|
30443105
|
2022-01-26T21:10:15
|
|
video: Don't add SDL_WINDOW_METAL to new windows if not using Cocoa or UIKit.
Fixes #4656.
|
|
77a9ca6b
|
2022-01-26T18:26:07
|
|
wayland: Fix SDL_SetWindowSize() being dropped right after exiting fullscreen
If we get a SDL_SetWindowSize() call right after SDL_SetWindowFullscreen() but
before we've gotten a new configure event from the compositor, the attempt to
set our window size will silently fail (when libdecor is enabled).
Fix this by remembering that we need to commit a new size, so we can do that
in decoration_frame_configure().
|
|
|
c6ec5a07
|
2022-01-02T15:43:35
|
|
d3d11: Respect SDL_HINT_RENDER_DIRECT3D_THREADSAFE when creating device
|
|
|
f37e4a94
|
2022-01-26T17:02:49
|
|
video: Don't check if we can use a "texture framebuffer" until needed.
This prevents SDL from making an OpenGL context and maybe throwing it away
immediately by default. It will now only do it when trying to request a
window framebuffer directly, or creating an SDL_Renderer with the "software"
backend, which makes that request itself.
The way SDL decides if it should use a "texture framebuffer" needs dramatic
updating, but this solves the immediate problem.
Reference Issue #4624.
|
|
71e06a53
|
2021-11-26T18:27:34
|
|
cmake: fix Metal detection
Confirmed to work on:
macOS 10.15.7
XCode Version 12.4 (12D4e)
|
|
8ceba27d
|
2021-04-17T12:07:38
|
|
video: Prefer Wayland over X11
|
|
|
68a71f91
|
2022-01-25T11:18:04
|
|
wayland: Try to avoid committing before the window is shown
|
|
84320266
|
2022-01-25T12:37:43
|
|
Fixed the queue filling up with sentinel events when the WaitEvent call is passed NULL for the event
The use case is an application that waits for events on the main thread and dispatches them on a separate thread.
|
|
|
f0e768da
|
2022-01-25T13:07:00
|
|
wayland: Call SetFullscreen directly in ShowWindow.
This cuts out an extra flush when getting the first configure event.
|
|
8c660ccb
|
2022-01-25T17:14:01
|
|
Fixed bug #5256: X11 Segmentation fault with multiple windows and renderers
First window is created and it triggers and 'EnterNotify' event
which calls SDL_SetMouseFocus() and X11_ShowCursor() while the second
windows hasn't finished to be created (eg window->driverdata isn't set)
Just check for a valid 'driverdata'
|
|
|
e2d74bcb
|
2022-01-25T11:16:09
|
|
wayland: Detach hidden surfaces in HideWindow, not ShowWindow
|
|
|
ca22b30e
|
2022-01-24T22:32:58
|
|
testgles2: Use vertex buffer objects instead of client-side arrays.
GLES2 always has them, and they work without hacks on Emscripten, unlike
client-side arrays.
I cleaned it up slightly, but this patch was mostly written by @bing2008.
Fixes #5258.
|
|
57bc9040
|
2022-01-07T08:54:08
|
|
Add hint to optionally forcibly raise the window under MS Windows.
|
|
|
d7873e65
|
2022-01-04T15:47:29
|
|
Disable tablet flicks under MS Windows.
|
|
15ebad6e
|
2022-01-22T13:31:11
|
|
emscripten: Implement SDL_OpenURL
|
|
|
99af3281
|
2022-01-22T12:27:43
|
|
timer: Fix Emscripten declaration-after-statement error
|
|
|
53df0e66
|
2022-01-21T17:15:18
|
|
Fix the erroneous generation of mouse-down events from touch-move events.
The issue is that MS Windows synthesizes a mouse-move event in response
to touch-move events, and those mouse-move events are NOT labeled as
coming from a touch (e.g. GetMouseMessageSource() will not return
SDL_MOUSE_EVENT_SOURCE_TOUCH for those synthesized mouse-move events).
In addition, there seems to be no way to prevent this from happening;
https://gist.github.com/vbfox/1339671 claims to demonstrate a technique
to prevent it, but in my experience, it doesn't work.
Because of this, the "fallthrough" case can't test that the synthesized
mouse-move came from a touch-move, and starts erroneously pressing down
the mouse-button, leading to massive confusion in the client
application.
|
|
|
8f8b14cb
|
2022-01-21T17:10:09
|
|
Synthesize a missing touch-up event.
If a touch-down event is received for an existing touch-ID, that
probably means the operating system lost it, and that the missing
touch-up should be synthesized, to keep the client state coherent.
|
|
d2456b6e
|
2022-01-21T11:50:40
|
|
os2, geniconv: default functions pointers to Uni*() API,
In case they get accessed before SDL_Init() was called.
|
|
|
67cb548b
|
2022-01-20T19:31:54
|
|
Fixed build error with C90 compilers
|
|
a23b3c76
|
2022-01-20T17:21:28
|
|
Add support for correlating trigger input
|
|
|
e1b4761c
|
2022-01-20T14:10:56
|
|
wayland: Avoid calling SetFullscreen in libdecor ShowWindow.
This caused some weird stuff to happen in the libdecor path, probably because
the window hasn't actually been mapped yet. It ends up calling stuff that
should not yet apply, and so fullscreen in particular would have a really
messed up titlebar.
The good news is, libdecor is good about tracking fullscreen state, so we can
let the callback do this for us. Keep this for xdg_shell because we actually
map the window ourselves, so we know this call is valid for that path.
|
|
|
ed3442d7
|
2022-01-19T15:47:52
|
|
wayland: Fix building with SDL_OPENGL=OFF
|
|
|
426c1f4b
|
2022-01-19T14:27:58
|
|
opengles2: fix comment about client-side arrays.
|
|
b3984df1
|
2022-01-19T13:14:54
|
|
audio: pipewire: Use client config files instead of module names
Pipewire, as of 0.3.22, uses client config files to load modules instead of explicitly specifying them (PW_KEY_CONTEXT_PROFILE_MODULES is deprecated). Use the new method to load the realtime module to boost the audio thread priority.
|
|
|
381520ad
|
2022-01-18T14:32:24
|
|
updates for watcom optimization flags
|
|
|
a0e3c884
|
2022-01-17T15:32:27
|
|
Updated to version 2.0.21 for development
|
|
|
1bfefb5b
|
2022-01-17T20:56:50
|
|
regenerated configure script.
|
|
|
a575e2cf
|
2022-01-17T20:56:24
|
|
Enable alloca on mingw/cygwin builds:
This effectively reverts commit 391d73e37b26614e2e343ca8e9d4334e07380896
Fixes https://github.com/libsdl-org/SDL/issues/5234
|
|
|
33dadbf1
|
2022-01-17T20:56:20
|
|
'dnl' whitespace tidy-up in configure.ac
|
|
|
d3952a8a
|
2022-01-16T15:14:33
|
|
wayland: Avoid spurious resize events
|
|
|
b06866ef
|
2022-01-16T09:44:28
|
|
Entirely remove Xbox axis filtering, we don't want this at all.
|
|
|
09aed102
|
2022-01-16T09:23:58
|
|
Don't send output reports to the Victrix Pro FS v2
It'll hang on reboot if it gets output reports, and it doesn't have functionality that requires them, so skip it.
|
|
|
85961b61
|
2022-01-16T09:01:56
|
|
Disable Xbox axis deadzone handling by default
Fixes https://github.com/libsdl-org/SDL/issues/5227
|
|
2a0cde8f
|
2022-01-16T09:40:51
|
|
Always send SDL_WINDOWEVENT_SIZE_CHANGED when window is resized
SDL_WINDOWEVENT_SIZE_CHANGED is now sent even if the resulting size of the window does not match the requested size.
|
|
|
1a50334c
|
2022-01-14T15:04:26
|
|
Fixed Xbox One trigger axis zero point
|
|
|
e9511f71
|
2022-01-14T14:39:04
|
|
hidapi, libusb: don't use iconv on OS/2, it lacks wchar_t functionality.
Taken from a patch by Silvan Scherrer at bitwiseworks' OS/2 fork.
|
|
|
9a2bbd8a
|
2022-01-12T13:01:05
|
|
wayland: Convert URI to local path for DropFile
|
|
|
0abc0a6f
|
2022-01-12T20:37:32
|
|
SDL_string.c: tidy-up gcc2 va_copy defines after previous haiku commit.
|
|
d37c9777
|
2022-01-12T06:48:00
|
|
haiku: updated for Haiku
|
|
|
bff86701
|
2022-01-12T06:47:02
|
|
haiku: updated for Haiku
|
|
|
094e9440
|
2022-01-12T06:42:17
|
|
haiku: Remove BDirectWindow, fix OpenGL handling.
From e6cc4e7f4b8189be55dd3b0e13e54e59f73d7672 Mon Sep 17 00:00:00 2001
From: X512 <danger_mail@list.ru>
Date: Thu, 30 Jan 2020 04:01:58 +0900
Subject: libsdl2: Remove BDirectWindow, fix OpenGL handling.
* BDirectWindow changed to BWindow.
* Implemented fullscreen.
* Introduced view for non-OpenGL drawing.
* Drawing thread removed, window thread is used instead.
* Use BGLView as OpenGL context. Implement proper context switching and OpenGL
locking. Only one context per window is supported. BGLView should be not
deleted when window is closed, it deleted when deleting context.
|
|
|
407d4e47
|
2022-01-12T20:32:00
|
|
cmake: simplify after the previous os/2 commit.
|
|
|
f199f832
|
2022-01-12T20:22:28
|
|
cmake: added target_properties for os/2.
|
|
|
5b6e2d69
|
2022-01-12T11:01:02
|
|
SDL_os2messagebox.c (_makeDlg): remove unnecessary casts
|
|
|
ae9e2149
|
2022-01-11T21:32:07
|
|
os2-buildbot.sh: Use the 64-bit Watcom binaries.
GitHub Actions is a 64-bit Ubuntu instance. It was only using the 32-bit
binaries because our buildbot put this on the 32-bit Linux host to spread
the CPU load around more evenly.
|
|
|
42302d0a
|
2022-01-11T21:17:21
|
|
x11: Let apps specify a custom _NET_WM_WINDOW_TYPE.
Fixes #5185.
|
|
|
f9b918ff
|
2022-01-11T16:15:45
|
|
opengles2: Use client-side arrays on everything but Emscripten.
Turns out they're much faster!
Fixes #5206.
|
|
|
bb9ebad7
|
2022-01-11T09:34:14
|
|
Optionally disable generation of SDL2main and uninstall projects.
Mostly meant to allow SDL2 to be incorporated as a sub-project of
external CMake projects.
|
|
fcf0e524
|
2022-01-07T11:17:47
|
|
cmake: Remove libunwind-generic as hard dependency
when using alternate unwind implementations like LLVM libunwind
this library is not provided yet the libunwind features are fully
implemented in main libunwind, making this hard dependency assumes
a particular libunwind implementation, this patch makes it optional
which makes the builds to work with llvm libunwind
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
73434a96
|
2022-01-11T10:37:59
|
|
Removed debian directory
This isn't used directly by any distribution and isn't helpful at this point.
|
|
c3a7cc34
|
2022-01-10T21:52:02
|
|
cmake: sdlchecks.cmake: pass cflags to the appropriate cmake variable
If egl.pc sets at least two macros as the i.MX Vivante driver does, e.g.:
| Cflags: -I${includedir} -DLINUX -DWL_EGL_PLATFORM
then we get the following error during configuration:
| -- Performing Test HAVE_OPENGL_EGL
| CMake Error: Parse error in command line argument: WL_EGL_PLATFORM
| Should be: VAR:type=value
If one changes to add a value to the macro, e.g.
| Cflags: -I${includedir} -DLINUX=1 -DWL_EGL_PLATFORM=1
then cmake does not error out but the macro is not passed to the
C compiler.
CMAKE_REQUIRED_FLAGS is the wrong variable to pass the CFLAGS in,
CMAKE_REQUIRED_DEFINITIONS should be used.
Signed-off-by: Max Krummenacher <max.krummenacher@toradex.com>
|
|
|
ea0bca57
|
2022-01-11T14:56:56
|
|
SDL_cocoamodes.m: fix build against older SDKs.
Fixes https://github.com/libsdl-org/SDL/issues/5208
|
|
|
7527c8d3
|
2022-01-11T11:33:02
|
|
hidapi, libusb: provide a wcsdup implementation if it is not available.
similar to mac and android versions..
|
|
|
c6c4d421
|
2022-01-11T11:33:02
|
|
pthread: disable SetThreadPriority for OS/2 too
|
|
|
b2656776
|
2022-01-11T11:33:02
|
|
cmake: initial attempt at adding os/2 support.
|
|
b424665e
|
2022-01-10T20:40:05
|
|
Sync wiki -> header
|
|
|
5e3f4ccc
|
2022-01-10T20:35:04
|
|
Sync wiki -> header
|
|
|
d8129c56
|
2022-01-10T10:50:59
|
|
Mark internal function as static
|
|
|
3e1b3bc3
|
2022-01-10T10:07:44
|
|
wayland: Horizontal wheel values do not need to be inverted
|
|
|
a29d3acc
|
2022-01-09T11:11:34
|
|
Updated Bresenham line drawing to match software renderer output
|
|
|
f14c7bb4
|
2022-01-09T20:33:32
|
|
added missing hidapi stuff for os/2 with libusb
|
|
|
b9d98331
|
2022-01-09T20:33:32
|
|
os2joystick: fix uninitialized use of local var 'maxdevs'
|
|
|
17da6006
|
2022-01-09T20:33:32
|
|
os2joystick: fix -Wsequence-point issues.
|
|
|
1f32dd8e
|
2022-01-09T06:36:18
|
|
Fixed overdraw with duplicate points, fixed not drawing single point lines
|
|
|
5d07c036
|
2022-01-09T08:13:34
|
|
psp: Force channels to stereo if user requested anything other than mono.
Before it would only clamp to stereo if it also had to resample, which would
fail if the app specified 44100Hz and surround sound.
|
|
|
c275436f
|
2022-01-09T00:39:02
|
|
opengles2: Batching lines/points used wrong var for summing vertex counts.
Fixed this in the vitagxm backend too, but I don't know what the state of
that code is otherwise.
Reference Issue #5061.
|
|
9e6249fa
|
2022-01-08T19:24:47
|
|
wayland: Avoid spurious key repeats when not pumping events
Previous to this commit, key repeats events were typically generated when
pumping events, based on the time of when the events are pumped. However,
if an application doesn't call `SDL_PumpEvents` for some seconds, this time
can be multiple seconds in the future compared to the actual key up event time,
and generates key repeats even if a key was pressed only for an instant.
In practice, this can happen when the user presses a key which causes the
application to do something without pumping events (e.g. load a level).
In Crispy Doom & PrBoom+, when the user presses the key bound to "Restart
level/demo", the game doesn't pump events during the "screen melt" effect,
and the level is restarted multiple times due to spurious repeats.
To fix this, if the key up event is among the events to be pumped, we generate
the key repeats there, since in the Wayland callback we receive the time when
the key up event happened. Otherwise, we know no key up event happened and we
can generate as many repeats as necessary after pumping.
Signed-off-by: Joan Bruguera <joanbrugueram@gmail.com>
|
|
|
461724d2
|
2022-01-08T19:09:35
|
|
wayland: Refactor time fields in SDL_WaylandKeyboardRepeat
Refactorization with no functional changes.
Instead of `next_repeat_ms` containing a timestamp based on SDL ticks, we make
it zero-based relative to the key press time, and we store the key press time in
SDL ticks in a new field.
This refactorization is groundwork for future commits which need to use the
key press and release timestamps provided by the Wayland API, which are also
expressed in milliseconds, but whose base does not match the one for SDL ticks.
Signed-off-by: Joan Bruguera <joanbrugueram@gmail.com>
|
|
|
fb0c3040
|
2022-01-08T21:10:14
|
|
wayland: Avoid infinite loop in keyboard_repeat_handle
If `repeat_info->next_repeat_ms` overflows, many key presses will be generated.
In the worst case, `now = 0xFFFFFFFFU` and the loop will never terminate.
Rearrange the comparison in order to gracefully handle the overflow case.
Signed-off-by: Joan Bruguera <joanbrugueram@gmail.com>
|
|
|
3b083b99
|
2022-01-08T14:07:42
|
|
Fixed compile warning
|
|
|
b82af9df
|
2022-01-08T14:05:50
|
|
Improved OpenGL point drawing performance (thanks @slime73!)
We'll do proper call batching for the SDL 2.0.22 release
|
|
|
1a73c45a
|
2022-01-08T13:34:49
|
|
Fixed incorrect color in the OpenGL driver
|
|
|
3cdda8f8
|
2022-01-08T21:58:26
|
|
PSP: use 'data' variable which is alread the driver data
|
|
|
9f1e609a
|
2022-01-08T12:42:36
|
|
Document SDL_HINT_RENDER_LINE_METHOD in the release notes
|
|
|
4ef2529b
|
2022-01-08T12:37:57
|
|
More efficient calculation of render_count
|
|
|
5346c93b
|
2022-01-08T12:28:43
|
|
Fixed assertion on number of points rendered
|
|
|
4b719620
|
2022-01-08T12:02:08
|
|
Prevent overdraw with connected line segments
|
|
|
63d10a0a
|
2022-01-08T23:00:32
|
|
updates to config.guess and config.sub from mainstream.
|
|
|
09ece861
|
2022-01-08T11:36:29
|
|
Added the hint SDL_HINT_RENDER_LINE_METHOD to select the line rendering method
|
|
|
836a4ec7
|
2022-01-08T22:35:02
|
|
updates to os/2 build
|
|
|
c1e60798
|
2022-01-08T11:09:11
|
|
As an optimization, draw horizontal and vertical lines using rectangles
|
|
|
3da7c630
|
2022-01-08T11:07:03
|
|
Assert that we have drawn the expected number of points
|
|
|
9f56faee
|
2022-01-08T10:59:31
|
|
Added line drawing using Bresenham's line algorithm (thanks @rtrussell!)
|
|
|
fe3a33a0
|
2022-01-08T10:10:18
|
|
Use RenderGeometry for drawing lines at all scales
|
|
0696fc01
|
2022-01-07T14:30:26
|
|
cmake: Split SDL2-static and SDL2main into their own target exports
This makes it so that the generated targets are not interdependent,
which allows Linux distributions to split libraries into the
appropriate subpackages as needed.
|
|
|
1694782b
|
2022-01-08T09:32:23
|
|
Remove redundant work when render scale is 1.0
|
|
|
2026a78d
|
2022-01-08T09:23:58
|
|
Don't flush in RenderDraw*WithRects()
This allows these functions to be reused in the future for other primitives
|
|
|
18e4d9fe
|
2022-01-08T09:02:25
|
|
Re-enable line drawing path in render drivers
This is still used for scaled line drawing in RenderDrawLinesWithRects()
|
|
|
dca281e8
|
2022-01-08T08:49:34
|
|
Fixed getting different results for SDL_PollEvent(NULL) and SDL_PollEvent(&event)
|
|
|
93cab32d
|
2022-01-08T14:40:20
|
|
minor update to test/Makefile.os2
|
|
|
4472b831
|
2022-01-08T11:56:56
|
|
PSP_RunCommandQueue: fix argument to PSP_SetBlendState()
src/render/psp/SDL_render_psp.c: In function 'PSP_RunCommandQueue':
src/render/psp/SDL_render_psp.c:1200: warning: passing argument 1 of 'PSP_SetBlendState' from incompatible pointer type
|
|
|
df1474b4
|
2022-01-08T11:55:10
|
|
psp/SDL_sysfilesystem.c: fixed a minor compiler warning.
src/filesystem/psp/SDL_sysfilesystem.c: In function 'SDL_GetPrefPath':
src/filesystem/psp/SDL_sysfilesystem.c:71: warning: passing argument 1 of 'free' discards qualifiers from pointer target type
|
|
|
21914b4e
|
2022-01-08T11:55:10
|
|
updated Makefile.psp and SDL_config_psp.h
|
|
|
f62b8071
|
2022-01-07T21:31:38
|
|
wikiheaders.pl: Put the manpages in man/man3
|
|
|
92946348
|
2022-01-07T18:28:25
|
|
Updated to version 2.0.20 for release
|
|
|
cc201337
|
2022-01-07T20:20:57
|
|
render: Fixed some compiler warnings.
Fixes #5197.
|
|
|
289c3fbb
|
2022-01-07T17:07:22
|
|
Revert "We only need to add the sentinel at the top of SDL_WaitEventTimeout()"
This reverts commit c477768e6f926c38d4cdc236cab8376faf9f6789.
We want to add the sentinel anytime we pump inside SDL_WaitEventTimeout() to avoid pumping again the next time through, as a performance optimization.
|
|
|
1085c317
|
2022-01-07T16:43:53
|
|
Use the requested cursor size instead of the default cursor size
CopyImage() will scale based on the system accessibility settings automatically.
Fixes https://github.com/libsdl-org/SDL/issues/5198
|
|
|
6d3b5d66
|
2022-01-07T18:18:53
|
|
psp: draw lines with RenderGeometry, same as with the other renderers.
|