Our understanding of what's going on here might be incorrect, but it seems
like we're getting this callback at a point where we shouldn't be able to,
with a device we've already closed.
If we're on the wrong track, this code should still be harmless; it just
verifies a device is still in the open list before dereferencing it.
Reference Issue #10432.
Set thread name on Android the same way as we do on Linux.
Acording to Bionic source code this function is available since 2013 [1] and
hase the same signature.
[1] 2a1bb4e646
In reality, this condition will never occur, since the index is checked before calling the display retrieval function, but aggressive LTO with jump threading can generate a warning if this isn't explicitly checked.
The objfiles of different builds show various differences in
the calls to SDL_DYNAPI_entry elements. This is generated
dynamically by gendynapi.pl which uses an unordered opendir/readdir
pair. To make the build reproducible and thereby e.g. debugging
easier change this to be used in an ordered fashion.
Fixes#11565.
(cherry picked from commit 5418740c4a)
Fixes:
src/video/directfb/SDL_DirectFB_render.c: In function ‘DirectFB_CreateRenderer’:
src/video/directfb/SDL_DirectFB_render.c:1153:35: error: assignment to ‘void (*)(SDL_Renderer *, SDL_Texture *, SDL_ScaleMode)’ from incompatible pointer type ‘void (*)(void)’ [-Wincompatible-pointer-types]
1153 | renderer->SetTextureScaleMode = DirectFB_SetTextureScaleMode;
| ^
Signed-off-by: Peter Seiderer <ps.report@gmx.net>
---
Notes:
- declaration was just recently changed by commit 'Fix warning for Android
NDK compiler: "function declaration without a prototype is deprecated in
all versions of C [-Wstrict-prototypes]"', see
ccade50587
WIN_AddDisplay asserted that the returned index is the expected next
one in the list. This can be an unexpected number when displays are
add and removed in a batch (eg, Remote Desktop disconnect and reconnect).
Instead always just make use of the next index, even if it is not the
expected next index value. This fixes issue #9105
(cherry picked from commit bee8874629)
There are low-latency fixes in SDL3 that really need to be backported to SDL2
before AAudio will be nicer across a range of devices, so OpenSL ES is safer
in SDL2 at the moment. At least, we suspect as such.
Fixes#11457.
This reverts commit 1e016fd5ea.
In order to use low latency mode, you need to set low latency callbacks. This has always been recommended and is apparently required on Android 15. SDL3 has these implemented correctly but SDL2 does not, so we should revert this for now.
Fixes https://github.com/libsdl-org/SDL/issues/11376
