- No more tapdance to either join the audio device thread or have it detach
itself. Significant simplication of and fixes to the locking code to prevent
deadlocks.
- Physical devices now keep a refcount. Each logical device increments it,
as does the existence of a device thread, etc. Last unref destroys the
device and takes it out of the device_hash. Since there's a lot of moving
parts that might be holding a reference to a physical device, this seemed
like a safer way to protect the object.
- Disconnected devices now continue to function as zombie devices. Playback
devices will still consume data (and just throw it away), and capture devices
will continue to produce data (which always be silence). This helps apps
that don't handle disconnect events; the device still stops playing/capturing,
but bound audio streams will still consume data so they don't allocate more
data infinitely, and apps that depend on an audio callback firing regularly
to make progress won't hang.
Please note that disconnected audio devices must now be explicitly closed!
They always _should_ have been, but before this commit, SDL3 would destroy the
disconnected device for you (and manually closing afterwards was a safe no-op).
Reference Issue #8331.
Fixes#8386.
(and probably others).
This is a race condition if the hashtable isn't protected by a mutex, and it
makes a read/write operation out of something what appears to be read-only,
which is dangerously surprising from an interface viewpoint.
The downside is that if you have an item that is frequently accessed that
isn't in the first slot of a bucket, each find operation will take longer
instead of common items bubbling to the front of the bucket. Then again,
if you have several common things being looked up in rotation, they'll just
be doing unnecessary shuffling here. In this case, it might be better to
just use a larger hashtable or a better hashing function (or just look up the
thing you need once instead of multiple times).
Fixes#8391.
All devices are in a single hash, whether playback or capture, or physical
or logical. Lookups are keyed on device ID and map to either
`SDL_AudioDevice *` for physical devices or `SDL_LogicalAudioDevice *` for
logical devices (as an implementation detail, you can determine which object
type you have by checking a specific bit in the device ID).
This simplifies a bunch of code, makes some cases significantly more
efficient, and solves the problem of having to lock each physical
device while the device list rwlock is held to find logical devices by ID.
Device IDs hash perfectly evenly, too, being incrementing integers.
And SDL_IterateHashTableKey is only necessary for stackable hashtables, since
non-stackable ones can either iterate each unique key/value pair with
SDL_IterateHashTable, or get a specific key/value pair by using
SDL_FindInHashTable.
The primary selection protocol is optional, so the function pointers to the internal Wayland functions should only be set if the protocol is supported. This allows graceful fall-back to the generic SDL implementation in other cases.
Fixes the clipboard tests under Weston.
The following objects now have properties that can be user modified:
* SDL_AudioStream
* SDL_Gamepad
* SDL_Joystick
* SDL_RWops
* SDL_Renderer
* SDL_Sensor
* SDL_Surface
* SDL_Texture
* SDL_Window
This specifically fixes a crash in X11_WarpMouseInternal if XInput2 was
missing at runtime, but also cleans up a few other existing checks.
Fixes#8378.
We need to do this early in the file, so that it will be taken into
account when deciding whether to define NEED_SCALAR_CONVERTER_FALLBACKS
and therefore provide a non-SIMD fallback.
Mitigates: https://github.com/libsdl-org/SDL/issues/8352
Signed-off-by: Simon McVittie <smcv@collabora.com>
Also switched the D3D11 and D3D12 renderers to use real NV12 textures for NV12 data.
The combination of these two changes allows us to implement 0-copy video decode and playback for D3D11 in testffmpeg without any access to the renderer internals.
This avoids assuming that the pixels are suitably aligned for direct
access, which there's no guarantee that they are; in particular,
3-bytes-per-pixel RGB images are likely to have 3 out of 4 pixels
misaligned. On x86, dereferencing a misaligned pointer does what you
would expect, but on other architectures it's undefined whether it will
work, crash with SIGBUS, or silently give a wrong answer.
Signed-off-by: Simon McVittie <smcv@collabora.com>
This shares its implementation with SDLTest_ReadSurfacePixel: the same
code is compiled twice, to get it into the static test library and also
the public shared library.
Signed-off-by: Simon McVittie <smcv@collabora.com>
This is essentially the same as was added in d95d2d70, but with clearer
error handling. It's implemented in a private header file so that it
can be shared with SDL_shape, which also wants this functionality.
Resolves: https://github.com/libsdl-org/SDL/issues/8319
Signed-off-by: Simon McVittie <smcv@collabora.com>
When initializing the Wayland driver, check if the application is being started in, or trying to connect to, a Wayland session and skip to another driver if not. If neither WAYLAND_DISPLAY nor XDG_SESSION_TYPE are set, try to start anyway, and if the Wayland library is missing or no Wayland sessions are started, initialization will fail later in the process as it previously did.
This fixes the case where a Wayland session is running on a different VT, but an application wishes to run via KMSDRM on the current VT.
Ryan C. Gordon
Anonymous Maarten
Frank Praznik
kanjitalk755
Sam Lantinga
Simon McVittie
Etaash Mathamsetty
Mathieu Eyraud