Commit Graph

1773 Commits (88543cb5453f74ac611759ff483b6a413f367e1b)
 

Author SHA1 Message Date
Romain Vimont 0ee9e2ff51 Expose function to find a USB device
The device was automatically found by sc_usb_connect(). Instead, expose
a function to find a device from a serial, and let the caller connect to
the device found (if any).

This will allow to list all devices first, then select one device to
connect to.

PR #2974 <https://github.com/Genymobile/scrcpy/pull/2974>
2 years ago
Romain Vimont 1ab3692f3d Add util function to read USB descriptor string
Use it from accept_device() to simplify (at the cost an additional
allocation for each serial, but it is not important).

It will also be useful in other functions in further commits.

PR #2974 <https://github.com/Genymobile/scrcpy/pull/2974>
2 years ago
Romain Vimont bbef426a4b Split USB initialization and connection
This will allow to execute other USB calls (retrieving the device list
for example) before connecting to the selected device.

PR #2974 <https://github.com/Genymobile/scrcpy/pull/2974>
2 years ago
Romain Vimont 2114f48185 Find device with USB context
An explicit context was used everywhere except for listing the devices.

PR #2974 <https://github.com/Genymobile/scrcpy/pull/2974>
2 years ago
Romain Vimont b779eca8d3 Remove libusb_device field
It is possible to retrieve the device instance from the handle via
libusb_get_device(), so we don't need to reference the device one more
time.

PR #2974 <https://github.com/Genymobile/scrcpy/pull/2974>
2 years ago
Romain Vimont adda47b0f7 Move sc_usb out of sc_aoa
This will allow to initialize a USB device separately and pass it to
sc_aoa.

PR #2974 <https://github.com/Genymobile/scrcpy/pull/2974>
2 years ago
Romain Vimont 48e3ff284f Make serial mandatory for sc_usb
In practice, it is already mandatory.

PR #2974 <https://github.com/Genymobile/scrcpy/pull/2974>
2 years ago
Romain Vimont 1d6f9952ee Extract USB handling from AOA
The AOA code handled both USB initialization and AOA commands/events.
Extract USB-related code to a separate file and structure.

PR #2974 <https://github.com/Genymobile/scrcpy/pull/2974>
2 years ago
Romain Vimont d48d191262 Rename HAVE_AOA_HID to HAVE_USB
The condition actually determines whether scrcpy can use libusb or not.

PR #2974 <https://github.com/Genymobile/scrcpy/pull/2974>
2 years ago
Romain Vimont 2762f5d183 Move AOA/HID code to usb/
PR #2974 <https://github.com/Genymobile/scrcpy/pull/2974>
2 years ago
Romain Vimont c996a6d462 Fix socket close race condition
The server needs to interrupt the sockets on stop, but it must not close
them while other threads may attempt to read from or write to them.

In particular, the video_socket is read by the stream thread, and the
control_socket is written by the controller and read by receiver.

Therefore, close the socket only on sc_server_destroy(), which is called
after all other threads are joined.

Reported by TSAN on close:

    WARNING: ThreadSanitizer: data race (pid=3287612)
      Write of size 8 at 0x7ba000000080 by thread T1:
        #0 close ../../../../src/libsanitizer/tsan/tsan_interceptors_posix.cpp:1690 (libtsan.so.0+0x359d8)
        #1 net_close ../app/src/util/net.c:280 (scrcpy+0x23643)
        #2 run_server ../app/src/server.c:772 (scrcpy+0x20047)
        #3 <null> <null> (libSDL2-2.0.so.0+0x905a0)

      Previous read of size 8 at 0x7ba000000080 by thread T16:
        #0 recv ../../../../src/libsanitizer/sanitizer_common/sanitizer_common_interceptors.inc:6603 (libtsan.so.0+0x4f4a6)
        #1 net_recv_all ../app/src/util/net.c:228 (scrcpy+0x234a9)
        #2 stream_recv_packet ../app/src/stream.c:33 (scrcpy+0x2045c)
        #3 run_stream ../app/src/stream.c:228 (scrcpy+0x21169)
        #4 <null> <null> (libSDL2-2.0.so.0+0x905a0)

Refs ddb9396743
2 years ago
Romain Vimont 8ea6fb1f0f Print version on stdout
Refs b25404ee4b
2 years ago
Romain Vimont b546c33eff Do not print scrcpy version twice on --version
Refs 6da6d905c2
2 years ago
Tomás Raiti 8615813005 Update README.sp.md to v1.21
PR #2962 <https://github.com/Genymobile/scrcpy/pull/2962>

Signed-off-by: Romain Vimont <rom@rom1v.com>
2 years ago
Romain Vimont f51c53e913 Mention "screen copy" in README
Help users make sense of the app name :)
2 years ago
Romain Vimont c0de365f67 Upgrade platform-tools (32.0.0) for Windows
Include the latest version of adb in Windows releases.
2 years ago
Romain Vimont 34e19dcc57 Upgrade SDL (2.0.20) for Windows
Include the latest version of SDL in Windows releases.
2 years ago
Romain Vimont 4817cadd09 Fix code style
Align function parameters.
2 years ago
Romain Vimont 02b5e87802 Slightly reduce lock usage
Locking the frame_buffer mutex to reference the input frame into the
tmp_frame is unnecessary.

This also fixes the missing unlock on error.
2 years ago
Romain Vimont 8e4d3beb01 Fix return value on adb commands error 2 years ago
Romain Vimont 50f4f1639c Add a shorcut "open a terminal here" on Windows
On Windows, the file explorer does not provide any "open a terminal
here" shortcut.

Add a bat file which just runs `cmd` to easily open a terminal directly
in the scrcpy directory (so there is no need to `cd C:\path\…`).

PR #2970 <https://github.com/Genymobile/scrcpy/pull/2970>
2 years ago
Romain Vimont c8dc1917f4 Do not restore power mode if --no-control
This totally disables deferred cleanup when --no-control is passed.

Refs f289d206ea
2 years ago
Romain Vimont 9d2e00697e Use sc_ prefix for control_msg enums
Refs afa4a1b728
2 years ago
Romain Vimont 2faf9715be Add server option raw_video_stream
For convenience, this new option forces the 3 following options:
 - send_device_meta=false
 - send_frame_meta=false
 - send_dummy_byte=false

This allows to send a raw H.264 stream on the video socket.

Concretely:

    adb push scrcpy-server /data/local/tmp/scrcpy-server.jar
    adb forward tcp:1234 localabstract:scrcpy
    adb shell CLASSPATH=/data/local/tmp/scrcpy-server.jar \
        app_process / com.genymobile.scrcpy.Server 1.21 \
        raw_video_stream=true tunnel_forward=true control=false

As soon as a client connects via TCP to localhost:1234, it will receive
the raw H.264 stream.

Refs #1419 comment <https://github.com/Genymobile/scrcpy/pull/1419#issuecomment-1013964650>
PR #2971 <https://github.com/Genymobile/scrcpy/pull/2971>
2 years ago
Romain Vimont 45a5e560df Add server option send_dummy_byte
If set to false, no dummy byte is written to detect a connection error.

PR #2971 <https://github.com/Genymobile/scrcpy/pull/2971>
2 years ago
Romain Vimont 3ba32c2a0d Add server option send_device_meta
Similar to send_device_frame, this option allows to disable sending the
device name and size on start.

This is only useful when using the scrcpy-server alone to get a raw
H.264 stream, without using the scrcpy client.

PR #2971 <https://github.com/Genymobile/scrcpy/pull/2971>
2 years ago
Romain Vimont 6b21f4ae13 Reorder scrcpy-server options
Move the options unused by the scrcpy client at the end.

These options may be useful to use scrcpy-server directly (to get a raw
H.264 stream for example).

PR #2971 <https://github.com/Genymobile/scrcpy/pull/2971>
2 years ago
Romain Vimont 31a5d0c2bf Move call to send device name and size
This will allow to optionally disable it.

PR #2971 <https://github.com/Genymobile/scrcpy/pull/2971>
2 years ago
Romain Vimont f289d206ea Disable more actions if --no-control
If control is disabled, then do not enable "show touches" or
automatically power off the device on close.
2 years ago
Romain Vimont ca516f4318 Refactor if-block in cli
Several tests must be performed if opts->control is false.
2 years ago
Romain Vimont 5d6076bffd Move misplaced break statements
With ifdefs, the resulting code could contain both a return statement
and a break.
2 years ago
Romain Vimont e0bce1725b Fix header guard prefix 2 years ago
Romain Vimont ae8fdda09e Improve FAQ explanations 2 years ago
Romain Vimont 1ff69e21c2 Update error messages in FAQ
With the recent versions, scrcpy first executes "adb get-serialno", so
the adb error messages are not exactly the same.
2 years ago
Romain Vimont a9429efa34 Fix downsize on error before first frame
Retry with a lower definition if MediaCodec fails before the first
frame, not the first packet.

In practice, the first packet is a config packet without any frame, and
MediaCodec might fail just after.

Refs 2eb6fe7d81
Refs #2963 <https://github.com/Genymobile/scrcpy/issues/2963>
2 years ago
Romain Vimont 063d103dd6 Capture mouse on start for --hid-mouse
If relative mode is enabled, capture the mouse immediately.
2 years ago
Romain Vimont 4bf9c057fe Extract relative mode check to an inline function
This will allow to reuse the condition in another function.
2 years ago
Romain Vimont 17c97820b2 Never forward capture keys
In relative mode, Alt and Super are "capture keys". Never forward them
to the input manager, to avoid inconsistencies between UP and DOWN
events.
2 years ago
Romain Vimont 8c7f0ed5ea Fix warning message
Make the message consistent for HID keyboard and HID mouse.
2 years ago
Romain Vimont ac038f276e Add missing break statement
This was harmless because this is the last "case" of the switch, but for
consistency, add the missing break.
2 years ago
Romain Vimont 1f65b1bf87 Remove inline hint
There is no reason to request inlining here.
2 years ago
Romain Vimont d41a46dc95 Handle libusb_get_device_descriptor() error
The function libusb_get_device_descriptor() might return an error.
Handle it.
2 years ago
Romain Vimont 308a1f8192 Simplify error handling in sc_aoa_init()
Use goto to avoid many repetitions.
2 years ago
Romain Vimont 241a587e61 Fix missing HID mouse destructor call
The destructor unregisters the HID mouse, so it was not reported as a
leak, but it must still be called.
2 years ago
Romain Vimont 7e35bfe382 Refactor if-blocks
Group all conditions requiring a controller in a single if-block.
2 years ago
Romain Vimont 855819bbd8 Remove redundant control boolean
The controller is NULL if and only if control is disabled, so an
additional control boolean is redundant.
2 years ago
Romain Vimont 557daf280e Pass NULL controller if control is disabled
If --no-control is requested, then the controller instance is not
initialized. However, its reference was still passed to screen and
input_manager.

Instead, pass NULL if no controller is available.
2 years ago
Romain Vimont 0b8e926330 Do not process finger events if no control
If --no-control is passed, then im->mp is NULL, so processing touches
would crash.
2 years ago
Romain Vimont 0ec3361bc9 Fix crash on --no-control
Relative mouse mode assumed that a mouse processor was always available,
but this is not the case if --no-control is passed.
2 years ago
Romain Vimont 81ff7ebd06 Simplify event loop
Merge single event handling with the event loop function.
2 years ago