qemu/ui/vnc.c

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

4316 lines
128 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 "qemu/osdep.h"
#include "vnc.h"
#include "vnc-jobs.h"
#include "trace.h"
trace: Do not include qom/cpu.h into generated trace.h docs/devel/tracing.txt explains "since many source files include trace.h, [the generated trace.h use] a minimum of types and other header files included to keep the namespace clean and compile times and dependencies down." Commit 4815185902 "trace: Add per-vCPU tracing states for events with the 'vcpu' property" made them all include qom/cpu.h via control-internal.h. qom/cpu.h in turn includes about thirty headers. Ouch. Per-vCPU tracing is currently not supported in sub-directories' trace-events. In other words, qom/cpu.h can only be used in trace-root.h, not in any trace.h. Split trace/control-vcpu.h off trace/control.h and trace/control-internal.h. Have the generated trace.h include trace/control.h (which no longer includes qom/cpu.h), and trace-root.h include trace/control-vcpu.h (which includes it). The resulting improvement is a bit disappointing: in my "build everything" tree, some 1100 out of 6600 objects (not counting tests and objects that don't depend on qemu/osdep.h) depend on a trace.h, and about 600 of them no longer depend on qom/cpu.h. But more than 1300 others depend on trace-root.h. More work is clearly needed. Left for another day. Cc: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-Id: <20190812052359.30071-8-armbru@redhat.com>
2019-08-12 08:23:37 +03:00
#include "hw/qdev-core.h"
#include "sysemu/sysemu.h"
#include "sysemu/runstate.h"
#include "qemu/error-report.h"
#include "qemu/main-loop.h"
#include "qemu/module.h"
#include "qemu/option.h"
#include "qemu/sockets.h"
#include "qemu/timer.h"
#include "authz/list.h"
#include "qemu/config-file.h"
#include "qapi/qapi-emit-events.h"
#include "qapi/qapi-events-ui.h"
#include "qapi/error.h"
#include "qapi/qapi-commands-ui.h"
#include "ui/input.h"
#include "crypto/hash.h"
#include "crypto/tlscreds.h"
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
#include "crypto/tlscredsanon.h"
#include "crypto/tlscredsx509.h"
#include "crypto/random.h"
#include "crypto/secret_common.h"
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
#include "qom/object_interfaces.h"
#include "qemu/cutils.h"
#include "qemu/help_option.h"
#include "io/dns-resolver.h"
#include "monitor/monitor.h"
#define VNC_REFRESH_INTERVAL_BASE GUI_REFRESH_INTERVAL_DEFAULT
#define VNC_REFRESH_INTERVAL_INC 50
#define VNC_REFRESH_INTERVAL_MAX GUI_REFRESH_INTERVAL_IDLE
static const struct timeval VNC_REFRESH_STATS = { 0, 500000 };
static const struct timeval VNC_REFRESH_LOSSY = { 2, 0 };
#include "vnc_keysym.h"
#include "crypto/cipher.h"
static QTAILQ_HEAD(, VncDisplay) vnc_displays =
QTAILQ_HEAD_INITIALIZER(vnc_displays);
static int vnc_cursor_define(VncState *vs);
ui: fix VNC client throttling when audio capture is active The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check must be disabled if audio capture is enabled, because when streaming audio the output buffer offset will rarely be zero due to queued audio data, and so this would starve framebuffer updates. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then enable audio capture, and simply never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle when audio capture is active too. To determine how to throttle incremental updates or audio data, we calculate a size threshold. Normally the threshold is the approximate number of bytes associated with a single complete framebuffer update. ie width * height * bytes per pixel. We'll send incremental updates until we hit this threshold, at which point we'll stop sending updates until data has been written to the wire, causing the output buffer offset to fall back below the threshold. If audio capture is enabled, we increase the size of the threshold to also allow for upto 1 seconds worth of audio data samples. ie nchannels * bytes per sample * frequency. This allows the output buffer to have a mixture of incremental framebuffer updates and audio data queued, but once the threshold is exceeded, audio data will be dropped and incremental updates will be throttled. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-10-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:24 +03:00
static void vnc_update_throttle_offset(VncState *vs);
static void vnc_set_share_mode(VncState *vs, VncShareMode mode)
{
#ifdef _VNC_DEBUG
static const char *mn[] = {
[0] = "undefined",
[VNC_SHARE_MODE_CONNECTING] = "connecting",
[VNC_SHARE_MODE_SHARED] = "shared",
[VNC_SHARE_MODE_EXCLUSIVE] = "exclusive",
[VNC_SHARE_MODE_DISCONNECTED] = "disconnected",
};
fprintf(stderr, "%s/%p: %s -> %s\n", __func__,
vs->ioc, mn[vs->share_mode], mn[mode]);
#endif
switch (vs->share_mode) {
case VNC_SHARE_MODE_CONNECTING:
vs->vd->num_connecting--;
break;
case VNC_SHARE_MODE_SHARED:
vs->vd->num_shared--;
break;
case VNC_SHARE_MODE_EXCLUSIVE:
vs->vd->num_exclusive--;
break;
default:
break;
}
vs->share_mode = mode;
switch (vs->share_mode) {
case VNC_SHARE_MODE_CONNECTING:
vs->vd->num_connecting++;
break;
case VNC_SHARE_MODE_SHARED:
vs->vd->num_shared++;
break;
case VNC_SHARE_MODE_EXCLUSIVE:
vs->vd->num_exclusive++;
break;
default:
break;
}
}
static void vnc_init_basic_info(SocketAddress *addr,
VncBasicInfo *info,
Error **errp)
{
switch (addr->type) {
case SOCKET_ADDRESS_TYPE_INET:
info->host = g_strdup(addr->u.inet.host);
info->service = g_strdup(addr->u.inet.port);
if (addr->u.inet.ipv6) {
info->family = NETWORK_ADDRESS_FAMILY_IPV6;
} else {
info->family = NETWORK_ADDRESS_FAMILY_IPV4;
}
break;
case SOCKET_ADDRESS_TYPE_UNIX:
info->host = g_strdup("");
info->service = g_strdup(addr->u.q_unix.path);
info->family = NETWORK_ADDRESS_FAMILY_UNIX;
break;
case SOCKET_ADDRESS_TYPE_VSOCK:
case SOCKET_ADDRESS_TYPE_FD:
error_setg(errp, "Unsupported socket address type %s",
SocketAddressType_str(addr->type));
break;
default:
abort();
}
return;
}
static void vnc_init_basic_info_from_server_addr(QIOChannelSocket *ioc,
VncBasicInfo *info,
Error **errp)
{
SocketAddress *addr = NULL;
vnc: don't crash getting server info if lsock is NULL When VNC is started with '-vnc none' there will be no listener socket present. When we try to populate the VncServerInfo we'll crash accessing a NULL 'lsock' field. #0 qio_channel_socket_get_local_address (ioc=0x0, errp=errp@entry=0x7ffd5b8aa0f0) at io/channel-socket.c:33 #1 0x00007f4b9a297d6f in vnc_init_basic_info_from_server_addr (errp=0x7ffd5b8aa0f0, info=0x7f4b9d425460, ioc=<optimized out>) at ui/vnc.c:146 #2 vnc_server_info_get (vd=0x7f4b9e858000) at ui/vnc.c:223 #3 0x00007f4b9a29d318 in vnc_qmp_event (vs=0x7f4b9ef82000, vs=0x7f4b9ef82000, event=QAPI_EVENT_VNC_CONNECTED) at ui/vnc.c:279 #4 vnc_connect (vd=vd@entry=0x7f4b9e858000, sioc=sioc@entry=0x7f4b9e8b3a20, skipauth=skipauth@entry=true, websocket=websocket @entry=false) at ui/vnc.c:2994 #5 0x00007f4b9a29e8c8 in vnc_display_add_client (id=<optimized out>, csock=<optimized out>, skipauth=<optimized out>) at ui/v nc.c:3825 #6 0x00007f4b9a18d8a1 in qmp_marshal_add_client (args=<optimized out>, ret=<optimized out>, errp=0x7ffd5b8aa230) at qmp-marsh al.c:123 #7 0x00007f4b9a0b53f5 in handle_qmp_command (parser=<optimized out>, tokens=<optimized out>) at /usr/src/debug/qemu-2.6.0/mon itor.c:3922 #8 0x00007f4b9a348580 in json_message_process_token (lexer=0x7f4b9c78dfe8, input=0x7f4b9c7350e0, type=JSON_RCURLY, x=111, y=5 9) at qobject/json-streamer.c:94 #9 0x00007f4b9a35cfeb in json_lexer_feed_char (lexer=lexer@entry=0x7f4b9c78dfe8, ch=125 '}', flush=flush@entry=false) at qobj ect/json-lexer.c:310 #10 0x00007f4b9a35d0ae in json_lexer_feed (lexer=0x7f4b9c78dfe8, buffer=<optimized out>, size=<optimized out>) at qobject/json -lexer.c:360 #11 0x00007f4b9a348679 in json_message_parser_feed (parser=<optimized out>, buffer=<optimized out>, size=<optimized out>) at q object/json-streamer.c:114 #12 0x00007f4b9a0b3a1b in monitor_qmp_read (opaque=<optimized out>, buf=<optimized out>, size=<optimized out>) at /usr/src/deb ug/qemu-2.6.0/monitor.c:3938 #13 0x00007f4b9a186751 in tcp_chr_read (chan=<optimized out>, cond=<optimized out>, opaque=0x7f4b9c7add40) at qemu-char.c:2895 #14 0x00007f4b92b5c79a in g_main_context_dispatch () from /lib64/libglib-2.0.so.0 #15 0x00007f4b9a2bb0c0 in glib_pollfds_poll () at main-loop.c:213 #16 os_host_main_loop_wait (timeout=<optimized out>) at main-loop.c:258 #17 main_loop_wait (nonblocking=<optimized out>) at main-loop.c:506 #18 0x00007f4b9a0835cf in main_loop () at vl.c:1934 #19 main (argc=<optimized out>, argv=<optimized out>, envp=<optimized out>) at vl.c:4667 Do an upfront check for a NULL lsock and report an error to the caller, which matches behaviour from before commit 04d2529da27db512dcbd5e99d0e26d333f16efcc Author: Daniel P. Berrange <berrange@redhat.com> Date: Fri Feb 27 16:20:57 2015 +0000 ui: convert VNC server to use QIOChannelSocket where getsockname() would be given a FD value -1 and thus report an error to the caller. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 1470134726-15697-2-git-send-email-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2016-08-02 13:45:24 +03:00
if (!ioc) {
error_setg(errp, "No listener socket available");
return;
}
addr = qio_channel_socket_get_local_address(ioc, errp);
if (!addr) {
return;
}
vnc_init_basic_info(addr, info, errp);
qapi_free_SocketAddress(addr);
}
static void vnc_init_basic_info_from_remote_addr(QIOChannelSocket *ioc,
VncBasicInfo *info,
Error **errp)
{
SocketAddress *addr = NULL;
addr = qio_channel_socket_get_remote_address(ioc, errp);
if (!addr) {
return;
}
vnc_init_basic_info(addr, info, errp);
qapi_free_SocketAddress(addr);
}
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:
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";
}
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 VncServerInfo *vnc_server_info_get(VncDisplay *vd)
{
VncServerInfo *info;
Error *err = NULL;
if (!vd->listener || !vd->listener->nsioc) {
return NULL;
}
vnc: fix crash when vnc_server_info_get has an error The vnc_server_info_get will allocate the VncServerInfo struct and then call vnc_init_basic_info_from_server_addr to populate the basic fields. If this returns an error though, the qapi_free_VncServerInfo call will then crash because the VncServerInfo struct instance was not properly NULL-initialized and thus contains random stack garbage. #0 0x00007f1987c8e6f5 in raise () at /lib64/libc.so.6 #1 0x00007f1987c902fa in abort () at /lib64/libc.so.6 #2 0x00007f1987ccf600 in __libc_message () at /lib64/libc.so.6 #3 0x00007f1987cd7d4a in _int_free () at /lib64/libc.so.6 #4 0x00007f1987cdb2ac in free () at /lib64/libc.so.6 #5 0x00007f198b654f6e in g_free () at /lib64/libglib-2.0.so.0 #6 0x0000559193cdcf54 in visit_type_str (v=v@entry= 0x5591972f14b0, name=name@entry=0x559193de1e29 "host", obj=obj@entry=0x5591961dbfa0, errp=errp@entry=0x7fffd7899d80) at qapi/qapi-visit-core.c:255 #7 0x0000559193cca8f3 in visit_type_VncBasicInfo_members (v=v@entry= 0x5591972f14b0, obj=obj@entry=0x5591961dbfa0, errp=errp@entry=0x7fffd7899dc0) at qapi-visit.c:12307 #8 0x0000559193ccb523 in visit_type_VncServerInfo_members (v=v@entry= 0x5591972f14b0, obj=0x5591961dbfa0, errp=errp@entry=0x7fffd7899e00) at qapi-visit.c:12632 #9 0x0000559193ccb60b in visit_type_VncServerInfo (v=v@entry= 0x5591972f14b0, name=name@entry=0x0, obj=obj@entry=0x7fffd7899e48, errp=errp@entry=0x0) at qapi-visit.c:12658 #10 0x0000559193cb53d8 in qapi_free_VncServerInfo (obj=<optimized out>) at qapi-types.c:3970 #11 0x0000559193c1e6ba in vnc_server_info_get (vd=0x7f1951498010) at ui/vnc.c:233 #12 0x0000559193c24275 in vnc_connect (vs=0x559197b2f200, vs=0x559197b2f200, event=QAPI_EVENT_VNC_CONNECTED) at ui/vnc.c:284 #13 0x0000559193c24275 in vnc_connect (vd=vd@entry=0x7f1951498010, sioc=sioc@entry=0x559196bf9c00, skipauth=skipauth@entry=tru e, websocket=websocket@entry=false) at ui/vnc.c:3039 #14 0x0000559193c25806 in vnc_display_add_client (id=<optimized out>, csock=<optimized out>, skipauth=<optimized out>) at ui/vnc.c:3877 #15 0x0000559193a90c28 in qmp_marshal_add_client (args=<optimized out>, ret=<optimized out>, errp=0x7fffd7899f90) at qmp-marshal.c:105 #16 0x000055919399c2b7 in handle_qmp_command (parser=<optimized out>, tokens=<optimized out>) at /home/berrange/src/virt/qemu/monitor.c:3971 #17 0x0000559193ce3307 in json_message_process_token (lexer=0x559194ab0838, input=0x559194a6d940, type=JSON_RCURLY, x=111, y=1 2) at qobject/json-streamer.c:105 #18 0x0000559193cfa90d in json_lexer_feed_char (lexer=lexer@entry=0x559194ab0838, ch=125 '}', flush=flush@entry=false) at qobject/json-lexer.c:319 #19 0x0000559193cfaa1e in json_lexer_feed (lexer=0x559194ab0838, buffer=<optimized out>, size=<optimized out>) at qobject/json-lexer.c:369 #20 0x0000559193ce33c9 in json_message_parser_feed (parser=<optimized out>, buffer=<optimized out>, size=<optimized out>) at qobject/json-streamer.c:124 #21 0x000055919399a85b in monitor_qmp_read (opaque=<optimized out>, buf=<optimized out>, size=<optimized out>) at /home/berrange/src/virt/qemu/monitor.c:3987 #22 0x0000559193a87d00 in tcp_chr_read (chan=<optimized out>, cond=<optimized out>, opaque=0x559194a7d900) at qemu-char.c:2895 #23 0x00007f198b64f703 in g_main_context_dispatch () at /lib64/libglib-2.0.so.0 #24 0x0000559193c484b3 in main_loop_wait () at main-loop.c:213 #25 0x0000559193c484b3 in main_loop_wait (timeout=<optimized out>) at main-loop.c:258 #26 0x0000559193c484b3 in main_loop_wait (nonblocking=<optimized out>) at main-loop.c:506 #27 0x0000559193964c55 in main () at vl.c:1908 #28 0x0000559193964c55 in main (argc=<optimized out>, argv=<optimized out>, envp=<optimized out>) at vl.c:4603 This was introduced in commit 98481bfcd661daa3c160cc87a297b0e60a307788 Author: Eric Blake <eblake@redhat.com> Date: Mon Oct 26 16:34:45 2015 -0600 vnc: Hoist allocation of VncBasicInfo to callers which added error reporting for vnc_init_basic_info_from_server_addr but didn't change the g_malloc calls to g_malloc0. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 1470134726-15697-3-git-send-email-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2016-08-02 13:45:25 +03:00
info = g_malloc0(sizeof(*info));
vnc_init_basic_info_from_server_addr(vd->listener->sioc[0],
qapi: Unbox base members Rather than storing a base class as a pointer to a box, just store the fields of that base class in the same order, so that a child struct can be directly cast to its parent. This gives less malloc overhead, less pointer dereferencing, and even less generated code. Compare to the earlier commit 1e6c1616a "qapi: Generate a nicer struct for flat unions" (although that patch had fewer places to change, as less of qemu was directly using qapi structs for flat unions). It also allows us to turn on automatic type-safe wrappers for upcasting to the base class of a struct. Changes to the generated code look like this in qapi-types.h: | struct SpiceChannel { |- SpiceBasicInfo *base; |+ /* Members inherited from SpiceBasicInfo: */ |+ char *host; |+ char *port; |+ NetworkAddressFamily family; |+ /* Own members: */ | int64_t connection_id; as well as additional upcast functions like qapi_SpiceChannel_base(). Meanwhile, changes to qapi-visit.c look like: | static void visit_type_SpiceChannel_fields(Visitor *v, SpiceChannel **obj, Error **errp) | { | Error *err = NULL; | |- visit_type_implicit_SpiceBasicInfo(v, &(*obj)->base, &err); |+ visit_type_SpiceBasicInfo_fields(v, (SpiceBasicInfo **)obj, &err); | if (err) { (the cast is necessary, since our upcast wrappers only deal with a single pointer, not pointer-to-pointer); plus the wholesale elimination of some now-unused visit_type_implicit_FOO() functions. Without boxing, the corner case of one empty struct having another empty struct as its base type now requires inserting a dummy member (previously, the 'Base *base' member sufficed). And now that we no longer consume a 'base' member in the generated C struct, we can delete the former negative struct-base-clash-base test. Signed-off-by: Eric Blake <eblake@redhat.com> Message-Id: <1445898903-12082-11-git-send-email-eblake@redhat.com> [Commit message tweaked slightly] Signed-off-by: Markus Armbruster <armbru@redhat.com>
2015-10-27 01:34:49 +03:00
qapi_VncServerInfo_base(info), &err);
info->auth = g_strdup(vnc_auth_name(vd));
if (err) {
qapi_free_VncServerInfo(info);
info = NULL;
error_free(err);
}
return info;
}
static void vnc_client_cache_auth(VncState *client)
{
if (!client->info) {
return;
}
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
if (client->tls) {
client->info->x509_dname =
qcrypto_tls_session_get_peer_name(client->tls);
}
#ifdef CONFIG_VNC_SASL
if (client->sasl.conn &&
client->sasl.username) {
client->info->sasl_username = g_strdup(client->sasl.username);
}
#endif
}
static void vnc_client_cache_addr(VncState *client)
{
Error *err = NULL;
client->info = g_malloc0(sizeof(*client->info));
vnc_init_basic_info_from_remote_addr(client->sioc,
qapi: Unbox base members Rather than storing a base class as a pointer to a box, just store the fields of that base class in the same order, so that a child struct can be directly cast to its parent. This gives less malloc overhead, less pointer dereferencing, and even less generated code. Compare to the earlier commit 1e6c1616a "qapi: Generate a nicer struct for flat unions" (although that patch had fewer places to change, as less of qemu was directly using qapi structs for flat unions). It also allows us to turn on automatic type-safe wrappers for upcasting to the base class of a struct. Changes to the generated code look like this in qapi-types.h: | struct SpiceChannel { |- SpiceBasicInfo *base; |+ /* Members inherited from SpiceBasicInfo: */ |+ char *host; |+ char *port; |+ NetworkAddressFamily family; |+ /* Own members: */ | int64_t connection_id; as well as additional upcast functions like qapi_SpiceChannel_base(). Meanwhile, changes to qapi-visit.c look like: | static void visit_type_SpiceChannel_fields(Visitor *v, SpiceChannel **obj, Error **errp) | { | Error *err = NULL; | |- visit_type_implicit_SpiceBasicInfo(v, &(*obj)->base, &err); |+ visit_type_SpiceBasicInfo_fields(v, (SpiceBasicInfo **)obj, &err); | if (err) { (the cast is necessary, since our upcast wrappers only deal with a single pointer, not pointer-to-pointer); plus the wholesale elimination of some now-unused visit_type_implicit_FOO() functions. Without boxing, the corner case of one empty struct having another empty struct as its base type now requires inserting a dummy member (previously, the 'Base *base' member sufficed). And now that we no longer consume a 'base' member in the generated C struct, we can delete the former negative struct-base-clash-base test. Signed-off-by: Eric Blake <eblake@redhat.com> Message-Id: <1445898903-12082-11-git-send-email-eblake@redhat.com> [Commit message tweaked slightly] Signed-off-by: Markus Armbruster <armbru@redhat.com>
2015-10-27 01:34:49 +03:00
qapi_VncClientInfo_base(client->info),
&err);
client->info->websocket = client->websocket;
if (err) {
qapi_free_VncClientInfo(client->info);
client->info = NULL;
error_free(err);
}
}
static void vnc_qmp_event(VncState *vs, QAPIEvent event)
{
VncServerInfo *si;
if (!vs->info) {
return;
}
si = vnc_server_info_get(vs->vd);
if (!si) {
return;
}
switch (event) {
case QAPI_EVENT_VNC_CONNECTED:
qapi_event_send_vnc_connected(si, qapi_VncClientInfo_base(vs->info));
break;
case QAPI_EVENT_VNC_INITIALIZED:
qapi_event_send_vnc_initialized(si, vs->info);
break;
case QAPI_EVENT_VNC_DISCONNECTED:
qapi_event_send_vnc_disconnected(si, vs->info);
break;
default:
break;
}
qapi_free_VncServerInfo(si);
}
static VncClientInfo *qmp_query_vnc_client(const VncState *client)
{
VncClientInfo *info;
Error *err = NULL;
info = g_malloc0(sizeof(*info));
vnc_init_basic_info_from_remote_addr(client->sioc,
qapi_VncClientInfo_base(info),
&err);
if (err) {
error_free(err);
qapi_free_VncClientInfo(info);
return NULL;
}
qapi: Unbox base members Rather than storing a base class as a pointer to a box, just store the fields of that base class in the same order, so that a child struct can be directly cast to its parent. This gives less malloc overhead, less pointer dereferencing, and even less generated code. Compare to the earlier commit 1e6c1616a "qapi: Generate a nicer struct for flat unions" (although that patch had fewer places to change, as less of qemu was directly using qapi structs for flat unions). It also allows us to turn on automatic type-safe wrappers for upcasting to the base class of a struct. Changes to the generated code look like this in qapi-types.h: | struct SpiceChannel { |- SpiceBasicInfo *base; |+ /* Members inherited from SpiceBasicInfo: */ |+ char *host; |+ char *port; |+ NetworkAddressFamily family; |+ /* Own members: */ | int64_t connection_id; as well as additional upcast functions like qapi_SpiceChannel_base(). Meanwhile, changes to qapi-visit.c look like: | static void visit_type_SpiceChannel_fields(Visitor *v, SpiceChannel **obj, Error **errp) | { | Error *err = NULL; | |- visit_type_implicit_SpiceBasicInfo(v, &(*obj)->base, &err); |+ visit_type_SpiceBasicInfo_fields(v, (SpiceBasicInfo **)obj, &err); | if (err) { (the cast is necessary, since our upcast wrappers only deal with a single pointer, not pointer-to-pointer); plus the wholesale elimination of some now-unused visit_type_implicit_FOO() functions. Without boxing, the corner case of one empty struct having another empty struct as its base type now requires inserting a dummy member (previously, the 'Base *base' member sufficed). And now that we no longer consume a 'base' member in the generated C struct, we can delete the former negative struct-base-clash-base test. Signed-off-by: Eric Blake <eblake@redhat.com> Message-Id: <1445898903-12082-11-git-send-email-eblake@redhat.com> [Commit message tweaked slightly] Signed-off-by: Markus Armbruster <armbru@redhat.com>
2015-10-27 01:34:49 +03:00
info->websocket = client->websocket;
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
if (client->tls) {
info->x509_dname = qcrypto_tls_session_get_peer_name(client->tls);
}
#ifdef CONFIG_VNC_SASL
if (client->sasl.conn && client->sasl.username) {
info->sasl_username = g_strdup(client->sasl.username);
}
#endif
return info;
}
static VncDisplay *vnc_display_find(const char *id)
{
VncDisplay *vd;
if (id == NULL) {
return QTAILQ_FIRST(&vnc_displays);
}
QTAILQ_FOREACH(vd, &vnc_displays, next) {
if (strcmp(id, vd->id) == 0) {
return vd;
}
}
return NULL;
}
static VncClientInfoList *qmp_query_client_list(VncDisplay *vd)
{
VncClientInfoList *prev = NULL;
VncState *client;
QTAILQ_FOREACH(client, &vd->clients, next) {
QAPI_LIST_PREPEND(prev, qmp_query_vnc_client(client));
}
return prev;
}
VncInfo *qmp_query_vnc(Error **errp)
{
VncInfo *info = g_malloc0(sizeof(*info));
VncDisplay *vd = vnc_display_find(NULL);
SocketAddress *addr = NULL;
if (vd == NULL || !vd->listener || !vd->listener->nsioc) {
info->enabled = false;
} else {
info->enabled = true;
/* for compatibility with the original command */
info->has_clients = true;
info->clients = qmp_query_client_list(vd);
addr = qio_channel_socket_get_local_address(vd->listener->sioc[0],
errp);
if (!addr) {
goto out_error;
}
switch (addr->type) {
case SOCKET_ADDRESS_TYPE_INET:
info->host = g_strdup(addr->u.inet.host);
info->service = g_strdup(addr->u.inet.port);
if (addr->u.inet.ipv6) {
info->family = NETWORK_ADDRESS_FAMILY_IPV6;
} else {
info->family = NETWORK_ADDRESS_FAMILY_IPV4;
}
break;
case SOCKET_ADDRESS_TYPE_UNIX:
info->host = g_strdup("");
info->service = g_strdup(addr->u.q_unix.path);
info->family = NETWORK_ADDRESS_FAMILY_UNIX;
break;
case SOCKET_ADDRESS_TYPE_VSOCK:
case SOCKET_ADDRESS_TYPE_FD:
error_setg(errp, "Unsupported socket address type %s",
SocketAddressType_str(addr->type));
goto out_error;
default:
abort();
}
info->has_family = true;
info->auth = g_strdup(vnc_auth_name(vd));
}
qapi_free_SocketAddress(addr);
return info;
out_error:
qapi_free_SocketAddress(addr);
qapi_free_VncInfo(info);
return NULL;
}
static void qmp_query_auth(int auth, int subauth,
VncPrimaryAuth *qmp_auth,
VncVencryptSubAuth *qmp_vencrypt,
bool *qmp_has_vencrypt);
static VncServerInfo2List *qmp_query_server_entry(QIOChannelSocket *ioc,
bool websocket,
int auth,
int subauth,
VncServerInfo2List *prev)
{
VncServerInfo2 *info;
Error *err = NULL;
SocketAddress *addr;
addr = qio_channel_socket_get_local_address(ioc, NULL);
if (!addr) {
return prev;
}
info = g_new0(VncServerInfo2, 1);
vnc_init_basic_info(addr, qapi_VncServerInfo2_base(info), &err);
qapi_free_SocketAddress(addr);
if (err) {
qapi_free_VncServerInfo2(info);
error_free(err);
return prev;
}
info->websocket = websocket;
qmp_query_auth(auth, subauth, &info->auth,
&info->vencrypt, &info->has_vencrypt);
QAPI_LIST_PREPEND(prev, info);
return prev;
}
static void qmp_query_auth(int auth, int subauth,
VncPrimaryAuth *qmp_auth,
VncVencryptSubAuth *qmp_vencrypt,
bool *qmp_has_vencrypt)
{
switch (auth) {
case VNC_AUTH_VNC:
*qmp_auth = VNC_PRIMARY_AUTH_VNC;
break;
case VNC_AUTH_RA2:
*qmp_auth = VNC_PRIMARY_AUTH_RA2;
break;
case VNC_AUTH_RA2NE:
*qmp_auth = VNC_PRIMARY_AUTH_RA2NE;
break;
case VNC_AUTH_TIGHT:
*qmp_auth = VNC_PRIMARY_AUTH_TIGHT;
break;
case VNC_AUTH_ULTRA:
*qmp_auth = VNC_PRIMARY_AUTH_ULTRA;
break;
case VNC_AUTH_TLS:
*qmp_auth = VNC_PRIMARY_AUTH_TLS;
break;
case VNC_AUTH_VENCRYPT:
*qmp_auth = VNC_PRIMARY_AUTH_VENCRYPT;
*qmp_has_vencrypt = true;
switch (subauth) {
case VNC_AUTH_VENCRYPT_PLAIN:
*qmp_vencrypt = VNC_VENCRYPT_SUB_AUTH_PLAIN;
break;
case VNC_AUTH_VENCRYPT_TLSNONE:
*qmp_vencrypt = VNC_VENCRYPT_SUB_AUTH_TLS_NONE;
break;
case VNC_AUTH_VENCRYPT_TLSVNC:
*qmp_vencrypt = VNC_VENCRYPT_SUB_AUTH_TLS_VNC;
break;
case VNC_AUTH_VENCRYPT_TLSPLAIN:
*qmp_vencrypt = VNC_VENCRYPT_SUB_AUTH_TLS_PLAIN;
break;
case VNC_AUTH_VENCRYPT_X509NONE:
*qmp_vencrypt = VNC_VENCRYPT_SUB_AUTH_X509_NONE;
break;
case VNC_AUTH_VENCRYPT_X509VNC:
*qmp_vencrypt = VNC_VENCRYPT_SUB_AUTH_X509_VNC;
break;
case VNC_AUTH_VENCRYPT_X509PLAIN:
*qmp_vencrypt = VNC_VENCRYPT_SUB_AUTH_X509_PLAIN;
break;
case VNC_AUTH_VENCRYPT_TLSSASL:
*qmp_vencrypt = VNC_VENCRYPT_SUB_AUTH_TLS_SASL;
break;
case VNC_AUTH_VENCRYPT_X509SASL:
*qmp_vencrypt = VNC_VENCRYPT_SUB_AUTH_X509_SASL;
break;
default:
*qmp_has_vencrypt = false;
break;
}
break;
case VNC_AUTH_SASL:
*qmp_auth = VNC_PRIMARY_AUTH_SASL;
break;
case VNC_AUTH_NONE:
default:
*qmp_auth = VNC_PRIMARY_AUTH_NONE;
break;
}
}
VncInfo2List *qmp_query_vnc_servers(Error **errp)
{
VncInfo2List *prev = NULL;
VncInfo2 *info;
VncDisplay *vd;
DeviceState *dev;
size_t i;
QTAILQ_FOREACH(vd, &vnc_displays, next) {
info = g_new0(VncInfo2, 1);
info->id = g_strdup(vd->id);
info->clients = qmp_query_client_list(vd);
qmp_query_auth(vd->auth, vd->subauth, &info->auth,
&info->vencrypt, &info->has_vencrypt);
if (vd->dcl.con) {
dev = DEVICE(object_property_get_link(OBJECT(vd->dcl.con),
"device", &error_abort));
info->display = g_strdup(dev->id);
}
for (i = 0; vd->listener != NULL && i < vd->listener->nsioc; i++) {
info->server = qmp_query_server_entry(
vd->listener->sioc[i], false, vd->auth, vd->subauth,
info->server);
}
for (i = 0; vd->wslistener != NULL && i < vd->wslistener->nsioc; i++) {
info->server = qmp_query_server_entry(
vd->wslistener->sioc[i], true, vd->ws_auth,
vd->ws_subauth, info->server);
}
QAPI_LIST_PREPEND(prev, info);
}
return prev;
}
bool vnc_display_reload_certs(const char *id, Error **errp)
{
VncDisplay *vd = vnc_display_find(id);
QCryptoTLSCredsClass *creds = NULL;
if (!vd) {
error_setg(errp, "Can not find vnc display");
return false;
}
if (!vd->tlscreds) {
error_setg(errp, "vnc tls is not enabled");
return false;
}
creds = QCRYPTO_TLS_CREDS_GET_CLASS(OBJECT(vd->tlscreds));
if (creds->reload == NULL) {
error_setg(errp, "%s doesn't support to reload TLS credential",
object_get_typename(OBJECT(vd->tlscreds)));
return false;
}
if (!creds->reload(vd->tlscreds, errp)) {
return false;
}
return true;
}
/* 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 void vnc_disconnect_start(VncState *vs);
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(DisplayChangeListener *dcl);
static int vnc_refresh_server_surface(VncDisplay *vd);
static int vnc_width(VncDisplay *vd)
{
return MIN(VNC_MAX_WIDTH, ROUND_UP(surface_width(vd->ds),
VNC_DIRTY_PIXELS_PER_BIT));
}
ui: honour the actual guest display dimensions without rounding A long time ago the VNC server code had some memory corruption fixes done in: commit bea60dd7679364493a0d7f5b54316c767cf894ef Author: Peter Lieven <pl@kamp.de> Date: Mon Jun 30 10:57:51 2014 +0200 ui/vnc: fix potential memory corruption issues One of the implications of the fix was that the VNC server would have a thin black bad down the right hand side if the guest desktop width was not a multiple of 16. In practice this was a non-issue since the VNC server was always honouring a guest specified resolution and guests essentially always pick from a small set of sane resolutions likely in real world hardware. We recently introduced support for the extended desktop resize extension and as a result the VNC client has ability to specify an arbitrary desktop size and the guest OS may well honour it exactly. As a result we no longer have any guarantee that the width will be a multiple of 16, and so when resizing the desktop we have a 93% chance of getting the black bar on the right hand size. The VNC server maintains three different desktop dimensions 1. The guest surface 2. The server surface 3. The client desktop The requirement for the width to be a multiple of 16 only applies to item 2, the server surface, for the purpose of doing dirty bitmap tracking. Normally we will set the client desktop size to always match the server surface size, but that's not a strict requirement. In order to cope with clients that don't support the desktop size encoding, we already allow for the client desktop to be a different size that the server surface. Thus we can trivially eliminate the black bar, but setting the client desktop size to be the un-rounded server surface size - the so called "true width". Signed-off-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20210311182957.486939-5-berrange@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2021-03-11 21:29:57 +03:00
static int vnc_true_width(VncDisplay *vd)
{
return MIN(VNC_MAX_WIDTH, surface_width(vd->ds));
}
static int vnc_height(VncDisplay *vd)
{
return MIN(VNC_MAX_HEIGHT, surface_height(vd->ds));
}
static void vnc_set_area_dirty(DECLARE_BITMAP(dirty[VNC_MAX_HEIGHT],
VNC_MAX_WIDTH / VNC_DIRTY_PIXELS_PER_BIT),
VncDisplay *vd,
int x, int y, int w, int h)
{
int width = vnc_width(vd);
int height = vnc_height(vd);
/* this is needed this to ensure we updated all affected
* blocks if x % VNC_DIRTY_PIXELS_PER_BIT != 0 */
w += (x % VNC_DIRTY_PIXELS_PER_BIT);
x -= (x % VNC_DIRTY_PIXELS_PER_BIT);
x = MIN(x, width);
y = MIN(y, height);
w = MIN(x + w, width) - x;
h = MIN(y + h, height);
for (; y < h; y++) {
bitmap_set(dirty[y], x / VNC_DIRTY_PIXELS_PER_BIT,
DIV_ROUND_UP(w, VNC_DIRTY_PIXELS_PER_BIT));
}
}
static void vnc_dpy_update(DisplayChangeListener *dcl,
int x, int y, int w, int h)
{
VncDisplay *vd = container_of(dcl, VncDisplay, dcl);
struct VncSurface *s = &vd->guest;
vnc_set_area_dirty(s->dirty, vd, x, y, w, h);
}
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);
}
static void vnc_desktop_resize_ext(VncState *vs, int reject_reason)
{
trace_vnc_msg_server_ext_desktop_resize(
vs, vs->ioc, vs->client_width, vs->client_height, reject_reason);
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,
reject_reason ? 1 : 0,
reject_reason,
vs->client_width, vs->client_height,
VNC_ENCODING_DESKTOP_RESIZE_EXT);
vnc_write_u8(vs, 1); /* number of screens */
vnc_write_u8(vs, 0); /* padding */
vnc_write_u8(vs, 0); /* padding */
vnc_write_u8(vs, 0); /* padding */
vnc_write_u32(vs, 0); /* screen id */
vnc_write_u16(vs, 0); /* screen x-pos */
vnc_write_u16(vs, 0); /* screen y-pos */
vnc_write_u16(vs, vs->client_width);
vnc_write_u16(vs, vs->client_height);
vnc_write_u32(vs, 0); /* screen flags */
vnc_unlock_output(vs);
vnc_flush(vs);
}
static void vnc_desktop_resize(VncState *vs)
{
if (vs->ioc == NULL || (!vnc_has_feature(vs, VNC_FEATURE_RESIZE) &&
!vnc_has_feature(vs, VNC_FEATURE_RESIZE_EXT))) {
return;
}
ui: honour the actual guest display dimensions without rounding A long time ago the VNC server code had some memory corruption fixes done in: commit bea60dd7679364493a0d7f5b54316c767cf894ef Author: Peter Lieven <pl@kamp.de> Date: Mon Jun 30 10:57:51 2014 +0200 ui/vnc: fix potential memory corruption issues One of the implications of the fix was that the VNC server would have a thin black bad down the right hand side if the guest desktop width was not a multiple of 16. In practice this was a non-issue since the VNC server was always honouring a guest specified resolution and guests essentially always pick from a small set of sane resolutions likely in real world hardware. We recently introduced support for the extended desktop resize extension and as a result the VNC client has ability to specify an arbitrary desktop size and the guest OS may well honour it exactly. As a result we no longer have any guarantee that the width will be a multiple of 16, and so when resizing the desktop we have a 93% chance of getting the black bar on the right hand size. The VNC server maintains three different desktop dimensions 1. The guest surface 2. The server surface 3. The client desktop The requirement for the width to be a multiple of 16 only applies to item 2, the server surface, for the purpose of doing dirty bitmap tracking. Normally we will set the client desktop size to always match the server surface size, but that's not a strict requirement. In order to cope with clients that don't support the desktop size encoding, we already allow for the client desktop to be a different size that the server surface. Thus we can trivially eliminate the black bar, but setting the client desktop size to be the un-rounded server surface size - the so called "true width". Signed-off-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20210311182957.486939-5-berrange@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2021-03-11 21:29:57 +03:00
if (vs->client_width == vs->vd->true_width &&
vs->client_height == pixman_image_get_height(vs->vd->server)) {
return;
}
ui: avoid sign extension using client width/height Pixman returns a signed int for the image width/height, but the VNC protocol only permits a unsigned int16. Effective framebuffer size is determined by the guest, limited by the video RAM size, so the dimensions are unlikely to exceed the range of an unsigned int16, but this is not currently validated. With the current use of 'int' for client width/height, the calculation of offsets in vnc_update_throttle_offset() suffers from integer size promotion and sign extension, causing coverity warnings *** CID 1385147: Integer handling issues (SIGN_EXTENSION) /ui/vnc.c: 979 in vnc_update_throttle_offset() 973 * than that the client would already suffering awful audio 974 * glitches, so dropping samples is no worse really). 975 */ 976 static void vnc_update_throttle_offset(VncState *vs) 977 { 978 size_t offset = >>> CID 1385147: Integer handling issues (SIGN_EXTENSION) >>> Suspicious implicit sign extension: "vs->client_pf.bytes_per_pixel" with type "unsigned char" (8 bits, unsigned) is promoted in "vs->client_width * vs->client_height * vs->client_pf.bytes_per_pixel" to type "int" (32 bits, signed), then sign-extended to type "unsigned long" (64 bits, unsigned). If "vs->client_width * vs->client_height * vs->client_pf.bytes_per_pixel" is greater than 0x7FFFFFFF, the upper bits of the result will all be 1. 979 vs->client_width * vs->client_height * vs->client_pf.bytes_per_pixel; Change client_width / client_height to be a size_t to avoid sign extension and integer promotion. Then validate that dimensions are in range wrt the RFB protocol u16 limits. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 20180118155254.17053-1-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2018-01-18 18:52:54 +03:00
ui: honour the actual guest display dimensions without rounding A long time ago the VNC server code had some memory corruption fixes done in: commit bea60dd7679364493a0d7f5b54316c767cf894ef Author: Peter Lieven <pl@kamp.de> Date: Mon Jun 30 10:57:51 2014 +0200 ui/vnc: fix potential memory corruption issues One of the implications of the fix was that the VNC server would have a thin black bad down the right hand side if the guest desktop width was not a multiple of 16. In practice this was a non-issue since the VNC server was always honouring a guest specified resolution and guests essentially always pick from a small set of sane resolutions likely in real world hardware. We recently introduced support for the extended desktop resize extension and as a result the VNC client has ability to specify an arbitrary desktop size and the guest OS may well honour it exactly. As a result we no longer have any guarantee that the width will be a multiple of 16, and so when resizing the desktop we have a 93% chance of getting the black bar on the right hand size. The VNC server maintains three different desktop dimensions 1. The guest surface 2. The server surface 3. The client desktop The requirement for the width to be a multiple of 16 only applies to item 2, the server surface, for the purpose of doing dirty bitmap tracking. Normally we will set the client desktop size to always match the server surface size, but that's not a strict requirement. In order to cope with clients that don't support the desktop size encoding, we already allow for the client desktop to be a different size that the server surface. Thus we can trivially eliminate the black bar, but setting the client desktop size to be the un-rounded server surface size - the so called "true width". Signed-off-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20210311182957.486939-5-berrange@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2021-03-11 21:29:57 +03:00
assert(vs->vd->true_width < 65536 &&
vs->vd->true_width >= 0);
ui: avoid sign extension using client width/height Pixman returns a signed int for the image width/height, but the VNC protocol only permits a unsigned int16. Effective framebuffer size is determined by the guest, limited by the video RAM size, so the dimensions are unlikely to exceed the range of an unsigned int16, but this is not currently validated. With the current use of 'int' for client width/height, the calculation of offsets in vnc_update_throttle_offset() suffers from integer size promotion and sign extension, causing coverity warnings *** CID 1385147: Integer handling issues (SIGN_EXTENSION) /ui/vnc.c: 979 in vnc_update_throttle_offset() 973 * than that the client would already suffering awful audio 974 * glitches, so dropping samples is no worse really). 975 */ 976 static void vnc_update_throttle_offset(VncState *vs) 977 { 978 size_t offset = >>> CID 1385147: Integer handling issues (SIGN_EXTENSION) >>> Suspicious implicit sign extension: "vs->client_pf.bytes_per_pixel" with type "unsigned char" (8 bits, unsigned) is promoted in "vs->client_width * vs->client_height * vs->client_pf.bytes_per_pixel" to type "int" (32 bits, signed), then sign-extended to type "unsigned long" (64 bits, unsigned). If "vs->client_width * vs->client_height * vs->client_pf.bytes_per_pixel" is greater than 0x7FFFFFFF, the upper bits of the result will all be 1. 979 vs->client_width * vs->client_height * vs->client_pf.bytes_per_pixel; Change client_width / client_height to be a size_t to avoid sign extension and integer promotion. Then validate that dimensions are in range wrt the RFB protocol u16 limits. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 20180118155254.17053-1-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2018-01-18 18:52:54 +03:00
assert(pixman_image_get_height(vs->vd->server) < 65536 &&
pixman_image_get_height(vs->vd->server) >= 0);
ui: honour the actual guest display dimensions without rounding A long time ago the VNC server code had some memory corruption fixes done in: commit bea60dd7679364493a0d7f5b54316c767cf894ef Author: Peter Lieven <pl@kamp.de> Date: Mon Jun 30 10:57:51 2014 +0200 ui/vnc: fix potential memory corruption issues One of the implications of the fix was that the VNC server would have a thin black bad down the right hand side if the guest desktop width was not a multiple of 16. In practice this was a non-issue since the VNC server was always honouring a guest specified resolution and guests essentially always pick from a small set of sane resolutions likely in real world hardware. We recently introduced support for the extended desktop resize extension and as a result the VNC client has ability to specify an arbitrary desktop size and the guest OS may well honour it exactly. As a result we no longer have any guarantee that the width will be a multiple of 16, and so when resizing the desktop we have a 93% chance of getting the black bar on the right hand size. The VNC server maintains three different desktop dimensions 1. The guest surface 2. The server surface 3. The client desktop The requirement for the width to be a multiple of 16 only applies to item 2, the server surface, for the purpose of doing dirty bitmap tracking. Normally we will set the client desktop size to always match the server surface size, but that's not a strict requirement. In order to cope with clients that don't support the desktop size encoding, we already allow for the client desktop to be a different size that the server surface. Thus we can trivially eliminate the black bar, but setting the client desktop size to be the un-rounded server surface size - the so called "true width". Signed-off-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20210311182957.486939-5-berrange@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2021-03-11 21:29:57 +03:00
vs->client_width = vs->vd->true_width;
vs->client_height = pixman_image_get_height(vs->vd->server);
if (vnc_has_feature(vs, VNC_FEATURE_RESIZE_EXT)) {
vnc_desktop_resize_ext(vs, 0);
return;
}
trace_vnc_msg_server_desktop_resize(
vs, vs->ioc, vs->client_width, vs->client_height);
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);
}
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);
if (vs->update == VNC_STATE_UPDATE_NONE &&
vs->job_update != VNC_STATE_UPDATE_NONE) {
/* job aborted before completion */
vs->update = vs->job_update;
vs->job_update = VNC_STATE_UPDATE_NONE;
}
vs->abort = false;
vnc_unlock_output(vs);
}
}
int vnc_server_fb_stride(VncDisplay *vd)
{
return pixman_image_get_stride(vd->server);
}
void *vnc_server_fb_ptr(VncDisplay *vd, int x, int y)
{
uint8_t *ptr;
ptr = (uint8_t *)pixman_image_get_data(vd->server);
ptr += y * vnc_server_fb_stride(vd);
ptr += x * VNC_SERVER_FB_BYTES;
return ptr;
}
static void vnc_update_server_surface(VncDisplay *vd)
{
int width, height;
qemu_pixman_image_unref(vd->server);
vd->server = NULL;
if (QTAILQ_EMPTY(&vd->clients)) {
return;
}
width = vnc_width(vd);
height = vnc_height(vd);
ui: honour the actual guest display dimensions without rounding A long time ago the VNC server code had some memory corruption fixes done in: commit bea60dd7679364493a0d7f5b54316c767cf894ef Author: Peter Lieven <pl@kamp.de> Date: Mon Jun 30 10:57:51 2014 +0200 ui/vnc: fix potential memory corruption issues One of the implications of the fix was that the VNC server would have a thin black bad down the right hand side if the guest desktop width was not a multiple of 16. In practice this was a non-issue since the VNC server was always honouring a guest specified resolution and guests essentially always pick from a small set of sane resolutions likely in real world hardware. We recently introduced support for the extended desktop resize extension and as a result the VNC client has ability to specify an arbitrary desktop size and the guest OS may well honour it exactly. As a result we no longer have any guarantee that the width will be a multiple of 16, and so when resizing the desktop we have a 93% chance of getting the black bar on the right hand size. The VNC server maintains three different desktop dimensions 1. The guest surface 2. The server surface 3. The client desktop The requirement for the width to be a multiple of 16 only applies to item 2, the server surface, for the purpose of doing dirty bitmap tracking. Normally we will set the client desktop size to always match the server surface size, but that's not a strict requirement. In order to cope with clients that don't support the desktop size encoding, we already allow for the client desktop to be a different size that the server surface. Thus we can trivially eliminate the black bar, but setting the client desktop size to be the un-rounded server surface size - the so called "true width". Signed-off-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20210311182957.486939-5-berrange@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2021-03-11 21:29:57 +03:00
vd->true_width = vnc_true_width(vd);
vd->server = pixman_image_create_bits(VNC_SERVER_FB_FORMAT,
width, height,
NULL, 0);
memset(vd->guest.dirty, 0x00, sizeof(vd->guest.dirty));
vnc_set_area_dirty(vd->guest.dirty, vd, 0, 0,
width, height);
}
static bool vnc_check_pageflip(DisplaySurface *s1,
DisplaySurface *s2)
{
return (s1 != NULL &&
s2 != NULL &&
surface_width(s1) == surface_width(s2) &&
surface_height(s1) == surface_height(s2) &&
surface_format(s1) == surface_format(s2));
}
static void vnc_dpy_switch(DisplayChangeListener *dcl,
DisplaySurface *surface)
{
VncDisplay *vd = container_of(dcl, VncDisplay, dcl);
bool pageflip = vnc_check_pageflip(vd->ds, surface);
VncState *vs;
vnc_abort_display_jobs(vd);
vd->ds = surface;
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 */
qemu_pixman_image_unref(vd->guest.fb);
vd->guest.fb = pixman_image_ref(surface->image);
vd->guest.format = surface_format(surface);
ui: honour the actual guest display dimensions without rounding A long time ago the VNC server code had some memory corruption fixes done in: commit bea60dd7679364493a0d7f5b54316c767cf894ef Author: Peter Lieven <pl@kamp.de> Date: Mon Jun 30 10:57:51 2014 +0200 ui/vnc: fix potential memory corruption issues One of the implications of the fix was that the VNC server would have a thin black bad down the right hand side if the guest desktop width was not a multiple of 16. In practice this was a non-issue since the VNC server was always honouring a guest specified resolution and guests essentially always pick from a small set of sane resolutions likely in real world hardware. We recently introduced support for the extended desktop resize extension and as a result the VNC client has ability to specify an arbitrary desktop size and the guest OS may well honour it exactly. As a result we no longer have any guarantee that the width will be a multiple of 16, and so when resizing the desktop we have a 93% chance of getting the black bar on the right hand size. The VNC server maintains three different desktop dimensions 1. The guest surface 2. The server surface 3. The client desktop The requirement for the width to be a multiple of 16 only applies to item 2, the server surface, for the purpose of doing dirty bitmap tracking. Normally we will set the client desktop size to always match the server surface size, but that's not a strict requirement. In order to cope with clients that don't support the desktop size encoding, we already allow for the client desktop to be a different size that the server surface. Thus we can trivially eliminate the black bar, but setting the client desktop size to be the un-rounded server surface size - the so called "true width". Signed-off-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20210311182957.486939-5-berrange@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2021-03-11 21:29:57 +03:00
if (pageflip) {
ui: honour the actual guest display dimensions without rounding A long time ago the VNC server code had some memory corruption fixes done in: commit bea60dd7679364493a0d7f5b54316c767cf894ef Author: Peter Lieven <pl@kamp.de> Date: Mon Jun 30 10:57:51 2014 +0200 ui/vnc: fix potential memory corruption issues One of the implications of the fix was that the VNC server would have a thin black bad down the right hand side if the guest desktop width was not a multiple of 16. In practice this was a non-issue since the VNC server was always honouring a guest specified resolution and guests essentially always pick from a small set of sane resolutions likely in real world hardware. We recently introduced support for the extended desktop resize extension and as a result the VNC client has ability to specify an arbitrary desktop size and the guest OS may well honour it exactly. As a result we no longer have any guarantee that the width will be a multiple of 16, and so when resizing the desktop we have a 93% chance of getting the black bar on the right hand size. The VNC server maintains three different desktop dimensions 1. The guest surface 2. The server surface 3. The client desktop The requirement for the width to be a multiple of 16 only applies to item 2, the server surface, for the purpose of doing dirty bitmap tracking. Normally we will set the client desktop size to always match the server surface size, but that's not a strict requirement. In order to cope with clients that don't support the desktop size encoding, we already allow for the client desktop to be a different size that the server surface. Thus we can trivially eliminate the black bar, but setting the client desktop size to be the un-rounded server surface size - the so called "true width". Signed-off-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20210311182957.486939-5-berrange@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2021-03-11 21:29:57 +03:00
trace_vnc_server_dpy_pageflip(vd,
surface_width(surface),
surface_height(surface),
surface_format(surface));
vnc_set_area_dirty(vd->guest.dirty, vd, 0, 0,
surface_width(surface),
surface_height(surface));
return;
}
ui: honour the actual guest display dimensions without rounding A long time ago the VNC server code had some memory corruption fixes done in: commit bea60dd7679364493a0d7f5b54316c767cf894ef Author: Peter Lieven <pl@kamp.de> Date: Mon Jun 30 10:57:51 2014 +0200 ui/vnc: fix potential memory corruption issues One of the implications of the fix was that the VNC server would have a thin black bad down the right hand side if the guest desktop width was not a multiple of 16. In practice this was a non-issue since the VNC server was always honouring a guest specified resolution and guests essentially always pick from a small set of sane resolutions likely in real world hardware. We recently introduced support for the extended desktop resize extension and as a result the VNC client has ability to specify an arbitrary desktop size and the guest OS may well honour it exactly. As a result we no longer have any guarantee that the width will be a multiple of 16, and so when resizing the desktop we have a 93% chance of getting the black bar on the right hand size. The VNC server maintains three different desktop dimensions 1. The guest surface 2. The server surface 3. The client desktop The requirement for the width to be a multiple of 16 only applies to item 2, the server surface, for the purpose of doing dirty bitmap tracking. Normally we will set the client desktop size to always match the server surface size, but that's not a strict requirement. In order to cope with clients that don't support the desktop size encoding, we already allow for the client desktop to be a different size that the server surface. Thus we can trivially eliminate the black bar, but setting the client desktop size to be the un-rounded server surface size - the so called "true width". Signed-off-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20210311182957.486939-5-berrange@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2021-03-11 21:29:57 +03:00
trace_vnc_server_dpy_recreate(vd,
surface_width(surface),
surface_height(surface),
surface_format(surface));
/* server surface */
vnc_update_server_surface(vd);
QTAILQ_FOREACH(vs, &vd->clients, next) {
vnc_colordepth(vs);
vnc_desktop_resize(vs);
vnc_cursor_define(vs);
memset(vs->dirty, 0x00, sizeof(vs->dirty));
vnc_set_area_dirty(vs->dirty, vd, 0, 0,
vnc_width(vd),
vnc_height(vd));
ui: fix VNC client throttling when audio capture is active The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check must be disabled if audio capture is enabled, because when streaming audio the output buffer offset will rarely be zero due to queued audio data, and so this would starve framebuffer updates. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then enable audio capture, and simply never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle when audio capture is active too. To determine how to throttle incremental updates or audio data, we calculate a size threshold. Normally the threshold is the approximate number of bytes associated with a single complete framebuffer update. ie width * height * bytes per pixel. We'll send incremental updates until we hit this threshold, at which point we'll stop sending updates until data has been written to the wire, causing the output buffer offset to fall back below the threshold. If audio capture is enabled, we increase the size of the threshold to also allow for upto 1 seconds worth of audio data samples. ie nchannels * bytes per sample * frequency. This allows the output buffer to have a mixture of incremental framebuffer updates and audio data queued, but once the threshold is exceeded, audio data will be dropped and incremental updates will be throttled. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-10-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:24 +03:00
vnc_update_throttle_offset(vs);
}
}
/* fastest code */
static void vnc_write_pixels_copy(VncState *vs,
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;
#if VNC_SERVER_FB_FORMAT == PIXMAN_FORMAT(32, PIXMAN_TYPE_ARGB, 0, 8, 8, 8)
r = (((v & 0x00ff0000) >> 16) << vs->client_pf.rbits) >> 8;
g = (((v & 0x0000ff00) >> 8) << vs->client_pf.gbits) >> 8;
b = (((v & 0x000000ff) >> 0) << vs->client_pf.bbits) >> 8;
#else
# error need some bits here if you change VNC_SERVER_FB_FORMAT
#endif
v = (r << vs->client_pf.rshift) |
(g << vs->client_pf.gshift) |
(b << vs->client_pf.bshift);
switch (vs->client_pf.bytes_per_pixel) {
case 1:
buf[0] = v;
break;
case 2:
if (vs->client_be) {
buf[0] = v >> 8;
buf[1] = v;
} else {
buf[1] = v >> 8;
buf[0] = v;
}
break;
default:
case 4:
if (vs->client_be) {
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,
void *pixels1, int size)
{
uint8_t buf[4];
if (VNC_SERVER_FB_BYTES == 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->client_pf.bytes_per_pixel);
}
}
}
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 = vnc_server_fb_ptr(vd, x, y);
for (i = 0; i < h; i++) {
vs->write_pixels(vs, row, w * VNC_SERVER_FB_BYTES);
row += vnc_server_fb_stride(vd);
}
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;
case VNC_ENCODING_ZRLE:
n = vnc_zrle_send_framebuffer_update(vs, x, y, w, h);
break;
case VNC_ENCODING_ZYWRLE:
n = vnc_zywrle_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_mouse_set(DisplayChangeListener *dcl,
int x, int y, bool visible)
{
/* can we ask the client(s) to move the pointer ??? */
}
static int vnc_cursor_define(VncState *vs)
{
QEMUCursor *c = qemu_console_get_cursor(vs->vd->dcl.con);
int isize;
if (!c) {
return -1;
}
if (vnc_has_feature(vs, VNC_FEATURE_ALPHA_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_ALPHA_CURSOR);
vnc_write_s32(vs, VNC_ENCODING_RAW);
vnc_write(vs, c->data, c->width * c->height * 4);
vnc_unlock_output(vs);
return 0;
}
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->client_pf.bytes_per_pixel;
vnc_write_pixels_generic(vs, 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(DisplayChangeListener *dcl,
QEMUCursor *c)
{
VncDisplay *vd = container_of(dcl, VncDisplay, dcl);
VncState *vs;
g_free(vd->cursor_mask);
vd->cursor_msize = cursor_get_mono_bpl(c) * c->height;
vd->cursor_mask = g_malloc0(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(VncState *vs,
int y, int last_x, int x, int height)
{
int h;
for (h = 1; h < (height - y); h++) {
if (!test_bit(last_x, vs->dirty[y + h])) {
break;
}
bitmap_clear(vs->dirty[y + h], last_x, x - last_x);
}
return h;
}
ui: fix VNC client throttling when audio capture is active The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check must be disabled if audio capture is enabled, because when streaming audio the output buffer offset will rarely be zero due to queued audio data, and so this would starve framebuffer updates. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then enable audio capture, and simply never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle when audio capture is active too. To determine how to throttle incremental updates or audio data, we calculate a size threshold. Normally the threshold is the approximate number of bytes associated with a single complete framebuffer update. ie width * height * bytes per pixel. We'll send incremental updates until we hit this threshold, at which point we'll stop sending updates until data has been written to the wire, causing the output buffer offset to fall back below the threshold. If audio capture is enabled, we increase the size of the threshold to also allow for upto 1 seconds worth of audio data samples. ie nchannels * bytes per sample * frequency. This allows the output buffer to have a mixture of incremental framebuffer updates and audio data queued, but once the threshold is exceeded, audio data will be dropped and incremental updates will be throttled. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-10-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:24 +03:00
/*
* Figure out how much pending data we should allow in the output
* buffer before we throttle incremental display updates, and/or
* drop audio samples.
*
* We allow for equiv of 1 full display's worth of FB updates,
* and 1 second of audio samples. If audio backlog was larger
* than that the client would already suffering awful audio
* glitches, so dropping samples is no worse really).
*/
static void vnc_update_throttle_offset(VncState *vs)
{
size_t offset =
vs->client_width * vs->client_height * vs->client_pf.bytes_per_pixel;
if (vs->audio_cap) {
int bps;
switch (vs->as.fmt) {
default:
case AUDIO_FORMAT_U8:
case AUDIO_FORMAT_S8:
ui: fix VNC client throttling when audio capture is active The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check must be disabled if audio capture is enabled, because when streaming audio the output buffer offset will rarely be zero due to queued audio data, and so this would starve framebuffer updates. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then enable audio capture, and simply never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle when audio capture is active too. To determine how to throttle incremental updates or audio data, we calculate a size threshold. Normally the threshold is the approximate number of bytes associated with a single complete framebuffer update. ie width * height * bytes per pixel. We'll send incremental updates until we hit this threshold, at which point we'll stop sending updates until data has been written to the wire, causing the output buffer offset to fall back below the threshold. If audio capture is enabled, we increase the size of the threshold to also allow for upto 1 seconds worth of audio data samples. ie nchannels * bytes per sample * frequency. This allows the output buffer to have a mixture of incremental framebuffer updates and audio data queued, but once the threshold is exceeded, audio data will be dropped and incremental updates will be throttled. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-10-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:24 +03:00
bps = 1;
break;
case AUDIO_FORMAT_U16:
case AUDIO_FORMAT_S16:
ui: fix VNC client throttling when audio capture is active The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check must be disabled if audio capture is enabled, because when streaming audio the output buffer offset will rarely be zero due to queued audio data, and so this would starve framebuffer updates. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then enable audio capture, and simply never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle when audio capture is active too. To determine how to throttle incremental updates or audio data, we calculate a size threshold. Normally the threshold is the approximate number of bytes associated with a single complete framebuffer update. ie width * height * bytes per pixel. We'll send incremental updates until we hit this threshold, at which point we'll stop sending updates until data has been written to the wire, causing the output buffer offset to fall back below the threshold. If audio capture is enabled, we increase the size of the threshold to also allow for upto 1 seconds worth of audio data samples. ie nchannels * bytes per sample * frequency. This allows the output buffer to have a mixture of incremental framebuffer updates and audio data queued, but once the threshold is exceeded, audio data will be dropped and incremental updates will be throttled. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-10-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:24 +03:00
bps = 2;
break;
case AUDIO_FORMAT_U32:
case AUDIO_FORMAT_S32:
ui: fix VNC client throttling when audio capture is active The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check must be disabled if audio capture is enabled, because when streaming audio the output buffer offset will rarely be zero due to queued audio data, and so this would starve framebuffer updates. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then enable audio capture, and simply never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle when audio capture is active too. To determine how to throttle incremental updates or audio data, we calculate a size threshold. Normally the threshold is the approximate number of bytes associated with a single complete framebuffer update. ie width * height * bytes per pixel. We'll send incremental updates until we hit this threshold, at which point we'll stop sending updates until data has been written to the wire, causing the output buffer offset to fall back below the threshold. If audio capture is enabled, we increase the size of the threshold to also allow for upto 1 seconds worth of audio data samples. ie nchannels * bytes per sample * frequency. This allows the output buffer to have a mixture of incremental framebuffer updates and audio data queued, but once the threshold is exceeded, audio data will be dropped and incremental updates will be throttled. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-10-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:24 +03:00
bps = 4;
break;
}
offset += vs->as.freq * bps * vs->as.nchannels;
ui: fix VNC client throttling when audio capture is active The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check must be disabled if audio capture is enabled, because when streaming audio the output buffer offset will rarely be zero due to queued audio data, and so this would starve framebuffer updates. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then enable audio capture, and simply never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle when audio capture is active too. To determine how to throttle incremental updates or audio data, we calculate a size threshold. Normally the threshold is the approximate number of bytes associated with a single complete framebuffer update. ie width * height * bytes per pixel. We'll send incremental updates until we hit this threshold, at which point we'll stop sending updates until data has been written to the wire, causing the output buffer offset to fall back below the threshold. If audio capture is enabled, we increase the size of the threshold to also allow for upto 1 seconds worth of audio data samples. ie nchannels * bytes per sample * frequency. This allows the output buffer to have a mixture of incremental framebuffer updates and audio data queued, but once the threshold is exceeded, audio data will be dropped and incremental updates will be throttled. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-10-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:24 +03:00
}
/* Put a floor of 1MB on offset, so that if we have a large pending
* buffer and the display is resized to a small size & back again
* we don't suddenly apply a tiny send limit
*/
offset = MAX(offset, 1024 * 1024);
if (vs->throttle_output_offset != offset) {
trace_vnc_client_throttle_threshold(
vs, vs->ioc, vs->throttle_output_offset, offset, vs->client_width,
vs->client_height, vs->client_pf.bytes_per_pixel, vs->audio_cap);
}
ui: fix VNC client throttling when audio capture is active The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check must be disabled if audio capture is enabled, because when streaming audio the output buffer offset will rarely be zero due to queued audio data, and so this would starve framebuffer updates. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then enable audio capture, and simply never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle when audio capture is active too. To determine how to throttle incremental updates or audio data, we calculate a size threshold. Normally the threshold is the approximate number of bytes associated with a single complete framebuffer update. ie width * height * bytes per pixel. We'll send incremental updates until we hit this threshold, at which point we'll stop sending updates until data has been written to the wire, causing the output buffer offset to fall back below the threshold. If audio capture is enabled, we increase the size of the threshold to also allow for upto 1 seconds worth of audio data samples. ie nchannels * bytes per sample * frequency. This allows the output buffer to have a mixture of incremental framebuffer updates and audio data queued, but once the threshold is exceeded, audio data will be dropped and incremental updates will be throttled. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-10-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:24 +03:00
vs->throttle_output_offset = offset;
}
static bool vnc_should_update(VncState *vs)
{
switch (vs->update) {
case VNC_STATE_UPDATE_NONE:
break;
case VNC_STATE_UPDATE_INCREMENTAL:
ui: fix VNC client throttling when audio capture is active The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check must be disabled if audio capture is enabled, because when streaming audio the output buffer offset will rarely be zero due to queued audio data, and so this would starve framebuffer updates. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then enable audio capture, and simply never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle when audio capture is active too. To determine how to throttle incremental updates or audio data, we calculate a size threshold. Normally the threshold is the approximate number of bytes associated with a single complete framebuffer update. ie width * height * bytes per pixel. We'll send incremental updates until we hit this threshold, at which point we'll stop sending updates until data has been written to the wire, causing the output buffer offset to fall back below the threshold. If audio capture is enabled, we increase the size of the threshold to also allow for upto 1 seconds worth of audio data samples. ie nchannels * bytes per sample * frequency. This allows the output buffer to have a mixture of incremental framebuffer updates and audio data queued, but once the threshold is exceeded, audio data will be dropped and incremental updates will be throttled. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-10-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:24 +03:00
/* Only allow incremental updates if the pending send queue
ui: fix VNC client throttling when forced update is requested The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check is disabled if the client has requested a forced update, because we want to send these as soon as possible. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then repeatedly send full framebuffer update requests, but never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle full updates. When we get a forced update request, we keep track of exactly how much data we put on the output buffer. We will not process a subsequent forced update request until this data has been fully sent on the wire. We always allow one forced update request to be in flight, regardless of what data is queued for incremental updates or audio data. The slight complication is that we do not initially know how much data an update will send, as this is done in the background by the VNC job thread. So we must track the fact that the job thread has an update pending, and not process any further updates until this job is has been completed & put data on the output buffer. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-11-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:25 +03:00
* is less than the permitted threshold, and the job worker
* is completely idle.
*/
ui: fix VNC client throttling when forced update is requested The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check is disabled if the client has requested a forced update, because we want to send these as soon as possible. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then repeatedly send full framebuffer update requests, but never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle full updates. When we get a forced update request, we keep track of exactly how much data we put on the output buffer. We will not process a subsequent forced update request until this data has been fully sent on the wire. We always allow one forced update request to be in flight, regardless of what data is queued for incremental updates or audio data. The slight complication is that we do not initially know how much data an update will send, as this is done in the background by the VNC job thread. So we must track the fact that the job thread has an update pending, and not process any further updates until this job is has been completed & put data on the output buffer. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-11-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:25 +03:00
if (vs->output.offset < vs->throttle_output_offset &&
vs->job_update == VNC_STATE_UPDATE_NONE) {
return true;
}
trace_vnc_client_throttle_incremental(
vs, vs->ioc, vs->job_update, vs->output.offset);
break;
case VNC_STATE_UPDATE_FORCE:
ui: fix VNC client throttling when forced update is requested The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check is disabled if the client has requested a forced update, because we want to send these as soon as possible. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then repeatedly send full framebuffer update requests, but never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle full updates. When we get a forced update request, we keep track of exactly how much data we put on the output buffer. We will not process a subsequent forced update request until this data has been fully sent on the wire. We always allow one forced update request to be in flight, regardless of what data is queued for incremental updates or audio data. The slight complication is that we do not initially know how much data an update will send, as this is done in the background by the VNC job thread. So we must track the fact that the job thread has an update pending, and not process any further updates until this job is has been completed & put data on the output buffer. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-11-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:25 +03:00
/* Only allow forced updates if the pending send queue
* does not contain a previous forced update, and the
* job worker is completely idle.
*
* Note this means we'll queue a forced update, even if
* the output buffer size is otherwise over the throttle
* output limit.
*/
if (vs->force_update_offset == 0 &&
vs->job_update == VNC_STATE_UPDATE_NONE) {
return true;
}
trace_vnc_client_throttle_forced(
vs, vs->ioc, vs->job_update, vs->force_update_offset);
ui: fix VNC client throttling when forced update is requested The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check is disabled if the client has requested a forced update, because we want to send these as soon as possible. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then repeatedly send full framebuffer update requests, but never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle full updates. When we get a forced update request, we keep track of exactly how much data we put on the output buffer. We will not process a subsequent forced update request until this data has been fully sent on the wire. We always allow one forced update request to be in flight, regardless of what data is queued for incremental updates or audio data. The slight complication is that we do not initially know how much data an update will send, as this is done in the background by the VNC job thread. So we must track the fact that the job thread has an update pending, and not process any further updates until this job is has been completed & put data on the output buffer. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-11-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:25 +03:00
break;
}
return false;
}
static int vnc_update_client(VncState *vs, int has_dirty)
{
VncDisplay *vd = vs->vd;
VncJob *job;
int y;
int height, width;
int n = 0;
if (vs->disconnecting) {
vnc_disconnect_finish(vs);
return 0;
}
vs->has_dirty += has_dirty;
if (!vnc_should_update(vs)) {
return 0;
}
if (!vs->has_dirty && vs->update != VNC_STATE_UPDATE_FORCE) {
return 0;
}
/*
* 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);
height = pixman_image_get_height(vd->server);
width = pixman_image_get_width(vd->server);
y = 0;
for (;;) {
int x, h;
unsigned long x2;
unsigned long offset = find_next_bit((unsigned long *) &vs->dirty,
height * VNC_DIRTY_BPL(vs),
y * VNC_DIRTY_BPL(vs));
if (offset == height * VNC_DIRTY_BPL(vs)) {
/* no more dirty bits */
break;
}
y = offset / VNC_DIRTY_BPL(vs);
x = offset % VNC_DIRTY_BPL(vs);
x2 = find_next_zero_bit((unsigned long *) &vs->dirty[y],
VNC_DIRTY_BPL(vs), x);
bitmap_clear(vs->dirty[y], x, x2 - x);
h = find_and_clear_dirty_height(vs, y, x, x2, height);
x2 = MIN(x2, width / VNC_DIRTY_PIXELS_PER_BIT);
if (x2 > x) {
n += vnc_job_add_rect(job, x * VNC_DIRTY_PIXELS_PER_BIT, y,
(x2 - x) * VNC_DIRTY_PIXELS_PER_BIT, h);
}
if (!x && x2 == width / VNC_DIRTY_PIXELS_PER_BIT) {
y += h;
if (y == height) {
break;
}
}
}
ui: fix VNC client throttling when forced update is requested The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check is disabled if the client has requested a forced update, because we want to send these as soon as possible. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then repeatedly send full framebuffer update requests, but never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle full updates. When we get a forced update request, we keep track of exactly how much data we put on the output buffer. We will not process a subsequent forced update request until this data has been fully sent on the wire. We always allow one forced update request to be in flight, regardless of what data is queued for incremental updates or audio data. The slight complication is that we do not initially know how much data an update will send, as this is done in the background by the VNC job thread. So we must track the fact that the job thread has an update pending, and not process any further updates until this job is has been completed & put data on the output buffer. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-11-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:25 +03:00
vs->job_update = vs->update;
vs->update = VNC_STATE_UPDATE_NONE;
ui: fix VNC client throttling when forced update is requested The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check is disabled if the client has requested a forced update, because we want to send these as soon as possible. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then repeatedly send full framebuffer update requests, but never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle full updates. When we get a forced update request, we keep track of exactly how much data we put on the output buffer. We will not process a subsequent forced update request until this data has been fully sent on the wire. We always allow one forced update request to be in flight, regardless of what data is queued for incremental updates or audio data. The slight complication is that we do not initially know how much data an update will send, as this is done in the background by the VNC job thread. So we must track the fact that the job thread has an update pending, and not process any further updates until this job is has been completed & put data on the output buffer. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-11-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:25 +03:00
vnc_job_push(job);
vs->has_dirty = 0;
return n;
}
/* audio */
static void audio_capture_notify(void *opaque, audcnotification_e cmd)
{
VncState *vs = opaque;
assert(vs->magic == VNC_MAGIC);
switch (cmd) {
case AUD_CNOTIFY_DISABLE:
trace_vnc_msg_server_audio_end(vs, vs->ioc);
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:
trace_vnc_msg_server_audio_begin(vs, vs->ioc);
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, const void *buf, int size)
{
VncState *vs = opaque;
assert(vs->magic == VNC_MAGIC);
trace_vnc_msg_server_audio_data(vs, vs->ioc, buf, size);
vnc_lock_output(vs);
ui: fix VNC client throttling when audio capture is active The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check must be disabled if audio capture is enabled, because when streaming audio the output buffer offset will rarely be zero due to queued audio data, and so this would starve framebuffer updates. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then enable audio capture, and simply never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle when audio capture is active too. To determine how to throttle incremental updates or audio data, we calculate a size threshold. Normally the threshold is the approximate number of bytes associated with a single complete framebuffer update. ie width * height * bytes per pixel. We'll send incremental updates until we hit this threshold, at which point we'll stop sending updates until data has been written to the wire, causing the output buffer offset to fall back below the threshold. If audio capture is enabled, we increase the size of the threshold to also allow for upto 1 seconds worth of audio data samples. ie nchannels * bytes per sample * frequency. This allows the output buffer to have a mixture of incremental framebuffer updates and audio data queued, but once the threshold is exceeded, audio data will be dropped and incremental updates will be throttled. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-10-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:24 +03:00
if (vs->output.offset < vs->throttle_output_offset) {
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);
} else {
trace_vnc_client_throttle_audio(vs, vs->ioc, vs->output.offset);
ui: fix VNC client throttling when audio capture is active The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check must be disabled if audio capture is enabled, because when streaming audio the output buffer offset will rarely be zero due to queued audio data, and so this would starve framebuffer updates. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then enable audio capture, and simply never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle when audio capture is active too. To determine how to throttle incremental updates or audio data, we calculate a size threshold. Normally the threshold is the approximate number of bytes associated with a single complete framebuffer update. ie width * height * bytes per pixel. We'll send incremental updates until we hit this threshold, at which point we'll stop sending updates until data has been written to the wire, causing the output buffer offset to fall back below the threshold. If audio capture is enabled, we increase the size of the threshold to also allow for upto 1 seconds worth of audio data samples. ie nchannels * bytes per sample * frequency. This allows the output buffer to have a mixture of incremental framebuffer updates and audio data queued, but once the threshold is exceeded, audio data will be dropped and incremental updates will be throttled. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-10-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:24 +03:00
}
vnc_unlock_output(vs);
vnc_flush(vs);
}
static void audio_add(VncState *vs)
{
struct audio_capture_ops ops;
if (vs->audio_cap) {
error_report("audio already running");
return;
}
ops.notify = audio_capture_notify;
ops.destroy = audio_capture_destroy;
ops.capture = audio_capture;
vs->audio_cap = AUD_add_capture(vs->vd->audio_state, &vs->as, &ops, vs);
if (!vs->audio_cap) {
error_report("Failed to add audio capture");
}
}
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->disconnecting) {
return;
}
trace_vnc_client_disconnect_start(vs, vs->ioc);
vnc_set_share_mode(vs, VNC_SHARE_MODE_DISCONNECTED);
if (vs->ioc_tag) {
g_source_remove(vs->ioc_tag);
vs->ioc_tag = 0;
}
qio_channel_close(vs->ioc, NULL);
vs->disconnecting = TRUE;
}
void vnc_disconnect_finish(VncState *vs)
{
int i;
trace_vnc_client_disconnect_finish(vs, vs->ioc);
vnc_jobs_join(vs); /* Wait encoding jobs */
vnc_lock_output(vs);
vnc_qmp_event(vs, QAPI_EVENT_VNC_DISCONNECTED);
buffer_free(&vs->input);
buffer_free(&vs->output);
qapi_free_VncClientInfo(vs->info);
vnc_zlib_clear(vs);
vnc_tight_clear(vs);
vnc_zrle_clear(vs);
#ifdef CONFIG_VNC_SASL
vnc_sasl_client_cleanup(vs);
#endif /* CONFIG_VNC_SASL */
audio_del(vs);
qkbd_state_lift_all_keys(vs->vd->kbd);
if (vs->mouse_mode_notifier.notify != NULL) {
qemu_remove_mouse_mode_change_notifier(&vs->mouse_mode_notifier);
}
QTAILQ_REMOVE(&vs->vd->clients, vs, next);
if (QTAILQ_EMPTY(&vs->vd->clients)) {
/* last client gone */
vnc_update_server_surface(vs->vd);
}
ui/vnc.c: Fixed a deadlock bug. The GDB statck is as follows: (gdb) bt 0 __lll_lock_wait (futex=futex@entry=0x56211df20360, private=0) at lowlevellock.c:52 1 0x00007f263caf20a3 in __GI___pthread_mutex_lock (mutex=0x56211df20360) at ../nptl/pthread_mutex_lock.c:80 2 0x000056211a757364 in qemu_mutex_lock_impl (mutex=0x56211df20360, file=0x56211a804857 "../ui/vnc-jobs.h", line=60) at ../util/qemu-thread-posix.c:80 3 0x000056211a0ef8c7 in vnc_lock_output (vs=0x56211df14200) at ../ui/vnc-jobs.h:60 4 0x000056211a0efcb7 in vnc_clipboard_send (vs=0x56211df14200, count=1, dwords=0x7ffdf1701338) at ../ui/vnc-clipboard.c:138 5 0x000056211a0f0129 in vnc_clipboard_notify (notifier=0x56211df244c8, data=0x56211dd1bbf0) at ../ui/vnc-clipboard.c:209 6 0x000056211a75dde8 in notifier_list_notify (list=0x56211afa17d0 <clipboard_notifiers>, data=0x56211dd1bbf0) at ../util/notify.c:39 7 0x000056211a0bf0e6 in qemu_clipboard_update (info=0x56211dd1bbf0) at ../ui/clipboard.c:50 8 0x000056211a0bf05d in qemu_clipboard_peer_release (peer=0x56211df244c0, selection=QEMU_CLIPBOARD_SELECTION_CLIPBOARD) at ../ui/clipboard.c:41 9 0x000056211a0bef9b in qemu_clipboard_peer_unregister (peer=0x56211df244c0) at ../ui/clipboard.c:19 10 0x000056211a0d45f3 in vnc_disconnect_finish (vs=0x56211df14200) at ../ui/vnc.c:1358 11 0x000056211a0d4c9d in vnc_client_read (vs=0x56211df14200) at ../ui/vnc.c:1611 12 0x000056211a0d4df8 in vnc_client_io (ioc=0x56211ce70690, condition=G_IO_IN, opaque=0x56211df14200) at ../ui/vnc.c:1649 13 0x000056211a5b976c in qio_channel_fd_source_dispatch (source=0x56211ce50a00, callback=0x56211a0d4d71 <vnc_client_io>, user_data=0x56211df14200) at ../io/channel-watch.c:84 14 0x00007f263ccede8e in g_main_context_dispatch () at /lib/x86_64-linux-gnu/libglib-2.0.so.0 15 0x000056211a77d4a1 in glib_pollfds_poll () at ../util/main-loop.c:232 16 0x000056211a77d51f in os_host_main_loop_wait (timeout=958545) at ../util/main-loop.c:255 17 0x000056211a77d630 in main_loop_wait (nonblocking=0) at ../util/main-loop.c:531 18 0x000056211a45bc8e in qemu_main_loop () at ../softmmu/runstate.c:726 19 0x000056211a0b45fa in main (argc=69, argv=0x7ffdf1701778, envp=0x7ffdf17019a8) at ../softmmu/main.c:50 From the call trace, we can see it is a deadlock bug. vnc_disconnect_finish will acquire the output_mutex. But, the output_mutex will be acquired again in vnc_clipboard_send. Repeated locking will cause deadlock. So, I move qemu_clipboard_peer_unregister() behind vnc_unlock_output(); Fixes: 0bf41cab93e ("ui/vnc: clipboard support") Signed-off-by: Lei Rao <lei.rao@intel.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20220105020808.597325-1-lei.rao@intel.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2022-01-05 05:08:08 +03:00
vnc_unlock_output(vs);
if (vs->cbpeer.notifier.notify) {
qemu_clipboard_peer_unregister(&vs->cbpeer);
}
qemu_mutex_destroy(&vs->output_mutex);
if (vs->bh != NULL) {
qemu_bh_delete(vs->bh);
}
buffer_free(&vs->jobs_buffer);
for (i = 0; i < VNC_STAT_ROWS; ++i) {
g_free(vs->lossy_rect[i]);
}
g_free(vs->lossy_rect);
object_unref(OBJECT(vs->ioc));
vs->ioc = NULL;
object_unref(OBJECT(vs->sioc));
vs->sioc = NULL;
vs->magic = 0;
vnc: fix memory leak when vnc disconnect Currently when qemu receives a vnc connect, it creates a 'VncState' to represent this connection. In 'vnc_worker_thread_loop' it creates a local 'VncState'. The connection 'VcnState' and local 'VncState' exchange data in 'vnc_async_encoding_start' and 'vnc_async_encoding_end'. In 'zrle_compress_data' it calls 'deflateInit2' to allocate the libz library opaque data. The 'VncState' used in 'zrle_compress_data' is the local 'VncState'. In 'vnc_zrle_clear' it calls 'deflateEnd' to free the libz library opaque data. The 'VncState' used in 'vnc_zrle_clear' is the connection 'VncState'. In currently implementation there will be a memory leak when the vnc disconnect. Following is the asan output backtrack: Direct leak of 29760 byte(s) in 5 object(s) allocated from: 0 0xffffa67ef3c3 in __interceptor_calloc (/lib64/libasan.so.4+0xd33c3) 1 0xffffa65071cb in g_malloc0 (/lib64/libglib-2.0.so.0+0x571cb) 2 0xffffa5e968f7 in deflateInit2_ (/lib64/libz.so.1+0x78f7) 3 0xaaaacec58613 in zrle_compress_data ui/vnc-enc-zrle.c:87 4 0xaaaacec58613 in zrle_send_framebuffer_update ui/vnc-enc-zrle.c:344 5 0xaaaacec34e77 in vnc_send_framebuffer_update ui/vnc.c:919 6 0xaaaacec5e023 in vnc_worker_thread_loop ui/vnc-jobs.c:271 7 0xaaaacec5e5e7 in vnc_worker_thread ui/vnc-jobs.c:340 8 0xaaaacee4d3c3 in qemu_thread_start util/qemu-thread-posix.c:502 9 0xffffa544e8bb in start_thread (/lib64/libpthread.so.0+0x78bb) 10 0xffffa53965cb in thread_start (/lib64/libc.so.6+0xd55cb) This is because the opaque allocated in 'deflateInit2' is not freed in 'deflateEnd'. The reason is that the 'deflateEnd' calls 'deflateStateCheck' and in the latter will check whether 's->strm != strm'(libz's data structure). This check will be true so in 'deflateEnd' it just return 'Z_STREAM_ERROR' and not free the data allocated in 'deflateInit2'. The reason this happens is that the 'VncState' contains the whole 'VncZrle', so when calling 'deflateInit2', the 's->strm' will be the local address. So 's->strm != strm' will be true. To fix this issue, we need to make 'zrle' of 'VncState' to be a pointer. Then the connection 'VncState' and local 'VncState' exchange mechanism will work as expection. The 'tight' of 'VncState' has the same issue, let's also turn it to a pointer. Reported-by: Ying Fang <fangying1@huawei.com> Signed-off-by: Li Qiang <liq3ea@163.com> Message-id: 20190831153922.121308-1-liq3ea@163.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2019-08-31 18:39:22 +03:00
g_free(vs->zrle);
g_free(vs->tight);
g_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
size_t vnc_client_io_error(VncState *vs, ssize_t ret, Error *err)
{
if (ret <= 0) {
if (ret == 0) {
trace_vnc_client_eof(vs, vs->ioc);
vnc: do not disconnect on EAGAIN When qemu vnc server is trying to send large update to clients, there might be a situation when system responds with something like EAGAIN, indicating that there's no system memory to send that much data (depending on the network speed, client and server and what is happening). In this case, something like this happens on qemu side (from strace): sendmsg(16, {msg_name(0)=NULL, msg_iov(1)=[{"\244\"..., 729186}], msg_controllen=0, msg_flags=0}, 0) = 103950 sendmsg(16, {msg_name(0)=NULL, msg_iov(1)=[{"lz\346"..., 1559618}], msg_controllen=0, msg_flags=0}, 0) = -1 EAGAIN sendmsg(-1, {msg_name(0)=NULL, msg_iov(1)=[{"lz\346"..., 1559618}], msg_controllen=0, msg_flags=0}, 0) = -1 EBADF qemu closes the socket before the retry, and obviously it gets EBADF when trying to send to -1. This is because there WAS a special handling for EAGAIN, but now it doesn't work anymore, after commit 04d2529da27db512dcbd5e99d0e26d333f16efcc, because now in all error-like cases we initiate vnc disconnect. This change were introduced in qemu 2.6, and caused numerous grief for many people, resulting in their vnc clients reporting sporadic random disconnects from vnc server. Fix that by doing the disconnect only when necessary, i.e. omitting this very case of EAGAIN. Hopefully the existing condition (comparing with QIO_CHANNEL_ERR_BLOCK) is sufficient, as the original code (before the above commit) were checking for other errno values too. Apparently there's another (semi?)bug exist somewhere here, since the code tries to write to fd# -1, it probably should check if the connection is open before. But this isn't important. Signed-off-by: Michael Tokarev <mjt@tls.msk.ru> Reviewed-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 1486115549-9398-1-git-send-email-mjt@msgid.tls.msk.ru Fixes: 04d2529da27db512dcbd5e99d0e26d333f16efcc Cc: Daniel P. Berrange <berrange@redhat.com> Cc: Gerd Hoffmann <kraxel@redhat.com> Cc: qemu-stable@nongnu.org Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-02-03 12:52:29 +03:00
vnc_disconnect_start(vs);
} else if (ret != QIO_CHANNEL_ERR_BLOCK) {
trace_vnc_client_io_error(vs, vs->ioc,
err ? error_get_pretty(err) : "Unknown");
vnc: do not disconnect on EAGAIN When qemu vnc server is trying to send large update to clients, there might be a situation when system responds with something like EAGAIN, indicating that there's no system memory to send that much data (depending on the network speed, client and server and what is happening). In this case, something like this happens on qemu side (from strace): sendmsg(16, {msg_name(0)=NULL, msg_iov(1)=[{"\244\"..., 729186}], msg_controllen=0, msg_flags=0}, 0) = 103950 sendmsg(16, {msg_name(0)=NULL, msg_iov(1)=[{"lz\346"..., 1559618}], msg_controllen=0, msg_flags=0}, 0) = -1 EAGAIN sendmsg(-1, {msg_name(0)=NULL, msg_iov(1)=[{"lz\346"..., 1559618}], msg_controllen=0, msg_flags=0}, 0) = -1 EBADF qemu closes the socket before the retry, and obviously it gets EBADF when trying to send to -1. This is because there WAS a special handling for EAGAIN, but now it doesn't work anymore, after commit 04d2529da27db512dcbd5e99d0e26d333f16efcc, because now in all error-like cases we initiate vnc disconnect. This change were introduced in qemu 2.6, and caused numerous grief for many people, resulting in their vnc clients reporting sporadic random disconnects from vnc server. Fix that by doing the disconnect only when necessary, i.e. omitting this very case of EAGAIN. Hopefully the existing condition (comparing with QIO_CHANNEL_ERR_BLOCK) is sufficient, as the original code (before the above commit) were checking for other errno values too. Apparently there's another (semi?)bug exist somewhere here, since the code tries to write to fd# -1, it probably should check if the connection is open before. But this isn't important. Signed-off-by: Michael Tokarev <mjt@tls.msk.ru> Reviewed-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 1486115549-9398-1-git-send-email-mjt@msgid.tls.msk.ru Fixes: 04d2529da27db512dcbd5e99d0e26d333f16efcc Cc: Daniel P. Berrange <berrange@redhat.com> Cc: Gerd Hoffmann <kraxel@redhat.com> Cc: qemu-stable@nongnu.org Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-02-03 12:52:29 +03:00
vnc_disconnect_start(vs);
}
error_free(err);
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);
}
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
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
* 0 on I/O error, and disconnects the client socket.
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
*/
size_t vnc_client_write_buf(VncState *vs, const uint8_t *data, size_t datalen)
{
Error *err = NULL;
ssize_t ret;
ret = qio_channel_write(vs->ioc, (const char *)data, datalen, &err);
VNC_DEBUG("Wrote wire %p %zd -> %ld\n", data, datalen, ret);
return vnc_client_io_error(vs, ret, err);
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 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
* 0 on I/O error, and disconnects the client socket.
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
*/
static size_t vnc_client_write_plain(VncState *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
{
size_t offset;
size_t 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
#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;
ui: fix VNC client throttling when forced update is requested The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check is disabled if the client has requested a forced update, because we want to send these as soon as possible. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then repeatedly send full framebuffer update requests, but never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle full updates. When we get a forced update request, we keep track of exactly how much data we put on the output buffer. We will not process a subsequent forced update request until this data has been fully sent on the wire. We always allow one forced update request to be in flight, regardless of what data is queued for incremental updates or audio data. The slight complication is that we do not initially know how much data an update will send, as this is done in the background by the VNC job thread. So we must track the fact that the job thread has an update pending, and not process any further updates until this job is has been completed & put data on the output buffer. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-11-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:25 +03:00
if (ret >= vs->force_update_offset) {
if (vs->force_update_offset != 0) {
trace_vnc_client_unthrottle_forced(vs, vs->ioc);
}
ui: fix VNC client throttling when forced update is requested The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check is disabled if the client has requested a forced update, because we want to send these as soon as possible. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then repeatedly send full framebuffer update requests, but never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle full updates. When we get a forced update request, we keep track of exactly how much data we put on the output buffer. We will not process a subsequent forced update request until this data has been fully sent on the wire. We always allow one forced update request to be in flight, regardless of what data is queued for incremental updates or audio data. The slight complication is that we do not initially know how much data an update will send, as this is done in the background by the VNC job thread. So we must track the fact that the job thread has an update pending, and not process any further updates until this job is has been completed & put data on the output buffer. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-11-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:25 +03:00
vs->force_update_offset = 0;
} else {
vs->force_update_offset -= ret;
}
offset = vs->output.offset;
buffer_advance(&vs->output, ret);
if (offset >= vs->throttle_output_offset &&
vs->output.offset < vs->throttle_output_offset) {
trace_vnc_client_unthrottle_incremental(vs, vs->ioc, vs->output.offset);
}
if (vs->output.offset == 0) {
if (vs->ioc_tag) {
g_source_remove(vs->ioc_tag);
}
vs->ioc_tag = qio_channel_add_watch(
vs->ioc, G_IO_IN | G_IO_HUP | G_IO_ERR,
vnc_client_io, vs, 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
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(VncState *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
{
#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);
}
}
static void vnc_client_write(VncState *vs)
{
assert(vs->magic == VNC_MAGIC);
vnc_lock_output(vs);
if (vs->output.offset) {
vnc_client_write_locked(vs);
} else if (vs->ioc != NULL) {
if (vs->ioc_tag) {
g_source_remove(vs->ioc_tag);
}
vs->ioc_tag = qio_channel_add_watch(
vs->ioc, G_IO_IN | G_IO_HUP | G_IO_ERR,
vnc_client_io, vs, NULL);
}
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
* 0 on I/O error or EOF, and disconnects the client socket.
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
*/
size_t vnc_client_read_buf(VncState *vs, uint8_t *data, size_t datalen)
{
ssize_t ret;
Error *err = NULL;
ret = qio_channel_read(vs->ioc, (char *)data, datalen, &err);
VNC_DEBUG("Read wire %p %zd -> %ld\n", data, datalen, ret);
return vnc_client_io_error(vs, ret, err);
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
}
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, which may be less than
* the requested 'datalen' if the socket would block. Returns
* 0 on I/O error or EOF, and disconnects the client socket.
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
*/
static size_t vnc_client_read_plain(VncState *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
{
size_t 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;
}
static void vnc_jobs_bh(void *opaque)
{
VncState *vs = opaque;
assert(vs->magic == VNC_MAGIC);
vnc_jobs_consume_buffer(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
/*
* 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)
* Returns 0 on success, -1 if client disconnected
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
*/
static int vnc_client_read(VncState *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
{
size_t sz;
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 (vs->sasl.conn && vs->sasl.runSSF)
sz = vnc_client_read_sasl(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
else
#endif /* CONFIG_VNC_SASL */
sz = vnc_client_read_plain(vs);
if (!sz) {
if (vs->disconnecting) {
vnc_disconnect_finish(vs);
return -1;
}
return 0;
}
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->disconnecting) {
vnc_disconnect_finish(vs);
return -1;
}
if (!ret) {
buffer_advance(&vs->input, len);
} else {
vs->read_handler_expect = ret;
}
}
return 0;
}
gboolean vnc_client_io(QIOChannel *ioc G_GNUC_UNUSED,
GIOCondition condition, void *opaque)
{
VncState *vs = opaque;
assert(vs->magic == VNC_MAGIC);
if (condition & (G_IO_HUP | G_IO_ERR)) {
vnc_disconnect_start(vs);
return TRUE;
}
if (condition & G_IO_IN) {
if (vnc_client_read(vs) < 0) {
/* vs is free()ed here */
return TRUE;
}
}
if (condition & G_IO_OUT) {
vnc_client_write(vs);
}
vnc: fix segfault in closed connection handling On one of our client's node, due to trying to read from closed ioc, a segmentation fault occured. Corresponding backtrace: 0 object_get_class (obj=obj@entry=0x0) 1 qio_channel_readv_full (ioc=0x0, iov=0x7ffe55277180 ... 2 qio_channel_read (ioc=<optimized out> ... 3 vnc_client_read_buf (vs=vs@entry=0x55625f3c6000, ... 4 vnc_client_read_plain (vs=0x55625f3c6000) 5 vnc_client_read (vs=0x55625f3c6000) 6 vnc_client_io (ioc=<optimized out>, condition=G_IO_IN, ... 7 g_main_dispatch (context=0x556251568a50) 8 g_main_context_dispatch (context=context@entry=0x556251568a50) 9 glib_pollfds_poll () 10 os_host_main_loop_wait (timeout=<optimized out>) 11 main_loop_wait (nonblocking=nonblocking@entry=0) 12 main_loop () at vl.c:1909 13 main (argc=<optimized out>, argv=<optimized out>, ... Having analyzed the coredump, I understood that the reason is that ioc_tag is reset on vnc_disconnect_start and ioc is cleaned in vnc_disconnect_finish. Between these two events due to some reasons the ioc_tag was set again and after vnc_disconnect_finish the handler is running with freed ioc, which led to the segmentation fault. The patch checks vs->disconnecting in places where we call qio_channel_add_watch and resets handler if disconnecting == TRUE to prevent such an occurrence. Signed-off-by: Klim Kireev <klim.kireev@virtuozzo.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> Message-id: 20180207094844.21402-1-klim.kireev@virtuozzo.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2018-02-07 12:48:44 +03:00
if (vs->disconnecting) {
if (vs->ioc_tag != 0) {
g_source_remove(vs->ioc_tag);
}
vs->ioc_tag = 0;
}
return TRUE;
}
ui: place a hard cap on VNC server output buffer size The previous patches fix problems with throttling of forced framebuffer updates and audio data capture that would cause the QEMU output buffer size to grow without bound. Those fixes are graceful in that once the client catches up with reading data from the server, everything continues operating normally. There is some data which the server sends to the client that is impractical to throttle. Specifically there are various pseudo framebuffer update encodings to inform the client of things like desktop resizes, pointer changes, audio playback start/stop, LED state and so on. These generally only involve sending a very small amount of data to the client, but a malicious guest might be able to do things that trigger these changes at a very high rate. Throttling them is not practical as missed or delayed events would cause broken behaviour for the client. This patch thus takes a more forceful approach of setting an absolute upper bound on the amount of data we permit to be present in the output buffer at any time. The previous patch set a threshold for throttling the output buffer by allowing an amount of data equivalent to one complete framebuffer update and one seconds worth of audio data. On top of this it allowed for one further forced framebuffer update to be queued. To be conservative, we thus take that throttling threshold and multiply it by 5 to form an absolute upper bound. If this bound is hit during vnc_write() we forceably disconnect the client, refusing to queue further data. This limit is high enough that it should never be hit unless a malicious client is trying to exploit the sever, or the network is completely saturated preventing any sending of data on the socket. This completes the fix for CVE-2017-15124 started in the previous patches. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-12-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:26 +03:00
/*
* Scale factor to apply to vs->throttle_output_offset when checking for
* hard limit. Worst case normal usage could be x2, if we have a complete
* incremental update and complete forced update in the output buffer.
* So x3 should be good enough, but we pick x5 to be conservative and thus
* (hopefully) never trigger incorrectly.
*/
#define VNC_THROTTLE_OUTPUT_LIMIT_SCALE 5
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)
{
assert(vs->magic == VNC_MAGIC);
ui: place a hard cap on VNC server output buffer size The previous patches fix problems with throttling of forced framebuffer updates and audio data capture that would cause the QEMU output buffer size to grow without bound. Those fixes are graceful in that once the client catches up with reading data from the server, everything continues operating normally. There is some data which the server sends to the client that is impractical to throttle. Specifically there are various pseudo framebuffer update encodings to inform the client of things like desktop resizes, pointer changes, audio playback start/stop, LED state and so on. These generally only involve sending a very small amount of data to the client, but a malicious guest might be able to do things that trigger these changes at a very high rate. Throttling them is not practical as missed or delayed events would cause broken behaviour for the client. This patch thus takes a more forceful approach of setting an absolute upper bound on the amount of data we permit to be present in the output buffer at any time. The previous patch set a threshold for throttling the output buffer by allowing an amount of data equivalent to one complete framebuffer update and one seconds worth of audio data. On top of this it allowed for one further forced framebuffer update to be queued. To be conservative, we thus take that throttling threshold and multiply it by 5 to form an absolute upper bound. If this bound is hit during vnc_write() we forceably disconnect the client, refusing to queue further data. This limit is high enough that it should never be hit unless a malicious client is trying to exploit the sever, or the network is completely saturated preventing any sending of data on the socket. This completes the fix for CVE-2017-15124 started in the previous patches. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-12-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:26 +03:00
if (vs->disconnecting) {
return;
}
/* Protection against malicious client/guest to prevent our output
* buffer growing without bound if client stops reading data. This
* should rarely trigger, because we have earlier throttling code
* which stops issuing framebuffer updates and drops audio data
* if the throttle_output_offset value is exceeded. So we only reach
* this higher level if a huge number of pseudo-encodings get
* triggered while data can't be sent on the socket.
*
* NB throttle_output_offset can be zero during early protocol
* handshake, or from the job thread's VncState clone
*/
if (vs->throttle_output_offset != 0 &&
(vs->output.offset / VNC_THROTTLE_OUTPUT_LIMIT_SCALE) >
vs->throttle_output_offset) {
trace_vnc_client_output_limit(vs, vs->ioc, vs->output.offset,
vs->throttle_output_offset);
ui: place a hard cap on VNC server output buffer size The previous patches fix problems with throttling of forced framebuffer updates and audio data capture that would cause the QEMU output buffer size to grow without bound. Those fixes are graceful in that once the client catches up with reading data from the server, everything continues operating normally. There is some data which the server sends to the client that is impractical to throttle. Specifically there are various pseudo framebuffer update encodings to inform the client of things like desktop resizes, pointer changes, audio playback start/stop, LED state and so on. These generally only involve sending a very small amount of data to the client, but a malicious guest might be able to do things that trigger these changes at a very high rate. Throttling them is not practical as missed or delayed events would cause broken behaviour for the client. This patch thus takes a more forceful approach of setting an absolute upper bound on the amount of data we permit to be present in the output buffer at any time. The previous patch set a threshold for throttling the output buffer by allowing an amount of data equivalent to one complete framebuffer update and one seconds worth of audio data. On top of this it allowed for one further forced framebuffer update to be queued. To be conservative, we thus take that throttling threshold and multiply it by 5 to form an absolute upper bound. If this bound is hit during vnc_write() we forceably disconnect the client, refusing to queue further data. This limit is high enough that it should never be hit unless a malicious client is trying to exploit the sever, or the network is completely saturated preventing any sending of data on the socket. This completes the fix for CVE-2017-15124 started in the previous patches. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-12-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:26 +03:00
vnc_disconnect_start(vs);
return;
}
buffer_reserve(&vs->output, len);
if (vs->ioc != NULL && buffer_empty(&vs->output)) {
if (vs->ioc_tag) {
g_source_remove(vs->ioc_tag);
}
vs->ioc_tag = qio_channel_add_watch(
vs->ioc, G_IO_IN | G_IO_HUP | G_IO_ERR | G_IO_OUT,
vnc_client_io, vs, NULL);
}
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->ioc != NULL && vs->output.offset) {
vnc_client_write_locked(vs);
}
vnc: fix segfault in closed connection handling On one of our client's node, due to trying to read from closed ioc, a segmentation fault occured. Corresponding backtrace: 0 object_get_class (obj=obj@entry=0x0) 1 qio_channel_readv_full (ioc=0x0, iov=0x7ffe55277180 ... 2 qio_channel_read (ioc=<optimized out> ... 3 vnc_client_read_buf (vs=vs@entry=0x55625f3c6000, ... 4 vnc_client_read_plain (vs=0x55625f3c6000) 5 vnc_client_read (vs=0x55625f3c6000) 6 vnc_client_io (ioc=<optimized out>, condition=G_IO_IN, ... 7 g_main_dispatch (context=0x556251568a50) 8 g_main_context_dispatch (context=context@entry=0x556251568a50) 9 glib_pollfds_poll () 10 os_host_main_loop_wait (timeout=<optimized out>) 11 main_loop_wait (nonblocking=nonblocking@entry=0) 12 main_loop () at vl.c:1909 13 main (argc=<optimized out>, argv=<optimized out>, ... Having analyzed the coredump, I understood that the reason is that ioc_tag is reset on vnc_disconnect_start and ioc is cleaned in vnc_disconnect_finish. Between these two events due to some reasons the ioc_tag was set again and after vnc_disconnect_finish the handler is running with freed ioc, which led to the segmentation fault. The patch checks vs->disconnecting in places where we call qio_channel_add_watch and resets handler if disconnecting == TRUE to prevent such an occurrence. Signed-off-by: Klim Kireev <klim.kireev@virtuozzo.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> Message-id: 20180207094844.21402-1-klim.kireev@virtuozzo.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2018-02-07 12:48:44 +03:00
if (vs->disconnecting) {
if (vs->ioc_tag != 0) {
g_source_remove(vs->ioc_tag);
}
vs->ioc_tag = 0;
}
vnc_unlock_output(vs);
}
static uint8_t read_u8(uint8_t *data, size_t offset)
{
return data[offset];
}
static uint16_t read_u16(uint8_t *data, size_t offset)
{
return ((data[offset] & 0xFF) << 8) | (data[offset + 1] & 0xFF);
}
static 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 check_pointer_type_change(Notifier *notifier, void *data)
{
VncState *vs = container_of(notifier, VncState, mouse_mode_notifier);
int absolute = qemu_input_is_absolute(vs->vd->dcl.con);
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,
pixman_image_get_width(vs->vd->server),
pixman_image_get_height(vs->vd->server),
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)
{
static uint32_t bmap[INPUT_BUTTON__MAX] = {
[INPUT_BUTTON_LEFT] = 0x01,
[INPUT_BUTTON_MIDDLE] = 0x02,
[INPUT_BUTTON_RIGHT] = 0x04,
[INPUT_BUTTON_WHEEL_UP] = 0x08,
[INPUT_BUTTON_WHEEL_DOWN] = 0x10,
};
QemuConsole *con = vs->vd->dcl.con;
int width = pixman_image_get_width(vs->vd->server);
int height = pixman_image_get_height(vs->vd->server);
if (vs->last_bmask != button_mask) {
qemu_input_update_buttons(con, bmap, vs->last_bmask, button_mask);
vs->last_bmask = button_mask;
}
if (vs->absolute) {
qemu_input_queue_abs(con, INPUT_AXIS_X, x, 0, width);
qemu_input_queue_abs(con, INPUT_AXIS_Y, y, 0, height);
} else if (vnc_has_feature(vs, VNC_FEATURE_POINTER_TYPE_CHANGE)) {
qemu_input_queue_rel(con, INPUT_AXIS_X, x - 0x7FFF);
qemu_input_queue_rel(con, INPUT_AXIS_Y, y - 0x7FFF);
} else {
if (vs->last_x != -1) {
qemu_input_queue_rel(con, INPUT_AXIS_X, x - vs->last_x);
qemu_input_queue_rel(con, INPUT_AXIS_Y, y - vs->last_y);
}
vs->last_x = x;
vs->last_y = y;
}
qemu_input_event_sync();
}
static void press_key(VncState *vs, QKeyCode qcode)
{
qkbd_state_key_event(vs->vd->kbd, qcode, true);
qkbd_state_key_event(vs->vd->kbd, qcode, false);
}
static void vnc_led_state_change(VncState *vs)
{
if (!vnc_has_feature(vs, VNC_FEATURE_LED_STATE)) {
return;
}
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, 1, 1, VNC_ENCODING_LED_STATE);
vnc_write_u8(vs, vs->vd->ledstate);
vnc_unlock_output(vs);
vnc_flush(vs);
}
static void kbd_leds(void *opaque, int ledstate)
{
VncDisplay *vd = opaque;
VncState *client;
trace_vnc_key_guest_leds((ledstate & QEMU_CAPS_LOCK_LED),
(ledstate & QEMU_NUM_LOCK_LED),
(ledstate & QEMU_SCROLL_LOCK_LED));
if (ledstate == vd->ledstate) {
return;
}
vd->ledstate = ledstate;
QTAILQ_FOREACH(client, &vd->clients, next) {
vnc_led_state_change(client);
}
}
static void do_key_event(VncState *vs, int down, int keycode, int sym)
{
QKeyCode qcode = qemu_input_key_number_to_qcode(keycode);
/* QEMU console switch */
switch (qcode) {
case Q_KEY_CODE_1 ... Q_KEY_CODE_9: /* '1' to '9' keys */
if (down &&
qkbd_state_modifier_get(vs->vd->kbd, QKBD_MOD_CTRL) &&
qkbd_state_modifier_get(vs->vd->kbd, QKBD_MOD_ALT)) {
QemuConsole *con = qemu_console_lookup_by_index(qcode - Q_KEY_CODE_1);
if (con) {
unregister_displaychangelistener(&vs->vd->dcl);
qkbd_state_switch_console(vs->vd->kbd, con);
vs->vd->dcl.con = con;
register_displaychangelistener(&vs->vd->dcl);
}
return;
}
default:
break;
}
/* Turn off the lock state sync logic if the client support the led
state extension.
*/
if (down && vs->vd->lock_key_sync &&
!vnc_has_feature(vs, VNC_FEATURE_LED_STATE) &&
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 (!qkbd_state_modifier_get(vs->vd->kbd, QKBD_MOD_NUMLOCK)) {
trace_vnc_key_sync_numlock(true);
press_key(vs, Q_KEY_CODE_NUM_LOCK);
}
} else {
if (qkbd_state_modifier_get(vs->vd->kbd, QKBD_MOD_NUMLOCK)) {
trace_vnc_key_sync_numlock(false);
press_key(vs, Q_KEY_CODE_NUM_LOCK);
}
}
}
if (down && vs->vd->lock_key_sync &&
!vnc_has_feature(vs, VNC_FEATURE_LED_STATE) &&
((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');
bool shift = qkbd_state_modifier_get(vs->vd->kbd, QKBD_MOD_SHIFT);
bool capslock = qkbd_state_modifier_get(vs->vd->kbd, QKBD_MOD_CAPSLOCK);
if (capslock) {
if (uppercase == shift) {
trace_vnc_key_sync_capslock(false);
press_key(vs, Q_KEY_CODE_CAPS_LOCK);
}
} else {
if (uppercase != shift) {
trace_vnc_key_sync_capslock(true);
press_key(vs, Q_KEY_CODE_CAPS_LOCK);
}
}
}
qkbd_state_key_event(vs->vd->kbd, qcode, down);
if (!qemu_console_is_graphic(vs->vd->dcl.con)) {
QemuTextConsole *con = QEMU_TEXT_CONSOLE(vs->vd->dcl.con);
bool numlock = qkbd_state_modifier_get(vs->vd->kbd, QKBD_MOD_NUMLOCK);
bool control = qkbd_state_modifier_get(vs->vd->kbd, QKBD_MOD_CTRL);
/* QEMU console emulation */
if (down) {
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:
qemu_text_console_put_keysym(con, QEMU_KEY_UP);
break;
case 0xd0:
qemu_text_console_put_keysym(con, QEMU_KEY_DOWN);
break;
case 0xcb:
qemu_text_console_put_keysym(con, QEMU_KEY_LEFT);
break;
case 0xcd:
qemu_text_console_put_keysym(con, QEMU_KEY_RIGHT);
break;
case 0xd3:
qemu_text_console_put_keysym(con, QEMU_KEY_DELETE);
break;
case 0xc7:
qemu_text_console_put_keysym(con, QEMU_KEY_HOME);
break;
case 0xcf:
qemu_text_console_put_keysym(con, QEMU_KEY_END);
break;
case 0xc9:
qemu_text_console_put_keysym(con, QEMU_KEY_PAGEUP);
break;
case 0xd1:
qemu_text_console_put_keysym(con, QEMU_KEY_PAGEDOWN);
break;
case 0x47:
qemu_text_console_put_keysym(con, numlock ? '7' : QEMU_KEY_HOME);
break;
case 0x48:
qemu_text_console_put_keysym(con, numlock ? '8' : QEMU_KEY_UP);
break;
case 0x49:
qemu_text_console_put_keysym(con, numlock ? '9' : QEMU_KEY_PAGEUP);
break;
case 0x4b:
qemu_text_console_put_keysym(con, numlock ? '4' : QEMU_KEY_LEFT);
break;
case 0x4c:
qemu_text_console_put_keysym(con, '5');
break;
case 0x4d:
qemu_text_console_put_keysym(con, numlock ? '6' : QEMU_KEY_RIGHT);
break;
case 0x4f:
qemu_text_console_put_keysym(con, numlock ? '1' : QEMU_KEY_END);
break;
case 0x50:
qemu_text_console_put_keysym(con, numlock ? '2' : QEMU_KEY_DOWN);
break;
case 0x51:
qemu_text_console_put_keysym(con, numlock ? '3' : QEMU_KEY_PAGEDOWN);
break;
case 0x52:
qemu_text_console_put_keysym(con, '0');
break;
case 0x53:
qemu_text_console_put_keysym(con, numlock ? '.' : QEMU_KEY_DELETE);
break;
case 0xb5:
qemu_text_console_put_keysym(con, '/');
break;
case 0x37:
qemu_text_console_put_keysym(con, '*');
break;
case 0x4a:
qemu_text_console_put_keysym(con, '-');
break;
case 0x4e:
qemu_text_console_put_keysym(con, '+');
break;
case 0x9c:
qemu_text_console_put_keysym(con, '\n');
break;
default:
if (control) {
qemu_text_console_put_keysym(con, sym & 0x1f);
} else {
qemu_text_console_put_keysym(con, sym);
}
break;
}
}
}
}
static const char *code2name(int keycode)
{
return QKeyCode_str(qemu_input_key_number_to_qcode(keycode));
}
static void key_event(VncState *vs, int down, uint32_t sym)
{
int keycode;
int lsym = sym;
if (lsym >= 'A' && lsym <= 'Z' && qemu_console_is_graphic(vs->vd->dcl.con)) {
lsym = lsym - 'A' + 'a';
}
keycode = keysym2scancode(vs->vd->kbd_layout, lsym & 0xFFFF,
vs->vd->kbd, down) & SCANCODE_KEYMASK;
trace_vnc_key_event_map(down, sym, keycode, code2name(keycode));
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 {
trace_vnc_key_event_ext(down, sym, keycode, code2name(keycode));
do_key_event(vs, down, keycode, sym);
}
}
static void framebuffer_update_request(VncState *vs, int incremental,
int x, int y, int w, int h)
{
if (incremental) {
if (vs->update != VNC_STATE_UPDATE_FORCE) {
vs->update = VNC_STATE_UPDATE_INCREMENTAL;
}
} else {
vs->update = VNC_STATE_UPDATE_FORCE;
vnc_set_area_dirty(vs->dirty, vs->vd, x, y, w, h);
if (vnc_has_feature(vs, VNC_FEATURE_RESIZE_EXT)) {
vnc_desktop_resize_ext(vs, 0);
}
}
}
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,
pixman_image_get_width(vs->vd->server),
pixman_image_get_height(vs->vd->server),
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,
pixman_image_get_width(vs->vd->server),
pixman_image_get_height(vs->vd->server),
VNC_ENCODING_AUDIO);
vnc_unlock_output(vs);
vnc_flush(vs);
}
static void send_xvp_message(VncState *vs, int code)
{
vnc_lock_output(vs);
vnc_write_u8(vs, VNC_MSG_SERVER_XVP);
vnc_write_u8(vs, 0); /* pad */
vnc_write_u8(vs, 1); /* version */
vnc_write_u8(vs, code);
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;
vnc: fix memory leak when vnc disconnect Currently when qemu receives a vnc connect, it creates a 'VncState' to represent this connection. In 'vnc_worker_thread_loop' it creates a local 'VncState'. The connection 'VcnState' and local 'VncState' exchange data in 'vnc_async_encoding_start' and 'vnc_async_encoding_end'. In 'zrle_compress_data' it calls 'deflateInit2' to allocate the libz library opaque data. The 'VncState' used in 'zrle_compress_data' is the local 'VncState'. In 'vnc_zrle_clear' it calls 'deflateEnd' to free the libz library opaque data. The 'VncState' used in 'vnc_zrle_clear' is the connection 'VncState'. In currently implementation there will be a memory leak when the vnc disconnect. Following is the asan output backtrack: Direct leak of 29760 byte(s) in 5 object(s) allocated from: 0 0xffffa67ef3c3 in __interceptor_calloc (/lib64/libasan.so.4+0xd33c3) 1 0xffffa65071cb in g_malloc0 (/lib64/libglib-2.0.so.0+0x571cb) 2 0xffffa5e968f7 in deflateInit2_ (/lib64/libz.so.1+0x78f7) 3 0xaaaacec58613 in zrle_compress_data ui/vnc-enc-zrle.c:87 4 0xaaaacec58613 in zrle_send_framebuffer_update ui/vnc-enc-zrle.c:344 5 0xaaaacec34e77 in vnc_send_framebuffer_update ui/vnc.c:919 6 0xaaaacec5e023 in vnc_worker_thread_loop ui/vnc-jobs.c:271 7 0xaaaacec5e5e7 in vnc_worker_thread ui/vnc-jobs.c:340 8 0xaaaacee4d3c3 in qemu_thread_start util/qemu-thread-posix.c:502 9 0xffffa544e8bb in start_thread (/lib64/libpthread.so.0+0x78bb) 10 0xffffa53965cb in thread_start (/lib64/libc.so.6+0xd55cb) This is because the opaque allocated in 'deflateInit2' is not freed in 'deflateEnd'. The reason is that the 'deflateEnd' calls 'deflateStateCheck' and in the latter will check whether 's->strm != strm'(libz's data structure). This check will be true so in 'deflateEnd' it just return 'Z_STREAM_ERROR' and not free the data allocated in 'deflateInit2'. The reason this happens is that the 'VncState' contains the whole 'VncZrle', so when calling 'deflateInit2', the 's->strm' will be the local address. So 's->strm != strm' will be true. To fix this issue, we need to make 'zrle' of 'VncState' to be a pointer. Then the connection 'VncState' and local 'VncState' exchange mechanism will work as expection. The 'tight' of 'VncState' has the same issue, let's also turn it to a pointer. Reported-by: Ying Fang <fangying1@huawei.com> Signed-off-by: Li Qiang <liq3ea@163.com> Message-id: 20190831153922.121308-1-liq3ea@163.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2019-08-31 18:39:22 +03:00
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 preferred 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_HEXTILE:
vnc_set_feature(vs, VNC_FEATURE_HEXTILE);
vs->vnc_encoding = enc;
break;
case VNC_ENCODING_TIGHT:
vnc_set_feature(vs, VNC_FEATURE_TIGHT);
vs->vnc_encoding = enc;
break;
#ifdef CONFIG_PNG
case VNC_ENCODING_TIGHT_PNG:
vnc_set_feature(vs, VNC_FEATURE_TIGHT_PNG);
vs->vnc_encoding = enc;
break;
#endif
case VNC_ENCODING_ZLIB:
/*
* VNC_ENCODING_ZRLE compresses better than VNC_ENCODING_ZLIB.
* So prioritize ZRLE, even if the client hints that it prefers
* ZLIB.
*/
if (!vnc_has_feature(vs, VNC_FEATURE_ZRLE)) {
vnc_set_feature(vs, VNC_FEATURE_ZLIB);
vs->vnc_encoding = enc;
}
break;
case VNC_ENCODING_ZRLE:
vnc_set_feature(vs, VNC_FEATURE_ZRLE);
vs->vnc_encoding = enc;
break;
case VNC_ENCODING_ZYWRLE:
vnc_set_feature(vs, VNC_FEATURE_ZYWRLE);
vs->vnc_encoding = enc;
break;
case VNC_ENCODING_DESKTOPRESIZE:
vnc_set_feature(vs, VNC_FEATURE_RESIZE);
break;
case VNC_ENCODING_DESKTOP_RESIZE_EXT:
vnc_set_feature(vs, VNC_FEATURE_RESIZE_EXT);
break;
case VNC_ENCODING_POINTER_TYPE_CHANGE:
vnc_set_feature(vs, VNC_FEATURE_POINTER_TYPE_CHANGE);
break;
case VNC_ENCODING_RICH_CURSOR:
vnc_set_feature(vs, VNC_FEATURE_RICH_CURSOR);
break;
case VNC_ENCODING_ALPHA_CURSOR:
vnc_set_feature(vs, VNC_FEATURE_ALPHA_CURSOR);
break;
case VNC_ENCODING_EXT_KEY_EVENT:
send_ext_key_event_ack(vs);
break;
case VNC_ENCODING_AUDIO:
if (vs->vd->audio_state) {
vnc_set_feature(vs, VNC_FEATURE_AUDIO);
send_ext_audio_ack(vs);
}
break;
case VNC_ENCODING_WMVi:
vnc_set_feature(vs, VNC_FEATURE_WMVI);
break;
case VNC_ENCODING_LED_STATE:
vnc_set_feature(vs, VNC_FEATURE_LED_STATE);
break;
case VNC_ENCODING_XVP:
if (vs->vd->power_control) {
vnc_set_feature(vs, VNC_FEATURE_XVP);
send_xvp_message(vs, VNC_XVP_CODE_INIT);
}
break;
case VNC_ENCODING_CLIPBOARD_EXT:
vnc_set_feature(vs, VNC_FEATURE_CLIPBOARD_EXT);
vnc_server_cut_text_caps(vs);
break;
case VNC_ENCODING_COMPRESSLEVEL0 ... VNC_ENCODING_COMPRESSLEVEL0 + 9:
vnc: fix memory leak when vnc disconnect Currently when qemu receives a vnc connect, it creates a 'VncState' to represent this connection. In 'vnc_worker_thread_loop' it creates a local 'VncState'. The connection 'VcnState' and local 'VncState' exchange data in 'vnc_async_encoding_start' and 'vnc_async_encoding_end'. In 'zrle_compress_data' it calls 'deflateInit2' to allocate the libz library opaque data. The 'VncState' used in 'zrle_compress_data' is the local 'VncState'. In 'vnc_zrle_clear' it calls 'deflateEnd' to free the libz library opaque data. The 'VncState' used in 'vnc_zrle_clear' is the connection 'VncState'. In currently implementation there will be a memory leak when the vnc disconnect. Following is the asan output backtrack: Direct leak of 29760 byte(s) in 5 object(s) allocated from: 0 0xffffa67ef3c3 in __interceptor_calloc (/lib64/libasan.so.4+0xd33c3) 1 0xffffa65071cb in g_malloc0 (/lib64/libglib-2.0.so.0+0x571cb) 2 0xffffa5e968f7 in deflateInit2_ (/lib64/libz.so.1+0x78f7) 3 0xaaaacec58613 in zrle_compress_data ui/vnc-enc-zrle.c:87 4 0xaaaacec58613 in zrle_send_framebuffer_update ui/vnc-enc-zrle.c:344 5 0xaaaacec34e77 in vnc_send_framebuffer_update ui/vnc.c:919 6 0xaaaacec5e023 in vnc_worker_thread_loop ui/vnc-jobs.c:271 7 0xaaaacec5e5e7 in vnc_worker_thread ui/vnc-jobs.c:340 8 0xaaaacee4d3c3 in qemu_thread_start util/qemu-thread-posix.c:502 9 0xffffa544e8bb in start_thread (/lib64/libpthread.so.0+0x78bb) 10 0xffffa53965cb in thread_start (/lib64/libc.so.6+0xd55cb) This is because the opaque allocated in 'deflateInit2' is not freed in 'deflateEnd'. The reason is that the 'deflateEnd' calls 'deflateStateCheck' and in the latter will check whether 's->strm != strm'(libz's data structure). This check will be true so in 'deflateEnd' it just return 'Z_STREAM_ERROR' and not free the data allocated in 'deflateInit2'. The reason this happens is that the 'VncState' contains the whole 'VncZrle', so when calling 'deflateInit2', the 's->strm' will be the local address. So 's->strm != strm' will be true. To fix this issue, we need to make 'zrle' of 'VncState' to be a pointer. Then the connection 'VncState' and local 'VncState' exchange mechanism will work as expection. The 'tight' of 'VncState' has the same issue, let's also turn it to a pointer. Reported-by: Ying Fang <fangying1@huawei.com> Signed-off-by: Li Qiang <liq3ea@163.com> Message-id: 20190831153922.121308-1-liq3ea@163.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2019-08-31 18:39:22 +03:00
vs->tight->compression = (enc & 0x0F);
break;
case VNC_ENCODING_QUALITYLEVEL0 ... VNC_ENCODING_QUALITYLEVEL0 + 9:
if (vs->vd->lossy) {
vnc: fix memory leak when vnc disconnect Currently when qemu receives a vnc connect, it creates a 'VncState' to represent this connection. In 'vnc_worker_thread_loop' it creates a local 'VncState'. The connection 'VcnState' and local 'VncState' exchange data in 'vnc_async_encoding_start' and 'vnc_async_encoding_end'. In 'zrle_compress_data' it calls 'deflateInit2' to allocate the libz library opaque data. The 'VncState' used in 'zrle_compress_data' is the local 'VncState'. In 'vnc_zrle_clear' it calls 'deflateEnd' to free the libz library opaque data. The 'VncState' used in 'vnc_zrle_clear' is the connection 'VncState'. In currently implementation there will be a memory leak when the vnc disconnect. Following is the asan output backtrack: Direct leak of 29760 byte(s) in 5 object(s) allocated from: 0 0xffffa67ef3c3 in __interceptor_calloc (/lib64/libasan.so.4+0xd33c3) 1 0xffffa65071cb in g_malloc0 (/lib64/libglib-2.0.so.0+0x571cb) 2 0xffffa5e968f7 in deflateInit2_ (/lib64/libz.so.1+0x78f7) 3 0xaaaacec58613 in zrle_compress_data ui/vnc-enc-zrle.c:87 4 0xaaaacec58613 in zrle_send_framebuffer_update ui/vnc-enc-zrle.c:344 5 0xaaaacec34e77 in vnc_send_framebuffer_update ui/vnc.c:919 6 0xaaaacec5e023 in vnc_worker_thread_loop ui/vnc-jobs.c:271 7 0xaaaacec5e5e7 in vnc_worker_thread ui/vnc-jobs.c:340 8 0xaaaacee4d3c3 in qemu_thread_start util/qemu-thread-posix.c:502 9 0xffffa544e8bb in start_thread (/lib64/libpthread.so.0+0x78bb) 10 0xffffa53965cb in thread_start (/lib64/libc.so.6+0xd55cb) This is because the opaque allocated in 'deflateInit2' is not freed in 'deflateEnd'. The reason is that the 'deflateEnd' calls 'deflateStateCheck' and in the latter will check whether 's->strm != strm'(libz's data structure). This check will be true so in 'deflateEnd' it just return 'Z_STREAM_ERROR' and not free the data allocated in 'deflateInit2'. The reason this happens is that the 'VncState' contains the whole 'VncZrle', so when calling 'deflateInit2', the 's->strm' will be the local address. So 's->strm != strm' will be true. To fix this issue, we need to make 'zrle' of 'VncState' to be a pointer. Then the connection 'VncState' and local 'VncState' exchange mechanism will work as expection. The 'tight' of 'VncState' has the same issue, let's also turn it to a pointer. Reported-by: Ying Fang <fangying1@huawei.com> Signed-off-by: Li Qiang <liq3ea@163.com> Message-id: 20190831153922.121308-1-liq3ea@163.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2019-08-31 18:39:22 +03:00
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, NULL);
vnc_led_state_change(vs);
vnc_cursor_define(vs);
}
static void set_pixel_conversion(VncState *vs)
{
pixman_format_code_t fmt = qemu_pixman_get_format(&vs->client_pf);
if (fmt == VNC_SERVER_FB_FORMAT) {
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 send_color_map(VncState *vs)
{
int i;
vnc_lock_output(vs);
vnc_write_u8(vs, VNC_MSG_SERVER_SET_COLOUR_MAP_ENTRIES);
vnc_write_u8(vs, 0); /* padding */
vnc_write_u16(vs, 0); /* first color */
vnc_write_u16(vs, 256); /* # of colors */
for (i = 0; i < 256; i++) {
PixelFormat *pf = &vs->client_pf;
vnc_write_u16(vs, (((i >> pf->rshift) & pf->rmax) << (16 - pf->rbits)));
vnc_write_u16(vs, (((i >> pf->gshift) & pf->gmax) << (16 - pf->gbits)));
vnc_write_u16(vs, (((i >> pf->bshift) & pf->bmax) << (16 - pf->bbits)));
}
vnc_unlock_output(vs);
}
static void set_pixel_format(VncState *vs, int bits_per_pixel,
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) {
/* Expose a reasonable default 256 color map */
bits_per_pixel = 8;
red_max = 7;
green_max = 7;
blue_max = 3;
red_shift = 0;
green_shift = 3;
blue_shift = 6;
}
switch (bits_per_pixel) {
case 8:
case 16:
case 32:
break;
default:
vnc_client_error(vs);
return;
}
vs->client_pf.rmax = red_max ? red_max : 0xFF;
vs->client_pf.rbits = ctpopl(red_max);
vs->client_pf.rshift = red_shift;
vs->client_pf.rmask = red_max << red_shift;
vs->client_pf.gmax = green_max ? green_max : 0xFF;
vs->client_pf.gbits = ctpopl(green_max);
vs->client_pf.gshift = green_shift;
vs->client_pf.gmask = green_max << green_shift;
vs->client_pf.bmax = blue_max ? blue_max : 0xFF;
vs->client_pf.bbits = ctpopl(blue_max);
vs->client_pf.bshift = blue_shift;
vs->client_pf.bmask = blue_max << blue_shift;
vs->client_pf.bits_per_pixel = bits_per_pixel;
vs->client_pf.bytes_per_pixel = bits_per_pixel / 8;
vs->client_pf.depth = bits_per_pixel == 32 ? 24 : bits_per_pixel;
vs->client_be = big_endian_flag;
if (!true_color_flag) {
send_color_map(vs);
}
set_pixel_conversion(vs);
graphic_hw_invalidate(vs->vd->dcl.con);
graphic_hw_update(vs->vd->dcl.con);
}
static void pixel_format_message (VncState *vs) {
char pad[3] = { 0, 0, 0 };
vs->client_pf = qemu_default_pixelformat(32);
vnc_write_u8(vs, vs->client_pf.bits_per_pixel); /* bits-per-pixel */
vnc_write_u8(vs, vs->client_pf.depth); /* depth */
#if HOST_BIG_ENDIAN
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->client_pf.rmax); /* red-max */
vnc_write_u16(vs, vs->client_pf.gmax); /* green-max */
vnc_write_u16(vs, vs->client_pf.bmax); /* blue-max */
vnc_write_u8(vs, vs->client_pf.rshift); /* red-shift */
vnc_write_u8(vs, vs->client_pf.gshift); /* green-shift */
vnc_write_u8(vs, vs->client_pf.bshift); /* blue-shift */
vnc_write(vs, pad, 3); /* padding */
vnc_hextile_set_pixel_conversion(vs, 0);
vs->write_pixels = vnc_write_pixels_copy;
}
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,
vs->client_width,
vs->client_height,
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;
uint32_t freq;
VncDisplay *vd = vs->vd;
if (data[0] > 3) {
update_displaychangelistener(&vd->dcl, VNC_REFRESH_INTERVAL_BASE);
}
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, 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;
}
uint32_t dlen = abs(read_s32(data, 4));
if (len == 8) {
if (dlen > (1 << 20)) {
error_report("vnc: client_cut_text msg payload has %u bytes"
" which exceeds our limit of 1MB.", dlen);
vnc_client_error(vs);
break;
}
if (dlen > 0) {
return 8 + dlen;
}
}
if (read_s32(data, 4) < 0) {
if (!vnc_has_feature(vs, VNC_FEATURE_CLIPBOARD_EXT)) {
error_report("vnc: extended clipboard message while disabled");
vnc_client_error(vs);
break;
}
if (dlen < 4) {
error_report("vnc: malformed payload (header less than 4 bytes)"
" in extended clipboard pseudo-encoding.");
vnc_client_error(vs);
break;
}
vnc_client_cut_text_ext(vs, dlen, read_u32(data, 8), data + 12);
break;
}
vnc_client_cut_text(vs, read_u32(data, 4), data + 8);
break;
case VNC_MSG_CLIENT_XVP:
if (!vnc_has_feature(vs, VNC_FEATURE_XVP)) {
error_report("vnc: xvp client message while disabled");
vnc_client_error(vs);
break;
}
if (len == 1) {
return 4;
}
if (len == 4) {
uint8_t version = read_u8(data, 2);
uint8_t action = read_u8(data, 3);
if (version != 1) {
error_report("vnc: xvp client message version %d != 1",
version);
vnc_client_error(vs);
break;
}
switch (action) {
case VNC_XVP_ACTION_SHUTDOWN:
qemu_system_powerdown_request();
break;
case VNC_XVP_ACTION_REBOOT:
send_xvp_message(vs, VNC_XVP_CODE_FAIL);
break;
case VNC_XVP_ACTION_RESET:
qemu_system_reset_request(SHUTDOWN_CAUSE_HOST_QMP_SYSTEM_RESET);
break;
default:
send_xvp_message(vs, VNC_XVP_CODE_FAIL);
break;
}
}
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 (!vnc_has_feature(vs, VNC_FEATURE_AUDIO)) {
error_report("Audio message %d with audio disabled", read_u8(data, 2));
vnc_client_error(vs);
break;
}
if (len == 2)
return 4;
switch (read_u16 (data, 2)) {
case VNC_MSG_CLIENT_QEMU_AUDIO_ENABLE:
trace_vnc_msg_client_audio_enable(vs, vs->ioc);
audio_add(vs);
break;
case VNC_MSG_CLIENT_QEMU_AUDIO_DISABLE:
trace_vnc_msg_client_audio_disable(vs, vs->ioc);
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 = AUDIO_FORMAT_U8; break;
case 1: vs->as.fmt = AUDIO_FORMAT_S8; break;
case 2: vs->as.fmt = AUDIO_FORMAT_U16; break;
case 3: vs->as.fmt = AUDIO_FORMAT_S16; break;
case 4: vs->as.fmt = AUDIO_FORMAT_U32; break;
case 5: vs->as.fmt = AUDIO_FORMAT_S32; break;
default:
VNC_DEBUG("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) {
VNC_DEBUG("Invalid audio channel count %d\n",
read_u8(data, 5));
vnc_client_error(vs);
break;
}
freq = read_u32(data, 6);
/* No official limit for protocol, but 48khz is a sensible
* upper bound for trustworthy clients, and this limit
* protects calculations involving 'vs->as.freq' later.
*/
if (freq > 48000) {
VNC_DEBUG("Invalid audio frequency %u > 48000", freq);
vnc_client_error(vs);
break;
}
vs->as.freq = freq;
trace_vnc_msg_client_audio_format(
vs, vs->ioc, vs->as.fmt, vs->as.nchannels, vs->as.freq);
break;
default:
VNC_DEBUG("Invalid audio message %d\n", read_u8(data, 2));
vnc_client_error(vs);
break;
}
break;
default:
VNC_DEBUG("Msg: %d\n", read_u16(data, 0));
vnc_client_error(vs);
break;
}
break;
case VNC_MSG_CLIENT_SET_DESKTOP_SIZE:
{
size_t size;
uint8_t screens;
int w, h;
if (len < 8) {
return 8;
}
screens = read_u8(data, 6);
size = 8 + screens * 16;
if (len < size) {
return size;
}
w = read_u16(data, 2);
h = read_u16(data, 4);
trace_vnc_msg_client_set_desktop_size(vs, vs->ioc, w, h, screens);
if (dpy_ui_info_supported(vs->vd->dcl.con)) {
QemuUIInfo info;
memset(&info, 0, sizeof(info));
info.width = w;
info.height = h;
dpy_set_ui_info(vs->vd->dcl.con, &info, false);
vnc_desktop_resize_ext(vs, 4 /* Request forwarded */);
} else {
vnc_desktop_resize_ext(vs, 3 /* Invalid screen layout */);
}
break;
}
default:
VNC_DEBUG("Msg: %d\n", data[0]);
vnc_client_error(vs);
break;
}
ui: fix VNC client throttling when audio capture is active The VNC server must throttle data sent to the client to prevent the 'output' buffer size growing without bound, if the client stops reading data off the socket (either maliciously or due to stalled/slow network connection). The current throttling is very crude because it simply checks whether the output buffer offset is zero. This check must be disabled if audio capture is enabled, because when streaming audio the output buffer offset will rarely be zero due to queued audio data, and so this would starve framebuffer updates. As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM. They can first start something in the guest that triggers lots of framebuffer updates eg play a youtube video. Then enable audio capture, and simply never read data back from the server. This can easily make QEMU's VNC server send buffer consume 100MB of RAM per second, until the OOM killer starts reaping processes (hopefully the rogue QEMU process, but it might pick others...). To address this we make the throttling more intelligent, so we can throttle when audio capture is active too. To determine how to throttle incremental updates or audio data, we calculate a size threshold. Normally the threshold is the approximate number of bytes associated with a single complete framebuffer update. ie width * height * bytes per pixel. We'll send incremental updates until we hit this threshold, at which point we'll stop sending updates until data has been written to the wire, causing the output buffer offset to fall back below the threshold. If audio capture is enabled, we increase the size of the threshold to also allow for upto 1 seconds worth of audio data samples. ie nchannels * bytes per sample * frequency. This allows the output buffer to have a mixture of incremental framebuffer updates and audio data queued, but once the threshold is exceeded, audio data will be dropped and incremental updates will be throttled. This unbounded memory growth affects all VNC server configurations supported by QEMU, with no workaround possible. The mitigating factor is that it can only be triggered by a client that has authenticated with the VNC server, and who is able to trigger a large quantity of framebuffer updates or audio samples from the guest OS. Mostly they'll just succeed in getting the OOM killer to kill their own QEMU process, but its possible other processes can get taken out as collateral damage. This is a more general variant of the similar unbounded memory usage flaw in the websockets server, that was previously assigned CVE-2017-15268, and fixed in 2.11 by: commit a7b20a8efa28e5f22c26c06cd06c2f12bc863493 Author: Daniel P. Berrange <berrange@redhat.com> Date: Mon Oct 9 14:43:42 2017 +0100 io: monitor encoutput buffer size from websocket GSource This new general memory usage flaw has been assigned CVE-2017-15124, and is partially fixed by this patch. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Darren Kenny <darren.kenny@oracle.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-id: 20171218191228.31018-10-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2017-12-18 22:12:24 +03:00
vnc_update_throttle_offset(vs);
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];
VncShareMode mode;
int size;
mode = data[0] ? VNC_SHARE_MODE_SHARED : VNC_SHARE_MODE_EXCLUSIVE;
switch (vs->vd->share_policy) {
case VNC_SHARE_POLICY_IGNORE:
/*
* Ignore the shared flag. Nothing to do here.
*
* Doesn't conform to the rfb spec but is traditional qemu
* behavior, thus left here as option for compatibility
* reasons.
*/
break;
case VNC_SHARE_POLICY_ALLOW_EXCLUSIVE:
/*
* Policy: Allow clients ask for exclusive access.
*
* Implementation: When a client asks for exclusive access,
* disconnect all others. Shared connects are allowed as long
* as no exclusive connection exists.
*
* This is how the rfb spec suggests to handle the shared flag.
*/
if (mode == VNC_SHARE_MODE_EXCLUSIVE) {
VncState *client;
QTAILQ_FOREACH(client, &vs->vd->clients, next) {
if (vs == client) {
continue;
}
if (client->share_mode != VNC_SHARE_MODE_EXCLUSIVE &&
client->share_mode != VNC_SHARE_MODE_SHARED) {
continue;
}
vnc_disconnect_start(client);
}
}
if (mode == VNC_SHARE_MODE_SHARED) {
if (vs->vd->num_exclusive > 0) {
vnc_disconnect_start(vs);
return 0;
}
}
break;
case VNC_SHARE_POLICY_FORCE_SHARED:
/*
* Policy: Shared connects only.
* Implementation: Disallow clients asking for exclusive access.
*
* Useful for shared desktop sessions where you don't want
* someone forgetting to say -shared when running the vnc
* client disconnect everybody else.
*/
if (mode == VNC_SHARE_MODE_EXCLUSIVE) {
vnc_disconnect_start(vs);
return 0;
}
break;
}
vnc_set_share_mode(vs, mode);
if (vs->vd->num_shared > vs->vd->connections_limit) {
vnc_disconnect_start(vs);
return 0;
}
ui: avoid sign extension using client width/height Pixman returns a signed int for the image width/height, but the VNC protocol only permits a unsigned int16. Effective framebuffer size is determined by the guest, limited by the video RAM size, so the dimensions are unlikely to exceed the range of an unsigned int16, but this is not currently validated. With the current use of 'int' for client width/height, the calculation of offsets in vnc_update_throttle_offset() suffers from integer size promotion and sign extension, causing coverity warnings *** CID 1385147: Integer handling issues (SIGN_EXTENSION) /ui/vnc.c: 979 in vnc_update_throttle_offset() 973 * than that the client would already suffering awful audio 974 * glitches, so dropping samples is no worse really). 975 */ 976 static void vnc_update_throttle_offset(VncState *vs) 977 { 978 size_t offset = >>> CID 1385147: Integer handling issues (SIGN_EXTENSION) >>> Suspicious implicit sign extension: "vs->client_pf.bytes_per_pixel" with type "unsigned char" (8 bits, unsigned) is promoted in "vs->client_width * vs->client_height * vs->client_pf.bytes_per_pixel" to type "int" (32 bits, signed), then sign-extended to type "unsigned long" (64 bits, unsigned). If "vs->client_width * vs->client_height * vs->client_pf.bytes_per_pixel" is greater than 0x7FFFFFFF, the upper bits of the result will all be 1. 979 vs->client_width * vs->client_height * vs->client_pf.bytes_per_pixel; Change client_width / client_height to be a size_t to avoid sign extension and integer promotion. Then validate that dimensions are in range wrt the RFB protocol u16 limits. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 20180118155254.17053-1-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2018-01-18 18:52:54 +03:00
assert(pixman_image_get_width(vs->vd->server) < 65536 &&
pixman_image_get_width(vs->vd->server) >= 0);
assert(pixman_image_get_height(vs->vd->server) < 65536 &&
pixman_image_get_height(vs->vd->server) >= 0);
vs->client_width = pixman_image_get_width(vs->vd->server);
vs->client_height = pixman_image_get_height(vs->vd->server);
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);
if (size > sizeof(buf)) {
size = sizeof(buf);
}
} 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, QAPI_EVENT_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 authentication_failed(VncState *vs)
{
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);
}
static void
vnc_munge_des_rfb_key(unsigned char *key, size_t nkey)
{
size_t i;
for (i = 0; i < nkey; i++) {
uint8_t r = key[i];
r = (r & 0xf0) >> 4 | (r & 0x0f) << 4;
r = (r & 0xcc) >> 2 | (r & 0x33) << 2;
r = (r & 0xaa) >> 1 | (r & 0x55) << 1;
key[i] = r;
}
}
static int protocol_client_auth_vnc(VncState *vs, uint8_t *data, size_t len)
{
unsigned char response[VNC_AUTH_CHALLENGE_SIZE];
size_t i, pwlen;
unsigned char key[8];
time_t now = time(NULL);
QCryptoCipher *cipher = NULL;
Error *err = NULL;
if (!vs->vd->password) {
trace_vnc_auth_fail(vs, vs->auth, "password is not set", "");
goto reject;
}
if (vs->vd->expires < now) {
trace_vnc_auth_fail(vs, vs->auth, "password is expired", "");
goto reject;
}
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;
vnc_munge_des_rfb_key(key, sizeof(key));
cipher = qcrypto_cipher_new(
QCRYPTO_CIPHER_ALG_DES,
QCRYPTO_CIPHER_MODE_ECB,
key, G_N_ELEMENTS(key),
&err);
if (!cipher) {
trace_vnc_auth_fail(vs, vs->auth, "cannot create cipher",
error_get_pretty(err));
error_free(err);
goto reject;
}
if (qcrypto_cipher_encrypt(cipher,
vs->challenge,
response,
VNC_AUTH_CHALLENGE_SIZE,
&err) < 0) {
trace_vnc_auth_fail(vs, vs->auth, "cannot encrypt challenge response",
error_get_pretty(err));
error_free(err);
goto reject;
}
/* Compare expected vs actual challenge response */
if (memcmp(response, data, VNC_AUTH_CHALLENGE_SIZE) != 0) {
trace_vnc_auth_fail(vs, vs->auth, "mis-matched challenge response", "");
goto reject;
} else {
trace_vnc_auth_pass(vs, vs->auth);
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);
}
qcrypto_cipher_free(cipher);
return 0;
reject:
authentication_failed(vs);
qcrypto_cipher_free(cipher);
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)
{
Error *err = NULL;
if (qcrypto_random_bytes(vs->challenge, sizeof(vs->challenge), &err)) {
trace_vnc_auth_fail(vs, vs->auth, "cannot get random bytes",
error_get_pretty(err));
error_free(err);
authentication_failed(vs);
return;
}
/* 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->auth) { /* Reject auth */
trace_vnc_auth_reject(vs, vs->auth, (int)data[0]);
authentication_failed(vs);
} else { /* Accept requested auth */
trace_vnc_auth_start(vs, vs->auth);
switch (vs->auth) {
case VNC_AUTH_NONE:
if (vs->minor >= 8) {
vnc_write_u32(vs, 0); /* Accept auth completion */
vnc_flush(vs);
}
trace_vnc_auth_pass(vs, vs->auth);
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:
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;
case VNC_AUTH_VENCRYPT:
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;
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:
start_auth_sasl(vs);
break;
#endif /* CONFIG_VNC_SASL */
default: /* Should not be possible, but just in case */
trace_vnc_auth_fail(vs, vs->auth, "Unhandled auth method", "");
authentication_failed(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) {
trace_vnc_auth_start(vs, vs->auth);
if (vs->auth == VNC_AUTH_NONE) {
vnc_write_u32(vs, vs->auth);
vnc_flush(vs);
trace_vnc_auth_pass(vs, vs->auth);
start_client_init(vs);
} else if (vs->auth == VNC_AUTH_VNC) {
VNC_DEBUG("Tell client VNC auth\n");
vnc_write_u32(vs, vs->auth);
vnc_flush(vs);
start_auth_vnc(vs);
} else {
trace_vnc_auth_fail(vs, vs->auth,
"Unsupported auth method for v3.3", "");
vnc_write_u32(vs, VNC_AUTH_INVALID);
vnc_flush(vs);
vnc_client_error(vs);
}
} else {
vnc_write_u8(vs, 1); /* num auth */
vnc_write_u8(vs, vs->auth);
vnc_read_when(vs, protocol_client_auth, 1);
vnc_flush(vs);
}
return 0;
}
static VncRectStat *vnc_stat_rect(VncDisplay *vd, int x, int y)
{
struct VncSurface *vs = &vd->guest;
return &vs->stats[y / VNC_STAT_RECT][x / VNC_STAT_RECT];
}
void vnc_sent_lossy_rect(VncState *vs, int x, int y, int w, int h)
{
int i, j;
w = (x + w) / VNC_STAT_RECT;
h = (y + h) / VNC_STAT_RECT;
x /= VNC_STAT_RECT;
y /= VNC_STAT_RECT;
for (j = y; j <= h; j++) {
for (i = x; i <= w; i++) {
vs->lossy_rect[j][i] = 1;
}
}
}
static int vnc_refresh_lossy_rect(VncDisplay *vd, int x, int y)
{
VncState *vs;
int sty = y / VNC_STAT_RECT;
int stx = x / VNC_STAT_RECT;
int has_dirty = 0;
y = QEMU_ALIGN_DOWN(y, VNC_STAT_RECT);
x = QEMU_ALIGN_DOWN(x, VNC_STAT_RECT);
QTAILQ_FOREACH(vs, &vd->clients, next) {
int j;
/* kernel send buffers are full -> refresh later */
if (vs->output.offset) {
continue;
}
if (!vs->lossy_rect[sty][stx]) {
continue;
}
vs->lossy_rect[sty][stx] = 0;
for (j = 0; j < VNC_STAT_RECT; ++j) {
bitmap_set(vs->dirty[y + j],
x / VNC_DIRTY_PIXELS_PER_BIT,
VNC_STAT_RECT / VNC_DIRTY_PIXELS_PER_BIT);
}
has_dirty++;
}
return has_dirty;
}
static int vnc_update_stats(VncDisplay *vd, struct timeval * tv)
{
int width = MIN(pixman_image_get_width(vd->guest.fb),
pixman_image_get_width(vd->server));
int height = MIN(pixman_image_get_height(vd->guest.fb),
pixman_image_get_height(vd->server));
int x, y;
struct timeval res;
int has_dirty = 0;
for (y = 0; y < height; y += VNC_STAT_RECT) {
for (x = 0; x < width; x += VNC_STAT_RECT) {
VncRectStat *rect = vnc_stat_rect(vd, x, y);
rect->updated = false;
}
}
qemu_timersub(tv, &VNC_REFRESH_STATS, &res);
if (timercmp(&vd->guest.last_freq_check, &res, >)) {
return has_dirty;
}
vd->guest.last_freq_check = *tv;
for (y = 0; y < height; y += VNC_STAT_RECT) {
for (x = 0; x < width; x += VNC_STAT_RECT) {
VncRectStat *rect= vnc_stat_rect(vd, x, y);
int count = ARRAY_SIZE(rect->times);
struct timeval min, max;
if (!timerisset(&rect->times[count - 1])) {
continue ;
}
max = rect->times[(rect->idx + count - 1) % count];
qemu_timersub(tv, &max, &res);
if (timercmp(&res, &VNC_REFRESH_LOSSY, >)) {
rect->freq = 0;
has_dirty += vnc_refresh_lossy_rect(vd, x, y);
memset(rect->times, 0, sizeof (rect->times));
continue ;
}
min = rect->times[rect->idx];
max = rect->times[(rect->idx + count - 1) % count];
qemu_timersub(&max, &min, &res);
rect->freq = res.tv_sec + res.tv_usec / 1000000.;
rect->freq /= count;
rect->freq = 1. / rect->freq;
}
}
return has_dirty;
}
double vnc_update_freq(VncState *vs, int x, int y, int w, int h)
{
int i, j;
double total = 0;
int num = 0;
x = QEMU_ALIGN_DOWN(x, VNC_STAT_RECT);
y = QEMU_ALIGN_DOWN(y, VNC_STAT_RECT);
for (j = y; j <= y + h; j += VNC_STAT_RECT) {
for (i = x; i <= x + w; i += VNC_STAT_RECT) {
total += vnc_stat_rect(vs->vd, i, j)->freq;
num++;
}
}
if (num) {
return total / num;
} else {
return 0;
}
}
static void vnc_rect_updated(VncDisplay *vd, int x, int y, struct timeval * tv)
{
VncRectStat *rect;
rect = vnc_stat_rect(vd, x, y);
if (rect->updated) {
return;
}
rect->times[rect->idx] = *tv;
rect->idx = (rect->idx + 1) % ARRAY_SIZE(rect->times);
rect->updated = true;
}
static int vnc_refresh_server_surface(VncDisplay *vd)
{
int width = MIN(pixman_image_get_width(vd->guest.fb),
pixman_image_get_width(vd->server));
int height = MIN(pixman_image_get_height(vd->guest.fb),
pixman_image_get_height(vd->server));
int cmp_bytes, server_stride, line_bytes, guest_ll, guest_stride, y = 0;
uint8_t *guest_row0 = NULL, *server_row0;
VncState *vs;
int has_dirty = 0;
pixman_image_t *tmpbuf = NULL;
unsigned long offset;
int x;
uint8_t *guest_ptr, *server_ptr;
struct timeval tv = { 0, 0 };
if (!vd->non_adaptive) {
gettimeofday(&tv, NULL);
has_dirty = vnc_update_stats(vd, &tv);
}
offset = find_next_bit((unsigned long *) &vd->guest.dirty,
height * VNC_DIRTY_BPL(&vd->guest), 0);
if (offset == height * VNC_DIRTY_BPL(&vd->guest)) {
/* no dirty bits in guest surface */
return has_dirty;
}
/*
* Walk through the guest dirty map.
* Check and copy modified bits from guest to server surface.
* Update server dirty map.
*/
server_row0 = (uint8_t *)pixman_image_get_data(vd->server);
server_stride = guest_stride = guest_ll =
pixman_image_get_stride(vd->server);
cmp_bytes = MIN(VNC_DIRTY_PIXELS_PER_BIT * VNC_SERVER_FB_BYTES,
server_stride);
if (vd->guest.format != VNC_SERVER_FB_FORMAT) {
int w = pixman_image_get_width(vd->server);
tmpbuf = qemu_pixman_linebuf_create(VNC_SERVER_FB_FORMAT, w);
} else {
int guest_bpp =
PIXMAN_FORMAT_BPP(pixman_image_get_format(vd->guest.fb));
guest_row0 = (uint8_t *)pixman_image_get_data(vd->guest.fb);
guest_stride = pixman_image_get_stride(vd->guest.fb);
guest_ll = pixman_image_get_width(vd->guest.fb)
* DIV_ROUND_UP(guest_bpp, 8);
}
line_bytes = MIN(server_stride, guest_ll);
for (;;) {
y = offset / VNC_DIRTY_BPL(&vd->guest);
x = offset % VNC_DIRTY_BPL(&vd->guest);
server_ptr = server_row0 + y * server_stride + x * cmp_bytes;
if (vd->guest.format != VNC_SERVER_FB_FORMAT) {
qemu_pixman_linebuf_fill(tmpbuf, vd->guest.fb, width, 0, y);
guest_ptr = (uint8_t *)pixman_image_get_data(tmpbuf);
} else {
guest_ptr = guest_row0 + y * guest_stride;
}
guest_ptr += x * cmp_bytes;
for (; x < DIV_ROUND_UP(width, VNC_DIRTY_PIXELS_PER_BIT);
x++, guest_ptr += cmp_bytes, server_ptr += cmp_bytes) {
int _cmp_bytes = cmp_bytes;
if (!test_and_clear_bit(x, vd->guest.dirty[y])) {
continue;
}
if ((x + 1) * cmp_bytes > line_bytes) {
_cmp_bytes = line_bytes - x * cmp_bytes;
}
assert(_cmp_bytes >= 0);
if (memcmp(server_ptr, guest_ptr, _cmp_bytes) == 0) {
continue;
}
memcpy(server_ptr, guest_ptr, _cmp_bytes);
if (!vd->non_adaptive) {
vnc_rect_updated(vd, x * VNC_DIRTY_PIXELS_PER_BIT,
y, &tv);
}
QTAILQ_FOREACH(vs, &vd->clients, next) {
set_bit(x, vs->dirty[y]);
}
has_dirty++;
}
y++;
offset = find_next_bit((unsigned long *) &vd->guest.dirty,
height * VNC_DIRTY_BPL(&vd->guest),
y * VNC_DIRTY_BPL(&vd->guest));
if (offset == height * VNC_DIRTY_BPL(&vd->guest)) {
/* no more dirty bits */
break;
}
}
qemu_pixman_image_unref(tmpbuf);
return has_dirty;
}
static void vnc_refresh(DisplayChangeListener *dcl)
{
VncDisplay *vd = container_of(dcl, VncDisplay, dcl);
VncState *vs, *vn;
int has_dirty, rects = 0;
if (QTAILQ_EMPTY(&vd->clients)) {
update_displaychangelistener(&vd->dcl, VNC_REFRESH_INTERVAL_MAX);
return;
}
graphic_hw_update(vd->dcl.con);
if (vnc_trylock_display(vd)) {
update_displaychangelistener(&vd->dcl, VNC_REFRESH_INTERVAL_BASE);
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 */
}
if (has_dirty && rects) {
vd->dcl.update_interval /= 2;
if (vd->dcl.update_interval < VNC_REFRESH_INTERVAL_BASE) {
vd->dcl.update_interval = VNC_REFRESH_INTERVAL_BASE;
}
} else {
vd->dcl.update_interval += VNC_REFRESH_INTERVAL_INC;
if (vd->dcl.update_interval > VNC_REFRESH_INTERVAL_MAX) {
vd->dcl.update_interval = VNC_REFRESH_INTERVAL_MAX;
}
}
}
static void vnc_connect(VncDisplay *vd, QIOChannelSocket *sioc,
bool skipauth, bool websocket)
{
VncState *vs = g_new0(VncState, 1);
bool first_client = QTAILQ_EMPTY(&vd->clients);
int i;
trace_vnc_client_connect(vs, sioc);
vnc: fix memory leak when vnc disconnect Currently when qemu receives a vnc connect, it creates a 'VncState' to represent this connection. In 'vnc_worker_thread_loop' it creates a local 'VncState'. The connection 'VcnState' and local 'VncState' exchange data in 'vnc_async_encoding_start' and 'vnc_async_encoding_end'. In 'zrle_compress_data' it calls 'deflateInit2' to allocate the libz library opaque data. The 'VncState' used in 'zrle_compress_data' is the local 'VncState'. In 'vnc_zrle_clear' it calls 'deflateEnd' to free the libz library opaque data. The 'VncState' used in 'vnc_zrle_clear' is the connection 'VncState'. In currently implementation there will be a memory leak when the vnc disconnect. Following is the asan output backtrack: Direct leak of 29760 byte(s) in 5 object(s) allocated from: 0 0xffffa67ef3c3 in __interceptor_calloc (/lib64/libasan.so.4+0xd33c3) 1 0xffffa65071cb in g_malloc0 (/lib64/libglib-2.0.so.0+0x571cb) 2 0xffffa5e968f7 in deflateInit2_ (/lib64/libz.so.1+0x78f7) 3 0xaaaacec58613 in zrle_compress_data ui/vnc-enc-zrle.c:87 4 0xaaaacec58613 in zrle_send_framebuffer_update ui/vnc-enc-zrle.c:344 5 0xaaaacec34e77 in vnc_send_framebuffer_update ui/vnc.c:919 6 0xaaaacec5e023 in vnc_worker_thread_loop ui/vnc-jobs.c:271 7 0xaaaacec5e5e7 in vnc_worker_thread ui/vnc-jobs.c:340 8 0xaaaacee4d3c3 in qemu_thread_start util/qemu-thread-posix.c:502 9 0xffffa544e8bb in start_thread (/lib64/libpthread.so.0+0x78bb) 10 0xffffa53965cb in thread_start (/lib64/libc.so.6+0xd55cb) This is because the opaque allocated in 'deflateInit2' is not freed in 'deflateEnd'. The reason is that the 'deflateEnd' calls 'deflateStateCheck' and in the latter will check whether 's->strm != strm'(libz's data structure). This check will be true so in 'deflateEnd' it just return 'Z_STREAM_ERROR' and not free the data allocated in 'deflateInit2'. The reason this happens is that the 'VncState' contains the whole 'VncZrle', so when calling 'deflateInit2', the 's->strm' will be the local address. So 's->strm != strm' will be true. To fix this issue, we need to make 'zrle' of 'VncState' to be a pointer. Then the connection 'VncState' and local 'VncState' exchange mechanism will work as expection. The 'tight' of 'VncState' has the same issue, let's also turn it to a pointer. Reported-by: Ying Fang <fangying1@huawei.com> Signed-off-by: Li Qiang <liq3ea@163.com> Message-id: 20190831153922.121308-1-liq3ea@163.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2019-08-31 18:39:22 +03:00
vs->zrle = g_new0(VncZrle, 1);
vs->tight = g_new0(VncTight, 1);
vs->magic = VNC_MAGIC;
vs->sioc = sioc;
object_ref(OBJECT(vs->sioc));
vs->ioc = QIO_CHANNEL(sioc);
object_ref(OBJECT(vs->ioc));
vs->vd = vd;
buffer_init(&vs->input, "vnc-input/%p", sioc);
buffer_init(&vs->output, "vnc-output/%p", sioc);
buffer_init(&vs->jobs_buffer, "vnc-jobs_buffer/%p", sioc);
vnc: fix memory leak when vnc disconnect Currently when qemu receives a vnc connect, it creates a 'VncState' to represent this connection. In 'vnc_worker_thread_loop' it creates a local 'VncState'. The connection 'VcnState' and local 'VncState' exchange data in 'vnc_async_encoding_start' and 'vnc_async_encoding_end'. In 'zrle_compress_data' it calls 'deflateInit2' to allocate the libz library opaque data. The 'VncState' used in 'zrle_compress_data' is the local 'VncState'. In 'vnc_zrle_clear' it calls 'deflateEnd' to free the libz library opaque data. The 'VncState' used in 'vnc_zrle_clear' is the connection 'VncState'. In currently implementation there will be a memory leak when the vnc disconnect. Following is the asan output backtrack: Direct leak of 29760 byte(s) in 5 object(s) allocated from: 0 0xffffa67ef3c3 in __interceptor_calloc (/lib64/libasan.so.4+0xd33c3) 1 0xffffa65071cb in g_malloc0 (/lib64/libglib-2.0.so.0+0x571cb) 2 0xffffa5e968f7 in deflateInit2_ (/lib64/libz.so.1+0x78f7) 3 0xaaaacec58613 in zrle_compress_data ui/vnc-enc-zrle.c:87 4 0xaaaacec58613 in zrle_send_framebuffer_update ui/vnc-enc-zrle.c:344 5 0xaaaacec34e77 in vnc_send_framebuffer_update ui/vnc.c:919 6 0xaaaacec5e023 in vnc_worker_thread_loop ui/vnc-jobs.c:271 7 0xaaaacec5e5e7 in vnc_worker_thread ui/vnc-jobs.c:340 8 0xaaaacee4d3c3 in qemu_thread_start util/qemu-thread-posix.c:502 9 0xffffa544e8bb in start_thread (/lib64/libpthread.so.0+0x78bb) 10 0xffffa53965cb in thread_start (/lib64/libc.so.6+0xd55cb) This is because the opaque allocated in 'deflateInit2' is not freed in 'deflateEnd'. The reason is that the 'deflateEnd' calls 'deflateStateCheck' and in the latter will check whether 's->strm != strm'(libz's data structure). This check will be true so in 'deflateEnd' it just return 'Z_STREAM_ERROR' and not free the data allocated in 'deflateInit2'. The reason this happens is that the 'VncState' contains the whole 'VncZrle', so when calling 'deflateInit2', the 's->strm' will be the local address. So 's->strm != strm' will be true. To fix this issue, we need to make 'zrle' of 'VncState' to be a pointer. Then the connection 'VncState' and local 'VncState' exchange mechanism will work as expection. The 'tight' of 'VncState' has the same issue, let's also turn it to a pointer. Reported-by: Ying Fang <fangying1@huawei.com> Signed-off-by: Li Qiang <liq3ea@163.com> Message-id: 20190831153922.121308-1-liq3ea@163.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2019-08-31 18:39:22 +03:00
buffer_init(&vs->tight->tight, "vnc-tight/%p", sioc);
buffer_init(&vs->tight->zlib, "vnc-tight-zlib/%p", sioc);
buffer_init(&vs->tight->gradient, "vnc-tight-gradient/%p", sioc);
#ifdef CONFIG_VNC_JPEG
vnc: fix memory leak when vnc disconnect Currently when qemu receives a vnc connect, it creates a 'VncState' to represent this connection. In 'vnc_worker_thread_loop' it creates a local 'VncState'. The connection 'VcnState' and local 'VncState' exchange data in 'vnc_async_encoding_start' and 'vnc_async_encoding_end'. In 'zrle_compress_data' it calls 'deflateInit2' to allocate the libz library opaque data. The 'VncState' used in 'zrle_compress_data' is the local 'VncState'. In 'vnc_zrle_clear' it calls 'deflateEnd' to free the libz library opaque data. The 'VncState' used in 'vnc_zrle_clear' is the connection 'VncState'. In currently implementation there will be a memory leak when the vnc disconnect. Following is the asan output backtrack: Direct leak of 29760 byte(s) in 5 object(s) allocated from: 0 0xffffa67ef3c3 in __interceptor_calloc (/lib64/libasan.so.4+0xd33c3) 1 0xffffa65071cb in g_malloc0 (/lib64/libglib-2.0.so.0+0x571cb) 2 0xffffa5e968f7 in deflateInit2_ (/lib64/libz.so.1+0x78f7) 3 0xaaaacec58613 in zrle_compress_data ui/vnc-enc-zrle.c:87 4 0xaaaacec58613 in zrle_send_framebuffer_update ui/vnc-enc-zrle.c:344 5 0xaaaacec34e77 in vnc_send_framebuffer_update ui/vnc.c:919 6 0xaaaacec5e023 in vnc_worker_thread_loop ui/vnc-jobs.c:271 7 0xaaaacec5e5e7 in vnc_worker_thread ui/vnc-jobs.c:340 8 0xaaaacee4d3c3 in qemu_thread_start util/qemu-thread-posix.c:502 9 0xffffa544e8bb in start_thread (/lib64/libpthread.so.0+0x78bb) 10 0xffffa53965cb in thread_start (/lib64/libc.so.6+0xd55cb) This is because the opaque allocated in 'deflateInit2' is not freed in 'deflateEnd'. The reason is that the 'deflateEnd' calls 'deflateStateCheck' and in the latter will check whether 's->strm != strm'(libz's data structure). This check will be true so in 'deflateEnd' it just return 'Z_STREAM_ERROR' and not free the data allocated in 'deflateInit2'. The reason this happens is that the 'VncState' contains the whole 'VncZrle', so when calling 'deflateInit2', the 's->strm' will be the local address. So 's->strm != strm' will be true. To fix this issue, we need to make 'zrle' of 'VncState' to be a pointer. Then the connection 'VncState' and local 'VncState' exchange mechanism will work as expection. The 'tight' of 'VncState' has the same issue, let's also turn it to a pointer. Reported-by: Ying Fang <fangying1@huawei.com> Signed-off-by: Li Qiang <liq3ea@163.com> Message-id: 20190831153922.121308-1-liq3ea@163.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2019-08-31 18:39:22 +03:00
buffer_init(&vs->tight->jpeg, "vnc-tight-jpeg/%p", sioc);
#endif
#ifdef CONFIG_PNG
vnc: fix memory leak when vnc disconnect Currently when qemu receives a vnc connect, it creates a 'VncState' to represent this connection. In 'vnc_worker_thread_loop' it creates a local 'VncState'. The connection 'VcnState' and local 'VncState' exchange data in 'vnc_async_encoding_start' and 'vnc_async_encoding_end'. In 'zrle_compress_data' it calls 'deflateInit2' to allocate the libz library opaque data. The 'VncState' used in 'zrle_compress_data' is the local 'VncState'. In 'vnc_zrle_clear' it calls 'deflateEnd' to free the libz library opaque data. The 'VncState' used in 'vnc_zrle_clear' is the connection 'VncState'. In currently implementation there will be a memory leak when the vnc disconnect. Following is the asan output backtrack: Direct leak of 29760 byte(s) in 5 object(s) allocated from: 0 0xffffa67ef3c3 in __interceptor_calloc (/lib64/libasan.so.4+0xd33c3) 1 0xffffa65071cb in g_malloc0 (/lib64/libglib-2.0.so.0+0x571cb) 2 0xffffa5e968f7 in deflateInit2_ (/lib64/libz.so.1+0x78f7) 3 0xaaaacec58613 in zrle_compress_data ui/vnc-enc-zrle.c:87 4 0xaaaacec58613 in zrle_send_framebuffer_update ui/vnc-enc-zrle.c:344 5 0xaaaacec34e77 in vnc_send_framebuffer_update ui/vnc.c:919 6 0xaaaacec5e023 in vnc_worker_thread_loop ui/vnc-jobs.c:271 7 0xaaaacec5e5e7 in vnc_worker_thread ui/vnc-jobs.c:340 8 0xaaaacee4d3c3 in qemu_thread_start util/qemu-thread-posix.c:502 9 0xffffa544e8bb in start_thread (/lib64/libpthread.so.0+0x78bb) 10 0xffffa53965cb in thread_start (/lib64/libc.so.6+0xd55cb) This is because the opaque allocated in 'deflateInit2' is not freed in 'deflateEnd'. The reason is that the 'deflateEnd' calls 'deflateStateCheck' and in the latter will check whether 's->strm != strm'(libz's data structure). This check will be true so in 'deflateEnd' it just return 'Z_STREAM_ERROR' and not free the data allocated in 'deflateInit2'. The reason this happens is that the 'VncState' contains the whole 'VncZrle', so when calling 'deflateInit2', the 's->strm' will be the local address. So 's->strm != strm' will be true. To fix this issue, we need to make 'zrle' of 'VncState' to be a pointer. Then the connection 'VncState' and local 'VncState' exchange mechanism will work as expection. The 'tight' of 'VncState' has the same issue, let's also turn it to a pointer. Reported-by: Ying Fang <fangying1@huawei.com> Signed-off-by: Li Qiang <liq3ea@163.com> Message-id: 20190831153922.121308-1-liq3ea@163.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2019-08-31 18:39:22 +03:00
buffer_init(&vs->tight->png, "vnc-tight-png/%p", sioc);
#endif
buffer_init(&vs->zlib.zlib, "vnc-zlib/%p", sioc);
vnc: fix memory leak when vnc disconnect Currently when qemu receives a vnc connect, it creates a 'VncState' to represent this connection. In 'vnc_worker_thread_loop' it creates a local 'VncState'. The connection 'VcnState' and local 'VncState' exchange data in 'vnc_async_encoding_start' and 'vnc_async_encoding_end'. In 'zrle_compress_data' it calls 'deflateInit2' to allocate the libz library opaque data. The 'VncState' used in 'zrle_compress_data' is the local 'VncState'. In 'vnc_zrle_clear' it calls 'deflateEnd' to free the libz library opaque data. The 'VncState' used in 'vnc_zrle_clear' is the connection 'VncState'. In currently implementation there will be a memory leak when the vnc disconnect. Following is the asan output backtrack: Direct leak of 29760 byte(s) in 5 object(s) allocated from: 0 0xffffa67ef3c3 in __interceptor_calloc (/lib64/libasan.so.4+0xd33c3) 1 0xffffa65071cb in g_malloc0 (/lib64/libglib-2.0.so.0+0x571cb) 2 0xffffa5e968f7 in deflateInit2_ (/lib64/libz.so.1+0x78f7) 3 0xaaaacec58613 in zrle_compress_data ui/vnc-enc-zrle.c:87 4 0xaaaacec58613 in zrle_send_framebuffer_update ui/vnc-enc-zrle.c:344 5 0xaaaacec34e77 in vnc_send_framebuffer_update ui/vnc.c:919 6 0xaaaacec5e023 in vnc_worker_thread_loop ui/vnc-jobs.c:271 7 0xaaaacec5e5e7 in vnc_worker_thread ui/vnc-jobs.c:340 8 0xaaaacee4d3c3 in qemu_thread_start util/qemu-thread-posix.c:502 9 0xffffa544e8bb in start_thread (/lib64/libpthread.so.0+0x78bb) 10 0xffffa53965cb in thread_start (/lib64/libc.so.6+0xd55cb) This is because the opaque allocated in 'deflateInit2' is not freed in 'deflateEnd'. The reason is that the 'deflateEnd' calls 'deflateStateCheck' and in the latter will check whether 's->strm != strm'(libz's data structure). This check will be true so in 'deflateEnd' it just return 'Z_STREAM_ERROR' and not free the data allocated in 'deflateInit2'. The reason this happens is that the 'VncState' contains the whole 'VncZrle', so when calling 'deflateInit2', the 's->strm' will be the local address. So 's->strm != strm' will be true. To fix this issue, we need to make 'zrle' of 'VncState' to be a pointer. Then the connection 'VncState' and local 'VncState' exchange mechanism will work as expection. The 'tight' of 'VncState' has the same issue, let's also turn it to a pointer. Reported-by: Ying Fang <fangying1@huawei.com> Signed-off-by: Li Qiang <liq3ea@163.com> Message-id: 20190831153922.121308-1-liq3ea@163.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2019-08-31 18:39:22 +03:00
buffer_init(&vs->zrle->zrle, "vnc-zrle/%p", sioc);
buffer_init(&vs->zrle->fb, "vnc-zrle-fb/%p", sioc);
buffer_init(&vs->zrle->zlib, "vnc-zrle-zlib/%p", sioc);
if (skipauth) {
avoid TABs in files that only contain a few Most files that have TABs only contain a handful of them. Change them to spaces so that we don't confuse people. disas, standard-headers, linux-headers and libdecnumber are imported from other projects and probably should be exempted from the check. Outside those, after this patch the following files still contain both 8-space and TAB sequences at the beginning of the line. Many of them have a majority of TABs, or were initially committed with all tabs. bsd-user/i386/target_syscall.h bsd-user/x86_64/target_syscall.h crypto/aes.c hw/audio/fmopl.c hw/audio/fmopl.h hw/block/tc58128.c hw/display/cirrus_vga.c hw/display/xenfb.c hw/dma/etraxfs_dma.c hw/intc/sh_intc.c hw/misc/mst_fpga.c hw/net/pcnet.c hw/sh4/sh7750.c hw/timer/m48t59.c hw/timer/sh_timer.c include/crypto/aes.h include/disas/bfd.h include/hw/sh4/sh.h libdecnumber/decNumber.c linux-headers/asm-generic/unistd.h linux-headers/linux/kvm.h linux-user/alpha/target_syscall.h linux-user/arm/nwfpe/double_cpdo.c linux-user/arm/nwfpe/fpa11_cpdt.c linux-user/arm/nwfpe/fpa11_cprt.c linux-user/arm/nwfpe/fpa11.h linux-user/flat.h linux-user/flatload.c linux-user/i386/target_syscall.h linux-user/ppc/target_syscall.h linux-user/sparc/target_syscall.h linux-user/syscall.c linux-user/syscall_defs.h linux-user/x86_64/target_syscall.h slirp/cksum.c slirp/if.c slirp/ip.h slirp/ip_icmp.c slirp/ip_icmp.h slirp/ip_input.c slirp/ip_output.c slirp/mbuf.c slirp/misc.c slirp/sbuf.c slirp/socket.c slirp/socket.h slirp/tcp_input.c slirp/tcpip.h slirp/tcp_output.c slirp/tcp_subr.c slirp/tcp_timer.c slirp/tftp.c slirp/udp.c slirp/udp.h target/cris/cpu.h target/cris/mmu.c target/cris/op_helper.c target/sh4/helper.c target/sh4/op_helper.c target/sh4/translate.c tcg/sparc/tcg-target.inc.c tests/tcg/cris/check_addo.c tests/tcg/cris/check_moveq.c tests/tcg/cris/check_swap.c tests/tcg/multiarch/test-mmap.c ui/vnc-enc-hextile-template.h ui/vnc-enc-zywrle.h util/envlist.c util/readline.c The following have only TABs: bsd-user/i386/target_signal.h bsd-user/sparc64/target_signal.h bsd-user/sparc64/target_syscall.h bsd-user/sparc/target_signal.h bsd-user/sparc/target_syscall.h bsd-user/x86_64/target_signal.h crypto/desrfb.c hw/audio/intel-hda-defs.h hw/core/uboot_image.h hw/sh4/sh7750_regnames.c hw/sh4/sh7750_regs.h include/hw/cris/etraxfs_dma.h linux-user/alpha/termbits.h linux-user/arm/nwfpe/fpopcode.h linux-user/arm/nwfpe/fpsr.h linux-user/arm/syscall_nr.h linux-user/arm/target_signal.h linux-user/cris/target_signal.h linux-user/i386/target_signal.h linux-user/linux_loop.h linux-user/m68k/target_signal.h linux-user/microblaze/target_signal.h linux-user/mips64/target_signal.h linux-user/mips/target_signal.h linux-user/mips/target_syscall.h linux-user/mips/termbits.h linux-user/ppc/target_signal.h linux-user/sh4/target_signal.h linux-user/sh4/termbits.h linux-user/sparc64/target_syscall.h linux-user/sparc/target_signal.h linux-user/x86_64/target_signal.h linux-user/x86_64/termbits.h pc-bios/optionrom/optionrom.h slirp/mbuf.h slirp/misc.h slirp/sbuf.h slirp/tcp.h slirp/tcp_timer.h slirp/tcp_var.h target/i386/svm.h target/sparc/asi.h target/xtensa/core-dc232b/xtensa-modules.inc.c target/xtensa/core-dc233c/xtensa-modules.inc.c target/xtensa/core-de212/core-isa.h target/xtensa/core-de212/xtensa-modules.inc.c target/xtensa/core-fsf/xtensa-modules.inc.c target/xtensa/core-sample_controller/core-isa.h target/xtensa/core-sample_controller/xtensa-modules.inc.c target/xtensa/core-test_kc705_be/core-isa.h target/xtensa/core-test_kc705_be/xtensa-modules.inc.c tests/tcg/cris/check_abs.c tests/tcg/cris/check_addc.c tests/tcg/cris/check_addcm.c tests/tcg/cris/check_addoq.c tests/tcg/cris/check_bound.c tests/tcg/cris/check_ftag.c tests/tcg/cris/check_int64.c tests/tcg/cris/check_lz.c tests/tcg/cris/check_openpf5.c tests/tcg/cris/check_sigalrm.c tests/tcg/cris/crisutils.h tests/tcg/cris/sys.c tests/tcg/i386/test-i386-ssse3.c ui/vgafont.h Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Message-Id: <20181213223737.11793-3-pbonzini@redhat.com> Reviewed-by: Aleksandar Markovic <amarkovic@wavecomp.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com> Acked-by: Richard Henderson <richard.henderson@linaro.org> Acked-by: Eric Blake <eblake@redhat.com> Acked-by: David Gibson <david@gibson.dropbear.id.au> Reviewed-by: Stefan Markovic <smarkovic@wavecomp.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2018-12-14 01:37:37 +03:00
vs->auth = VNC_AUTH_NONE;
vs->subauth = VNC_AUTH_INVALID;
} else {
if (websocket) {
vs->auth = vd->ws_auth;
vs->subauth = VNC_AUTH_INVALID;
} else {
vs->auth = vd->auth;
vs->subauth = vd->subauth;
}
}
VNC_DEBUG("Client sioc=%p ws=%d auth=%d subauth=%d\n",
sioc, websocket, vs->auth, vs->subauth);
vs->lossy_rect = g_malloc0(VNC_STAT_ROWS * sizeof (*vs->lossy_rect));
for (i = 0; i < VNC_STAT_ROWS; ++i) {
vs->lossy_rect[i] = g_new0(uint8_t, VNC_STAT_COLS);
}
VNC_DEBUG("New client on socket %p\n", vs->sioc);
update_displaychangelistener(&vd->dcl, VNC_REFRESH_INTERVAL_BASE);
qio_channel_set_blocking(vs->ioc, false, NULL);
if (vs->ioc_tag) {
g_source_remove(vs->ioc_tag);
}
if (websocket) {
vs->websocket = 1;
if (vd->tlscreds) {
vs->ioc_tag = qio_channel_add_watch(
vs->ioc, G_IO_IN | G_IO_HUP | G_IO_ERR,
vncws_tls_handshake_io, vs, NULL);
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
} else {
vs->ioc_tag = qio_channel_add_watch(
vs->ioc, G_IO_IN | G_IO_HUP | G_IO_ERR,
vncws_handshake_io, vs, NULL);
}
} else {
vs->ioc_tag = qio_channel_add_watch(
vs->ioc, G_IO_IN | G_IO_HUP | G_IO_ERR,
vnc_client_io, vs, NULL);
}
vnc_client_cache_addr(vs);
vnc_qmp_event(vs, QAPI_EVENT_VNC_CONNECTED);
vnc_set_share_mode(vs, VNC_SHARE_MODE_CONNECTING);
vs->last_x = -1;
vs->last_y = -1;
vs->as.freq = 44100;
vs->as.nchannels = 2;
vs->as.fmt = AUDIO_FORMAT_S16;
vs->as.endianness = 0;
qemu_mutex_init(&vs->output_mutex);
vs->bh = qemu_bh_new(vnc_jobs_bh, vs);
QTAILQ_INSERT_TAIL(&vd->clients, vs, next);
if (first_client) {
vnc_update_server_surface(vd);
}
graphic_hw_update(vd->dcl.con);
if (!vs->websocket) {
vnc_start_protocol(vs);
}
if (vd->num_connecting > vd->connections_limit) {
QTAILQ_FOREACH(vs, &vd->clients, next) {
if (vs->share_mode == VNC_SHARE_MODE_CONNECTING) {
vnc_disconnect_start(vs);
return;
}
}
}
}
void vnc_start_protocol(VncState *vs)
{
vnc_write(vs, "RFB 003.008\n", 12);
vnc_flush(vs);
vnc_read_when(vs, protocol_version, 12);
vs->mouse_mode_notifier.notify = check_pointer_type_change;
qemu_add_mouse_mode_change_notifier(&vs->mouse_mode_notifier);
}
static void vnc_listen_io(QIONetListener *listener,
QIOChannelSocket *cioc,
void *opaque)
{
VncDisplay *vd = opaque;
bool isWebsock = listener == vd->wslistener;
qio_channel_set_name(QIO_CHANNEL(cioc),
isWebsock ? "vnc-ws-server" : "vnc-server");
qio_channel_set_delay(QIO_CHANNEL(cioc), false);
vnc_connect(vd, cioc, false, isWebsock);
}
static const DisplayChangeListenerOps dcl_ops = {
.dpy_name = "vnc",
.dpy_refresh = vnc_refresh,
.dpy_gfx_update = vnc_dpy_update,
.dpy_gfx_switch = vnc_dpy_switch,
.dpy_gfx_check_format = qemu_pixman_check_format,
.dpy_mouse_set = vnc_mouse_set,
.dpy_cursor_define = vnc_dpy_cursor_define,
};
void vnc_display_init(const char *id, Error **errp)
{
VncDisplay *vd;
if (vnc_display_find(id) != NULL) {
return;
}
vd = g_malloc0(sizeof(*vd));
vd->id = strdup(id);
QTAILQ_INSERT_TAIL(&vnc_displays, vd, next);
QTAILQ_INIT(&vd->clients);
vd->expires = TIME_MAX;
if (keyboard_layout) {
trace_vnc_key_map_init(keyboard_layout);
vd->kbd_layout = init_keyboard_layout(name2keysym,
keyboard_layout, errp);
} else {
vd->kbd_layout = init_keyboard_layout(name2keysym, "en-us", errp);
}
if (!vd->kbd_layout) {
return;
}
vd->share_policy = VNC_SHARE_POLICY_ALLOW_EXCLUSIVE;
vd->connections_limit = 32;
qemu_mutex_init(&vd->mutex);
vnc_start_worker_thread();
vd->dcl.ops = &dcl_ops;
register_displaychangelistener(&vd->dcl);
vd->kbd = qkbd_state_init(vd->dcl.con);
}
static void vnc_display_close(VncDisplay *vd)
{
if (!vd) {
return;
}
vd->is_unix = false;
if (vd->listener) {
qio_net_listener_disconnect(vd->listener);
object_unref(OBJECT(vd->listener));
}
vd->listener = NULL;
if (vd->wslistener) {
qio_net_listener_disconnect(vd->wslistener);
object_unref(OBJECT(vd->wslistener));
}
vd->wslistener = NULL;
vd->auth = VNC_AUTH_INVALID;
vd->subauth = VNC_AUTH_INVALID;
if (vd->tlscreds) {
object_unref(OBJECT(vd->tlscreds));
vd->tlscreds = NULL;
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
}
if (vd->tlsauthz) {
object_unparent(OBJECT(vd->tlsauthz));
vd->tlsauthz = NULL;
}
g_free(vd->tlsauthzid);
vd->tlsauthzid = NULL;
if (vd->lock_key_sync) {
qemu_remove_led_event_handler(vd->led);
vd->led = NULL;
}
#ifdef CONFIG_VNC_SASL
if (vd->sasl.authz) {
object_unparent(OBJECT(vd->sasl.authz));
vd->sasl.authz = NULL;
}
g_free(vd->sasl.authzid);
vd->sasl.authzid = NULL;
#endif
}
int vnc_display_password(const char *id, const char *password)
{
VncDisplay *vd = vnc_display_find(id);
if (!vd) {
return -EINVAL;
}
if (vd->auth == VNC_AUTH_NONE) {
error_printf_unless_qmp("If you want use passwords please enable "
"password auth using '-vnc ${dpy},password'.\n");
return -EINVAL;
}
g_free(vd->password);
vd->password = g_strdup(password);
return 0;
}
int vnc_display_pw_expire(const char *id, time_t expires)
{
VncDisplay *vd = vnc_display_find(id);
if (!vd) {
return -EINVAL;
}
vd->expires = expires;
return 0;
}
static void vnc_display_print_local_addr(VncDisplay *vd)
{
SocketAddress *addr;
if (!vd->listener || !vd->listener->nsioc) {
return;
}
addr = qio_channel_socket_get_local_address(vd->listener->sioc[0], NULL);
if (!addr) {
return;
}
if (addr->type != SOCKET_ADDRESS_TYPE_INET) {
qapi_free_SocketAddress(addr);
return;
}
error_printf_unless_qmp("VNC server running on %s:%s\n",
addr->u.inet.host,
addr->u.inet.port);
qapi_free_SocketAddress(addr);
}
static QemuOptsList qemu_vnc_opts = {
.name = "vnc",
.head = QTAILQ_HEAD_INITIALIZER(qemu_vnc_opts.head),
.implied_opt_name = "vnc",
.desc = {
{
.name = "vnc",
.type = QEMU_OPT_STRING,
},{
.name = "websocket",
.type = QEMU_OPT_STRING,
},{
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
.name = "tls-creds",
.type = QEMU_OPT_STRING,
},{
.name = "share",
.type = QEMU_OPT_STRING,
},{
.name = "display",
.type = QEMU_OPT_STRING,
},{
.name = "head",
.type = QEMU_OPT_NUMBER,
},{
.name = "connections",
.type = QEMU_OPT_NUMBER,
},{
.name = "to",
.type = QEMU_OPT_NUMBER,
},{
.name = "ipv4",
.type = QEMU_OPT_BOOL,
},{
.name = "ipv6",
.type = QEMU_OPT_BOOL,
},{
.name = "password",
.type = QEMU_OPT_BOOL,
},{
.name = "password-secret",
.type = QEMU_OPT_STRING,
},{
.name = "reverse",
.type = QEMU_OPT_BOOL,
},{
.name = "lock-key-sync",
.type = QEMU_OPT_BOOL,
},{
.name = "key-delay-ms",
.type = QEMU_OPT_NUMBER,
},{
.name = "sasl",
.type = QEMU_OPT_BOOL,
vnc: allow specifying a custom authorization object name The VNC server has historically had support for ACLs to check both the SASL username and the TLS x509 distinguished name. The VNC server was responsible for creating the initial ACL, and the client app was then responsible for populating it with rules using the HMP 'acl_add' command. This is not satisfactory for a variety of reasons. There is no way to populate the ACLs from the command line, users are forced to use the HMP. With multiple network services all supporting TLS and ACLs now, it is desirable to be able to define a single ACL that is referenced by all services. To address these limitations, two new options are added to the VNC server CLI. The 'tls-authz' option takes the ID of a QAuthZ object to use for checking TLS x509 distinguished names, and the 'sasl-authz' option takes the ID of another object to use for checking SASL usernames. In this example, we setup two authorization rules. The first allows any client with a certificate issued by the 'RedHat' organization in the 'London' locality. The second ACL allows clients with either the 'joe@REDHAT.COM' or 'fred@REDHAT.COM' kerberos usernames. Both checks must pass for the user to be allowed. $QEMU -object tls-creds-x509,id=tls0,dir=/home/berrange/qemutls,\ endpoint=server,verify-peer=yes \ -object authz-simple,id=authz0,policy=deny,\ rules.0.match=O=RedHat,,L=London,rules.0.policy=allow \ -object authz-simple,id=authz1,policy=deny,\ rules.0.match=fred@REDHAT.COM,rules.0.policy=allow \ rules.0.match=joe@REDHAT.COM,rules.0.policy=allow \ -vnc 0.0.0.0:1,tls-creds=tls0,tls-authz=authz0, sasl,sasl-authz=authz1 \ ...other QEMU args... Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 20190227145755.26556-2-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2019-02-27 17:57:54 +03:00
},{
.name = "tls-authz",
.type = QEMU_OPT_STRING,
},{
.name = "sasl-authz",
.type = QEMU_OPT_STRING,
},{
.name = "lossy",
.type = QEMU_OPT_BOOL,
},{
.name = "non-adaptive",
.type = QEMU_OPT_BOOL,
},{
.name = "audiodev",
.type = QEMU_OPT_STRING,
},{
.name = "power-control",
.type = QEMU_OPT_BOOL,
},
{ /* end of list */ }
},
};
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
static int
vnc_display_setup_auth(int *auth,
int *subauth,
QCryptoTLSCreds *tlscreds,
bool password,
bool sasl,
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
bool websocket,
Error **errp)
{
/*
* We have a choice of 3 authentication options
*
* 1. none
* 2. vnc
* 3. sasl
*
* The channel can be run in 2 modes
*
* 1. clear
* 2. tls
*
* And TLS can use 2 types of credentials
*
* 1. anon
* 2. x509
*
* We thus have 9 possible logical combinations
*
* 1. clear + none
* 2. clear + vnc
* 3. clear + sasl
* 4. tls + anon + none
* 5. tls + anon + vnc
* 6. tls + anon + sasl
* 7. tls + x509 + none
* 8. tls + x509 + vnc
* 9. tls + x509 + sasl
*
* These need to be mapped into the VNC auth schemes
* in an appropriate manner. In regular VNC, all the
* TLS options get mapped into VNC_AUTH_VENCRYPT
* sub-auth types.
*
* In websockets, the https:// protocol already provides
* TLS support, so there is no need to make use of the
* VeNCrypt extension. Furthermore, websockets browser
* clients could not use VeNCrypt even if they wanted to,
* as they cannot control when the TLS handshake takes
* place. Thus there is no option but to rely on https://,
* meaning combinations 4->6 and 7->9 will be mapped to
* VNC auth schemes in the same way as combos 1->3.
*
* Regardless of fact that we have a different mapping to
* VNC auth mechs for plain VNC vs websockets VNC, the end
* result has the same security characteristics.
*/
if (websocket || !tlscreds) {
if (password) {
VNC_DEBUG("Initializing VNC server with password auth\n");
*auth = VNC_AUTH_VNC;
} else if (sasl) {
VNC_DEBUG("Initializing VNC server with SASL auth\n");
*auth = VNC_AUTH_SASL;
} else {
VNC_DEBUG("Initializing VNC server with no auth\n");
*auth = VNC_AUTH_NONE;
}
*subauth = VNC_AUTH_INVALID;
} else {
bool is_x509 = object_dynamic_cast(OBJECT(tlscreds),
TYPE_QCRYPTO_TLS_CREDS_X509) != NULL;
bool is_anon = object_dynamic_cast(OBJECT(tlscreds),
TYPE_QCRYPTO_TLS_CREDS_ANON) != NULL;
if (!is_x509 && !is_anon) {
error_setg(errp,
"Unsupported TLS cred type %s",
object_get_typename(OBJECT(tlscreds)));
return -1;
}
*auth = VNC_AUTH_VENCRYPT;
if (password) {
if (is_x509) {
VNC_DEBUG("Initializing VNC server with x509 password auth\n");
*subauth = VNC_AUTH_VENCRYPT_X509VNC;
} else {
VNC_DEBUG("Initializing VNC server with TLS password auth\n");
*subauth = VNC_AUTH_VENCRYPT_TLSVNC;
}
} else if (sasl) {
if (is_x509) {
VNC_DEBUG("Initializing VNC server with x509 SASL auth\n");
*subauth = VNC_AUTH_VENCRYPT_X509SASL;
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
} else {
VNC_DEBUG("Initializing VNC server with TLS SASL auth\n");
*subauth = VNC_AUTH_VENCRYPT_TLSSASL;
}
} else {
if (is_x509) {
VNC_DEBUG("Initializing VNC server with x509 no auth\n");
*subauth = VNC_AUTH_VENCRYPT_X509NONE;
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
} else {
VNC_DEBUG("Initializing VNC server with TLS no auth\n");
*subauth = VNC_AUTH_VENCRYPT_TLSNONE;
}
}
}
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
return 0;
}
static int vnc_display_get_address(const char *addrstr,
bool websocket,
bool reverse,
int displaynum,
int to,
bool has_ipv4,
bool has_ipv6,
bool ipv4,
bool ipv6,
SocketAddress **retaddr,
Error **errp)
{
int ret = -1;
SocketAddress *addr = NULL;
addr = g_new0(SocketAddress, 1);
if (strncmp(addrstr, "unix:", 5) == 0) {
addr->type = SOCKET_ADDRESS_TYPE_UNIX;
addr->u.q_unix.path = g_strdup(addrstr + 5);
if (to) {
error_setg(errp, "Port range not support with UNIX socket");
goto cleanup;
}
ret = 0;
} else {
const char *port;
size_t hostlen;
uint64_t baseport = 0;
InetSocketAddress *inet;
port = strrchr(addrstr, ':');
if (!port) {
if (websocket) {
hostlen = 0;
port = addrstr;
} else {
error_setg(errp, "no vnc port specified");
goto cleanup;
}
} else {
hostlen = port - addrstr;
port++;
if (*port == '\0') {
error_setg(errp, "vnc port cannot be empty");
goto cleanup;
}
}
addr->type = SOCKET_ADDRESS_TYPE_INET;
inet = &addr->u.inet;
if (hostlen && addrstr[0] == '[' && addrstr[hostlen - 1] == ']') {
inet->host = g_strndup(addrstr + 1, hostlen - 2);
} else {
inet->host = g_strndup(addrstr, hostlen);
}
/* plain VNC port is just an offset, for websocket
* port is absolute */
if (websocket) {
if (g_str_equal(addrstr, "") ||
g_str_equal(addrstr, "on")) {
if (displaynum == -1) {
error_setg(errp, "explicit websocket port is required");
goto cleanup;
}
inet->port = g_strdup_printf(
"%d", displaynum + 5700);
if (to) {
inet->has_to = true;
inet->to = to + 5700;
}
} else {
inet->port = g_strdup(port);
}
} else {
int offset = reverse ? 0 : 5900;
if (parse_uint_full(port, 10, &baseport) < 0) {
error_setg(errp, "can't convert to a number: %s", port);
goto cleanup;
}
if (baseport > 65535 ||
baseport + offset > 65535) {
error_setg(errp, "port %s out of range", port);
goto cleanup;
}
inet->port = g_strdup_printf(
"%d", (int)baseport + offset);
if (to) {
inet->has_to = true;
inet->to = to + offset;
}
}
inet->ipv4 = ipv4;
inet->has_ipv4 = has_ipv4;
inet->ipv6 = ipv6;
inet->has_ipv6 = has_ipv6;
ret = baseport;
}
*retaddr = addr;
cleanup:
if (ret < 0) {
qapi_free_SocketAddress(addr);
}
return ret;
}
static int vnc_display_get_addresses(QemuOpts *opts,
bool reverse,
SocketAddressList **saddr_list_ret,
SocketAddressList **wsaddr_list_ret,
Error **errp)
{
SocketAddress *saddr = NULL;
SocketAddress *wsaddr = NULL;
g_autoptr(SocketAddressList) saddr_list = NULL;
SocketAddressList **saddr_tail = &saddr_list;
SocketAddress *single_saddr = NULL;
g_autoptr(SocketAddressList) wsaddr_list = NULL;
SocketAddressList **wsaddr_tail = &wsaddr_list;
QemuOptsIter addriter;
const char *addr;
int to = qemu_opt_get_number(opts, "to", 0);
bool has_ipv4 = qemu_opt_get(opts, "ipv4");
bool has_ipv6 = qemu_opt_get(opts, "ipv6");
bool ipv4 = qemu_opt_get_bool(opts, "ipv4", false);
bool ipv6 = qemu_opt_get_bool(opts, "ipv6", false);
int displaynum = -1;
addr = qemu_opt_get(opts, "vnc");
if (addr == NULL || g_str_equal(addr, "none")) {
return 0;
}
if (qemu_opt_get(opts, "websocket") &&
!qcrypto_hash_supports(QCRYPTO_HASH_ALG_SHA1)) {
error_setg(errp,
"SHA1 hash support is required for websockets");
return -1;
}
qemu_opt_iter_init(&addriter, opts, "vnc");
while ((addr = qemu_opt_iter_next(&addriter)) != NULL) {
int rv;
rv = vnc_display_get_address(addr, false, reverse, 0, to,
has_ipv4, has_ipv6,
ipv4, ipv6,
&saddr, errp);
if (rv < 0) {
return -1;
}
/* Historical compat - first listen address can be used
* to set the default websocket port
*/
if (displaynum == -1) {
displaynum = rv;
}
QAPI_LIST_APPEND(saddr_tail, saddr);
}
if (saddr_list && !saddr_list->next) {
single_saddr = saddr_list->value;
} else {
/*
* If we had multiple primary displays, we don't do defaults
* for websocket, and require explicit config instead.
*/
displaynum = -1;
}
qemu_opt_iter_init(&addriter, opts, "websocket");
while ((addr = qemu_opt_iter_next(&addriter)) != NULL) {
if (vnc_display_get_address(addr, true, reverse, displaynum, to,
has_ipv4, has_ipv6,
ipv4, ipv6,
&wsaddr, errp) < 0) {
return -1;
}
/* Historical compat - if only a single listen address was
* provided, then this is used to set the default listen
* address for websocket too
*/
if (single_saddr &&
single_saddr->type == SOCKET_ADDRESS_TYPE_INET &&
wsaddr->type == SOCKET_ADDRESS_TYPE_INET &&
g_str_equal(wsaddr->u.inet.host, "") &&
!g_str_equal(single_saddr->u.inet.host, "")) {
g_free(wsaddr->u.inet.host);
wsaddr->u.inet.host = g_strdup(single_saddr->u.inet.host);
}
QAPI_LIST_APPEND(wsaddr_tail, wsaddr);
}
*saddr_list_ret = g_steal_pointer(&saddr_list);
*wsaddr_list_ret = g_steal_pointer(&wsaddr_list);
return 0;
}
static int vnc_display_connect(VncDisplay *vd,
SocketAddressList *saddr_list,
SocketAddressList *wsaddr_list,
Error **errp)
{
/* connect to viewer */
QIOChannelSocket *sioc = NULL;
if (wsaddr_list) {
error_setg(errp, "Cannot use websockets in reverse mode");
return -1;
}
if (!saddr_list || saddr_list->next) {
error_setg(errp, "Expected a single address in reverse mode");
return -1;
}
/* TODO SOCKET_ADDRESS_TYPE_FD when fd has AF_UNIX */
vd->is_unix = saddr_list->value->type == SOCKET_ADDRESS_TYPE_UNIX;
sioc = qio_channel_socket_new();
qio_channel_set_name(QIO_CHANNEL(sioc), "vnc-reverse");
if (qio_channel_socket_connect_sync(sioc, saddr_list->value, errp) < 0) {
object_unref(OBJECT(sioc));
return -1;
}
vnc_connect(vd, sioc, false, false);
object_unref(OBJECT(sioc));
return 0;
}
static int vnc_display_listen(VncDisplay *vd,
SocketAddressList *saddr_list,
SocketAddressList *wsaddr_list,
Error **errp)
{
SocketAddressList *el;
if (saddr_list) {
vd->listener = qio_net_listener_new();
qio_net_listener_set_name(vd->listener, "vnc-listen");
for (el = saddr_list; el; el = el->next) {
if (qio_net_listener_open_sync(vd->listener,
el->value, 1,
errp) < 0) {
return -1;
}
}
qio_net_listener_set_client_func(vd->listener,
vnc_listen_io, vd, NULL);
}
if (wsaddr_list) {
vd->wslistener = qio_net_listener_new();
qio_net_listener_set_name(vd->wslistener, "vnc-ws-listen");
for (el = wsaddr_list; el; el = el->next) {
if (qio_net_listener_open_sync(vd->wslistener,
el->value, 1,
errp) < 0) {
return -1;
}
}
qio_net_listener_set_client_func(vd->wslistener,
vnc_listen_io, vd, NULL);
}
return 0;
}
bool vnc_display_update(DisplayUpdateOptionsVNC *arg, Error **errp)
{
VncDisplay *vd = vnc_display_find(NULL);
if (!vd) {
error_setg(errp, "Can not find vnc display");
return false;
}
if (arg->has_addresses) {
if (vd->listener) {
qio_net_listener_disconnect(vd->listener);
object_unref(OBJECT(vd->listener));
vd->listener = NULL;
}
if (vnc_display_listen(vd, arg->addresses, NULL, errp) < 0) {
return false;
}
}
return true;
}
void vnc_display_open(const char *id, Error **errp)
{
VncDisplay *vd = vnc_display_find(id);
QemuOpts *opts = qemu_opts_find(&qemu_vnc_opts, id);
g_autoptr(SocketAddressList) saddr_list = NULL;
g_autoptr(SocketAddressList) wsaddr_list = NULL;
const char *share, *device_id;
QemuConsole *con;
bool password = false;
bool reverse = false;
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
const char *credid;
bool sasl = false;
vnc: allow specifying a custom authorization object name The VNC server has historically had support for ACLs to check both the SASL username and the TLS x509 distinguished name. The VNC server was responsible for creating the initial ACL, and the client app was then responsible for populating it with rules using the HMP 'acl_add' command. This is not satisfactory for a variety of reasons. There is no way to populate the ACLs from the command line, users are forced to use the HMP. With multiple network services all supporting TLS and ACLs now, it is desirable to be able to define a single ACL that is referenced by all services. To address these limitations, two new options are added to the VNC server CLI. The 'tls-authz' option takes the ID of a QAuthZ object to use for checking TLS x509 distinguished names, and the 'sasl-authz' option takes the ID of another object to use for checking SASL usernames. In this example, we setup two authorization rules. The first allows any client with a certificate issued by the 'RedHat' organization in the 'London' locality. The second ACL allows clients with either the 'joe@REDHAT.COM' or 'fred@REDHAT.COM' kerberos usernames. Both checks must pass for the user to be allowed. $QEMU -object tls-creds-x509,id=tls0,dir=/home/berrange/qemutls,\ endpoint=server,verify-peer=yes \ -object authz-simple,id=authz0,policy=deny,\ rules.0.match=O=RedHat,,L=London,rules.0.policy=allow \ -object authz-simple,id=authz1,policy=deny,\ rules.0.match=fred@REDHAT.COM,rules.0.policy=allow \ rules.0.match=joe@REDHAT.COM,rules.0.policy=allow \ -vnc 0.0.0.0:1,tls-creds=tls0,tls-authz=authz0, sasl,sasl-authz=authz1 \ ...other QEMU args... Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 20190227145755.26556-2-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2019-02-27 17:57:54 +03:00
const char *tlsauthz;
const char *saslauthz;
int lock_key_sync = 1;
int key_delay_ms;
const char *audiodev;
const char *passwordSecret;
if (!vd) {
error_setg(errp, "VNC display not active");
return;
}
vnc_display_close(vd);
if (!opts) {
return;
}
reverse = qemu_opt_get_bool(opts, "reverse", false);
if (vnc_display_get_addresses(opts, reverse, &saddr_list, &wsaddr_list,
errp) < 0) {
goto fail;
}
passwordSecret = qemu_opt_get(opts, "password-secret");
if (passwordSecret) {
if (qemu_opt_get(opts, "password")) {
error_setg(errp,
"'password' flag is redundant with 'password-secret'");
goto fail;
}
vd->password = qcrypto_secret_lookup_as_utf8(passwordSecret,
errp);
if (!vd->password) {
goto fail;
}
password = true;
} else {
password = qemu_opt_get_bool(opts, "password", false);
}
if (password) {
if (!qcrypto_cipher_supports(
QCRYPTO_CIPHER_ALG_DES, QCRYPTO_CIPHER_MODE_ECB)) {
error_setg(errp,
"Cipher backend does not support DES algorithm");
goto fail;
}
}
lock_key_sync = qemu_opt_get_bool(opts, "lock-key-sync", true);
key_delay_ms = qemu_opt_get_number(opts, "key-delay-ms", 10);
sasl = qemu_opt_get_bool(opts, "sasl", false);
#ifndef CONFIG_VNC_SASL
if (sasl) {
error_setg(errp, "VNC SASL auth requires cyrus-sasl support");
goto fail;
}
#endif /* CONFIG_VNC_SASL */
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
credid = qemu_opt_get(opts, "tls-creds");
if (credid) {
Object *creds;
creds = object_resolve_path_component(
object_get_objects_root(), credid);
if (!creds) {
error_setg(errp, "No TLS credentials with id '%s'",
credid);
goto fail;
}
vd->tlscreds = (QCryptoTLSCreds *)
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
object_dynamic_cast(creds,
TYPE_QCRYPTO_TLS_CREDS);
if (!vd->tlscreds) {
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
error_setg(errp, "Object with id '%s' is not TLS credentials",
credid);
goto fail;
}
object_ref(OBJECT(vd->tlscreds));
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
if (!qcrypto_tls_creds_check_endpoint(vd->tlscreds,
QCRYPTO_TLS_CREDS_ENDPOINT_SERVER,
errp)) {
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
goto fail;
}
}
vnc: allow specifying a custom authorization object name The VNC server has historically had support for ACLs to check both the SASL username and the TLS x509 distinguished name. The VNC server was responsible for creating the initial ACL, and the client app was then responsible for populating it with rules using the HMP 'acl_add' command. This is not satisfactory for a variety of reasons. There is no way to populate the ACLs from the command line, users are forced to use the HMP. With multiple network services all supporting TLS and ACLs now, it is desirable to be able to define a single ACL that is referenced by all services. To address these limitations, two new options are added to the VNC server CLI. The 'tls-authz' option takes the ID of a QAuthZ object to use for checking TLS x509 distinguished names, and the 'sasl-authz' option takes the ID of another object to use for checking SASL usernames. In this example, we setup two authorization rules. The first allows any client with a certificate issued by the 'RedHat' organization in the 'London' locality. The second ACL allows clients with either the 'joe@REDHAT.COM' or 'fred@REDHAT.COM' kerberos usernames. Both checks must pass for the user to be allowed. $QEMU -object tls-creds-x509,id=tls0,dir=/home/berrange/qemutls,\ endpoint=server,verify-peer=yes \ -object authz-simple,id=authz0,policy=deny,\ rules.0.match=O=RedHat,,L=London,rules.0.policy=allow \ -object authz-simple,id=authz1,policy=deny,\ rules.0.match=fred@REDHAT.COM,rules.0.policy=allow \ rules.0.match=joe@REDHAT.COM,rules.0.policy=allow \ -vnc 0.0.0.0:1,tls-creds=tls0,tls-authz=authz0, sasl,sasl-authz=authz1 \ ...other QEMU args... Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 20190227145755.26556-2-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2019-02-27 17:57:54 +03:00
tlsauthz = qemu_opt_get(opts, "tls-authz");
if (tlsauthz && !vd->tlscreds) {
error_setg(errp, "'tls-authz' provided but TLS is not enabled");
goto fail;
}
saslauthz = qemu_opt_get(opts, "sasl-authz");
if (saslauthz && !sasl) {
error_setg(errp, "'sasl-authz' provided but SASL auth is not enabled");
goto fail;
}
share = qemu_opt_get(opts, "share");
if (share) {
if (strcmp(share, "ignore") == 0) {
vd->share_policy = VNC_SHARE_POLICY_IGNORE;
} else if (strcmp(share, "allow-exclusive") == 0) {
vd->share_policy = VNC_SHARE_POLICY_ALLOW_EXCLUSIVE;
} else if (strcmp(share, "force-shared") == 0) {
vd->share_policy = VNC_SHARE_POLICY_FORCE_SHARED;
} else {
error_setg(errp, "unknown vnc share= option");
goto fail;
}
} else {
vd->share_policy = VNC_SHARE_POLICY_ALLOW_EXCLUSIVE;
}
vd->connections_limit = qemu_opt_get_number(opts, "connections", 32);
#ifdef CONFIG_VNC_JPEG
vd->lossy = qemu_opt_get_bool(opts, "lossy", false);
#endif
vd->non_adaptive = qemu_opt_get_bool(opts, "non-adaptive", false);
/* adaptive updates are only used with tight encoding and
* if lossy updates are enabled so we can disable all the
* calculations otherwise */
if (!vd->lossy) {
vd->non_adaptive = true;
}
vd->power_control = qemu_opt_get_bool(opts, "power-control", false);
vnc: allow specifying a custom authorization object name The VNC server has historically had support for ACLs to check both the SASL username and the TLS x509 distinguished name. The VNC server was responsible for creating the initial ACL, and the client app was then responsible for populating it with rules using the HMP 'acl_add' command. This is not satisfactory for a variety of reasons. There is no way to populate the ACLs from the command line, users are forced to use the HMP. With multiple network services all supporting TLS and ACLs now, it is desirable to be able to define a single ACL that is referenced by all services. To address these limitations, two new options are added to the VNC server CLI. The 'tls-authz' option takes the ID of a QAuthZ object to use for checking TLS x509 distinguished names, and the 'sasl-authz' option takes the ID of another object to use for checking SASL usernames. In this example, we setup two authorization rules. The first allows any client with a certificate issued by the 'RedHat' organization in the 'London' locality. The second ACL allows clients with either the 'joe@REDHAT.COM' or 'fred@REDHAT.COM' kerberos usernames. Both checks must pass for the user to be allowed. $QEMU -object tls-creds-x509,id=tls0,dir=/home/berrange/qemutls,\ endpoint=server,verify-peer=yes \ -object authz-simple,id=authz0,policy=deny,\ rules.0.match=O=RedHat,,L=London,rules.0.policy=allow \ -object authz-simple,id=authz1,policy=deny,\ rules.0.match=fred@REDHAT.COM,rules.0.policy=allow \ rules.0.match=joe@REDHAT.COM,rules.0.policy=allow \ -vnc 0.0.0.0:1,tls-creds=tls0,tls-authz=authz0, sasl,sasl-authz=authz1 \ ...other QEMU args... Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 20190227145755.26556-2-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2019-02-27 17:57:54 +03:00
if (tlsauthz) {
vd->tlsauthzid = g_strdup(tlsauthz);
}
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_SASL
vnc: allow specifying a custom authorization object name The VNC server has historically had support for ACLs to check both the SASL username and the TLS x509 distinguished name. The VNC server was responsible for creating the initial ACL, and the client app was then responsible for populating it with rules using the HMP 'acl_add' command. This is not satisfactory for a variety of reasons. There is no way to populate the ACLs from the command line, users are forced to use the HMP. With multiple network services all supporting TLS and ACLs now, it is desirable to be able to define a single ACL that is referenced by all services. To address these limitations, two new options are added to the VNC server CLI. The 'tls-authz' option takes the ID of a QAuthZ object to use for checking TLS x509 distinguished names, and the 'sasl-authz' option takes the ID of another object to use for checking SASL usernames. In this example, we setup two authorization rules. The first allows any client with a certificate issued by the 'RedHat' organization in the 'London' locality. The second ACL allows clients with either the 'joe@REDHAT.COM' or 'fred@REDHAT.COM' kerberos usernames. Both checks must pass for the user to be allowed. $QEMU -object tls-creds-x509,id=tls0,dir=/home/berrange/qemutls,\ endpoint=server,verify-peer=yes \ -object authz-simple,id=authz0,policy=deny,\ rules.0.match=O=RedHat,,L=London,rules.0.policy=allow \ -object authz-simple,id=authz1,policy=deny,\ rules.0.match=fred@REDHAT.COM,rules.0.policy=allow \ rules.0.match=joe@REDHAT.COM,rules.0.policy=allow \ -vnc 0.0.0.0:1,tls-creds=tls0,tls-authz=authz0, sasl,sasl-authz=authz1 \ ...other QEMU args... Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 20190227145755.26556-2-berrange@redhat.com Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2019-02-27 17:57:54 +03:00
if (sasl) {
if (saslauthz) {
vd->sasl.authzid = g_strdup(saslauthz);
}
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
if (vnc_display_setup_auth(&vd->auth, &vd->subauth,
vd->tlscreds, password,
sasl, false, errp) < 0) {
goto fail;
}
trace_vnc_auth_init(vd, 0, vd->auth, vd->subauth);
if (vnc_display_setup_auth(&vd->ws_auth, &vd->ws_subauth,
vd->tlscreds, password,
sasl, true, errp) < 0) {
ui: convert VNC server to use QCryptoTLSSession Switch VNC server over to using the QCryptoTLSSession object for the TLS session. This removes the direct use of gnutls from the VNC server code. It also removes most knowledge about TLS certificate handling from the VNC server code. This has the nice effect that all the CONFIG_VNC_TLS conditionals go away and the user gets an actual error message when requesting TLS instead of it being silently ignored. With this change, the existing configuration options for enabling TLS with -vnc are deprecated. Old syntax for anon-DH credentials: -vnc hostname:0,tls New syntax: -object tls-creds-anon,id=tls0,endpoint=server \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, no client certs: -vnc hostname:0,tls,x509=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=no \ -vnc hostname:0,tls-creds=tls0 Old syntax for x509 credentials, requiring client certs: -vnc hostname:0,tls,x509verify=/path/to/certs New syntax: -object tls-creds-x509,id=tls0,dir=/path/to/certs,endpoint=server,verify-peer=yes \ -vnc hostname:0,tls-creds=tls0 This aligns VNC with the way TLS credentials are to be configured in the future for chardev, nbd and migration backends. It also has the benefit that the same TLS credentials can be shared across multiple VNC server instances, if desired. If someone uses the deprecated syntax, it will internally result in the creation of a 'tls-creds' object with an ID based on the VNC server ID. This allows backwards compat with the CLI syntax, while still deleting all the original TLS code from the VNC server. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-08-06 16:39:32 +03:00
goto fail;
}
trace_vnc_auth_init(vd, 1, vd->ws_auth, vd->ws_subauth);
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 (sasl && !vnc_sasl_server_init(errp)) {
goto fail;
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
vd->lock_key_sync = lock_key_sync;
if (lock_key_sync) {
vd->led = qemu_add_led_event_handler(kbd_leds, vd);
}
vd->ledstate = 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
audiodev = qemu_opt_get(opts, "audiodev");
if (audiodev) {
vd->audio_state = audio_state_by_name(audiodev, errp);
if (!vd->audio_state) {
goto fail;
}
} else {
vd->audio_state = audio_get_default_audio_state(NULL);
}
device_id = qemu_opt_get(opts, "display");
if (device_id) {
int head = qemu_opt_get_number(opts, "head", 0);
Error *err = NULL;
con = qemu_console_lookup_by_device_name(device_id, head, &err);
if (err) {
error_propagate(errp, err);
goto fail;
}
} else {
con = qemu_console_lookup_default();
}
if (con != vd->dcl.con) {
qkbd_state_free(vd->kbd);
unregister_displaychangelistener(&vd->dcl);
vd->dcl.con = con;
register_displaychangelistener(&vd->dcl);
vd->kbd = qkbd_state_init(vd->dcl.con);
}
qkbd_state_set_delay(vd->kbd, key_delay_ms);
if (saddr_list == NULL) {
return;
}
if (reverse) {
if (vnc_display_connect(vd, saddr_list, wsaddr_list, errp) < 0) {
goto fail;
}
} else {
if (vnc_display_listen(vd, saddr_list, wsaddr_list, errp) < 0) {
goto fail;
}
}
if (qemu_opt_get(opts, "to")) {
vnc_display_print_local_addr(vd);
}
/* Success */
return;
fail:
vnc_display_close(vd);
}
void vnc_display_add_client(const char *id, int csock, bool skipauth)
{
VncDisplay *vd = vnc_display_find(id);
QIOChannelSocket *sioc;
if (!vd) {
return;
}
sioc = qio_channel_socket_new_fd(csock, NULL);
if (sioc) {
qio_channel_set_name(QIO_CHANNEL(sioc), "vnc-server");
vnc_connect(vd, sioc, skipauth, false);
object_unref(OBJECT(sioc));
}
}
static void vnc_auto_assign_id(QemuOptsList *olist, QemuOpts *opts)
{
int i = 2;
char *id;
id = g_strdup("default");
while (qemu_opts_find(olist, id)) {
g_free(id);
id = g_strdup_printf("vnc%d", i++);
}
qemu_opts_set_id(opts, id);
}
void vnc_parse(const char *str)
{
QemuOptsList *olist = qemu_find_opts("vnc");
QemuOpts *opts = qemu_opts_parse_noisily(olist, str, !is_help_option(str));
const char *id;
if (!opts) {
exit(1);
}
id = qemu_opts_id(opts);
if (!id) {
/* auto-assign id if not present */
vnc_auto_assign_id(olist, opts);
}
}
int vnc_init_func(void *opaque, QemuOpts *opts, Error **errp)
{
Error *local_err = NULL;
char *id = (char *)qemu_opts_id(opts);
assert(id);
vnc_display_init(id, &local_err);
if (local_err) {
error_propagate(errp, local_err);
return -1;
}
vnc_display_open(id, &local_err);
if (local_err != NULL) {
error_propagate(errp, local_err);
return -1;
}
return 0;
}
static void vnc_register_config(void)
{
qemu_add_opts(&qemu_vnc_opts);
}
opts_init(vnc_register_config);