ configure | 18 +++++ monitor.c | 95 ++++++++++++++++++++++++++++ qemu-doc.texi | 49 ++++++++++++++ vnc-auth-sasl.c | 16 +++- vnc-auth-sasl.h | 7 ++ vnc-tls.c | 19 +++++ vnc-tls.h | 3 vnc.c | 21 ++++++ vnc.h | 3 12 files changed, 491 insertions(+), 5 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6726 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:37 +03:00
}
#endif
#ifdef CONFIG_VNC_SASL
if (acl && sasl) {
if (!(vs->sasl.acl = qemu_acl_init("vnc.username"))) {
fprintf(stderr, "Failed to create username ACL\n");
exit(1);
}
Support ACLs for controlling VNC access ("Daniel P. Berrange") This patch introduces a generic internal API for access control lists to be used by network servers in QEMU. It adds support for checking these ACL in the VNC server, in two places. The first ACL is for the SASL authentication mechanism, checking the SASL username. This ACL is called 'vnc.username'. The second is for the TLS authentication mechanism, when x509 client certificates are turned on, checking against the Distinguished Name of the client. This ACL is called 'vnc.x509dname' The internal API provides for an ACL with the following characteristics - A unique name, eg vnc.username, and vnc.x509dname. - A default policy, allow or deny - An ordered series of match rules, with allow or deny policy If none of the match rules apply, then the default policy is used. There is a monitor API to manipulate the ACLs, which I'll describe via examples (qemu) acl show vnc.username policy: allow (qemu) acl policy vnc.username denya acl: policy set to 'deny' (qemu) acl allow vnc.username fred acl: added rule at position 1 (qemu) acl allow vnc.username bob acl: added rule at position 2 (qemu) acl allow vnc.username joe 1 acl: added rule at position 1 (qemu) acl show vnc.username policy: deny 0: allow fred 1: allow joe 2: allow bob (qemu) acl show vnc.x509dname policy: allow (qemu) acl policy vnc.x509dname deny acl: policy set to 'deny' (qemu) acl allow vnc.x509dname C=GB,O=ACME,L=London,CN=* acl: added rule at position 1 (qemu) acl allow vnc.x509dname C=GB,O=ACME,L=Boston,CN=bob acl: added rule at position 2 (qemu) acl show vnc.x509dname policy: deny 0: allow C=GB,O=ACME,L=London,CN=* 1: allow C=GB,O=ACME,L=Boston,CN=bob By default the VNC server will not use any ACLs, allowing access to the server if the user successfully authenticates. To enable use of ACLs to restrict user access, the ',acl' flag should be given when starting QEMU. The initial ACL activated will be a 'deny all' policy and should be customized using monitor commands. eg enable SASL auth and ACLs qemu .... -vnc localhost:1,sasl,acl The next patch will provide a way to load a pre-defined ACL when starting up Makefile | 6 + b/acl.c | 185 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ b/acl.h | 74 ++++++++++++++++++++++ configure | 18 +++++ monitor.c | 95 ++++++++++++++++++++++++++++ qemu-doc.texi | 49 ++++++++++++++ vnc-auth-sasl.c | 16 +++- vnc-auth-sasl.h | 7 ++ vnc-tls.c | 19 +++++ vnc-tls.h | 3 vnc.c | 21 ++++++ vnc.h | 3 12 files changed, 491 insertions(+), 5 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6726 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:37 +03:00
}
#endif
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
/*
* Combinations we support here:
*
* - no-auth (clear text, no auth)
* - password (clear text, weak auth)
* - sasl (encrypt, good auth *IF* using Kerberos via GSSAPI)
* - tls (encrypt, weak anonymous creds, no auth)
* - tls + password (encrypt, weak anonymous creds, weak auth)
* - tls + sasl (encrypt, weak anonymous creds, good auth)
* - tls + x509 (encrypt, good x509 creds, no auth)
* - tls + x509 + password (encrypt, good x509 creds, weak auth)
* - tls + x509 + sasl (encrypt, good x509 creds, good auth)
*
* NB1. TLS is a stackable auth scheme.
* NB2. the x509 schemes have option to validate a client cert dname
*/
if (password) {
#ifdef CONFIG_VNC_TLS
if (tls) {
vs->auth = VNC_AUTH_VENCRYPT;
if (x509) {
VNC_DEBUG("Initializing VNC server with x509 password auth\n");
vs->subauth = VNC_AUTH_VENCRYPT_X509VNC;
} else {
VNC_DEBUG("Initializing VNC server with TLS password auth\n");
vs->subauth = VNC_AUTH_VENCRYPT_TLSVNC;
}
} else {
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
#endif /* CONFIG_VNC_TLS */
VNC_DEBUG("Initializing VNC server with password auth\n");
vs->auth = VNC_AUTH_VNC;
#ifdef CONFIG_VNC_TLS
vs->subauth = VNC_AUTH_INVALID;
}
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
#endif /* CONFIG_VNC_TLS */
#ifdef CONFIG_VNC_SASL
} else if (sasl) {
#ifdef CONFIG_VNC_TLS
if (tls) {
vs->auth = VNC_AUTH_VENCRYPT;
if (x509) {
VNC_DEBUG("Initializing VNC server with x509 SASL auth\n");
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
vs->subauth = VNC_AUTH_VENCRYPT_X509SASL;
} else {
VNC_DEBUG("Initializing VNC server with TLS SASL auth\n");
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
vs->subauth = VNC_AUTH_VENCRYPT_TLSSASL;
}
} else {
#endif /* CONFIG_VNC_TLS */
VNC_DEBUG("Initializing VNC server with SASL auth\n");
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
vs->auth = VNC_AUTH_SASL;
#ifdef CONFIG_VNC_TLS
vs->subauth = VNC_AUTH_INVALID;
}
#endif /* CONFIG_VNC_TLS */
#endif /* CONFIG_VNC_SASL */
} else {
#ifdef CONFIG_VNC_TLS
if (tls) {
vs->auth = VNC_AUTH_VENCRYPT;
if (x509) {
VNC_DEBUG("Initializing VNC server with x509 no auth\n");
vs->subauth = VNC_AUTH_VENCRYPT_X509NONE;
} else {
VNC_DEBUG("Initializing VNC server with TLS no auth\n");
vs->subauth = VNC_AUTH_VENCRYPT_TLSNONE;
}
} else {
#endif
VNC_DEBUG("Initializing VNC server with no auth\n");
vs->auth = VNC_AUTH_NONE;
#ifdef CONFIG_VNC_TLS
vs->subauth = VNC_AUTH_INVALID;
}
#endif
}
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
#ifdef CONFIG_VNC_SASL
if ((saslErr = sasl_server_init(NULL, "qemu")) != SASL_OK) {
fprintf(stderr, "Failed to initialize SASL auth %s",
sasl_errstring(saslErr, NULL, NULL));
free(vs->display);
vs->display = NULL;
return -1;
}
#endif
vs->lock_key_sync = lock_key_sync;
Add SASL authentication support ("Daniel P. Berrange") This patch adds the new SASL authentication protocol to the VNC server. It is enabled by setting the 'sasl' flag when launching VNC. SASL can optionally provide encryption via its SSF layer, if a suitable mechanism is configured (eg, GSSAPI/Kerberos, or Digest-MD5). If an SSF layer is not available, then it should be combined with the x509 VNC authentication protocol which provides encryption. eg, if using GSSAPI qemu -vnc localhost:1,sasl eg if using TLS/x509 for encryption qemu -vnc localhost:1,sasl,tls,x509 By default the Cyrus SASL library will look for its configuration in the file /etc/sasl2/qemu.conf. For non-root users, this can be overridden by setting the SASL_CONF_PATH environment variable, eg to make it look in $HOME/.sasl2. NB unprivileged users may not have access to the full range of SASL mechanisms, since some of them require some administrative privileges to configure. The patch includes an example SASL configuration file which illustrates config for GSSAPI and Digest-MD5, though it should be noted that the latter is not really considered secure any more. Most of the SASL authentication code is located in a separate source file, vnc-auth-sasl.c. The main vnc.c file only contains minimal integration glue, specifically parsing of command line flags / setup, and calls to start the SASL auth process, to do encoding/decoding for data. There are several possible stacks for reading & writing of data, depending on the combo of VNC authentication methods in use - Clear. read/write straight to socket - TLS. read/write via GNUTLS helpers - SASL. encode/decode via SASL SSF layer, then read/write to socket - SASL+TLS. encode/decode via SASL SSF layer, then read/write via GNUTLS Hence, the vnc_client_read & vnc_client_write methods have been refactored a little. vnc_client_read: main entry point for reading, calls either - vnc_client_read_plain reading, with no intermediate decoding - vnc_client_read_sasl reading, with SASL SSF decoding These two methods, then call vnc_client_read_buf(). This decides whether to write to the socket directly or write via GNUTLS. The situation is the same for writing data. More extensive comments have been added in the code / patch. The vnc_client_read_sasl and vnc_client_write_sasl method implementations live in the separate vnc-auth-sasl.c file. The state required for the SASL auth mechanism is kept in a separate VncStateSASL struct, defined in vnc-auth-sasl.h and included in the main VncState. The configure script probes for SASL and automatically enables it if found, unless --disable-vnc-sasl was given to override it. Makefile | 7 Makefile.target | 5 b/qemu.sasl | 34 ++ b/vnc-auth-sasl.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++++ b/vnc-auth-sasl.h | 67 +++++ configure | 34 ++ qemu-doc.texi | 97 ++++++++ vnc-auth-vencrypt.c | 12 vnc.c | 249 ++++++++++++++++++-- vnc.h | 31 ++ 10 files changed, 1129 insertions(+), 33 deletions(-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com> git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6724 c046a42c-6fe2-441c-8c8c-71466251a162
2009-03-06 23:27:28 +03:00
if (reverse) {
/* connect to viewer */
if (strncmp(display, "unix:", 5) == 0)
vs->lsock = unix_connect(display+5);
else
vs->lsock = inet_connect(display, SOCK_STREAM);
if (-1 == vs->lsock) {
free(vs->display);
vs->display = NULL;
return -1;
} else {
int csock = vs->lsock;
vs->lsock = -1;
vnc_connect(vs, csock);
}
return 0;
} else {
/* listen for connects */
char *dpy;
dpy = qemu_malloc(256);
if (strncmp(display, "unix:", 5) == 0) {
pstrcpy(dpy, 256, "unix:");
vs->lsock = unix_listen(display+5, dpy+5, 256-5);
} else {
vs->lsock = inet_listen(display, dpy, 256, SOCK_STREAM, 5900);
}
if (-1 == vs->lsock) {
free(dpy);
return -1;
} else {
free(vs->display);
vs->display = dpy;
}
}
return qemu_set_fd_handler2(vs->lsock, NULL, vnc_listen_read, NULL, vs);
}