|
ed6eb07e
|
2021-08-12T01:40:50
|
|
SDL_wasapi.c: fixed build against older SDKs.
|
|
68ca35c3
|
2021-08-11T21:40:46
|
|
wayland: fix memleak in output listener
|
|
|
e9179314
|
2021-08-11T21:27:42
|
|
joystick/linux: fix memleaks; streamline joylist item removal
|
|
|
44a76710
|
2021-08-11T12:55:50
|
|
Don't disable clang avx instrinsics on win32 if __AVX__ is defined.
C.f.: https://github.com/libsdl-org/SDL/issues/4533
|
|
ae7ee2a5
|
2021-08-11T09:59:43
|
|
wayland: Ignore stateless/sizeless configs when starting in fullscreen mode
|
|
ac32c522
|
2021-08-10T18:11:09
|
|
Try using the built-in WASAPI audio rate conversion
Fixes https://github.com/libsdl-org/SDL/issues/4608
|
|
|
cb1e20b0
|
2021-08-10T17:50:17
|
|
Added KMOD_SCROLL to track the scroll lock state
Fixes https://github.com/libsdl-org/SDL/issues/4566
|
|
|
609cea1e
|
2021-08-10T17:14:06
|
|
Enable SSE3 intrinsics when building with Visual Studio
Fixes https://github.com/libsdl-org/SDL/issues/4530
|
|
|
d5ad6f6e
|
2021-08-10T16:51:03
|
|
Clarified that you should never have side-effects in the parameters to SDL_min/SDL_max
|
|
|
bb91d7a5
|
2021-08-11T01:56:40
|
|
regenerated configure script
|
|
056a42f1
|
2021-08-10T22:20:06
|
|
Sync wiki -> header
|
|
|
f5794f9e
|
2021-08-10T15:17:59
|
|
Added SDL_SetTextureUserData() and SDL_GetTextureUserData() to associate a user-specified pointer with an SDL texture
|
|
|
a0af7ce7
|
2021-08-10T15:05:49
|
|
OSS is no longer the preferred audio backend on modern UNIX systems
Fixes https://github.com/libsdl-org/SDL/issues/4207
|
|
|
3f6ebfff
|
2021-08-10T15:02:36
|
|
Updated to version 2.0.17 for development
|
|
|
68103993
|
2021-08-10T14:50:43
|
|
Enable AAUDIO driver for Android
I thought this was already enabled for 2.0.16, but apparently not.
Fixes https://github.com/libsdl-org/SDL/issues/3710
|
|
|
016b02f2
|
2021-08-10T12:16:55
|
|
Fixed build, C89 doesn't allow non-constant static initializers
|
|
|
77c8d111
|
2021-08-10T20:55:50
|
|
configuration updates for dlopen:
- cmake, configure (CheckDLOPEN): --enable-sdl-dlopen is now history..
detach the dl api discovery from SDL_LOADSO_DLOPEN functionality.
define HAVE_DLOPEN. also define DYNAPI_NEEDS_DLOPEN (CheckDLOPEN is
called only for relevant platforms.)
- update SDL_config.in and SDL_config.cmake accordingly.
- SDL_dynapi.h: set SDL_DYNAMIC_API to 0 if DYNAPI_NEEDS_DLOPEN is
defined, but HAVE_DLOPEN is not.
- pthread/SDL_systhread.c: conditionalize dl api use to HAVE_DLOPEN
- SDL_x11opengl.c, SDL_DirectFB_opengl.c, SDL_naclopengles.c: rely
on HAVE_DLOPEN, not SDL_LOADSO_DLOPEN.
- SDL_config_android.h, SDL_config_iphoneos.h, SDL_config_macosx.h,
SDL_config_pandora.h, and SDL_config_wiz.h: define HAVE_DLOPEN.
Closes: https://github.com/libsdl-org/SDL/pull/4351
|
|
cb0fd05e
|
2021-08-07T22:41:54
|
|
Adding a clarifying comment in case a programmer unfamiliar with UTF-16 and UTF-32's relationship chances upon the code.
|
|
|
34701129
|
2021-08-07T20:51:49
|
|
Optimizing the implementation.
|
|
|
712e0d1f
|
2021-08-07T16:23:15
|
|
Fixing WM_CHAR event handling for Unicode characters outside the Basic Multilingual Plane.
|
|
|
fcfd19db
|
2021-08-10T12:02:17
|
|
Added support for SDL_RENDERER_PRESENTVSYNC to the software renderer
This fixes https://github.com/libsdl-org/SDL/issues/4612
|
|
ea9bece5
|
2021-08-10T18:51:33
|
|
Fix breaking GNUInstallDirs cmake module $CMAKE_INSTALL_LIBDIR directory. Fix #4621
According to https://cmake.org/cmake/help/v3.0/module/GNUInstallDirs.html
`${CMAKE_INSTALL_LIBDIR} ` is only variable which should be used as base
directory. Without that patch on 64 bit archs libraries, cmake modules and
pkgconfig file are installed for example in /usr/lib6464 base directory.
This patch fixes #4621.
Signed-off-by: Tomasz Kłoczko <koczek@github.com>
|
|
25f9ed87
|
2021-08-10T13:10:36
|
|
audio: Fix false positives in driver name comparison
Without this change, driver names don't get matched correctly;
for example "a" can get matched with "alsa" since it only checks
whether the string matches up to the length of the requested
driver name.
|
|
|
b3a989d0
|
2021-08-10T13:08:27
|
|
video: Fix false positives in driver name comparison
Without this change, driver names don't get matched correctly;
for example "x" can get matched with "x11" since it only checks
whether the string matches up to the length of the requested
driver name.
|
|
|
de6ba40d
|
2021-08-09T11:38:22
|
|
Added support for the Flydigi Vader 2 controller in 2.4G Android mode
|
|
|
3f440daa
|
2021-08-09T10:40:36
|
|
Added support for the Flydigi Vader 2 controller in Bluetooth mode
|
|
6ae227d0
|
2021-08-08T23:27:08
|
|
x11/wayland: fix screensaver suspension via D-Bus
b08b1bde introduced a subtle bug. Despite not using D-Bus types directly,
the code used the SDL_USE_LIBDBUS definition set by SDL_dbus.h to conditionally
compile calls SDL_DBus_ScreensaverTickle() and SDL_DBus_ScreensaverInhibit().
As a result, it still compiled without SDL_dbus.h included, but screensaver
suspension silently failed to work.
The D-Bus stuff could probably use some tweaks to be harder to accidentally
break, but for now just restore the header includes.
|
|
773e1ba1
|
2021-08-08T11:00:07
|
|
testvulkan: Clamp the drawable size to the allowed range
SDL_Vulkan_GetDrawableSize() doesn't always return a size which is
within the Vulkan swapchain's allowed image extent range.
(This happens on X11 a lot when resizing, which is bug #3287)
Clamp the value we get back from SDL_Vulkan_GetDrawableSize() to this
range. Given the range usually is just a single value, this is almost
always equivalent to just using the min or max image extent, but this
seems logically most correct.
|
|
|
2e6dac87
|
2021-08-09T01:47:42
|
|
wayland: Add a hint to allow disabling libdecor use
Useful for testing xdg-shell path with compositors like Weston.
|
|
|
092a20d9
|
2021-08-08T16:47:34
|
|
wayland: Avoid busy waiting for vsync
|
|
|
791d9d3f
|
2021-08-07T12:41:15
|
|
Fixed DInput mapping for NVIDIA Virtual Gamepad
|
|
|
7be970db
|
2021-08-06T19:51:27
|
|
wayland: Avoid setting floating width/height when re-entering fullscreen
|
|
|
bec78357
|
2021-08-06T14:20:55
|
|
Better fix for compiling using clang on Windows
This fixes https://github.com/libsdl-org/SDL/issues/4533
|
|
|
a91ab883
|
2021-08-06T12:28:03
|
|
Fixed building on Windows with cmake, ninja, and clang
|
|
dd683073
|
2021-08-06T15:13:14
|
|
hidapi: Patched to compile on QNX.
Fixes #4591.
|
|
|
cae7bd9b
|
2021-08-06T12:01:24
|
|
Don't use AVX with clang if the compiler isn't building with AVX instructions
Fixes https://github.com/libsdl-org/SDL/issues/4533
|
|
|
cdac61ff
|
2021-08-06T12:00:53
|
|
Fixed typo
|
|
|
eec41d25
|
2021-08-06T11:30:22
|
|
Add info on how to avoid the "Prevented in-tree build" message with CMake
|
|
|
ac20167e
|
2021-08-06T14:08:50
|
|
configure: Fixes to make SDL link as C++ on Haiku.
Fixes #4590.
|
|
|
10979d37
|
2021-08-05T12:42:03
|
|
wayland: Always assume configure is wrong for non-resizable windows.
Configure events from compositors have an extremely annoying habit of giving us
completely bogus sizes, from all sorts of places. Thankfully, the protocol
gives us the ability to completely ignore the width/height and just stick with
what we know, so for all windows that are not meant to be resized, pretend we
never even got the width/height at all, the compositor is required to respect
our dimensions whether they match configure's suggestion or not.
|
|
520bc713
|
2021-08-05T14:33:54
|
|
Fix typo in CMake target name
|
|
|
b83a019d
|
2021-08-04T19:22:10
|
|
cmake: Force linker to treat input as a C++ program on Haiku.
Reference issue #4590.
|
|
|
88674b4a
|
2021-08-04T13:17:35
|
|
Fixed build
|
|
|
31726150
|
2021-08-04T13:13:22
|
|
Let Android know that we can handle USB devices
This is important on Android 29 and above if you don't want to be prompted for each device that you open
|
|
|
c14813a8
|
2021-08-04T13:00:14
|
|
Fixed spacing
|
|
|
3cad447e
|
2021-08-04T12:57:51
|
|
Only update the window size if setting the display mode succeeded
|
|
0eb6f791
|
2021-05-21T13:48:14
|
|
SDL_SetWindowDisplayMode(): If already fullscreen, adjust window size
Otherwise only the display resolution is changed, but the SDL window size
(and for example the window-surface size) aren't adjusted accordingly
and thus don't fill the whole screen.
See #3313
|
|
|
72d81285
|
2021-05-19T18:47:56
|
|
Improve SDL_CreateWindow() fullscreen support on Windows
.. and maybe other platforms as well (though X11 was not affected)?
The issue was that passing a higher resolution than the current desktop
resolution to SDL_CreateWindow() with SDL_WINDOW_FULLSCREEN didn't switch
to that resolution (even though it did switch to lower resolutions).
When creating a fullscreen window, window->fullscreen wasn't even set
at all (only zeroed out), setting it only happened if the user explicitly
called SDL_SetWindowDisplayMode(). So without that, SDL_CreateWindow()
-> SDL_UpdateFullscreenMode() -> SDL_GetWindowDisplayMode() used the
resolution from window->windowed.w/h which were limited to the desktop size
due to some weird combination of WIN_AdjustWindowRectWithStyle() and
WIN_WindowProc() being called after a call to SetWindowPos().
fixes #3313
|
|
|
9fdb0645
|
2021-08-04T11:33:28
|
|
Added support for the Joy-Con Charging Grip when SDL_HINT_JOYSTICK_HIDAPI_JOY_CONS is enabled
|
|
|
d4839702
|
2021-08-04T13:30:12
|
|
dbus: Wrap init in a spinlock.
This prevents a race if two threads that need d-bus try to init it at the
same time. Note that SDL_Init will likely handle this from a single thread
at startup, but there are places outside of init where one might trigger
D-Bus init, like setting thread priority (a common first thing for a new
thread to do) resulting in SDL trying to use RTKit.
Fixes #4587.
|
|
|
b08b1bde
|
2021-08-03T22:50:58
|
|
linux: remove d-bus lazy init dead code
Lazy init in X11/Wayland is dead code since dbdbae4
|
|
|
f5bc58bf
|
2021-08-03T18:30:08
|
|
Add note about keyboard grab to WhatsNew.txt for 2.0.16
|
|
|
5fc13fcb
|
2021-08-04T07:00:17
|
|
direct3d: Commit viewport state before clearing, attempt 2.
This reintroduces the fix from 0e16ee8330606603, but just marks
the viewport state as dirty after a clear that needs to expand the
viewport to fill the render target, as we'll need to also reset
the orthographic projection state elsewhere, and that won't
happen if we clear the dirty flag here.
Fixes #4210.
(again.)
(...sorry...!)
|
|
|
ed6ce5cc
|
2021-08-04T02:34:09
|
|
log: Fixed compiler warnings on 64-bit Visual Studio.
|
|
|
7e88e772
|
2021-08-04T02:30:47
|
|
winrt: Fixed incorrect code in a Windows Phone #ifdef.
|
|
|
8104c9e3
|
2021-08-04T00:02:39
|
|
doxygen: manually move SDL_CloseAudioDevice docs from wiki.
Apparently we don't fill these in from the wiki if the function doesn't
have _any_ documentation already. That's a bug, I think.
|
|
|
57bfed79
|
2021-08-03T22:15:04
|
|
README-dynapi.md: Google+ link is gone, remove it.
The Wayback Machine doesn't even have it. :(
|
|
|
178c95f8
|
2021-08-03T21:52:47
|
|
README-winrt.md: Cleaned up sample code
- Fixed the markdown.
- Code can now be exited by pressing ESC.
- Cleans up and returns from main()
- Mushed all the `if (x) { return 0; }` blocks into else ifs.
|
|
|
32f909f7
|
2021-08-03T13:30:57
|
|
wayland: Remove redundant waylanddyn.h includes.
All files including waylanddyn.h already include waylandvideo.h first.
|
|
|
124405a0
|
2021-08-03T13:17:19
|
|
wayland: Fix building without Vulkan support
|
|
|
02563f75
|
2021-08-03T20:12:56
|
|
regenerate configure
|
|
d2e8998f
|
2021-08-03T07:59:01
|
|
build: Fix the LINKER variable.
The LINKER variable is set in configure.ac as either 'CC' or 'CXX'
where it is then passed to the created Makefile. This fails with
slibtool which can't find the 'CC' file and can be fixed by correctly
setting the LINKER variable to an actual Makefile variable like '$(CC)'
or '$(CXX)' instead. Presumably GNU libtool does some magic here to
hide the issue.
|
|
|
4077f7a2
|
2021-08-03T20:55:45
|
|
Update the Renderer dpi_scale on SIZE_CHANGED event (fix #4580)
The Renderer logical scaling code scales mouse coordinates, and needs to
take the window DPI into account on HIGHDPI windows. However, the
variable which tracks this, renderer->dpi_scale, is set once when the
renderer is created, and then not updated. In the event that the window
is moved to another screen, or the screen DPI otherwise changes, this
will be outdates, and potentially the coordinates will be all wrong.
So let's update the dpi_scale on the SIZE_CHANGED event: it's at least a
possibility that this will be issued on some OSes when DPI changes, and
it's otherwise already handled by SDL_Renderer's event filter.
|
|
|
9b4884d5
|
2021-08-03T09:23:54
|
|
Warn if you're building a partially functional RAWINPUT driver without device notifications
|
|
|
886eb02e
|
2021-08-03T03:16:35
|
|
Fixed detection of Steam Virtual Gamepad with the WGI driver
|
|
|
7c63e1d3
|
2021-08-03T01:52:02
|
|
Added WGI mappings for the Amazon Luna and Google Stadia controllers
|
|
|
99c8972b
|
2021-08-03T01:39:07
|
|
Added WGI mappings for PS4 and PS5 controllers
|
|
|
2dfc92ca
|
2021-08-03T01:33:35
|
|
Added default mapping for WGI game controllers
|
|
|
3d747078
|
2021-08-03T01:25:54
|
|
Run the renderer so Steam can find the main window
|
|
|
1907463a
|
2021-08-03T18:04:50
|
|
regenerate configure
|
|
|
b74bcd5d
|
2021-08-03T22:25:10
|
|
configure.ac: mkdir $(objects)/ when building wayland protocols
The $(objects) directory (usually build/) may not have been created by
the time the wayland-scanner protocol files are being compiled. The
$(gen) directory is explicitly made with mkinstalldir, but the final
object file (and gcc dependency files) need to go into $(objects).
For whatever reason, this only ever seemed to occur if --disable-shared
was set.
Note that this commit doesn't regenerate ./configure, as there were a
few unexplained, unrelated differences my version of autoconf created,
as as an autotools novice, I didn't want to poke that bear just yet.
This hopefully should fix #3689
|
|
|
5b18ae29
|
2021-08-03T14:01:02
|
|
improvements to libdecor support in cmake
this makes it consistent with wayland support, and makes cmake side
in line with autotools' libdecor support.
|
|
|
350ca0f9
|
2021-08-03T04:57:53
|
|
winrt: Don't register orientation hint callback in startup code.
SDL_AddHintCallback() uses SDL_malloc(), which means this would
run before main(), so the app wouldn't be able to supply its own
replacement SDL_malloc() implementation in time.
This code was moved to under SDL_Init. Since the hint callback
already makes efforts to not override the app manifest's
orientation settings, this is safe to move until after pre-main()
startup.
Fixes #4449.
|
|
|
834a84fa
|
2021-08-03T05:02:32
|
|
winrt: Don't use LoadLibraryA on WinRT.
This removes the CM_Register_Notification code on WinRT. Note
that this API _is_ available to UWP apps as of Windows 10.0.17763
(version 1809, released October 2018), according to:
https://docs.microsoft.com/en-us/uwp/win32-and-com/win32-apis#apis-from-api-ms-win-devices-config-l1-1-1dll
So it might be worth readding with some sort of preprocessor check
for minimum targeted version, or whatever is appropriate for WinRT
development.
|
|
|
a15ec365
|
2021-08-03T11:33:50
|
|
attempt to fix cmake builds with libdecor
|
|
|
8726f500
|
2021-08-03T01:10:48
|
|
Allow quitting controllermap if there are no controllers attached
|
|
|
9d64e6b4
|
2021-08-03T00:29:33
|
|
Revert "direct3d: Commit dirty viewport state before clearing."
This reverts commit 0e16ee833060660318cc324cf99c9736f4d48499.
|
|
|
993a5676
|
2021-08-03T00:14:15
|
|
Only use the default Xbox mapping on Linux
On Windows, this is already covered by XInput and is the wrong mapping if we try to use it with DirectInput/RAWINPUT/WGI controllers
|
|
|
74aa4b63
|
2021-08-03T03:10:14
|
|
cmake: don't fail if Wayland is enabled but libdecor isn't found.
Instead, just disable libdecor support.
This matches what the configure script does.
|
|
|
c3084539
|
2021-08-03T02:49:08
|
|
cmake: Use CMAKE_INSTALL_FULL_* for libdir,bindir,etc.
This makes sure sdl2.pc gets the full paths, matching what the
configure script will generate.
Fixes #4569.
|
|
|
0e16ee83
|
2021-08-03T02:07:47
|
|
direct3d: Commit dirty viewport state before clearing.
Otherwise you might have set the viewport to the full size of
the render target in SDL's API but this change hasn't been
transmitted to Direct3D yet by the time we attempt to clear.
Fixes #4210.
|
|
|
7edc7d53
|
2021-08-02T22:44:38
|
|
Added a mapping for the Steam Virtual Gamepad when it's detected in DirectInput mode
|
|
|
7082fb04
|
2021-08-02T22:42:28
|
|
wayland: Only dispatch on fullscreen set when the window is visible.
Fixes hang-on-startup described in #4572.
|
|
|
0f5b8a99
|
2021-08-02T19:48:31
|
|
wayland: For libdecor, dispatch immediately after setting fullscreen
Reference: #4578
|
|
|
02223501
|
2021-08-02T16:40:58
|
|
Fixed crash if WGI didn't give us the name for a controller
|
|
|
4562da62
|
2021-08-02T16:40:56
|
|
wayland: Make libdecor configure match xdg_toplevel
|
|
|
1fb1aa19
|
2021-08-02T12:57:28
|
|
wayland: Keep the fullscreen check for maximize/restore events
|
|
|
584673a5
|
2021-08-02T12:51:23
|
|
wayland: Don't ignore configs for non-fullscreen, non-resizable windows.
This was causing configure events to not inform SDL of window size
changes, even when they were based on resizes that we fully expected. The
result was fullscreen->windowed not working at all, because it would
retain the desktop resolution instead of reverting to the floating size
that it had before moving to fullscreen mode.
Fixes Super Hexagon fullscreen toggling.
|
|
|
25067298
|
2021-08-02T08:15:10
|
|
wayland: Flush in SetWindowFullscreen
The flush has been removed in e5f9fae034df8ca305241c5a666e7702fcf9958f.
Unfortunately, even though ideally the flush shouldn't be necessary,
our resize sequence isn't... well, perfect, and removing that flush causes
tons of troubles.
We're also still flushing in other paths where the window size can be
changed by the compositor and where we may potentially have to obey that
change, like in Wayland_MaximizeWindow.
This also removes the hack introduced in 7f261d3b7667978ced93a4fdc6aec2380df3d87f,
which introduces problems with protocol violations and seems to not be
necessary when flushing.
|
|
|
ea28187c
|
2021-08-02T08:07:23
|
|
wayland: Hack surface resize into compliance with set_window_geometry
We have issues with correct resize sequence and happen to commit old-sized
buffers even after configure event for the new size has been already
acknowledged. While the reason for that stays unknown, let's at least
workaround the problem by faking window geometry into expected size.
This does not fix visual glitch on e.g. fullscreen toggling, but having
a split-second glitch is still a much better outcome than being
terminated by the compositor for protocol violation.
|
|
|
b5210cac
|
2021-08-02T13:47:32
|
|
wayland: Initialize floating size at window creation
This allows the windowed size to be restored for windows initially
created already in non-floating state.
|
|
|
dab33844
|
2021-08-02T11:58:47
|
|
wayland: Fix a warning in handle_configure_xdg_toplevel
|
|
|
754aa2d3
|
2021-08-02T11:55:02
|
|
Mostly revert cfcdfb7be9f5c97a51ddef9ad3677a604856266b.
This was causing window changes to completely break, resulting in broken
decorations and bizarre frame timing, I don't know what exactly it's doing
but it's not good. Kept the libdecor_frame_is_floating logic, at least.
|
|
|
8803589f
|
2021-08-02T12:05:06
|
|
wayland: Avoid redundant SetWindowBordered for libdecor
|
|
|
b4c4060a
|
2021-08-02T04:52:25
|
|
wayland: Clean up Wayland_HandlePendingResize
Commit 871c11191bfc7214061a3da37c112522a102ddf5 removed delayed
resize handling, but it left the whole structure untouched that
now became unnecessary. To help with code clarity, get rid
of the structure where pending resize state used to be stored
and pass all the data directly to Wayland_HandlePendingResize
(now renamed to Wayland_HandleResize, since it's not "pending"
anymore but applied immediately)
|
|
|
9bb24ad0
|
2021-08-02T05:49:35
|
|
wayland: Always call Wayland_SetWindowBordered when showing the window
Otherwise our windows have no window decoration on compositors that
support xdg-decoration-unstable-v1, but default to client-side mode.
Contrary to what the comment was stating, there is nothing in the protocol
that would make redundant calls to zxdg_toplevel_decoration_v1::set_mode
problematic.
|
|
cfcdfb7b
|
2021-08-01T18:14:53
|
|
libdecor: use same fullscreen/maximised restore logic as for xdg-toplevel
|
|
|
42452f8c
|
2021-08-01T13:51:30
|
|
wayland: store and restore floating states
Some Wayland compositors send (0,0) as "suggested" configure event sizes to
indicate that the client has to decide on its own which sizes to used. This
is commonly done when restoring from maximised, fullscreen or tiles states
to fullscreen.
We now store the last known floating states in a new set of variables and
restore them when we receive such a (0,0) configure event.
|
|
|
d4e1b497
|
2021-08-01T09:15:26
|
|
unix: OpenURL: Move unsetenv above vfork
From the vfork manpage:
> The vfork() function has the same effect as fork(2), except that
> the behavior is undefined if the process created by vfork() either
> modifies any data other than a variable of type pid_t used to store
> the return value from vfork(), or returns from the function in which
> vfork() was called, or calls any other function before successfully
> calling _exit(2) or one of the exec(3) family of functions.
unsetenv is still called inside a child process, so it does not
influence the rest of the application.
|
|
|
9b7b9287
|
2021-07-31T23:28:41
|
|
Clear LD_PRELOAD so Chrome opens correctly when this application is launched by Steam
This fixes https://github.com/libsdl-org/SDL/issues/4565
|