|
a6a2cdbd
|
2022-06-15T17:28:18
|
|
Added the name of the Qanba Dragon Arcade Joystick in PC mode
|
|
fc78d0ff
|
2022-06-16T00:36:52
|
|
Fix Compilation CI
|
|
|
52f00833
|
2022-06-13T20:58:18
|
|
Add Thread drivers
|
|
|
f0a05ed4
|
2022-06-06T00:04:56
|
|
Add some PS2 flags
|
|
|
580416d3
|
2022-06-06T00:02:08
|
|
Initial CMake configuration
|
|
|
217d3a6d
|
2022-06-13T20:35:27
|
|
Add systimer support
|
|
|
273d9e46
|
2022-06-06T00:00:45
|
|
Fix atomic support for PS2
|
|
|
161a316d
|
2022-06-05T23:52:53
|
|
Add PS2 readme
|
|
|
6b03c1ad
|
2022-06-13T20:59:18
|
|
Add filesystem driver
|
|
|
7343ece8
|
2022-06-06T00:05:43
|
|
Define available PS2 RAM
|
|
|
52f8d035
|
2022-03-20T18:51:37
|
|
Add PS2 CI config
|
|
|
e4409f67
|
2022-06-06T00:05:20
|
|
Include SDL_Log driver
|
|
|
5b4f53bc
|
2022-06-15T14:26:47
|
|
Added SDL_memset.c and SDL_memcpy.c to Xcode project
|
|
cd8f57ee
|
2022-05-29T10:38:56
|
|
cmake: freebsd: Don't REQUIRE libunwind.pc on FreeBSD
FreeBSD includes the libunwind APIs in in the base system libgcc_s and
does not install a .pc file for it.
This change fixes the build on FreeBSD for me.
|
|
ae7446a9
|
2022-06-15T20:48:54
|
|
stlib: Extract SDL_memcpy and SDL_memset to its own file respectively
This is done such that we can disable LTO for these 2 functions when
building with MSVC.
This is due to a limitation of Link Time Code Generation (LTCG).
Code generation might generate a new reference to memset after linking
has started. The LTCG must make assumptions about where memset is
defined which is normally the C runtime.
|
|
|
a5d338bd
|
2022-06-15T14:11:00
|
|
Added support for the Qanba Dragon Arcade Joystick (PS3) on macOS
|
|
|
7d7c4830
|
2022-06-15T14:10:22
|
|
Use python3 for sort_controllers.py
|
|
|
72fea0ed
|
2022-06-15T13:58:28
|
|
Added support for the Qanba Dragon Arcade Joystick
|
|
6bd3e0b1
|
2022-06-15T15:58:48
|
|
Test: Check sqrt and atan against the epsilon.
On i686-linux, the `sqrt_regularCases` and `atan_limitCases` tests would
fail as the result was not precise enough.
|
|
|
b06eda55
|
2022-05-16T20:02:24
|
|
Test: Fix math suite build on Win32.
The cosine precision test now uses an array of double and the result
gets truncated instead of casted to signed int64.
|
|
|
b72b5d0f
|
2022-05-04T12:39:49
|
|
Test: Add math automation test suite.
|
|
|
adb445ea
|
2022-05-16T18:29:08
|
|
Test: Add Cos tests to math suite.
|
|
|
a864180c
|
2022-05-15T19:36:30
|
|
Test: Add float header for FLT_RADIX definition.
|
|
|
7a55fa4e
|
2022-05-15T19:03:53
|
|
Test: Add Scalbn tests to math suite.
|
|
|
afd81237
|
2022-05-15T17:43:38
|
|
Test: Add Sqrt tests to math suite.
|
|
|
c389c32d
|
2022-05-15T17:00:12
|
|
Test: Change assertion type in range tests.
Changes SDLTest_AssertPass(...) to SDLTest_AssertCheck(SDL_FALSE, ...)
for failed assertions so the internal counter gets updated properly.
|
|
|
5ecc75a4
|
2022-05-15T16:30:38
|
|
Test: Add Pow tests to math suite.
|
|
|
6349ad73
|
2022-05-15T11:52:12
|
|
Test: Add Log10 tests to math suite.
|
|
|
cee47a9e
|
2022-06-15T14:36:13
|
|
Test: Use inexact helper for log10 regular cases.
On ARMv6, the result is not precise enough for this function.
|
|
|
7041bbaf
|
2022-05-15T11:36:12
|
|
Test: Add Log tests to math suite.
|
|
|
dd30ff2e
|
2022-05-12T22:19:59
|
|
Test: Add Exp tests to math suite.
|
|
|
a52b8580
|
2022-05-27T23:49:42
|
|
Test: Tidy up test descriptions and documentation.
Test function documentation now lists the input(s) and expected output(s).
Descriptions in TestCaseReference were updated.
|
|
|
aacb5e17
|
2022-05-11T22:17:18
|
|
Test: Extract range test parameters into defines.
|
|
|
75b9aab6
|
2022-05-11T22:04:28
|
|
Test: Add Fmod tests to math suite.
|
|
|
4d7f12f6
|
2022-05-25T01:28:03
|
|
Test: Add Atan2 tests to math suite.
|
|
|
0dbdf90e
|
2022-05-11T12:54:16
|
|
Test: Use SDLCALL in typedefs instead of ifdefs.
Thanks to @sezero for the tip.
|
|
|
2ec48b36
|
2022-05-10T22:55:41
|
|
Test: Fix OS/2 compilation in math suite.
|
|
|
8ebe640a
|
2022-05-25T00:41:06
|
|
Test: Change inexact tests to use an epsilon.
Instead of using `trunc` to check the first ten digits, inexact test now
relies on an epsilon defining an acceptable range for the expected
result to be in.
|
|
|
43c22e5d
|
2022-05-10T22:07:28
|
|
Test: Refactor math suite duplicate code.
Many tests were using the same underlying routine, as such three helper
functions were added:
- A wrapper to test double -> double functions.
- A wrapper to test (double, double) -> double functions,
- A wrapper for range tests on double -> double functions.
|
|
|
3d720ba3
|
2022-05-10T21:05:05
|
|
Test: Refactor math suites edge cases.
Split infinity and zero checks in their own functions.
The result of NAN tests is now logged.
The SDL_TestCaseReference structure were renamed to be more explicit.
|
|
|
62fd6aad
|
2022-05-23T20:12:28
|
|
Test: Add Atan tests to math suite.
|
|
|
a530fc91
|
2022-05-09T22:56:44
|
|
Test: Add Copysign tests to math suite.
|
|
|
48a406db
|
2022-05-06T14:32:50
|
|
Test: Add Fabs tests to math suite.
|
|
|
6a6e93bc
|
2022-05-21T23:11:40
|
|
Test: Add +/-0.0 tests to Acos.
|
|
|
2130dff0
|
2022-05-06T14:19:44
|
|
Test: Add Round tests to math suite.
|
|
|
324b9615
|
2022-05-06T13:57:02
|
|
Test: Add Trunc tests to math suite.
|
|
|
6b4b6d8e
|
2022-05-21T23:01:50
|
|
Test: Add Asin tests to math suite.
|
|
|
b09266a4
|
2022-05-06T13:51:04
|
|
Test: Add Ceil tests to math suite.
|
|
|
9eb09d23
|
2022-05-06T13:39:44
|
|
Test: Refactored and formatted math test suite.
|
|
|
43f6983a
|
2022-05-21T21:51:48
|
|
Test: Add Acos tests to math suite.
|
|
|
0f630e91
|
2022-05-04T21:37:40
|
|
Test: fix indentation in Makefile.in.
|
|
|
4e3a26e2
|
2022-05-04T14:36:49
|
|
Test: Add math test suite to autotools.
|
|
|
95f6edb9
|
2022-05-21T20:07:14
|
|
Test: Refactor trigonometric tests into a helper.
The precision test of these functions need a special helper, it can also
be used for their arc functions down the line.
|
|
|
e64acb61
|
2022-05-04T14:25:10
|
|
Test: Fix for-loop for OS/2 in math suite.
|
|
|
a3a852e9
|
2022-05-04T13:51:42
|
|
Test: Unrolled the array of cases in math suite.
On OS/2, `INFINITY` is a `const double` which cannot be used to
instantiate an array.
|
|
|
3b9f47b8
|
2022-05-20T21:45:31
|
|
Test: Remove early return in pow test.
|
|
|
c23216bf
|
2022-05-04T13:20:09
|
|
Test: Removed static from local variables (Math).
Static would break compilation on OS/2.
|
|
|
6208ed86
|
2022-05-04T13:18:13
|
|
Test: Add math suite to Visual Studio solution.
|
|
|
6be430c7
|
2022-05-16T20:54:59
|
|
Test: Add Tan tests to math suite.
|
|
|
6646edf6
|
2022-05-04T13:10:41
|
|
Test: Fix Windows build for math suite.
|
|
|
d9ff6380
|
2022-05-04T12:51:13
|
|
Test: Add math test suite to Watcom Makefile.
|
|
|
af79b46f
|
2022-05-16T20:34:07
|
|
Test: Add Sin tests to math suite.
|
|
|
7389eba9
|
2022-05-04T12:49:25
|
|
Test: Fix C89 declaration and math include.
|
|
|
b3faebb8
|
2022-05-04T12:40:24
|
|
Test: Add Floor tests to math.
|
|
|
86f50ae7
|
2022-06-15T06:10:25
|
|
ci: Test cmake build script on Android
android
|
|
|
18206446
|
2022-06-15T17:10:31
|
|
cmake: collect SDL_LIBS as list + export PKG_DIRECTFB_LDFLAGS
|
|
|
b5f1fe50
|
2022-06-15T06:17:59
|
|
cmake: Allow integers for SDL_ASSERTIONS + fix -DSDL_ASSERTIONS=0
|
|
|
97711e99
|
2022-06-15T18:57:13
|
|
cmake: no need to explicitly add a depency (target_link_libraries does this implicitly)
|
|
|
ad21c704
|
2022-06-15T05:36:28
|
|
cmake: fix CMP0072 warning by prefering GLVND
|
|
c90e1ec3
|
2021-08-15T21:51:31
|
|
Remove redundant sizeof "size_t" check
The result variables: HAVE_${VARIABLE}, ${VARIABLE}, ${VARIABLE}_CODE,
etc. do not seem to be referenced anywhere in the CMake build script.
|
|
12f9bcd2
|
2022-05-08T17:08:41
|
|
Build tests for all CMake platforms on CI
|
|
f8ae3ef1
|
2022-06-15T22:01:59
|
|
wayland: use libdecor resize edge enums for libdecor
|
|
163f6697
|
2022-06-15T17:00:02
|
|
minor tidy-up to configury.
|
|
|
960b86dc
|
2022-06-14T01:56:40
|
|
update SDL_render_d3d12.c after DPI scaling/highdpi support patch.
See https://github.com/libsdl-org/SDL/pull/5778#issuecomment-1153006938
|
|
|
6d1dfc83
|
2022-06-15T04:29:14
|
|
cmake: make sure SDL2::SDL2 is always available
|
|
|
69df26a0
|
2022-06-15T03:51:46
|
|
ci: test clang+MinGW on CI
|
|
|
32df5aeb
|
2022-06-15T03:16:28
|
|
Check result of FindLibraryAndSONAME before use
|
|
|
8a2c6b60
|
2022-06-15T03:11:33
|
|
cmake: don't link to libibus (not needed, I think)
|
|
|
c950018f
|
2022-06-15T03:07:41
|
|
cmake: don't link to libdbus (see note in SDL_dbus.c)
|
|
|
fb7c03eb
|
2022-06-15T03:04:10
|
|
cmake: install LICENSE.txt
|
|
|
afc350ee
|
2022-06-15T03:03:33
|
|
cmake: do 's#set(SOURCE_FILES #list(APPEND SOURCE_FILES #g'
|
|
ded6c22e
|
2022-06-14T14:09:13
|
|
render: Corrected wrong SDL_memcpy() sizes for viewport, cliprect.
Fixes #5786.
|
|
e427e80b
|
2022-06-14T10:41:18
|
|
wayland: Use the output descriptions from xdg-output when available
Some compositors will provide 'nicer' / 'human readable' output descriptions via the xdg-output protocol. Use these description strings, when available, instead of the model name provided by wl-output. On compositors such as GNOME where this is provided, the display names provided to applications by SDL will now match those in the desktop display settings panel. On compositors where this data isn't provided, the old behavior of using the model string provided by wl-output will remain unchanged.
Additionally, per the protocol spec, output data provided by xdg-output should supersede wl-output data, so this is the recommended behavior in general.
|
|
|
714502d3
|
2022-06-13T22:15:56
|
|
minor windows warning fixes.
|
|
e9749859
|
2022-06-13T16:54:42
|
|
test: Add test coverage for surface size overflows
Signed-off-by: Simon McVittie <smcv@collabora.com>
|
|
|
507ce36d
|
2022-05-10T10:35:06
|
|
video: Note unused SDL_surface creation parameters for removal in SDL 3
Signed-off-by: Simon McVittie <smcv@collabora.com>
|
|
|
535fdc3a
|
2022-05-10T10:34:41
|
|
video: Detect and reject nonsense SDL_surface dimensions
Signed-off-by: Simon McVittie <smcv@collabora.com>
|
|
|
8c9f7104
|
2022-05-10T10:33:54
|
|
video: Harden calculation of SDL_surface pitch and size against overflow
If the width is sufficiently ludicrous, then the calculated pitch or
the image size could conceivably be a signed integer overflow, which
is undefined behaviour. Calculate in the unsigned size_t domain, with
overflow checks.
Signed-off-by: Simon McVittie <smcv@collabora.com>
|
|
|
6606ab8b
|
2022-06-13T18:35:32
|
|
Introduce EXTRA_TEST_LIBS variable for test-only libraries
|
|
|
e99a7144
|
2022-06-13T18:04:51
|
|
Use list(APPEND EXTRA_LIBS) everywhere
|
|
|
851eecdf
|
2022-06-13T18:04:10
|
|
Don't add extra build linker flags to SDL2_static
These leak to users, when SDL2 is vendored.
|
|
|
63b3b9a5
|
2022-06-13T16:02:40
|
|
Fix some typos in diagnostic messages
Detected by Debian's packaging QA tool, Lintian.
Signed-off-by: Simon McVittie <smcv@collabora.com>
|
|
|
c95edd94
|
2022-06-13T06:36:39
|
|
Fixed spacing
|
|
33f8a4dd
|
2022-06-13T13:29:04
|
|
Sync wiki -> header
|
|
|
6258f2ef
|
2022-06-13T06:27:01
|
|
Fixed spacing
|
|
|
73215379
|
2022-06-13T11:51:41
|
|
cmake: Fix static linking to dependencies with "-" in library name
When SDL is built with Wayland support on Linux, and Wayland libraries
are linked as dependencies instead of being loaded with dlopen(), its
dependencies will include libraries whose names contain a dash, like
`-lwayland-client` and `-ldecor-0`. Don't replace such libraries with
`-lwayland` and `-ldecor`: those don't exist and linking them will fail.
Signed-off-by: Simon McVittie <smcv@collabora.com>
|
|
|
bdf14133
|
2022-06-12T23:22:20
|
|
autotools: fixed d3d12 detection result report.
|
|
|
9a036767
|
2022-06-12T15:28:49
|
|
windows: Get better name for the physical display, for Vista and later.
Fixes #5321.
|
|
1963cccc
|
2022-06-12T09:48:57
|
|
CMake: Fix XCode CMake config file. (#5787)
As mentionned in libsdl-org/SDL_net#48 and libsdl-org/SDL_ttf#213:
- Options needs to use `SHELL:` to avoid aggressive option de-duplication
- Framework path needs to be quoted to support paths with spaces.
|
|
b84ce0d4
|
2022-06-11T20:07:26
|
|
Fix typo
platform was spelled like plattform
|