Most dynamic channels share the same copied and pasted code for the XXX_CHANNEL_CALLBACK
and XXX_LISTENER_CALLBACK types. This patch introduce GENERIC_CHANNEL_CALLBACK and
GENERIC_LISTENER_CALLBACK that fits for most channels and discard custom type definitions.
The current server sided channel handling of RDPSND/AUDIO_PLAYBACK_DVC
is currently very constrained.
So, solve this. This means:
- Add the missing Training/Training Confirm PDUs
- Stop overriding the average bytes per second values, when submitting
the audio formats, as this currently makes the usage of codecs
impossible
- Add a way to send the server formats manually again, to be able to
restart the protocol after a Close PDU was sent
- Add a way to send already encoded audio data to let server
implementations to take care of the encoding process and to set
custom audio timestamps for the Video Optimized Remoting channel
- Add public attributes to let server implementations know the initial
volume and pitch values
- Add public attribute to let server implementations know the quality
mode setting
Currently, all FreeRDP-based clients don't send any WaveConfirm PDUs
for received samples, when using a dynamic channel for audio output
redirection.
[MS-RDPEA] 2.2.3.8 Wave Confirm PDU mentions, that a WaveConfirm PDU
MUST be sent, when a WaveInfo PDU + Wave PDU, or Wave2 PDU is received
and when the audio data sample is emitted to completion by the client.
The first WaveConfirm PDU is used by the server to determine the
network latency and the second WaveConfirm PDU is used by the server to
determine the render latency.
So, fix the current behaviour, where FreeRDP currently does not send
any WaveConfirm PDU, when using the dynamic channel, or only sends one
WaveConfirm PDU for the sample.
For the first WaveConfirm PDU, use the same timestamp, that was
included in the first WaveInfo/Wave2 PDU.
For the second WaveConfirm PDU, add the render latency on top of the
arrival timestamp.
_align_XXX functions aren't widely available depending on the C runtime. That causes
problems with mingw where we can easily have some runtime mixes (which lead to mysterious
segfaults most of the time). This patch introduce winpr_aligned_XXX functions that will
either use the function available, or use an emulation layer.
* Fixed remdesk settings pointer
* Fixed sign warnings in display_write_monitor_layout_pdu
* Use freerdp_abort_connect_context and freerdp_shall_disconnect_context
* Added and updates settings
* info assert/dynamic timezone
* mcs assert/log/flags
* Fixed and added assertions for wStream
* Unified stream length checks
* Added new function to check for lenght and log
* Replace all usages with this new function
* Cleaned up PER, added parser logging
* Cleaned up BER, added parser logging
* log messages
* Modified Stream_CheckAndLogRequiredLengthEx
* Allow custom format and options
* Add Stream_CheckAndLogRequiredLengthExVa for prepared va_list
* Improved Stream_CheckAndLogRequiredLength
* Now have log level adjustable
* Added function equivalents for existing logger
* Added a backtrace in case of a failure is detected
* Fixed public API input checks
* Fixed#7753: Step calculation in YUV decoder
* Fixed rectangle_is_empty
* Added rectangle intersection check in YUV decoder
* Skip intersecting rectangles in YUV decoder
* Refactored YUV work object handling
* Allocate / free in yuv_context_new and yuv_context_free
* WINPR_ASSERT all function arguments
* Pass yuv_context_reset result from h264_context_reset
* Reset PTP_WORK buffer to NULL after use
Building client channels out of tree does not work reliable as the
dependencies can not be easily split. (exceptions may be simple ones
as echo channel)
It does complicate the build system and code though, so remove this
for maintainability reasons.
The server sided handling of the display control channel currently
treats the physical monitor size values as obligatory values.
When the physical monitor size is invalid according to the
documentation, it is supposed to be ignored.
However, this is currently not done, leading to an abort of the display
control channel handling, when using Remmina, as Remmina does not submit
physical-monitor- widths and -heights.
Fix this issue by sanitizing the physical monitor size values, instead
of handling them as obligatory values, where both the physical with and
the physical height must be constrained to the values in the
documentation.
{channels/tsmf/client/ffmpeg/tsmf,libfreerdp/codec/dsp}_ffmpeg.c:
Guard calls to `avcodec_register_all()` against use beyond
`AV_VERSION_INT(58, 10, 100)`, where upstream ffmpeg made it obsolete.
* fix libusb libusb_device usage (manually unref required usb devices, versus freeing all when we still hold references to the ones we want)
* disabled detach_kernel_driver & attach_kernel_driver on win32 since libusb does not support them
* fixed libusb async event handling
* add log for transfer request error
* Update libusb_udevice.c
* refactor code
Add support for new graphics redirection protocol (aka VAIL) used by
WSLg. This protocol is currently being documented and the official
documentation will be posted in the near future.
* All channels inheriting from RDPDR_DRIVE base struct
* Add functions to create/free a new device of a type
* Fixed settings array resize code, prevent 0 sizes
* Properly initialize the stream buffer
* Add Stream_StaticConstInit accepting a const buffer
* Modify API to return a pointer to the stream initialized
* Do length/capacity checks in every read/write/seek function
if WINPR_ASSERT is defined.
* Ensure s->pointer is valid, e.g. within s->buffer + s->capacity
(Stream_Rewind, Stream_Seek, ...)
* Add return values to Stream_Set* functions so inalid arguments
can be reported to the caller
* Deprecated problematic stream manipulation functions
(Stream_SetBuffer, Stream_SetPointer, Stream_SetCapacity)
* Ensure length/capacity functions never return a value larger
than the actual length/capacity
When the server supports the protocol version RDPINPUT_PROTOCOL_V300,
the additional supportedFeatures field will be present.
The pduLength in the RDPINPUT_HEADER should, however, reflect this.
So, fix this error by writing the correct PDU length when the
supportedFeatures field is present.
* Split out proxy headers and moved to public API to allow external
modules to be built.
* Split proxy into proxy library and proxy binary. The library
can be used by other applications and provides a simple API
* Improved channel passthrough, now all channels including dynamic
channels work.
* Extended module API to hook more events, improved module samples
* Cleaned up proxy code, removed global static variables used,
added WINPR_ASSERT
RDPSND channel is special, as it has many names.
(e.g. static channel, dynamic channel and UDP one.
Use RDPSND_CHANNEL_NAME to identify the module name instad of
RDPSND_DVC_CHANNEL_NAME
If -DBUILTIN_CHANNELS=OFF is supplied to cmake, auxiliary channel modules
are built as plugins, that are loaded from $LIBDIR/freerdp as shared libs.
This patch fixes the linkage of these plugins by taking inter-channel dependencies
of this separation into account. If BUILTIN_CHANNELS=ON, these channel modules
are linked directly into the main modules. Therefore, the linking of the plugins
is conditioned accordingly.
* Fixed research for native PCM format
* Fixed compatibility detection for resampling
* dsp: Makes channel mix and encoding process use separated buffers
* Give FramesPerPacket to freerdp_dsp_context_reset function
* dsp: Fix DVI_ADPCM codec by respecting the required packet size
* Forbid mono channel in macos
* Fix audin command line channel option
* Disable ADPCM codec as it does not work
* Refactor valid native codec research
* Native codec research: Try different channels number
* Fix dsp channel mix
* Fix dsp resample
* Monitor coordinates are exclusive
* Remove force override of shadow resolution.
The client might ignore the server requested values, in that case
retry
* Added new function pointers to cancel a pending operation
* Added new function pointers to send custom event flags
* Added exposed feature mask to disable channel features
Covscan report contains various memory leak defects which were marked
as important. I have spent some time analyzing them and although they
were marked as important, most of them are in error cases, so probably
nothing serious. Let's fix most of them anyway. The rest are false
positives, or too complicated to fix, or already fixed in master, or
simply I am unsure about them.
Relates: https://github.com/FreeRDP/FreeRDP/issues/6981
In #6821 it has been reported that there are buggy smartcard
drivers that report a string size but fail to allocate the string
itself. This check avoids a crash with such input parameters
(some client side channels and all server side channels still need to be
ported to new api)
server: build fix, do not disable threads for rfx encoder
cliprdr client channel: implemented support for DisableThreads option
looks like thread does not make sense at all for this channel
do not initialize disabled image codecs (respect settings)
channels: client: rail: added support for DisableThreads setting
changed "BOOL DisableThreads" to "UINT32 ThreadingFlags"
dropped unnecessary apu changes
draft implementation of threading settings aware message handling api
for addins/channels
rail: use new messaging api
fixed memory leak
msgs handlers external api changes (as requested)
msgs_handlers: init fix
fixed memory leak
logic fix
resolved problems appeared after rebase to master, dropped unnecessary
changes
git clang-format origin/master
fixed TestFreeRDPCodecRemoteFX.c
"formatting, run `clang-format` please"
properly use new "rfx_context_new(BOOL, UINT32)" everywhere
passed Threading Flags to "rfx_context_new" where available
in older C standarts veriables declaration must be done before any code
requested changes
clang-format as requested
use broken signatures of standert C functions for m$ s**tos
clang-format
requested changes
requested changes
moved ThreadingFlags to stable api zone
define type for channel msg handler
typo fix
clang-format
build fix
us ThreadingFlags from server settings
git clang-format origin/master
clang-format
some hotplug implementations report the same drive multiple times.
to avoid redirecting the same drive multiple times check if it is
already in the list before adding.
The path was incorrectly cut of after the drive letter causing hot
plugged drives to show up empty and disconnecting the client if the
device was removed later on.
