Remove an alignment warning
vdagent/vdagent.cpp: In member function 'bool VDAgent::handle_announce_capabilities(const VDAgentAnnounceCapabilities*, uint32_t)':
vdagent/vdagent.cpp:1048:48: warning: taking address of packed member of 'VDAgentAnnounceCapabilities' may result in an unaligned pointer value [-Waddress-of-packed-member]
1048 | _client_caps.assign(announce_capabilities->caps, announce_capabilities->caps + caps_size);
| ~~~~~~~~~~~~~~~~~~~~~~~^~~~
Not a real issue, just annoying.
Signed-off-by: Frediano Ziglio <freddy77@gmail.com>
When the user wants to copy files, new spice-gtk can share those
files using the existing phodav server. In that case, it advertises
the VD_AGENT_CLIPBOARD_FILE_LIST type in the clipboard grab message.
Upon request of the clipboard data in the mentioned type, spice-gtk
provides a list of absolute paths in the phodav server - these are
the files that are supposed to be copied/moved.
The role of the vdagent is to adjust this data given the drive
letter of the mapped webdav share.
Files can be both copied and moved, although move was tested
only with Windows' File Explorer.
Copying files from the vdagnet side to the client is not
supported yet.
Signed-off-by: Jakub Janků <jjanku@redhat.com>
Acked-by: Frediano Ziglio <fziglio@redhat.com>
If the client closes the connection while having pending large messages
(split into multiple chunks) the state is not correctly reset causing
next client messages to be not correctly handled.
This can occur for instance if the client is closed (or killed) while
transferring a large file.
This fixes https://bugzilla.redhat.com/show_bug.cgi?id=1548419.
Signed-off-by: Frediano Ziglio <freddy77@gmail.com>
Acked-by: Victor Toso <victortoso@redhat.com>
The client is sending messages that are compatible with agent
capabilities but also encoded with capabilities it knows of so to
check them we need to consider final capabilities, not client ones.
This fixes a regression introduced with
"vdagent: Reuse agent_check_message" where some messages were wrongly
interpreted as wrong (particularly some clipboard messages).
Signed-off-by: Frediano Ziglio <freddy77@gmail.com>
Acked-by: Jakub Janků <jjanku@redhat.com>
Revert "try to remove other bugs", commit
57266085b9.
Revert "try to fix previous", commit
8e40377068.
Revert "attempt to fix more warnings!", commit
66b6933a38.
I put by mistake on the MR due to some automatic scripts.
Their state are pretty experimental.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Have the same behaviour of Linux machines.
The directory is not opened again and again, only is not already
opened.
This is done checking the directory opened in explorer.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
In 8251fa25, a check on the minimum size of a message was introduced.
For unsupported messages, the vdagent simply exited. This makes it
inconsistent with previous behavior and inconsistent with the behavior
of the linux vdagent. Instead, just print a warning indicating that an
unsupported message was received and ignore it.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Acked-by: Frediano Ziglio <fziglio@redhat.com>
The library is surely already loaded as providing clipboard and
other utilities we need.
user32 is one of the main win32 libraries.
Avoid using LoadLibrary that increment the reference so we don't
need to call FreeLibrary to cleanly decrement the reference.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
Multiple times while I was reading this code was not clear why the
synchronization of this event was not done in the main event loop.
Also document why we want to use it not relying just on
WTSRegisterSessionNotification and WM_WTSSESSION_CHANGE.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
The thread launched to detect desktop switches events is not stopped
correctly causing potentially dandling pointers.
Queue a APC to make the loop exit and wait for thread termination
from the main thread.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
Do not assume we allocated the handle doing the cleanup.
This utility will be reused in next patch.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe de Dinechin <dinechin@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
For some reason kernel handles in Windows have 2 invalid values
depending on the type. Files/devices use INVALID_HANDLE_VALUE (-1),
while others use NULL (0). As _vio_serial is a file, created
with CreateFile, use INVALID_HANDLE_VALUE for consistency.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
The only reason we call this function is to check if the
system should support some APIs.
Instead just check directly if these APIs are supported
calling GetProcAddress directly.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Jonathon Jongsma <jjongsma@redhat.com>
The check can be done a single time, no reason to cache
supported_system_version() value.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Jonathon Jongsma <jjongsma@redhat.com>
As the value is never used we can pass NULL in CreateThread.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Jonathon Jongsma <jjongsma@redhat.com>
Handle _control_event and _stop_event destruction in destructor.
Allows to call VDAgent::run multiple time if needed.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Jonathon Jongsma <jjongsma@redhat.com>
Make automatic the release of this pointer.
Also avoids having a leak if VDAgent::run is called twice.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Jonathon Jongsma <jjongsma@redhat.com>
Although source of these data should be safe, improve data checks
to avoid some overflows and make the code more robust.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Jonathon Jongsma <jjongsma@redhat.com>
send_input() may not be immediately called from handle_mouse_event() on
movement. INPUT structure is generated and stored and a timer may be set
instead. If subsequent call to handle_mouse_event() occurs before timer
expires, prepared INPUT structure gets overwritten and MOUSEEVENTF_MOVE
bit is lost. Windows doesn't see updated mouse position as the result.
Make handle_mouse_event() merely store the new mouse state, and move
INPUT structure generation to send_input(). Shuffle new mouse state to
previous only after mouse events are submitted to SendInput() Windows
API function.
This patch was sent to the mailing list by an anonymous contributor
with minimal style changes.
You can easily test increasing VD_INPUT_INTERVAL_MS (like 1000).
For instance you can try in a word processor to move the cursor
clicking the mouse on different positions.
Acked-by: Victor Toso <victortoso@redhat.com>
This will make easier to change code that handle images.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
Previously, when a client disconnects, the vdagent executable exited.
The agent was restarted immediately in order to be ready for the next
client to reconnect.
This is generally fine, but there are certain situations
where it causes issues. For example, if client A is connected to a guest
and client B connects to the same guest, client A will be forcibly
disconnected and causes the the vdagent to restart. This scenario is
racy because the agent can take some time to exit and restart. Client B
may think that the agent is connected at startup and may send agent
messages to the guest. At some point the server will recieve
notification that the agent has exited and send an AGENT_DISCONNECTED
message to client B. After the agent has been fully restarted, an
AGENT_CONNECTED message will be sent to the client, but any messages
sent between client connection and the AGENT_DISCONNECTED message will
be lost. This causes problems especially with fullscreen mode in
virt-viewer.
The solution is to not exit and restart the agent when a client
disconnects. This is how the linux vdagent behaves. Instead, we simply
cancel all ongoing file transfers and reset the clipboard when a client
is disconnected.
Fixes: rhbz#1064495
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Acked-by: Pavel Grunt <pgrunt@redhat.com>
Make clipboard_formats and image_types static and constant.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
This patch updates the display configurations only when needed.
Currently vdagent updates the display configurations almost on every
"VD_AGENT_MONITORS_CONFIG" request, this approach is redundant as some
"VD_AGENT_MONITORS_CONFIG" requests don't apply any change to the
display configurations.
A good example of why this is needed, is the 'resize-guest' feature of
the spice-widget which when enabled causes the client to send periodic
"VD_AGENT_MONITORS_CONFIG" requests in order to resize the guest even
when no change is required. Prior to this patch the screen would
periodically flicker when using Virt-manager or Spicy due to this issue.
Signed-off-by: Sameeh Jubran <sameeh@daynix.com>
Acked-by: Frediano Ziglio <fziglio@redhat.com>
As _stop_event is not mandatory, we are initializing the events array
with something that might be NULL. The problem is with the following
patch as I'm introducing a new non mandatory event and logic starts to
get a bit hard to follow.
So this patch makes explicit if we are setting or receiving a
_stop_event or not and uses an auxiliary enum and array 'actions' to
work around this non mandatory events and the return value of
MsgWaitForMultipleObjectsEx()
Signed-off-by: Victor Toso <victortoso@redhat.com>
Acked-by: Frediano Ziglio <fziglio@redhat.com>
