mirrors/SDL
0
0
Fork 0
mirror of https://github.com/libsdl-org/SDL.git synced 2024-11-27 05:43:29 +08:00
Code Issues Packages Projects Releases Wiki Activity
d0b8295c0d
SDL/CMakeLists.txt

2395 lines
81 KiB
CMake
Raw Normal View History

Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(${CMAKE_SOURCE_DIR} STREQUAL ${CMAKE_BINARY_DIR})
message(FATAL_ERROR "Prevented in-tree built. Please create a build directory outside of the SDL source code and call cmake from there")
endif()
cmake: Bump minimum required CMake to 3.0.0 and mark link libraries PRIVATE. Fixes Bugzilla #2992.
2020-06-27 08:11:32 +08:00
cmake_minimum_required(VERSION 3.0.0)
Fixed bug 4639 - CMake build does not generate libhidapi.so for Android Manuel Sabogal I noticed that the current Android.mk builds a libhidapi.so library for Android but the CMake build hasn't been updated to do so. I'll attach a patch that fixes this issue.
2019-05-22 08:33:31 +08:00
project(SDL2 C CXX)
CMake: use the old MACOSX_RPATH behavior (for now), silence the warning.
2016-01-05 14:25:54 +08:00
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
if(WINDOWS_STORE)
enable_language(CXX)
cmake_minimum_required(VERSION 3.11)
add_definitions(-DSDL_BUILDING_WINRT=1 -ZW)
link_libraries(
-nodefaultlib:vccorlib$<$<CONFIG:Debug>:d>
-nodefaultlib:msvcrt$<$<CONFIG:Debug>:d>
vccorlib$<$<CONFIG:Debug>:d>.lib
msvcrt$<$<CONFIG:Debug>:d>.lib
)
endif()
CMake: use the old MACOSX_RPATH behavior (for now), silence the warning.
2016-01-05 14:25:54 +08:00
# !!! FIXME: this should probably do "MACOSX_RPATH ON" as a target property
# !!! FIXME: for the SDL2 shared library (so you get an
# !!! FIXME: install_name ("soname") of "@rpath/libSDL-whatever.dylib"
# !!! FIXME: instead of "/usr/local/lib/libSDL-whatever.dylib"), but I'm
# !!! FIXME: punting for now and leaving the existing behavior. Until this
# !!! FIXME: properly resolved, this line silences a warning in CMake 3.0+.
# !!! FIXME: remove it and this comment entirely once the problem is
# !!! FIXME: properly resolved.
CMake: Turned off Mac OS X rpath warning kludge. Apparently CMake errors out if it doesn't know this policy, and we don't otherwise require CMake 3.0 yet. Sigh.
2016-01-05 18:44:32 +08:00
#cmake_policy(SET CMP0042 OLD)
CMake: use the old MACOSX_RPATH behavior (for now), silence the warning.
2016-01-05 14:25:54 +08:00
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
include(CheckLibraryExists)
include(CheckIncludeFiles)
include(CheckIncludeFile)
include(CheckSymbolExists)
vulkan: Initial Vulkan support! This work was done by Jacob Lifshay and Mark Callow; I'm just merging it into revision control.
2017-08-28 10:15:57 +08:00
include(CheckCSourceCompiles)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
include(CheckCSourceRuns)
include(CheckCCompilerFlag)
include(CheckTypeSize)
include(CheckStructHasMember)
include(CMakeDependentOption)
include(FindPkgConfig)
Fixed bug 3078 - cmake: fix sdl2.m4 install location on cross hosts Timo Gurr On cross hosts running autotools for SDL2_gfx-1.0.1 fails to find sdl2.m4: eautoreconf: running in /var/tmp/paludis/build/media-libs-SDL2_gfx-1.0.1/work/SDL2_gfx-1.0.1 ... aclocal aclocal-1.13: warning: autoconf input should be named 'configure.ac', not 'configure.in' configure.in:128: warning: macro 'AM_PATH_SDL2' not found in library libtoolize --copy --force --automake aclocal aclocal-1.13: warning: autoconf input should be named 'configure.ac', not 'configure.in' configure.in:128: warning: macro 'AM_PATH_SDL2' not found in library autoconf configure.in:128: error: possibly undefined macro: AM_PATH_SDL2 If this token and others are legitimate, please use m4_pattern_allow. See the Autoconf documentation. * Failed Running autoconf ! SDL2 installs it to /usr/x86_64-pc-linux-gnu/share/aclocal on cross hosts, attached patch makes use of CMAKE_INSTALL_FULL_DATAROOTDIR to support correctly installing to the arch independent location /usr/share/aclocal.
2016-10-02 06:11:18 +08:00
include(GNUInstallDirs)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(CMAKE_MODULE_PATH "${SDL2_SOURCE_DIR}/cmake")
include(${SDL2_SOURCE_DIR}/cmake/macros.cmake)
include(${SDL2_SOURCE_DIR}/cmake/sdlchecks.cmake)
# General settings
# Edit include/SDL_version.h and change the version, then:
# SDL_MICRO_VERSION += 1;
# SDL_INTERFACE_AGE += 1;
# SDL_BINARY_AGE += 1;
# if any functions have been added, set SDL_INTERFACE_AGE to 0.
# if backwards compatibility has been broken,
# set SDL_BINARY_AGE and SDL_INTERFACE_AGE to 0.
set(SDL_MAJOR_VERSION 2)
set(SDL_MINOR_VERSION 0)
Updated SDL to version 2.0.15 for development
2020-12-23 02:36:15 +08:00
set(SDL_MICRO_VERSION 15)
set(SDL_INTERFACE_AGE 1)
set(SDL_BINARY_AGE 15)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SDL_VERSION "${SDL_MAJOR_VERSION}.${SDL_MINOR_VERSION}.${SDL_MICRO_VERSION}")
CMakeLists.txt: set dylib version numbers properly. (bug #2915.)
2018-10-25 16:11:02 +08:00
# the following should match the versions in Xcode project file:
CMakeLists.txt: sync DYLIB_CURRENT_VERSION to Xcode project
2020-12-23 03:10:50 +08:00
set(DYLIB_CURRENT_VERSION 15.1.0)
CMakeLists.txt: set dylib version numbers properly. (bug #2915.)
2018-10-25 16:11:02 +08:00
set(DYLIB_COMPATIBILITY_VERSION 1.0.0)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
Fixed bug 3651 - CMake build does not install CMake package configuration tschwinger@elitemail.org Most ironically, although autoconf/automake-based builds install (pretty half-assed) CMake package configuration files, they're missing in installations resulting from CMake-based builds entirely. A proper configuration file typically also loads target exports (implemented in patch 3572, also fixing this issue - see my comment on that issue for details). I believe it would be best to let the dinosaurs go extinct and redirect all build efforts to the CMake end for two reasons: 1. It potentially provides the best user experience, but you'd have to give it some love and ship with less quirky buildfiles. 2. It would force distros to build SDL via CMake and thus would ensure target exports are actually available everywhere. Various CMake patches I submitted today in summary (directly converted from the HG commits and `am`d onto a fork of a git mirror that happened to be on `tip`). https://github.com/tschw/SDL/commits/patched Fixing #2576 #3572, #3613, and this fresh ticket, which is almost entirely advertisement ;). These already do to make SDL much less of a quirky fella to have in your dependency tree...
2017-08-10 10:03:10 +08:00
# Set defaults preventing destination file conflicts
cmake: Set debug library name suffix per target instead of setting it globally. Don't add a suffix on android
2018-02-26 01:22:47 +08:00
set(SDL_CMAKE_DEBUG_POSTFIX "d"
Fixed bug 3651 - CMake build does not install CMake package configuration tschwinger@elitemail.org Most ironically, although autoconf/automake-based builds install (pretty half-assed) CMake package configuration files, they're missing in installations resulting from CMake-based builds entirely. A proper configuration file typically also loads target exports (implemented in patch 3572, also fixing this issue - see my comment on that issue for details). I believe it would be best to let the dinosaurs go extinct and redirect all build efforts to the CMake end for two reasons: 1. It potentially provides the best user experience, but you'd have to give it some love and ship with less quirky buildfiles. 2. It would force distros to build SDL via CMake and thus would ensure target exports are actually available everywhere. Various CMake patches I submitted today in summary (directly converted from the HG commits and `am`d onto a fork of a git mirror that happened to be on `tip`). https://github.com/tschw/SDL/commits/patched Fixing #2576 #3572, #3613, and this fresh ticket, which is almost entirely advertisement ;). These already do to make SDL much less of a quirky fella to have in your dependency tree...
2017-08-10 10:03:10 +08:00
CACHE STRING "Name suffix for debug builds")
cmake: Set debug library name suffix per target instead of setting it globally. Don't add a suffix on android
2018-02-26 01:22:47 +08:00
mark_as_advanced(CMAKE_IMPORT_LIBRARY_SUFFIX SDL_CMAKE_DEBUG_POSTFIX)
Fixed bug 3651 - CMake build does not install CMake package configuration tschwinger@elitemail.org Most ironically, although autoconf/automake-based builds install (pretty half-assed) CMake package configuration files, they're missing in installations resulting from CMake-based builds entirely. A proper configuration file typically also loads target exports (implemented in patch 3572, also fixing this issue - see my comment on that issue for details). I believe it would be best to let the dinosaurs go extinct and redirect all build efforts to the CMake end for two reasons: 1. It potentially provides the best user experience, but you'd have to give it some love and ship with less quirky buildfiles. 2. It would force distros to build SDL via CMake and thus would ensure target exports are actually available everywhere. Various CMake patches I submitted today in summary (directly converted from the HG commits and `am`d onto a fork of a git mirror that happened to be on `tip`). https://github.com/tschw/SDL/commits/patched Fixing #2576 #3572, #3613, and this fresh ticket, which is almost entirely advertisement ;). These already do to make SDL much less of a quirky fella to have in your dependency tree...
2017-08-10 10:03:10 +08:00
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# Calculate a libtool-like version number
math(EXPR LT_CURRENT "${SDL_MICRO_VERSION} - ${SDL_INTERFACE_AGE}")
math(EXPR LT_AGE "${SDL_BINARY_AGE} - ${SDL_INTERFACE_AGE}")
math(EXPR LT_MAJOR "${LT_CURRENT}- ${LT_AGE}")
set(LT_REVISION "${SDL_INTERFACE_AGE}")
set(LT_RELEASE "${SDL_MAJOR_VERSION}.${SDL_MINOR_VERSION}")
set(LT_VERSION "${LT_MAJOR}.${LT_AGE}.${LT_REVISION}")
cmake: Comment out some debug logging that can upset build environments. Something about the dashes, colons, numbers makes some builders believe this is an error message.
2018-12-02 15:43:32 +08:00
#message(STATUS "${LT_VERSION} :: ${LT_AGE} :: ${LT_REVISION} :: ${LT_CURRENT} :: ${LT_RELEASE}")
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# General settings & flags
set(LIBRARY_OUTPUT_DIRECTORY "build")
# Check for 64 or 32 bit
set(SIZEOF_VOIDP ${CMAKE_SIZEOF_VOID_P})
if(CMAKE_SIZEOF_VOID_P EQUAL 8)
set(ARCH_64 TRUE)
set(PROCESSOR_ARCH "x64")
else()
set(ARCH_64 FALSE)
set(PROCESSOR_ARCH "x86")
endif()
set(LIBNAME SDL2)
if(NOT LIBTYPE)
set(LIBTYPE SHARED)
endif()
# Get the platform
if(WIN32)
if(NOT WINDOWS)
set(WINDOWS TRUE)
endif()
elseif(UNIX AND NOT APPLE)
if(CMAKE_SYSTEM_NAME MATCHES ".*Linux")
set(LINUX TRUE)
elseif(CMAKE_SYSTEM_NAME MATCHES "kFreeBSD.*")
set(FREEBSD TRUE)
elseif(CMAKE_SYSTEM_NAME MATCHES "kNetBSD.*|NetBSD.*")
set(NETBSD TRUE)
elseif(CMAKE_SYSTEM_NAME MATCHES "kOpenBSD.*|OpenBSD.*")
set(OPENBSD TRUE)
elseif(CMAKE_SYSTEM_NAME MATCHES ".*GNU.*")
set(GNU TRUE)
elseif(CMAKE_SYSTEM_NAME MATCHES ".*BSDI.*")
set(BSDI TRUE)
elseif(CMAKE_SYSTEM_NAME MATCHES "DragonFly.*|FreeBSD")
set(FREEBSD TRUE)
elseif(CMAKE_SYSTEM_NAME MATCHES "SYSV5.*")
set(SYSV5 TRUE)
elseif(CMAKE_SYSTEM_NAME MATCHES "Solaris.*")
set(SOLARIS TRUE)
elseif(CMAKE_SYSTEM_NAME MATCHES "HP-UX.*")
set(HPUX TRUE)
elseif(CMAKE_SYSTEM_NAME MATCHES "AIX.*")
set(AIX TRUE)
elseif(CMAKE_SYSTEM_NAME MATCHES "Minix.*")
set(MINIX TRUE)
endif()
elseif(APPLE)
if(CMAKE_SYSTEM_NAME MATCHES ".*Darwin.*")
set(DARWIN TRUE)
elseif(CMAKE_SYSTEM_NAME MATCHES ".*MacOS.*")
set(MACOSX TRUE)
CMake: tvOS support/fixes To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=tvOS - CMAKE_OSX_SYSROOT=<SDK> (examples: appletvos, appletvsimulator, appletvos12.4, /full/path/to/AppleTVOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;x86_64")
2019-08-28 00:30:20 +08:00
elseif(CMAKE_SYSTEM_NAME MATCHES ".*tvOS.*")
set(TVOS TRUE)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
# TODO: iOS?
elseif(CMAKE_SYSTEM_NAME MATCHES "BeOS.*")
message_error("BeOS support has been removed as of SDL 2.0.2.")
elseif(CMAKE_SYSTEM_NAME MATCHES "Haiku.*")
set(HAIKU TRUE)
endif()
# Don't mistake osx for unix
Add basic support for compiling on RISC OS
2020-02-14 04:50:47 +08:00
if(UNIX AND NOT APPLE AND NOT RISCOS)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(UNIX_SYS ON)
else()
set(UNIX_SYS OFF)
endif()
if(UNIX OR APPLE)
set(UNIX_OR_MAC_SYS ON)
else()
set(UNIX_OR_MAC_SYS OFF)
endif()
if (UNIX_OR_MAC_SYS AND NOT EMSCRIPTEN) # JavaScript does not yet have threading support, so disable pthreads when building for Emscripten.
set(SDL_PTHREADS_ENABLED_BY_DEFAULT ON)
else()
set(SDL_PTHREADS_ENABLED_BY_DEFAULT OFF)
endif()
# Default option knobs
if(APPLE OR ARCH_64)
Android cmake build fixed to work with the official android gradle plugin
2017-06-05 02:25:57 +08:00
if(NOT "${CMAKE_OSX_ARCHITECTURES}" MATCHES "arm")
set(OPT_DEF_SSEMATH ON)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
if(UNIX OR MINGW OR MSYS)
set(OPT_DEF_LIBC ON)
endif()
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
# The hidraw support doesn't catch Xbox, PS4 and Nintendo controllers,
hidapi: Add SDL_hidapi.c, allows support for multiple hidapi backends. This is currently supported on Linux and macOS. iOS and Android are not supported at all, Windows support could be added with some changes to the libusb backend. The Visual Studio and Xcode projects do not use this feature. Based on Valve Software's hid.cpp, written in collaboration with Andrew Eikum.
2019-08-01 00:20:55 +08:00
# so we'll just use libusb when it's available. libusb does not support iOS,
# so we default to yes on iOS.
# TODO: Windows can support libusb, the hid.c file just depends on Unix APIs
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/disable-hidapi-for-uwp.patch
2020-12-09 01:04:28 +08:00
if((WINDOWS AND NOT WINDOWS_STORE) OR IOS OR TVOS OR ANDROID)
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
set(HIDAPI_SKIP_LIBUSB TRUE)
else()
set(HIDAPI_SKIP_LIBUSB FALSE)
endif()
if (HIDAPI_SKIP_LIBUSB)
set(OPT_DEF_HIDAPI ON)
endif()
hidapi: Add SDL_hidapi.c, allows support for multiple hidapi backends. This is currently supported on Linux and macOS. iOS and Android are not supported at all, Windows support could be added with some changes to the libusb backend. The Visual Studio and Xcode projects do not use this feature. Based on Valve Software's hid.cpp, written in collaboration with Andrew Eikum.
2019-08-01 00:20:55 +08:00
# On the other hand, *BSD specifically uses libusb only, so we make a special
# case just for them.
if(FREEBSD OR NETBSD OR OPENBSD OR BSDI)
set(HIDAPI_ONLY_LIBUSB TRUE)
else()
set(HIDAPI_ONLY_LIBUSB FALSE)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# Compiler info
if(CMAKE_COMPILER_IS_GNUCC)
set(USE_GCC TRUE)
set(OPT_DEF_ASM TRUE)
elseif(CMAKE_C_COMPILER_ID MATCHES "Clang")
set(USE_CLANG TRUE)
set(OPT_DEF_ASM TRUE)
elseif(MSVC_VERSION GREATER 1400) # VisualStudio 8.0+
set(OPT_DEF_ASM TRUE)
#set(CMAKE_C_FLAGS "/ZI /WX- /
else()
set(OPT_DEF_ASM FALSE)
endif()
Android build fixes -Enabling checking for GCC_ATOMICS also on clang by default. This way all Android ABIs build successfully -Android cmake: Threading was not enabled correctly -Android cmake: Timers and dynamic lib loading were not included in the sources
2017-06-09 03:07:55 +08:00
if(USE_GCC OR USE_CLANG)
set(OPT_DEF_GCC_ATOMICS ON)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# Default flags, if not set otherwise
if("$ENV{CFLAGS}" STREQUAL "")
CMake: only set "-O3 -g" defaults if CMAKE_BUILD_TYPE wasn't set at all.
2016-01-08 20:32:51 +08:00
if(CMAKE_BUILD_TYPE STREQUAL "")
if(USE_GCC OR USE_CLANG)
set(CMAKE_C_FLAGS "-g -O3")
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
else()
set(CMAKE_C_FLAGS "$ENV{CFLAGS}")
list(APPEND EXTRA_CFLAGS "$ENV{CFLAGS}")
endif()
if(NOT ("$ENV{CFLAGS}" STREQUAL "")) # Hackish, but does the trick on Win32
list(APPEND EXTRA_LDFLAGS "$ENV{LDFLAGS}")
endif()
if(MSVC)
option(FORCE_STATIC_VCRT "Force /MT for static VC runtimes" OFF)
if(FORCE_STATIC_VCRT)
foreach(flag_var
CMAKE_C_FLAGS CMAKE_C_FLAGS_DEBUG CMAKE_C_FLAGS_RELEASE
CMAKE_C_FLAGS_MINSIZEREL CMAKE_C_FLAGS_RELWITHDEBINFO)
if(${flag_var} MATCHES "/MD")
string(REGEX REPLACE "/MD" "/MT" ${flag_var} "${${flag_var}}")
endif()
endforeach()
endif()
CMake: Fixed building for Windows with VS2015 (bug #3080). - Don't try to link with the Visual C runtime. - Avoid code generation that would use functions from the VC runtime.
2016-01-01 03:26:40 +08:00
# Make sure /RTC1 is disabled, otherwise it will use functions from the CRT
foreach(flag_var
CMAKE_C_FLAGS CMAKE_C_FLAGS_DEBUG CMAKE_C_FLAGS_RELEASE
CMAKE_C_FLAGS_MINSIZEREL CMAKE_C_FLAGS_RELWITHDEBINFO)
string(REGEX REPLACE "/RTC(su|[1su])" "" ${flag_var} "${${flag_var}}")
endforeach(flag_var)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
# Those are used for pkg-config and friends, so that the SDL2.pc, sdl2-config,
# etc. are created correctly.
set(SDL_LIBS "-lSDL2")
set(SDL_CFLAGS "")
Remove "lib" prefix from DLL file on MinGW builds closes bug #4209.
2018-06-29 23:29:17 +08:00
# When building shared lib for Windows with MinGW,
# avoid the DLL having a "lib" prefix
cmake: use WINDOWS instead of WIN32.
2018-08-03 04:21:43 +08:00
if(WINDOWS)
Remove "lib" prefix from DLL file on MinGW builds closes bug #4209.
2018-06-29 23:29:17 +08:00
set(CMAKE_SHARED_LIBRARY_PREFIX "")
endif()
Added debug postfix to install command and fixed library path
2018-04-10 23:03:54 +08:00
# Emscripten toolchain has a nonempty default value for this, and the checks
# in this file need to change that, so remember the original value, and
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# restore back to that afterwards. For check_function_exists() to work in
# Emscripten, this value must be at its default value.
set(ORIG_CMAKE_REQUIRED_FLAGS ${CMAKE_REQUIRED_FLAGS})
if(CYGWIN)
# We build SDL on cygwin without the UNIX emulation layer
include_directories("-I/usr/include/mingw")
Fix CMake CMAKE_REQUIRED_FLAGS handling to preserve existing CMake flags required by Emscripten toolchain.
2016-09-13 15:03:50 +08:00
set(CMAKE_REQUIRED_FLAGS "${CMAKE_REQUIRED_FLAGS} -mno-cygwin")
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
check_c_source_compiles("int main(int argc, char **argv) {}"
HAVE_GCC_NO_CYGWIN)
set(CMAKE_REQUIRED_FLAGS ${ORIG_CMAKE_REQUIRED_FLAGS})
if(HAVE_GCC_NO_CYGWIN)
list(APPEND EXTRA_LDFLAGS "-mno-cygwin")
list(APPEND SDL_LIBS "-mno-cygwin")
endif()
set(SDL_CFLAGS "${SDL_CFLAGS} -I/usr/include/mingw")
endif()
add_definitions(-DUSING_GENERATED_CONFIG_H)
# General includes
Fix QNX build - prioritize system EGL headers over the Khronos ones
2017-08-30 15:40:06 +08:00
include_directories(${SDL2_BINARY_DIR}/include ${SDL2_SOURCE_DIR}/include)
Fixed the include path for khronos with Visual Studio
2017-09-11 01:25:36 +08:00
if(USE_GCC OR USE_CLANG)
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/fix-space-in-path.patch
2020-12-09 01:16:34 +08:00
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -idirafter \"${SDL2_SOURCE_DIR}/src/video/khronos\"")
Fixed the include path for khronos with Visual Studio
2017-09-11 01:25:36 +08:00
else()
include_directories(${SDL2_SOURCE_DIR}/src/video/khronos)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# All these ENABLED_BY_DEFAULT vars will default to ON if not specified, so
# you only need to have a platform override them if they are disabling.
set(OPT_DEF_ASM TRUE)
if(EMSCRIPTEN)
# Set up default values for the currently supported set of subsystems:
Added debug postfix to install command and fixed library path
2018-04-10 23:03:54 +08:00
# Emscripten/Javascript does not have assembly support, a dynamic library
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# loading architecture, low-level CPU inspection or multithreading.
set(OPT_DEF_ASM FALSE)
set(SDL_SHARED_ENABLED_BY_DEFAULT OFF)
set(SDL_ATOMIC_ENABLED_BY_DEFAULT OFF)
set(SDL_THREADS_ENABLED_BY_DEFAULT OFF)
set(SDL_LOADSO_ENABLED_BY_DEFAULT OFF)
set(SDL_CPUINFO_ENABLED_BY_DEFAULT OFF)
set(SDL_DLOPEN_ENABLED_BY_DEFAULT OFF)
endif()
Fixed bug 4593 - Respect CMake's BUILD_SHARED_LIBS default behavior tschwinger Respect the BUILD_SHARED_LIBS variable when defined, and build either shared or static libs, which is CMake's default behavior (See https://cmake.org/cmake/help/latest/variable/BUILD_SHARED_LIBS.html). If the variable is not defined, the current behavior remains unchanged and both variants are built where the platform supports it. This way, it remains possible to build both in one shot, which seems convenient for distro builds and useful to promote some consistency between them.
2019-06-09 05:36:03 +08:00
# When defined, respect CMake's BUILD_SHARED_LIBS setting:
set(SDL_STATIC_ENABLED_BY_DEFAULT ON)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if (NOT DEFINED SDL_SHARED_ENABLED_BY_DEFAULT)
Fixed bug 4593 - Respect CMake's BUILD_SHARED_LIBS default behavior tschwinger Respect the BUILD_SHARED_LIBS variable when defined, and build either shared or static libs, which is CMake's default behavior (See https://cmake.org/cmake/help/latest/variable/BUILD_SHARED_LIBS.html). If the variable is not defined, the current behavior remains unchanged and both variants are built where the platform supports it. This way, it remains possible to build both in one shot, which seems convenient for distro builds and useful to promote some consistency between them.
2019-06-09 05:36:03 +08:00
# ...unless decided already (as for EMSCRIPTEN)
set(SDL_SHARED_ENABLED_BY_DEFAULT OFF)
if (NOT DEFINED BUILD_SHARED_LIBS)
# No preference? Build both, just like the AC/AM configure
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SDL_SHARED_ENABLED_BY_DEFAULT ON)
Fixed bug 4593 - Respect CMake's BUILD_SHARED_LIBS default behavior tschwinger Respect the BUILD_SHARED_LIBS variable when defined, and build either shared or static libs, which is CMake's default behavior (See https://cmake.org/cmake/help/latest/variable/BUILD_SHARED_LIBS.html). If the variable is not defined, the current behavior remains unchanged and both variants are built where the platform supports it. This way, it remains possible to build both in one shot, which seems convenient for distro builds and useful to promote some consistency between them.
2019-06-09 05:36:03 +08:00
elseif (BUILD_SHARED_LIBS)
# In this case, we assume the user wants a shared lib and don't build
# the static one
set(SDL_SHARED_ENABLED_BY_DEFAULT ON)
set(SDL_STATIC_ENABLED_BY_DEFAULT OFF)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
set(SDL_SUBSYSTEMS
Atomic Audio Video Render Events Joystick Haptic Power Threads Timers
locale: Implemented SDL_GetPreferredLocales(). This was something I proposed a long time ago, Sylvain Becker did additional work on it, then back to me. Fixes Bugzilla #2131.
2020-05-04 14:27:29 +08:00
File Loadso CPUinfo Filesystem Dlopen Sensor Locale)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
foreach(_SUB ${SDL_SUBSYSTEMS})
string(TOUPPER ${_SUB} _OPT)
if (NOT DEFINED SDL_${_OPT}_ENABLED_BY_DEFAULT)
set(SDL_${_OPT}_ENABLED_BY_DEFAULT ON)
endif()
option(SDL_${_OPT} "Enable the ${_SUB} subsystem" ${SDL_${_OPT}_ENABLED_BY_DEFAULT})
endforeach()
option_string(ASSERTIONS "Enable internal sanity checks (auto/disabled/release/enabled/paranoid)" "auto")
#set_option(DEPENDENCY_TRACKING "Use gcc -MMD -MT dependency tracking" ON)
set_option(LIBC "Use the system C library" ${OPT_DEF_LIBC})
Android build fixes -Enabling checking for GCC_ATOMICS also on clang by default. This way all Android ABIs build successfully -Android cmake: Threading was not enabled correctly -Android cmake: Timers and dynamic lib loading were not included in the sources
2017-06-09 03:07:55 +08:00
set_option(GCC_ATOMICS "Use gcc builtin atomics" ${OPT_DEF_GCC_ATOMICS})
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set_option(ASSEMBLY "Enable assembly routines" ${OPT_DEF_ASM})
set_option(SSEMATH "Allow GCC to use SSE floating point math" ${OPT_DEF_SSEMATH})
set_option(MMX "Use MMX assembly routines" ${OPT_DEF_ASM})
set_option(3DNOW "Use 3Dnow! MMX assembly routines" ${OPT_DEF_ASM})
set_option(SSE "Use SSE assembly routines" ${OPT_DEF_ASM})
set_option(SSE2 "Use SSE2 assembly routines" ${OPT_DEF_SSEMATH})
audio: Wired up new SSE code to build system.
2017-01-23 14:05:44 +08:00
set_option(SSE3 "Use SSE3 assembly routines" ${OPT_DEF_SSEMATH})
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set_option(ALTIVEC "Use Altivec assembly routines" ${OPT_DEF_ASM})
configure/cmake: Disable ARM SIMD and NEON by default. It's buggy at the moment.
2020-06-27 11:30:59 +08:00
set_option(ARMSIMD "use SIMD assembly blitters on ARM" OFF)
set_option(ARMNEON "use NEON assembly blitters on ARM" OFF)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set_option(DISKAUDIO "Support the disk writer audio driver" ON)
set_option(DUMMYAUDIO "Support the dummy audio driver" ON)
set_option(VIDEO_DIRECTFB "Use DirectFB video driver" OFF)
dep_option(DIRECTFB_SHARED "Dynamically load directfb support" ON "VIDEO_DIRECTFB" OFF)
set_option(VIDEO_DUMMY "Use dummy video driver" ON)
set_option(VIDEO_OPENGL "Include OpenGL support" ON)
set_option(VIDEO_OPENGLES "Include OpenGL ES support" ON)
set_option(PTHREADS "Use POSIX threads for multi-threading" ${SDL_PTHREADS_ENABLED_BY_DEFAULT})
dep_option(PTHREADS_SEM "Use pthread semaphores" ON "PTHREADS" OFF)
set_option(SDL_DLOPEN "Use dlopen for shared object loading" ${SDL_DLOPEN_ENABLED_BY_DEFAULT})
Add basic support for compiling on RISC OS
2020-02-14 04:50:47 +08:00
dep_option(OSS "Support the OSS audio API" ON "UNIX_SYS OR RISCOS" OFF)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set_option(ALSA "Support the ALSA audio API" ${UNIX_SYS})
dep_option(ALSA_SHARED "Dynamically load ALSA audio support" ON "ALSA" OFF)
jack: Initial shot at a JACK audio target. http://jackaudio.org/ Fixes Bugzilla #2163. (with several more commits following to improve this code.)
2017-06-09 01:27:58 +08:00
set_option(JACK "Support the JACK audio API" ${UNIX_SYS})
dep_option(JACK_SHARED "Dynamically load JACK audio support" ON "JACK" OFF)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set_option(ESD "Support the Enlightened Sound Daemon" ${UNIX_SYS})
dep_option(ESD_SHARED "Dynamically load ESD audio support" ON "ESD" OFF)
set_option(PULSEAUDIO "Use PulseAudio" ${UNIX_SYS})
dep_option(PULSEAUDIO_SHARED "Dynamically load PulseAudio support" ON "PULSEAUDIO" OFF)
set_option(ARTS "Support the Analog Real Time Synthesizer" ${UNIX_SYS})
dep_option(ARTS_SHARED "Dynamically load aRts audio support" ON "ARTS" OFF)
set_option(NAS "Support the NAS audio API" ${UNIX_SYS})
set_option(NAS_SHARED "Dynamically load NAS audio API" ${UNIX_SYS})
set_option(SNDIO "Support the sndio audio API" ${UNIX_SYS})
Fixed bug 5105 - sndio support not working in dynamic mode (dlopen) Giovanni Bajo The CMake build system supports several audio frameworks for Linux: one of them is sndio. All frameworks can be built with "runtime linking" (that is, using dlopen to load the library at runtime). In sdlchecks.cmake, there's code to do the same with sndio: ================================================================= # Requires: # - n/a # Optional: # - SNDIO_SHARED opt # - HAVE_DLOPEN opt macro(CheckSNDIO) if(SNDIO) # TODO: set include paths properly, so the sndio headers are found check_include_file(sndio.h HAVE_SNDIO_H) find_library(D_SNDIO_LIB sndio) if(HAVE_SNDIO_H AND D_SNDIO_LIB) set(HAVE_SNDIO TRUE) file(GLOB SNDIO_SOURCES ${SDL2_SOURCE_DIR}/src/audio/sndio/*.c) set(SOURCE_FILES ${SOURCE_FILES} ${SNDIO_SOURCES}) set(SDL_AUDIO_DRIVER_SNDIO 1) if(SNDIO_SHARED) if(NOT HAVE_DLOPEN) message_warn("You must have SDL_LoadObject() support for dynamic sndio loading") else() FindLibraryAndSONAME("sndio") set(SDL_AUDIO_DRIVER_SNDIO_DYNAMIC "\"${SNDIO_LIB_SONAME}\"") set(HAVE_SNDIO_SHARED TRUE) endif() else() list(APPEND EXTRA_LIBS ${D_SNDIO_LIB}) endif() set(HAVE_SDL_AUDIO TRUE) endif() endif() endmacro() ================================================================= The feature is gated by an option called SNDIO_SHARED. It is also fully implemented in SDL_sndioaudio.c Unfortunately, it seems there is a missing line in CMakeLists.txt, so SNDIO_SHARED is not defined: ====================================================================== set_option(ALSA "Support the ALSA audio API" ${UNIX_SYS}) dep_option(ALSA_SHARED "Dynamically load ALSA audio support" ON "ALSA" OFF) set_option(JACK "Support the JACK audio API" ${UNIX_SYS}) dep_option(JACK_SHARED "Dynamically load JACK audio support" ON "JACK" OFF) set_option(ESD "Support the Enlightened Sound Daemon" ${UNIX_SYS}) dep_option(ESD_SHARED "Dynamically load ESD audio support" ON "ESD" OFF) set_option(PULSEAUDIO "Use PulseAudio" ${UNIX_SYS}) dep_option(PULSEAUDIO_SHARED "Dynamically load PulseAudio support" ON "PULSEAUDIO" OFF) set_option(ARTS "Support the Analog Real Time Synthesizer" ${UNIX_SYS}) dep_option(ARTS_SHARED "Dynamically load aRts audio support" ON "ARTS" OFF) set_option(NAS "Support the NAS audio API" ${UNIX_SYS}) set_option(NAS_SHARED "Dynamically load NAS audio API" ${UNIX_SYS}) set_option(SNDIO "Support the sndio audio API" ${UNIX_SYS}) set_option(FUSIONSOUND "Use FusionSound audio driver" OFF) dep_option(FUSIONSOUND_SHARED "Dynamically load fusionsound audio support" ON "FUSIONSOUND" OFF) ====================================================================== You can see that all frameworks define a "dep_option" NAME_SHARED, and SNDIO is the only one where the option is missing. This means that runtime loading of sndio is never activated. If sndio is found at configuration time, it is always activated in "linked" mode, so that the final binary will have a load-time dependency with libsdnio. This is unfortunate. To fix the problem, it is sufficient to add this line: dep_option(SNDIO_SHARED "Dynamically load the sndio audio API" ${UNIX_SYS} ON "SNDIO" OFF) I've verified that this fixes the bug, and sndio can now be dynamically loaded as expected.
2020-05-30 07:31:05 +08:00
dep_option(SNDIO_SHARED "Dynamically load the sndio audio API" ${UNIX_SYS} ON "SNDIO" OFF)
Added configure and cmake support for libsamplerate
2017-01-07 12:43:53 +08:00
set_option(FUSIONSOUND "Use FusionSound audio driver" OFF)
dep_option(FUSIONSOUND_SHARED "Dynamically load fusionsound audio support" ON "FUSIONSOUND" OFF)
set_option(LIBSAMPLERATE "Use libsamplerate for audio rate conversion" ${UNIX_SYS})
dep_option(LIBSAMPLERATE_SHARED "Dynamically load libsamplerate" ON "LIBSAMPLERATE" OFF)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set_option(RPATH "Use an rpath when linking SDL" ${UNIX_SYS})
set_option(CLOCK_GETTIME "Use clock_gettime() instead of gettimeofday()" OFF)
set_option(VIDEO_X11 "Use X11 video driver" ${UNIX_SYS})
set_option(VIDEO_WAYLAND "Use Wayland video driver" ${UNIX_SYS})
dep_option(WAYLAND_SHARED "Dynamically load Wayland support" ON "VIDEO_WAYLAND" OFF)
dep_option(VIDEO_WAYLAND_QT_TOUCH "QtWayland server support for Wayland video driver" ON "VIDEO_WAYLAND" OFF)
set_option(VIDEO_RPI "Use Raspberry Pi video driver" ${UNIX_SYS})
dep_option(X11_SHARED "Dynamically load X11 support" ON "VIDEO_X11" OFF)
set(SDL_X11_OPTIONS Xcursor Xinerama XInput Xrandr Xscrnsaver XShape Xvm)
foreach(_SUB ${SDL_X11_OPTIONS})
string(TOUPPER "VIDEO_X11_${_SUB}" _OPT)
dep_option(${_OPT} "Enable ${_SUB} support" ON "VIDEO_X11" OFF)
endforeach()
set_option(VIDEO_COCOA "Use Cocoa video driver" ${APPLE})
set_option(DIRECTX "Use DirectX for Windows audio/video" ${WINDOWS})
configury, cmake: make wasapi option independent of directx.
2018-09-28 16:30:50 +08:00
set_option(WASAPI "Use the Windows WASAPI audio driver" ${WINDOWS})
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set_option(RENDER_D3D "Enable the Direct3D render driver" ${WINDOWS})
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
set_option(RENDER_METAL "Enable the Metal render driver" ${APPLE})
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set_option(VIDEO_VIVANTE "Use Vivante EGL video driver" ${UNIX_SYS})
Be clear that disabling Vulkan surface support disables the entire SDL Vulkan integration
2017-08-28 15:11:38 +08:00
dep_option(VIDEO_VULKAN "Enable Vulkan support" ON "ANDROID OR APPLE OR LINUX OR WINDOWS" OFF)
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
set_option(VIDEO_METAL "Enable Metal support" ${APPLE})
Fixed bug 3690 - SDL2 KMS/DRM render context support Manuel The attached patch adds support for KMS/DRM context graphics. It builds with no problem on X86_64 GNU/Linux systems, provided the needed libraries are present, and on ARM GNU/Linux systems that have KMS/DRM support and a GLES2 implementation. Tested on Raspberry Pi: KMS/DRM is what the Raspberry Pi will use as default in the near future, once the propietary DispmanX API by Broadcom is overtaken by open graphics stack, it's possible to boot current Raspbian system in KMS mode by adding "dtoverlay=vc4-kms-v3d" to config.txt on Raspbian's boot partition. X86 systems use KMS right away in every current GNU/Linux system. Simple build instructions: $./autogen.sh $./configure --enable-video-kmsdrm $make
2017-08-03 01:22:48 +08:00
set_option(VIDEO_KMSDRM "Use KMS DRM video driver" ${UNIX_SYS})
dep_option(KMSDRM_SHARED "Dynamically load KMS DRM support" ON "VIDEO_KMSDRM" OFF)
offscreen: Add new video driver backend Offscreen The Offscreen video driver is intended to be used for headless rendering as well as allows for multiple GPUs to be used for headless rendering Currently only supports EGL (OpenGL / ES) or Framebuffers Adds a hint to specifiy which EGL device to use: SDL_HINT_EGL_DEVICE Adds testoffscreen.c which can be used to test the backend out Disabled by default for now
2019-09-25 04:36:48 +08:00
set_option(VIDEO_OFFSCREEN "Use offscreen video driver" OFF)
events: Let arbitrary signals to simulate iOS/Android backgrounding events. This lets you build a custom embedded device that roughly offers the "this process is going to the background NOW" semantics of SDL on a mobile device.
2019-03-16 03:51:05 +08:00
option_string(BACKGROUNDING_SIGNAL "number to use for magic backgrounding signal or 'OFF'" "OFF")
option_string(FOREGROUNDING_SIGNAL "number to use for magic foregrounding signal or 'OFF'" "OFF")
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
set_option(HIDAPI "Use HIDAPI for low level joystick drivers" ${OPT_DEF_HIDAPI})
Fixed bug 5028 - Virtual Joysticks (new joystick backend) David Ludwig I have created a new driver for SDL's Joystick and Game-Controller subsystem: a Virtual driver. This driver allows one to create a software-based joystick, which to SDL applications will look and react like a real joystick, but whose state can be set programmatically. A primary use case for this is to help enable developers to add touch-screen joysticks to their apps. The driver comes with a set of new, public APIs, with functions to attach and detach joysticks, set virtual-joystick state, and to determine if a joystick is a virtual-one. Use of virtual joysticks goes as such: 1. Attach one or more virtual joysticks by calling SDL_JoystickAttachVirtual. If successful, this returns the virtual-device's joystick-index. 2. Open the virtual joysticks (using indicies returned by SDL_JoystickAttachVirtual). 3. Call any of the SDL_JoystickSetVirtual* functions when joystick-state changes. Please note that virtual-joystick state will only get applied on the next call to SDL_JoystickUpdate, or when pumping or polling for SDL events (via SDL_PumpEvents or SDL_PollEvent). Here is a listing of the new, public APIs, at present and subject to change: ------------------------------------------------------------ /** * Attaches a new virtual joystick. * Returns the joystick's device index, or -1 if an error occurred. */ extern DECLSPEC int SDLCALL SDL_JoystickAttachVirtual(SDL_JoystickType type, int naxes, int nballs, int nbuttons, int nhats); /** * Detaches a virtual joystick * Returns 0 on success, or -1 if an error occurred. */ extern DECLSPEC int SDLCALL SDL_JoystickDetachVirtual(int device_index); /** * Indicates whether or not a virtual-joystick is at a given device index. */ extern DECLSPEC SDL_bool SDLCALL SDL_JoystickIsVirtual(int device_index); /** * Set values on an opened, virtual-joystick's controls. * Returns 0 on success, -1 on error. */ extern DECLSPEC int SDLCALL SDL_JoystickSetVirtualAxis(SDL_Joystick * joystick, int axis, Sint16 value); extern DECLSPEC int SDLCALL SDL_JoystickSetVirtualBall(SDL_Joystick * joystick, int ball, Sint16 xrel, Sint16 yrel); extern DECLSPEC int SDLCALL SDL_JoystickSetVirtualButton(SDL_Joystick * joystick, int button, Uint8 value); extern DECLSPEC int SDLCALL SDL_JoystickSetVirtualHat(SDL_Joystick * joystick, int hat, Uint8 value); ------------------------------------------------------------ Miscellaneous notes on the initial patch, which are also subject to change: 1. no test code is present in SDL, yet. This should, perhaps, change. Initial development was done with an ImGui-based app, which potentially is too thick for use in SDL-official. If tests are to be added, what kind of tests? Automated? Graphical? 2. virtual game controllers can be created by calling SDL_JoystickAttachVirtual with a joystick-type of SDL_JOYSTICK_TYPE_GAME_CONTROLLER, with naxes (num axes) set to SDL_CONTROLLER_AXIS_MAX, and with nbuttons (num buttons) set to SDL_CONTROLLER_BUTTON_MAX. When updating their state, values of type SDL_GameControllerAxis or SDL_GameControllerButton can be casted to an int and used for the control-index (in calls to SDL_JoystickSetVirtual* functions). 3. virtual joysticks' guids are mostly all-zeros with the exception of the last two bytes, the first of which is a 'v', to indicate that the guid is a virtual one, and the second of which is a SDL_JoystickType that has been converted into a Uint8. 4. virtual joysticks are ONLY turned into virtual game-controllers if and when their joystick-type is set to SDL_JOYSTICK_TYPE_GAMECONTROLLER. This is controlled by having SDL's default list of game-controllers have a single entry for a virtual game controller (of guid, "00000000000000000000000000007601", which is subject to the guid-encoding described above). 5. regarding having to call SDL_JoystickUpdate, either directly or indirectly via SDL_PumpEvents or SDL_PollEvents, before new virtual-joystick state becomes active (as specified via SDL_JoystickSetVirtual* function-calls), this was done to match behavior found in SDL's other joystick drivers, almost all of which will only update SDL-state during SDL_JoystickUpdate. 6. the initial patch is based off of SDL 2.0.12 7. the virtual joystick subsystem is disabled by default. It should be possible to enable it by building with SDL_JOYSTICK_VIRTUAL=1 Questions, comments, suggestions, or bug reports very welcome!
2020-03-14 10:08:45 +08:00
set_option(JOYSTICK_VIRTUAL "Enable the virtual-joystick driver" ON)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SDL_SHARED ${SDL_SHARED_ENABLED_BY_DEFAULT} CACHE BOOL "Build a shared version of the library")
Fixed bug 4593 - Respect CMake's BUILD_SHARED_LIBS default behavior tschwinger Respect the BUILD_SHARED_LIBS variable when defined, and build either shared or static libs, which is CMake's default behavior (See https://cmake.org/cmake/help/latest/variable/BUILD_SHARED_LIBS.html). If the variable is not defined, the current behavior remains unchanged and both variants are built where the platform supports it. This way, it remains possible to build both in one shot, which seems convenient for distro builds and useful to promote some consistency between them.
2019-06-09 05:36:03 +08:00
set(SDL_STATIC ${SDL_STATIC_ENABLED_BY_DEFAULT} CACHE BOOL "Build a static version of the library")
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
CMake: Let static libraries optionally be built with Position Independent Code.
2016-01-05 14:26:32 +08:00
dep_option(SDL_STATIC_PIC "Static version of the library should be built with Position Independent Code" OFF "SDL_STATIC" OFF)
Fixed bug 4092 - CMake support for building everything in the "test" directory Eric Wasylishen Patch to support building the tests with cmake. Disabled by default, use: "cmake .. -DSDL_TEST=YES" to enable the tests. Tested on macOS 10.13 with the ninja, makefile, and Xcode generators, and Windows 10 with the Visual Studio 2017 generator.
2018-02-25 00:59:58 +08:00
set_option(SDL_TEST "Build the test directory" OFF)
CMake: Let static libraries optionally be built with Position Independent Code.
2016-01-05 14:26:32 +08:00
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# General source files
file(GLOB SOURCE_FILES
${SDL2_SOURCE_DIR}/src/*.c
${SDL2_SOURCE_DIR}/src/atomic/*.c
${SDL2_SOURCE_DIR}/src/audio/*.c
${SDL2_SOURCE_DIR}/src/cpuinfo/*.c
${SDL2_SOURCE_DIR}/src/dynapi/*.c
${SDL2_SOURCE_DIR}/src/events/*.c
${SDL2_SOURCE_DIR}/src/file/*.c
cmake: Fix building with -DSDL_HAPTIC=Off
2020-07-08 23:28:34 +08:00
${SDL2_SOURCE_DIR}/src/haptic/*.c
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
${SDL2_SOURCE_DIR}/src/libm/*.c
locale: Implemented SDL_GetPreferredLocales(). This was something I proposed a long time ago, Sylvain Becker did additional work on it, then back to me. Fixes Bugzilla #2131.
2020-05-04 14:27:29 +08:00
${SDL2_SOURCE_DIR}/src/locale/*.c
Removed SDL_AndroidOpenURL, added SDL_OpenURL. Still needs to be wired into Xcode and Visual Studio projects.
2020-10-05 23:30:33 +08:00
${SDL2_SOURCE_DIR}/src/misc/*.c
cmake: Fix building with -DSDL_POWER=Off Fixes Bugzilla #5193.
2020-06-27 03:04:57 +08:00
${SDL2_SOURCE_DIR}/src/power/*.c
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
${SDL2_SOURCE_DIR}/src/render/*.c
${SDL2_SOURCE_DIR}/src/render/*/*.c
cmake: Fix building with -DSDL_SENSOR=Off
2020-06-27 04:45:15 +08:00
${SDL2_SOURCE_DIR}/src/sensor/*.c
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
${SDL2_SOURCE_DIR}/src/stdlib/*.c
${SDL2_SOURCE_DIR}/src/thread/*.c
${SDL2_SOURCE_DIR}/src/timer/*.c
Fixed bug 3959 - cmake build broken by commit 11702 (7fdbffd47c0e) due to typo Mate Nagy There is a typo in CMakeLists.txt that makes CMake exit with failure. Change that causes the problem: (Notice the double ending brackets) ${SDL2_SOURCE_DIR}/src/video/*.c) + ${SDL2_SOURCE_DIR}/src/video/yuv2rgb/*.c) Fix: Just remove the first ending bracket resulting in: ${SDL2_SOURCE_DIR}/src/video/*.c ${SDL2_SOURCE_DIR}/src/video/yuv2rgb/*.c)
2017-11-20 16:03:23 +08:00
${SDL2_SOURCE_DIR}/src/video/*.c
Updated SDL's YUV support, many thanks to Adrien Descamps New functions get and set the YUV colorspace conversion mode: SDL_SetYUVConversionMode() SDL_GetYUVConversionMode() SDL_GetYUVConversionModeForResolution() SDL_ConvertPixels() converts between all supported RGB and YUV formats, with SSE acceleration for converting from planar YUV formats (YV12, NV12, etc) to common RGB/RGBA formats. Added a new test program, testyuv, to verify correctness and speed of YUV conversion functionality.
2017-11-13 14:51:12 +08:00
${SDL2_SOURCE_DIR}/src/video/yuv2rgb/*.c)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(ASSERTIONS STREQUAL "auto")
# Do nada - use optimization settings to determine the assertion level
elseif(ASSERTIONS STREQUAL "disabled")
set(SDL_DEFAULT_ASSERT_LEVEL 0)
elseif(ASSERTIONS STREQUAL "release")
set(SDL_DEFAULT_ASSERT_LEVEL 1)
elseif(ASSERTIONS STREQUAL "enabled")
set(SDL_DEFAULT_ASSERT_LEVEL 2)
elseif(ASSERTIONS STREQUAL "paranoid")
set(SDL_DEFAULT_ASSERT_LEVEL 3)
else()
message_error("unknown assertion level")
endif()
set(HAVE_ASSERTIONS ${ASSERTIONS})
events: Let arbitrary signals to simulate iOS/Android backgrounding events. This lets you build a custom embedded device that roughly offers the "this process is going to the background NOW" semantics of SDL on a mobile device.
2019-03-16 03:51:05 +08:00
if(NOT BACKGROUNDING_SIGNAL STREQUAL "OFF")
add_definitions("-DSDL_BACKGROUNDING_SIGNAL=${BACKGROUNDING_SIGNAL}")
endif()
if(NOT FOREGROUNDING_SIGNAL STREQUAL "OFF")
add_definitions("-DSDL_FOREGROUNDING_SIGNAL=${FOREGROUNDING_SIGNAL}")
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# Compiler option evaluation
if(USE_GCC OR USE_CLANG)
Try to build with GCC's -Wdeclaration-after-statement. This will help catch things that'll cause issues on C89 compilers before we send them on to fail on Buildbot.
2016-01-05 11:58:38 +08:00
# Check for -Wall first, so later things can override pieces of it.
check_c_compiler_flag(-Wall HAVE_GCC_WALL)
if(HAVE_GCC_WALL)
list(APPEND EXTRA_CFLAGS "-Wall")
if(HAIKU)
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -Wno-multichar")
endif()
endif()
Temporary fix for bug 4254 - a _lot_ of strict aliasing warnings Ozkan Sezer A horde of strict aliasing violation warnings are emitted from joystick layer, and also from a few other places. This happens with gcc-4.4.7 on Linux CentOS 6.10. Some other sysjoystick would possibly have the same warnings. Attached my full log here. Example entry: src/joystick/SDL_joystick.c: In function 'SDL_GetJoystickGUIDInfo': src/joystick/SDL_joystick.c:1094: warning: dereferencing pointer '({anonymous})' does break strict-aliasing rules
2019-06-09 09:32:29 +08:00
check_c_compiler_flag(-fno-strict-aliasing HAVE_GCC_NO_STRICT_ALIASING)
if(HAVE_GCC_NO_STRICT_ALIASING)
list(APPEND EXTRA_CFLAGS "-fno-strict-aliasing")
endif()
Try to build with GCC's -Wdeclaration-after-statement. This will help catch things that'll cause issues on C89 compilers before we send them on to fail on Buildbot.
2016-01-05 11:58:38 +08:00
check_c_compiler_flag(-Wdeclaration-after-statement HAVE_GCC_WDECLARATION_AFTER_STATEMENT)
if(HAVE_GCC_WDECLARATION_AFTER_STATEMENT)
check_c_compiler_flag(-Werror=declaration-after-statement HAVE_GCC_WERROR_DECLARATION_AFTER_STATEMENT)
if(HAVE_GCC_WERROR_DECLARATION_AFTER_STATEMENT)
list(APPEND EXTRA_CFLAGS "-Werror=declaration-after-statement")
endif()
list(APPEND EXTRA_CFLAGS "-Wdeclaration-after-statement")
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(DEPENDENCY_TRACKING)
check_c_source_compiles("
#if !defined(__GNUC__) || __GNUC__ < 3
#error Dependency tracking requires GCC 3.0 or newer
#endif
int main(int argc, char **argv) { }" HAVE_DEPENDENCY_TRACKING)
endif()
if(GCC_ATOMICS)
check_c_source_compiles("int main(int argc, char **argv) {
int a;
void *x, *y, *z;
__sync_lock_test_and_set(&a, 4);
__sync_lock_test_and_set(&x, y);
__sync_fetch_and_add(&a, 1);
__sync_bool_compare_and_swap(&a, 5, 10);
__sync_bool_compare_and_swap(&x, y, z); }" HAVE_GCC_ATOMICS)
if(NOT HAVE_GCC_ATOMICS)
check_c_source_compiles("int main(int argc, char **argv) {
int a;
__sync_lock_test_and_set(&a, 1);
__sync_lock_release(&a); }" HAVE_GCC_SYNC_LOCK_TEST_AND_SET)
endif()
endif()
set(CMAKE_REQUIRED_FLAGS "-mpreferred-stack-boundary=2")
check_c_source_compiles("int x = 0; int main(int argc, char **argv) {}"
HAVE_GCC_PREFERRED_STACK_BOUNDARY)
set(CMAKE_REQUIRED_FLAGS ${ORIG_CMAKE_REQUIRED_FLAGS})
set(CMAKE_REQUIRED_FLAGS "-fvisibility=hidden -Werror")
check_c_source_compiles("
#if !defined(__GNUC__) || __GNUC__ < 4
#error SDL only uses visibility attributes in GCC 4 or newer
#endif
int main(int argc, char **argv) {}" HAVE_GCC_FVISIBILITY)
if(HAVE_GCC_FVISIBILITY)
list(APPEND EXTRA_CFLAGS "-fvisibility=hidden")
endif()
set(CMAKE_REQUIRED_FLAGS ${ORIG_CMAKE_REQUIRED_FLAGS})
check_c_compiler_flag(-Wshadow HAVE_GCC_WSHADOW)
if(HAVE_GCC_WSHADOW)
list(APPEND EXTRA_CFLAGS "-Wshadow")
endif()
cmake: Fix building for macOS with Xcode generator (thanks, dungaipara!). Fixes Bugzilla #3625.
2017-08-19 06:52:25 +08:00
if(APPLE)
list(APPEND EXTRA_LDFLAGS "-Wl,-undefined,error")
CMakeLists.txt: set dylib version numbers properly. (bug #2915.)
2018-10-25 16:11:02 +08:00
list(APPEND EXTRA_LDFLAGS "-Wl,-compatibility_version,${DYLIB_COMPATIBILITY_VERSION}")
list(APPEND EXTRA_LDFLAGS "-Wl,-current_version,${DYLIB_CURRENT_VERSION}")
cmake: Fix building for macOS with Xcode generator (thanks, dungaipara!). Fixes Bugzilla #3625.
2017-08-19 06:52:25 +08:00
else()
set(CMAKE_REQUIRED_FLAGS "-Wl,--no-undefined")
check_c_compiler_flag("" HAVE_NO_UNDEFINED)
set(CMAKE_REQUIRED_FLAGS ${ORIG_CMAKE_REQUIRED_FLAGS})
if(HAVE_NO_UNDEFINED)
list(APPEND EXTRA_LDFLAGS "-Wl,--no-undefined")
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
endif()
if(ASSEMBLY)
if(USE_GCC OR USE_CLANG)
set(SDL_ASSEMBLY_ROUTINES 1)
# TODO: Those all seem to be quite GCC specific - needs to be
# reworked for better compiler support
set(HAVE_ASSEMBLY TRUE)
if(MMX)
set(CMAKE_REQUIRED_FLAGS "-mmmx")
check_c_source_compiles("
#ifdef __MINGW32__
#include <_mingw.h>
#ifdef __MINGW64_VERSION_MAJOR
#include <intrin.h>
#else
#include <mmintrin.h>
#endif
#else
#include <mmintrin.h>
#endif
#ifndef __MMX__
#error Assembler CPP flag not enabled
#endif
int main(int argc, char **argv) { }" HAVE_MMX)
if(HAVE_MMX)
list(APPEND EXTRA_CFLAGS "-mmmx")
endif()
set(CMAKE_REQUIRED_FLAGS ${ORIG_CMAKE_REQUIRED_FLAGS})
endif()
if(3DNOW)
set(CMAKE_REQUIRED_FLAGS "-m3dnow")
check_c_source_compiles("
#include <mm3dnow.h>
#ifndef __3dNOW__
#error Assembler CPP flag not enabled
#endif
int main(int argc, char **argv) {
void *p = 0;
_m_prefetch(p);
}" HAVE_3DNOW)
if(HAVE_3DNOW)
list(APPEND EXTRA_CFLAGS "-m3dnow")
endif()
set(CMAKE_REQUIRED_FLAGS ${ORIG_CMAKE_REQUIRED_FLAGS})
endif()
if(SSE)
set(CMAKE_REQUIRED_FLAGS "-msse")
check_c_source_compiles("
#ifdef __MINGW32__
#include <_mingw.h>
#ifdef __MINGW64_VERSION_MAJOR
#include <intrin.h>
#else
#include <xmmintrin.h>
#endif
#else
#include <xmmintrin.h>
#endif
#ifndef __SSE__
#error Assembler CPP flag not enabled
#endif
int main(int argc, char **argv) { }" HAVE_SSE)
if(HAVE_SSE)
list(APPEND EXTRA_CFLAGS "-msse")
endif()
set(CMAKE_REQUIRED_FLAGS ${ORIG_CMAKE_REQUIRED_FLAGS})
endif()
if(SSE2)
set(CMAKE_REQUIRED_FLAGS "-msse2")
check_c_source_compiles("
#ifdef __MINGW32__
#include <_mingw.h>
#ifdef __MINGW64_VERSION_MAJOR
#include <intrin.h>
#else
#include <emmintrin.h>
#endif
#else
#include <emmintrin.h>
#endif
#ifndef __SSE2__
#error Assembler CPP flag not enabled
#endif
int main(int argc, char **argv) { }" HAVE_SSE2)
if(HAVE_SSE2)
list(APPEND EXTRA_CFLAGS "-msse2")
endif()
set(CMAKE_REQUIRED_FLAGS ${ORIG_CMAKE_REQUIRED_FLAGS})
endif()
audio: Wired up new SSE code to build system.
2017-01-23 14:05:44 +08:00
if(SSE3)
set(CMAKE_REQUIRED_FLAGS "-msse3")
check_c_source_compiles("
#ifdef __MINGW32__
#include <_mingw.h>
#ifdef __MINGW64_VERSION_MAJOR
#include <intrin.h>
#else
#include <pmmintrin.h>
#endif
#else
#include <pmmintrin.h>
#endif
#ifndef __SSE3__
#error Assembler CPP flag not enabled
#endif
int main(int argc, char **argv) { }" HAVE_SSE3)
if(HAVE_SSE3)
list(APPEND EXTRA_CFLAGS "-msse3")
endif()
set(CMAKE_REQUIRED_FLAGS ${ORIG_CMAKE_REQUIRED_FLAGS})
endif()
CMake: fixed logic error in setting -mfpmath=387. Fixes Bugzilla #3565.
2017-01-25 01:20:41 +08:00
if(NOT SSEMATH)
audio: Wired up new SSE code to build system.
2017-01-23 14:05:44 +08:00
if(SSE OR SSE2 OR SSE3)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(USE_GCC)
Android cmake build fixed to work with the official android gradle plugin
2017-06-05 02:25:57 +08:00
check_c_compiler_flag(-mfpmath=387 HAVE_FP_387)
if(HAVE_FP_387)
list(APPEND EXTRA_CFLAGS "-mfpmath=387")
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
set(HAVE_SSEMATH TRUE)
endif()
endif()
Check for immintrin.h before using it in SDL_cpuinfo.h
2017-12-12 04:00:12 +08:00
check_include_file("immintrin.h" HAVE_IMMINTRIN_H)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(ALTIVEC)
set(CMAKE_REQUIRED_FLAGS "-maltivec")
check_c_source_compiles("
#include <altivec.h>
vector unsigned int vzero() {
return vec_splat_u32(0);
}
int main(int argc, char **argv) { }" HAVE_ALTIVEC_H_HDR)
check_c_source_compiles("
vector unsigned int vzero() {
return vec_splat_u32(0);
}
int main(int argc, char **argv) { }" HAVE_ALTIVEC)
set(CMAKE_REQUIRED_FLAGS ${ORIG_CMAKE_REQUIRED_FLAGS})
if(HAVE_ALTIVEC OR HAVE_ALTIVEC_H_HDR)
set(HAVE_ALTIVEC TRUE) # if only HAVE_ALTIVEC_H_HDR is set
list(APPEND EXTRA_CFLAGS "-maltivec")
set(SDL_ALTIVEC_BLITTERS 1)
if(HAVE_ALTIVEC_H_HDR)
set(HAVE_ALTIVEC_H 1)
endif()
endif()
endif()
cmake: added support for enabling the ARM SIMD/NEON code.
2019-10-25 11:17:19 +08:00
if(ARMSIMD)
set(ORIG_CMAKE_REQUIRED_FLAGS "${CMAKE_REQUIRED_FLAGS}")
set(CMAKE_REQUIRED_FLAGS "${CMAKE_REQUIRED_FLAGS} -x assembler-with-cpp")
check_c_source_compiles("
.text
.arch armv6
.object_arch armv4
.arm
.altmacro
#ifndef __ARM_EABI__
#error EABI is required (to be sure that calling conventions are compatible)
#endif
pld [r0]
uqadd8 r0, r0, r0
" ARMSIMD_FOUND)
set(CMAKE_REQUIRED_FLAGS "${ORIG_CMAKE_REQUIRED_FLAGS}")
if(ARMSIMD_FOUND)
set(HAVE_ARMSIMD TRUE)
set(SDL_ARM_SIMD_BLITTERS 1)
file(GLOB ARMSIMD_SOURCES ${SDL2_SOURCE_DIR}/src/video/arm/pixman-arm-simd*.S)
set(SOURCE_FILES ${SOURCE_FILES} ${ARMSIMD_SOURCES})
set(WARN_ABOUT_ARM_SIMD_ASM_MIT TRUE)
endif()
endif()
if(ARMNEON)
set(ORIG_CMAKE_REQUIRED_FLAGS "${CMAKE_REQUIRED_FLAGS}")
set(CMAKE_REQUIRED_FLAGS "${CMAKE_REQUIRED_FLAGS} -x assembler-with-cpp")
check_c_source_compiles("
.text
.fpu neon
.arch armv7a
.object_arch armv4
.eabi_attribute 10, 0
.arm
.altmacro
#ifndef __ARM_EABI__
#error EABI is required (to be sure that calling conventions are compatible)
#endif
pld [r0]
vmovn.u16 d0, q0
" ARMNEON_FOUND)
set(CMAKE_REQUIRED_FLAGS "${ORIG_CMAKE_REQUIRED_FLAGS}")
if(ARMNEON_FOUND)
set(HAVE_ARMNEON TRUE)
set(SDL_ARM_NEON_BLITTERS 1)
file(GLOB ARMNEON_SOURCES ${SDL2_SOURCE_DIR}/src/video/arm/pixman-arm-neon*.S)
set(SOURCE_FILES ${SOURCE_FILES} ${ARMNEON_SOURCES})
set(WARN_ABOUT_ARM_NEON_ASM_MIT TRUE)
endif()
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
elseif(MSVC_VERSION GREATER 1500)
# TODO: SDL_cpuinfo.h needs to support the user's configuration wish
# for MSVC - right now it is always activated
if(NOT ARCH_64)
set(HAVE_MMX TRUE)
set(HAVE_3DNOW TRUE)
endif()
set(HAVE_SSE TRUE)
set(HAVE_SSE2 TRUE)
audio: Wired up new SSE code to build system.
2017-01-23 14:05:44 +08:00
set(HAVE_SSE3 TRUE)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SDL_ASSEMBLY_ROUTINES 1)
endif()
# TODO:
#else()
# if(USE_GCC OR USE_CLANG)
audio: Wired up new SSE code to build system.
2017-01-23 14:05:44 +08:00
# list(APPEND EXTRA_CFLAGS "-mno-sse" "-mno-sse2" "-mno-sse3" "-mno-mmx")
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# endif()
endif()
# TODO: Can't deactivate on FreeBSD? w/o LIBC, SDL_stdinc.h can't define
# anything.
if(LIBC)
if(WINDOWS AND NOT MINGW)
set(HAVE_LIBC TRUE)
Fixed bug 3760 - RWops doesn't check for integer overflow when stdio_fseek only supports 32 bits Simon Hug When RWops seeks with fseek or fseeko it uses the types long or off_t which can be 32 bits on some platforms. stdio_seek does not check if the 64-bit integer for the offset fits into a 32-bit integer. Offsets equal or larger than 2 GiB will have implementation-defined behavior and failure states would be very confusing to debug. The attached patch adds range checking by using the macros from limits.h for long type and some bit shifting for off_t because POSIX couldn't be bothered to specify min and max macros. It also defines HAVE_FSEEKI64 in SDL_config_windows.h so that the Windows function gets picked up automatically with the default config. And there's an additional error message for when ftell fails.
2017-09-09 23:36:37 +08:00
foreach(_HEADER stdio.h string.h wchar.h ctype.h math.h limits.h)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
string(TOUPPER "HAVE_${_HEADER}" _UPPER)
string(REPLACE "." "_" _HAVE_H ${_UPPER})
set(${_HAVE_H} 1)
endforeach()
set(HAVE_SIGNAL_H 1)
foreach(_FN
malloc calloc realloc free qsort abs memset memcpy memmove memcmp
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/disable-wcslcpy-and-wcslcat-for-windows.patch
2020-12-09 01:07:21 +08:00
wcslen wcsdup wcsstr wcscmp wcsncmp _wcsicmp _wcsnicmp
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
strlen _strrev _strupr _strlwr strchr strrchr strstr itoa _ltoa
_ultoa strtol strtoul strtoll strtod atoi atof strcmp strncmp
added SDL_strtokr() as a replacement for POSIX strtok_r (bug #4046.)
2019-11-21 01:40:50 +08:00
_stricmp _strnicmp strtok_s sscanf
Added float versions of SDL's math functions
2017-11-05 06:34:14 +08:00
acos acosf asin asinf atan atanf atan2 atan2f ceil ceilf
SDL_exp
2018-08-04 23:52:46 +08:00
copysign copysignf cos cosf exp expf fabs fabsf floor floorf fmod fmodf
stdlib: Add SDL_trunc and SDL_truncf
2020-04-11 00:17:14 +08:00
log logf log10 log10f pow powf scalbn scalbnf sin sinf sqrt sqrtf tan tanf
trunc truncf)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
string(TOUPPER ${_FN} _UPPER)
set(HAVE_${_UPPER} 1)
endforeach()
if(NOT CYGWIN AND NOT MINGW)
set(HAVE_ALLOCA 1)
endif()
set(HAVE_M_PI 1)
add_definitions(-D_USE_MATH_DEFINES) # needed for M_PI
set(STDC_HEADERS 1)
else()
set(HAVE_LIBC TRUE)
check_include_file(sys/types.h HAVE_SYS_TYPES_H)
foreach(_HEADER
Fixed bug 3760 - RWops doesn't check for integer overflow when stdio_fseek only supports 32 bits Simon Hug When RWops seeks with fseek or fseeko it uses the types long or off_t which can be 32 bits on some platforms. stdio_seek does not check if the 64-bit integer for the offset fits into a 32-bit integer. Offsets equal or larger than 2 GiB will have implementation-defined behavior and failure states would be very confusing to debug. The attached patch adds range checking by using the macros from limits.h for long type and some bit shifting for off_t because POSIX couldn't be bothered to specify min and max macros. It also defines HAVE_FSEEKI64 in SDL_config_windows.h so that the Windows function gets picked up automatically with the default config. And there's an additional error message for when ftell fails.
2017-09-09 23:36:37 +08:00
stdio.h stdlib.h stddef.h stdarg.h malloc.h memory.h string.h limits.h
Added functions to query and set the SDL memory allocation functions: SDL_GetMemoryFunctions() SDL_SetMemoryFunctions() SDL_GetNumAllocations()
2017-10-13 04:44:28 +08:00
strings.h wchar.h inttypes.h stdint.h ctype.h math.h iconv.h signal.h libunwind.h)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
string(TOUPPER "HAVE_${_HEADER}" _UPPER)
string(REPLACE "." "_" _HAVE_H ${_UPPER})
check_include_file("${_HEADER}" ${_HAVE_H})
endforeach()
cmake: reduce STDC_HEADER_NAMES list to only relevant headers, i.e. stddef.h, stdarg.h, stdlib.h, string.h, stdio.h, wchar.h, float.h. Fixes issue described at: https://bugzilla.libsdl.org/show_bug.cgi?id=4885#c2
2020-10-12 07:40:00 +08:00
set(STDC_HEADER_NAMES "stddef.h;stdarg.h;stdlib.h;string.h;stdio.h;wchar.h;float.h")
cmake: use check_symbol_exists, not check_function_exists (thanks, Manuel!) This fixes the problem where we think iOS has fseeko64, etc, but doesn't. Fixes Bugzilla #4885.
2020-04-14 03:22:09 +08:00
check_include_files("${STDC_HEADER_NAMES}" STDC_HEADERS)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
check_type_size("size_t" SIZEOF_SIZE_T)
check_symbol_exists(M_PI math.h HAVE_M_PI)
# TODO: refine the mprotect check
check_c_source_compiles("#include <sys/types.h>
#include <sys/mman.h>
int main() { }" HAVE_MPROTECT)
foreach(_FN
strtod malloc calloc realloc free getenv setenv putenv unsetenv
qsort abs bcopy memset memcpy memmove memcmp strlen strlcpy strlcat
added SDL_strtokr() as a replacement for POSIX strtok_r (bug #4046.)
2019-11-21 01:40:50 +08:00
_strrev _strupr _strlwr strchr strrchr strstr strtok_r itoa _ltoa
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
_uitoa _ultoa strtol strtoul _i64toa _ui64toa strtoll strtoull
atoi atof strcmp strncmp _stricmp strcasecmp _strnicmp strncasecmp
CMakeLists.txt: add several missing function checks for unix case.
2019-11-21 15:33:56 +08:00
wcscmp wcsdup wcslcat wcslcpy wcslen wcsncmp wcsstr
cmake: add missing checks for wcscasecmp, _wcsicmp, wcsncasecmp, _wcsnicmp
2020-11-25 19:51:56 +08:00
wcscasecmp _wcsicmp wcsncasecmp _wcsnicmp
cmake: use check_symbol_exists, not check_function_exists (thanks, Manuel!) This fixes the problem where we think iOS has fseeko64, etc, but doesn't. Fixes Bugzilla #4885.
2020-04-14 03:22:09 +08:00
sscanf vsscanf vsnprintf fopen64 fseeko fseeko64 _Exit
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
)
string(TOUPPER ${_FN} _UPPER)
set(_HAVEVAR "HAVE_${_UPPER}")
cmake: use check_symbol_exists, not check_function_exists (thanks, Manuel!) This fixes the problem where we think iOS has fseeko64, etc, but doesn't. Fixes Bugzilla #4885.
2020-04-14 03:22:09 +08:00
check_symbol_exists("${_FN}" "${STDC_HEADER_NAMES}" ${_HAVEVAR})
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endforeach()
cmake: use check_symbol_exists, not check_function_exists (thanks, Manuel!) This fixes the problem where we think iOS has fseeko64, etc, but doesn't. Fixes Bugzilla #4885.
2020-04-14 03:22:09 +08:00
check_symbol_exists(sigaction "signal.h" HAVE_SIGACTION)
check_symbol_exists(setjmp "setjmp.h" HAVE_SETJMP)
check_symbol_exists(nanosleep "time.h" HAVE_NANOSLEEP)
check_symbol_exists(sysconf "unistd.h" HAVE_SYSCONF)
check_symbol_exists(sysctlbyname "sys/types.h;sys/sysctl.h" HAVE_SYSCTLBYNAME)
check_symbol_exists(getauxval "sys/auxv.h" HAVE_GETAUXVAL)
Fixed bug 5402 - ARM support little update proposal David Carlier No fix but mostly an update for ARM architecture.
2020-12-18 13:41:23 +08:00
check_symbol_exists(elf_aux_info "sys/auxv.h" HAVE_ELF_AUX_INFO)
cmake: use check_symbol_exists, not check_function_exists (thanks, Manuel!) This fixes the problem where we think iOS has fseeko64, etc, but doesn't. Fixes Bugzilla #4885.
2020-04-14 03:22:09 +08:00
check_symbol_exists(poll "poll.h" HAVE_POLL)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
check_library_exists(m pow "" HAVE_LIBM)
if(HAVE_LIBM)
set(CMAKE_REQUIRED_LIBRARIES m)
foreach(_FN
CMakeLists.txt: add several missing function checks for unix case.
2019-11-21 15:33:56 +08:00
atan atan2 atanf atan2f ceil ceilf copysign copysignf cos cosf
exp expf fabs fabsf floor floorf fmod fmodf log logf log10 log10f
pow powf scalbn scalbnf sin sinf sqrt sqrtf tan tanf acos acosf
stdlib: Add SDL_trunc and SDL_truncf
2020-04-11 00:17:14 +08:00
asin asinf trunc truncf)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
string(TOUPPER ${_FN} _UPPER)
set(_HAVEVAR "HAVE_${_UPPER}")
cmake: use check_symbol_exists, not check_function_exists (thanks, Manuel!) This fixes the problem where we think iOS has fseeko64, etc, but doesn't. Fixes Bugzilla #4885.
2020-04-14 03:22:09 +08:00
check_symbol_exists("${_FN}" "math.h" ${_HAVEVAR})
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endforeach()
set(CMAKE_REQUIRED_LIBRARIES)
list(APPEND EXTRA_LIBS m)
endif()
check_library_exists(iconv iconv_open "" HAVE_LIBICONV)
cmake: look for iconv in libc, too (bug #5316.)
2020-10-13 20:20:03 +08:00
check_library_exists(c iconv_open "" HAVE_BUILTIN_ICONV)
if(HAVE_BUILTIN_ICONV)
set(HAVE_ICONV 1)
elseif(HAVE_LIBICONV)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
list(APPEND EXTRA_LIBS iconv)
set(HAVE_ICONV 1)
endif()
if(NOT APPLE)
check_include_file(alloca.h HAVE_ALLOCA_H)
cmake: check for alloca() in stdlib.h and malloc.h, too (bug #5316)
2020-10-13 19:32:15 +08:00
check_symbol_exists(alloca "alloca.h" HAVE_ALLOCA1)
check_symbol_exists(alloca "stdlib.h" HAVE_ALLOCA2)
check_symbol_exists(alloca "malloc.h" HAVE_ALLOCA3)
if(HAVE_ALLOCA1 OR HAVE_ALLOCA2 OR HAVE_ALLOCA3)
set(HAVE_ALLOCA 1)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
else()
set(HAVE_ALLOCA_H 1)
set(HAVE_ALLOCA 1)
endif()
check_struct_has_member("struct sigaction" "sa_sigaction" "signal.h" HAVE_SA_SIGACTION)
endif()
else()
if(WINDOWS)
set(HAVE_STDARG_H 1)
set(HAVE_STDDEF_H 1)
endif()
endif()
# Enable/disable various subsystems of the SDL library
foreach(_SUB ${SDL_SUBSYSTEMS})
string(TOUPPER ${_SUB} _OPT)
if(NOT SDL_${_OPT})
set(SDL_${_OPT}_DISABLED 1)
endif()
endforeach()
if(SDL_JOYSTICK)
file(GLOB JOYSTICK_SOURCES ${SDL2_SOURCE_DIR}/src/joystick/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${JOYSTICK_SOURCES})
endif()
if(SDL_HAPTIC)
if(NOT SDL_JOYSTICK)
# Haptic requires some private functions from the joystick subsystem.
message_error("SDL_HAPTIC requires SDL_JOYSTICK, which is not enabled")
endif()
endif()
locale: Implemented SDL_GetPreferredLocales(). This was something I proposed a long time ago, Sylvain Becker did additional work on it, then back to me. Fixes Bugzilla #2131.
2020-05-04 14:27:29 +08:00
configure.in: Rename to configure.ac to fix an 'aclocal' warning Also rename references in related files.
2019-03-25 20:01:32 +08:00
# TODO: in configure.ac, the test for LOADSO and SDL_DLOPEN is a bit weird:
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# if LOADSO is not wanted, SDL_LOADSO_DISABLED is set
# If however on Unix or APPLE dlopen() is detected via CheckDLOPEN(),
# SDL_LOADSO_DISABLED will not be set, regardless of the LOADSO settings
# General SDL subsystem options, valid for all platforms
if(SDL_AUDIO)
# CheckDummyAudio/CheckDiskAudio - valid for all platforms
if(DUMMYAUDIO)
set(SDL_AUDIO_DRIVER_DUMMY 1)
file(GLOB DUMMYAUDIO_SOURCES ${SDL2_SOURCE_DIR}/src/audio/dummy/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${DUMMYAUDIO_SOURCES})
set(HAVE_DUMMYAUDIO TRUE)
endif()
if(DISKAUDIO)
set(SDL_AUDIO_DRIVER_DISK 1)
file(GLOB DISKAUDIO_SOURCES ${SDL2_SOURCE_DIR}/src/audio/disk/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${DISKAUDIO_SOURCES})
set(HAVE_DISKAUDIO TRUE)
endif()
endif()
if(SDL_DLOPEN)
# Relevant for Unix/Darwin only
if(UNIX OR APPLE)
CheckDLOPEN()
endif()
endif()
Fixed bug 5028 - Virtual Joysticks (new joystick backend) David Ludwig I have created a new driver for SDL's Joystick and Game-Controller subsystem: a Virtual driver. This driver allows one to create a software-based joystick, which to SDL applications will look and react like a real joystick, but whose state can be set programmatically. A primary use case for this is to help enable developers to add touch-screen joysticks to their apps. The driver comes with a set of new, public APIs, with functions to attach and detach joysticks, set virtual-joystick state, and to determine if a joystick is a virtual-one. Use of virtual joysticks goes as such: 1. Attach one or more virtual joysticks by calling SDL_JoystickAttachVirtual. If successful, this returns the virtual-device's joystick-index. 2. Open the virtual joysticks (using indicies returned by SDL_JoystickAttachVirtual). 3. Call any of the SDL_JoystickSetVirtual* functions when joystick-state changes. Please note that virtual-joystick state will only get applied on the next call to SDL_JoystickUpdate, or when pumping or polling for SDL events (via SDL_PumpEvents or SDL_PollEvent). Here is a listing of the new, public APIs, at present and subject to change: ------------------------------------------------------------ /** * Attaches a new virtual joystick. * Returns the joystick's device index, or -1 if an error occurred. */ extern DECLSPEC int SDLCALL SDL_JoystickAttachVirtual(SDL_JoystickType type, int naxes, int nballs, int nbuttons, int nhats); /** * Detaches a virtual joystick * Returns 0 on success, or -1 if an error occurred. */ extern DECLSPEC int SDLCALL SDL_JoystickDetachVirtual(int device_index); /** * Indicates whether or not a virtual-joystick is at a given device index. */ extern DECLSPEC SDL_bool SDLCALL SDL_JoystickIsVirtual(int device_index); /** * Set values on an opened, virtual-joystick's controls. * Returns 0 on success, -1 on error. */ extern DECLSPEC int SDLCALL SDL_JoystickSetVirtualAxis(SDL_Joystick * joystick, int axis, Sint16 value); extern DECLSPEC int SDLCALL SDL_JoystickSetVirtualBall(SDL_Joystick * joystick, int ball, Sint16 xrel, Sint16 yrel); extern DECLSPEC int SDLCALL SDL_JoystickSetVirtualButton(SDL_Joystick * joystick, int button, Uint8 value); extern DECLSPEC int SDLCALL SDL_JoystickSetVirtualHat(SDL_Joystick * joystick, int hat, Uint8 value); ------------------------------------------------------------ Miscellaneous notes on the initial patch, which are also subject to change: 1. no test code is present in SDL, yet. This should, perhaps, change. Initial development was done with an ImGui-based app, which potentially is too thick for use in SDL-official. If tests are to be added, what kind of tests? Automated? Graphical? 2. virtual game controllers can be created by calling SDL_JoystickAttachVirtual with a joystick-type of SDL_JOYSTICK_TYPE_GAME_CONTROLLER, with naxes (num axes) set to SDL_CONTROLLER_AXIS_MAX, and with nbuttons (num buttons) set to SDL_CONTROLLER_BUTTON_MAX. When updating their state, values of type SDL_GameControllerAxis or SDL_GameControllerButton can be casted to an int and used for the control-index (in calls to SDL_JoystickSetVirtual* functions). 3. virtual joysticks' guids are mostly all-zeros with the exception of the last two bytes, the first of which is a 'v', to indicate that the guid is a virtual one, and the second of which is a SDL_JoystickType that has been converted into a Uint8. 4. virtual joysticks are ONLY turned into virtual game-controllers if and when their joystick-type is set to SDL_JOYSTICK_TYPE_GAMECONTROLLER. This is controlled by having SDL's default list of game-controllers have a single entry for a virtual game controller (of guid, "00000000000000000000000000007601", which is subject to the guid-encoding described above). 5. regarding having to call SDL_JoystickUpdate, either directly or indirectly via SDL_PumpEvents or SDL_PollEvents, before new virtual-joystick state becomes active (as specified via SDL_JoystickSetVirtual* function-calls), this was done to match behavior found in SDL's other joystick drivers, almost all of which will only update SDL-state during SDL_JoystickUpdate. 6. the initial patch is based off of SDL 2.0.12 7. the virtual joystick subsystem is disabled by default. It should be possible to enable it by building with SDL_JOYSTICK_VIRTUAL=1 Questions, comments, suggestions, or bug reports very welcome!
2020-03-14 10:08:45 +08:00
if(SDL_JOYSTICK)
if(JOYSTICK_VIRTUAL)
set(SDL_JOYSTICK_VIRTUAL 1)
file(GLOB JOYSTICK_VIRTUAL_SOURCES ${SDL2_SOURCE_DIR}/src/joystick/virtual/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${JOYSTICK_VIRTUAL_SOURCES})
endif()
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_VIDEO)
if(VIDEO_DUMMY)
set(SDL_VIDEO_DRIVER_DUMMY 1)
file(GLOB VIDEO_DUMMY_SOURCES ${SDL2_SOURCE_DIR}/src/video/dummy/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${VIDEO_DUMMY_SOURCES})
set(HAVE_VIDEO_DUMMY TRUE)
set(HAVE_SDL_VIDEO TRUE)
endif()
offscreen: Add new video driver backend Offscreen The Offscreen video driver is intended to be used for headless rendering as well as allows for multiple GPUs to be used for headless rendering Currently only supports EGL (OpenGL / ES) or Framebuffers Adds a hint to specifiy which EGL device to use: SDL_HINT_EGL_DEVICE Adds testoffscreen.c which can be used to test the backend out Disabled by default for now
2019-09-25 04:36:48 +08:00
if(VIDEO_OFFSCREEN)
set(SDL_VIDEO_DRIVER_OFFSCREEN 1)
file(GLOB VIDEO_OFFSCREEN_SOURCES ${SDL2_SOURCE_DIR}/src/video/offscreen/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${VIDEO_OFFSCREEN_SOURCES})
set(HAVE_VIDEO_OFFSCREEN TRUE)
set(HAVE_SDL_VIDEO TRUE)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
# Platform-specific options and settings
CMake: Changes to get CMake project to work with Android (thanks, Martin!). Fixes Bugzilla #3194. (but note that Bugzilla #3200 still needs to be resolved to get this really going on Android, at a minimum.)
2015-12-29 13:57:24 +08:00
if(ANDROID)
file(GLOB ANDROID_CORE_SOURCES ${SDL2_SOURCE_DIR}/src/core/android/*.c)
Fixed bug 5100 - compilation CMake Android armeabi-v7a (Thanks Steve Robinson!) fatal error: 'cpu-features.h' file not found on CMake Android armeabi-v7a
2020-05-16 03:12:23 +08:00
set(SOURCE_FILES ${SOURCE_FILES} ${ANDROID_CORE_SOURCES} ${ANDROID_NDK}/sources/android/cpufeatures/cpu-features.c)
Android cmake build fixed to work with the official android gradle plugin
2017-06-05 02:25:57 +08:00
Removed SDL_AndroidOpenURL, added SDL_OpenURL. Still needs to be wired into Xcode and Visual Studio projects.
2020-10-05 23:30:33 +08:00
file(GLOB ANDROID_MISC_SOURCES ${SDL2_SOURCE_DIR}/src/misc/android/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${ANDROID_MISC_SOURCES})
set(HAVE_SDL_MISC TRUE)
Android cmake build fixed to work with the official android gradle plugin
2017-06-05 02:25:57 +08:00
# SDL_spinlock.c Needs to be compiled in ARM mode.
# There seems to be no better way currently to set the ARM mode.
# see: https://issuetracker.google.com/issues/62264618
# Another option would be to set ARM mode to all compiled files
check_c_compiler_flag(-marm HAVE_ARM_MODE)
if(HAVE_ARM_MODE)
set_source_files_properties(${SDL2_SOURCE_DIR}/src/atomic/SDL_spinlock.c PROPERTIES COMPILE_FLAGS -marm)
endif()
CMake: Changes to get CMake project to work with Android (thanks, Martin!). Fixes Bugzilla #3194. (but note that Bugzilla #3200 still needs to be resolved to get this really going on Android, at a minimum.)
2015-12-29 13:57:24 +08:00
file(GLOB ANDROID_MAIN_SOURCES ${SDL2_SOURCE_DIR}/src/main/android/*.c)
Android cmake build fixed to work with the official android gradle plugin
2017-06-05 02:25:57 +08:00
set(SDLMAIN_SOURCES ${SDLMAIN_SOURCES} ${ANDROID_MAIN_SOURCES})
CMake: Changes to get CMake project to work with Android (thanks, Martin!). Fixes Bugzilla #3194. (but note that Bugzilla #3200 still needs to be resolved to get this really going on Android, at a minimum.)
2015-12-29 13:57:24 +08:00
if(SDL_AUDIO)
set(SDL_AUDIO_DRIVER_ANDROID 1)
file(GLOB ANDROID_AUDIO_SOURCES ${SDL2_SOURCE_DIR}/src/audio/android/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${ANDROID_AUDIO_SOURCES})
set(HAVE_SDL_AUDIO TRUE)
endif()
if(SDL_FILESYSTEM)
set(SDL_FILESYSTEM_ANDROID 1)
file(GLOB ANDROID_FILESYSTEM_SOURCES ${SDL2_SOURCE_DIR}/src/filesystem/android/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${ANDROID_FILESYSTEM_SOURCES})
set(HAVE_SDL_FILESYSTEM TRUE)
endif()
Fixed bug 3774 - Cmake build fails for Android sfalexrog Android haptic code was not added to CMakeLists.txt, leading to build failures when targeting Android platform. Attached patch adds Android haptic driver to source sets and adds configuration parameter to SDL_config.h.cmake.
2017-08-27 12:17:12 +08:00
if(SDL_HAPTIC)
set(SDL_HAPTIC_ANDROID 1)
file(GLOB ANDROID_HAPTIC_SOURCES ${SDL2_SOURCE_DIR}/src/haptic/android/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${ANDROID_HAPTIC_SOURCES})
set(HAVE_SDL_HAPTIC TRUE)
endif()
CMake: Changes to get CMake project to work with Android (thanks, Martin!). Fixes Bugzilla #3194. (but note that Bugzilla #3200 still needs to be resolved to get this really going on Android, at a minimum.)
2015-12-29 13:57:24 +08:00
if(SDL_JOYSTICK)
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
CheckHIDAPI()
CMake: Changes to get CMake project to work with Android (thanks, Martin!). Fixes Bugzilla #3194. (but note that Bugzilla #3200 still needs to be resolved to get this really going on Android, at a minimum.)
2015-12-29 13:57:24 +08:00
set(SDL_JOYSTICK_ANDROID 1)
Added stub Steam Controller sources to Android and iOS command line builds
2017-09-23 07:33:34 +08:00
file(GLOB ANDROID_JOYSTICK_SOURCES ${SDL2_SOURCE_DIR}/src/joystick/android/*.c ${SDL2_SOURCE_DIR}/src/joystick/steam/*.c)
CMake: Changes to get CMake project to work with Android (thanks, Martin!). Fixes Bugzilla #3194. (but note that Bugzilla #3200 still needs to be resolved to get this really going on Android, at a minimum.)
2015-12-29 13:57:24 +08:00
set(SOURCE_FILES ${SOURCE_FILES} ${ANDROID_JOYSTICK_SOURCES})
set(HAVE_SDL_JOYSTICK TRUE)
endif()
Android build fixes -Enabling checking for GCC_ATOMICS also on clang by default. This way all Android ABIs build successfully -Android cmake: Threading was not enabled correctly -Android cmake: Timers and dynamic lib loading were not included in the sources
2017-06-09 03:07:55 +08:00
if(SDL_LOADSO)
set(SDL_LOADSO_DLOPEN 1)
file(GLOB LOADSO_SOURCES ${SDL2_SOURCE_DIR}/src/loadso/dlopen/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${LOADSO_SOURCES})
set(HAVE_SDL_LOADSO TRUE)
endif()
CMake: Changes to get CMake project to work with Android (thanks, Martin!). Fixes Bugzilla #3194. (but note that Bugzilla #3200 still needs to be resolved to get this really going on Android, at a minimum.)
2015-12-29 13:57:24 +08:00
if(SDL_POWER)
set(SDL_POWER_ANDROID 1)
file(GLOB ANDROID_POWER_SOURCES ${SDL2_SOURCE_DIR}/src/power/android/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${ANDROID_POWER_SOURCES})
set(HAVE_SDL_POWER TRUE)
endif()
locale: Implemented SDL_GetPreferredLocales(). This was something I proposed a long time ago, Sylvain Becker did additional work on it, then back to me. Fixes Bugzilla #2131.
2020-05-04 14:27:29 +08:00
if(SDL_LOCALE)
file(GLOB ANDROID_LOCALE_SOURCES ${SDL2_SOURCE_DIR}/src/locale/android/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${ANDROID_LOCALE_SOURCES})
set(HAVE_SDL_LOCALE TRUE)
endif()
Android build fixes -Enabling checking for GCC_ATOMICS also on clang by default. This way all Android ABIs build successfully -Android cmake: Threading was not enabled correctly -Android cmake: Timers and dynamic lib loading were not included in the sources
2017-06-09 03:07:55 +08:00
if(SDL_TIMERS)
set(SDL_TIMER_UNIX 1)
file(GLOB TIMER_SOURCES ${SDL2_SOURCE_DIR}/src/timer/unix/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${TIMER_SOURCES})
set(HAVE_SDL_TIMERS TRUE)
endif()
cmake: add sensor subsystem to the build.
2018-08-24 02:32:30 +08:00
if(SDL_SENSOR)
set(SDL_SENSOR_ANDROID 1)
set(HAVE_SDL_SENSORS TRUE)
file(GLOB ANDROID_SENSOR_SOURCES ${SDL2_SOURCE_DIR}/src/sensor/android/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${ANDROID_SENSOR_SOURCES})
endif()
CMake: Changes to get CMake project to work with Android (thanks, Martin!). Fixes Bugzilla #3194. (but note that Bugzilla #3200 still needs to be resolved to get this really going on Android, at a minimum.)
2015-12-29 13:57:24 +08:00
if(SDL_VIDEO)
set(SDL_VIDEO_DRIVER_ANDROID 1)
file(GLOB ANDROID_VIDEO_SOURCES ${SDL2_SOURCE_DIR}/src/video/android/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${ANDROID_VIDEO_SOURCES})
set(HAVE_SDL_VIDEO TRUE)
Android cmake build fixed to work with the official android gradle plugin
2017-06-05 02:25:57 +08:00
# Core stuff
Fixed bug 3918 - CMake support for android project
2019-08-11 21:23:37 +08:00
# find_library(ANDROID_DL_LIBRARY dl)
# FIXME failing dlopen https://github.com/android-ndk/ndk/issues/929
find_library(ANDROID_DL_LIBRARY NAMES libdl.so dl)
Android cmake build fixed to work with the official android gradle plugin
2017-06-05 02:25:57 +08:00
find_library(ANDROID_LOG_LIBRARY log)
find_library(ANDROID_LIBRARY_LIBRARY android)
list(APPEND EXTRA_LIBS ${ANDROID_DL_LIBRARY} ${ANDROID_LOG_LIBRARY} ${ANDROID_LIBRARY_LIBRARY})
add_definitions(-DGL_GLEXT_PROTOTYPES)
Fixed bug 3918 - HIDAPI, CMake support for android project
2019-08-27 17:38:43 +08:00
if (HAVE_HIDAPI)
list(APPEND EXTRA_LIBS hidapi)
endif()
CMake: Changes to get CMake project to work with Android (thanks, Martin!). Fixes Bugzilla #3194. (but note that Bugzilla #3200 still needs to be resolved to get this really going on Android, at a minimum.)
2015-12-29 13:57:24 +08:00
#enable gles
if(VIDEO_OPENGLES)
set(SDL_VIDEO_OPENGL_EGL 1)
set(HAVE_VIDEO_OPENGLES TRUE)
Fixed bug 4684 - GLES1 variables missing under Android with CMake Braden Obrzut https://hg.libsdl.org/SDL/file/7dc39b047055/CMakeLists.txt#l911 I believe the following should also be specified there: set(SDL_VIDEO_OPENGL_ES 1) set(SDL_VIDEO_RENDER_OGL_ES 1) As it is now GLES1 support is missing when building for Android despite it linking to the library.
2019-06-22 13:01:27 +08:00
set(SDL_VIDEO_OPENGL_ES 1)
set(SDL_VIDEO_RENDER_OGL_ES 1)
CMake: Changes to get CMake project to work with Android (thanks, Martin!). Fixes Bugzilla #3194. (but note that Bugzilla #3200 still needs to be resolved to get this really going on Android, at a minimum.)
2015-12-29 13:57:24 +08:00
set(SDL_VIDEO_OPENGL_ES2 1)
set(SDL_VIDEO_RENDER_OGL_ES2 1)
Android cmake build fixed to work with the official android gradle plugin
2017-06-05 02:25:57 +08:00
find_library(OpenGLES1_LIBRARY GLESv1_CM)
find_library(OpenGLES2_LIBRARY GLESv2)
list(APPEND EXTRA_LIBS ${OpenGLES1_LIBRARY} ${OpenGLES2_LIBRARY})
CMake: Changes to get CMake project to work with Android (thanks, Martin!). Fixes Bugzilla #3194. (but note that Bugzilla #3200 still needs to be resolved to get this really going on Android, at a minimum.)
2015-12-29 13:57:24 +08:00
endif()
vulkan: Initial Vulkan support! This work was done by Jacob Lifshay and Mark Callow; I'm just merging it into revision control.
2017-08-28 10:15:57 +08:00
CHECK_C_SOURCE_COMPILES("
We removed QuartzCore link dependency in commit 5be6badaf7e1
2017-08-29 13:42:41 +08:00
#if defined(__ARM_ARCH) && __ARM_ARCH < 7
vulkan: Initial Vulkan support! This work was done by Jacob Lifshay and Mark Callow; I'm just merging it into revision control.
2017-08-28 10:15:57 +08:00
#error Vulkan doesn't work on this configuration
#endif
int main()
{
return 0;
}
" VULKAN_PASSED_ANDROID_CHECKS)
if(NOT VULKAN_PASSED_ANDROID_CHECKS)
set(VIDEO_VULKAN OFF)
message(STATUS "Vulkan doesn't work on this configuration")
endif()
CMake: Changes to get CMake project to work with Android (thanks, Martin!). Fixes Bugzilla #3194. (but note that Bugzilla #3200 still needs to be resolved to get this really going on Android, at a minimum.)
2015-12-29 13:57:24 +08:00
endif()
Android build fixes -Enabling checking for GCC_ATOMICS also on clang by default. This way all Android ABIs build successfully -Android cmake: Threading was not enabled correctly -Android cmake: Timers and dynamic lib loading were not included in the sources
2017-06-09 03:07:55 +08:00
CheckPTHREAD()
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
elseif(EMSCRIPTEN)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# Hide noisy warnings that intend to aid mostly during initial stages of porting a new
# project. Uncomment at will for verbose cross-compiling -I/../ path info.
add_definitions(-Wno-warn-absolute-paths)
if(SDL_AUDIO)
set(SDL_AUDIO_DRIVER_EMSCRIPTEN 1)
file(GLOB EM_AUDIO_SOURCES ${SDL2_SOURCE_DIR}/src/audio/emscripten/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${EM_AUDIO_SOURCES})
set(HAVE_SDL_AUDIO TRUE)
endif()
if(SDL_FILESYSTEM)
set(SDL_FILESYSTEM_EMSCRIPTEN 1)
file(GLOB EM_FILESYSTEM_SOURCES ${SDL2_SOURCE_DIR}/src/filesystem/emscripten/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${EM_FILESYSTEM_SOURCES})
set(HAVE_SDL_FILESYSTEM TRUE)
endif()
if(SDL_JOYSTICK)
set(SDL_JOYSTICK_EMSCRIPTEN 1)
file(GLOB EM_JOYSTICK_SOURCES ${SDL2_SOURCE_DIR}/src/joystick/emscripten/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${EM_JOYSTICK_SOURCES})
set(HAVE_SDL_JOYSTICK TRUE)
endif()
if(SDL_POWER)
set(SDL_POWER_EMSCRIPTEN 1)
file(GLOB EM_POWER_SOURCES ${SDL2_SOURCE_DIR}/src/power/emscripten/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${EM_POWER_SOURCES})
set(HAVE_SDL_POWER TRUE)
endif()
locale: Implemented SDL_GetPreferredLocales(). This was something I proposed a long time ago, Sylvain Becker did additional work on it, then back to me. Fixes Bugzilla #2131.
2020-05-04 14:27:29 +08:00
if(SDL_LOCALE)
file(GLOB LOCALE_SOURCES ${SDL2_SOURCE_DIR}/src/locale/emscripten/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${LOCALE_SOURCES})
set(HAVE_SDL_LOCALE TRUE)
endif()
Include timers when using cmake (#29)
2016-09-13 15:03:40 +08:00
if(SDL_TIMERS)
set(SDL_TIMER_UNIX 1)
file(GLOB TIMER_SOURCES ${SDL2_SOURCE_DIR}/src/timer/unix/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${TIMER_SOURCES})
set(HAVE_SDL_TIMERS TRUE)
if(CLOCK_GETTIME)
set(HAVE_CLOCK_GETTIME 1)
endif()
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_VIDEO)
set(SDL_VIDEO_DRIVER_EMSCRIPTEN 1)
file(GLOB EM_VIDEO_SOURCES ${SDL2_SOURCE_DIR}/src/video/emscripten/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${EM_VIDEO_SOURCES})
set(HAVE_SDL_VIDEO TRUE)
#enable gles
if(VIDEO_OPENGLES)
set(SDL_VIDEO_OPENGL_EGL 1)
set(HAVE_VIDEO_OPENGLES TRUE)
set(SDL_VIDEO_OPENGL_ES2 1)
set(SDL_VIDEO_RENDER_OGL_ES2 1)
endif()
endif()
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
Add basic support for compiling on RISC OS
2020-02-14 04:50:47 +08:00
elseif(UNIX AND NOT APPLE AND NOT ANDROID AND NOT RISCOS)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_AUDIO)
if(SYSV5 OR SOLARIS OR HPUX)
set(SDL_AUDIO_DRIVER_SUNAUDIO 1)
file(GLOB SUN_AUDIO_SOURCES ${SDL2_SOURCE_DIR}/src/audio/sun/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${SUN_AUDIO_SOURCES})
set(HAVE_SDL_AUDIO TRUE)
audio: rename bsd target to netbsd. Apparently this is no longer a generic BSD audio target, and hasn't been for years, so rename it for NetBSD.
2017-05-25 07:56:59 +08:00
elseif(NETBSD)
set(SDL_AUDIO_DRIVER_NETBSD 1)
file(GLOB NETBSD_AUDIO_SOURCES ${SDL2_SOURCE_DIR}/src/audio/netbsd/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${NETBSD_AUDIO_SOURCES})
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(HAVE_SDL_AUDIO TRUE)
elseif(AIX)
set(SDL_AUDIO_DRIVER_PAUDIO 1)
file(GLOB AIX_AUDIO_SOURCES ${SDL2_SOURCE_DIR}/src/audio/paudio/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${AIX_AUDIO_SOURCES})
set(HAVE_SDL_AUDIO TRUE)
endif()
CheckOSS()
CheckALSA()
jack: Initial shot at a JACK audio target. http://jackaudio.org/ Fixes Bugzilla #2163. (with several more commits following to improve this code.)
2017-06-09 01:27:58 +08:00
CheckJACK()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
CheckPulseAudio()
CheckESD()
CheckARTS()
CheckNAS()
CheckSNDIO()
CheckFusionSound()
Added configure and cmake support for libsamplerate
2017-01-07 12:43:53 +08:00
CheckLibSampleRate()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
if(SDL_VIDEO)
# Need to check for Raspberry PI first and add platform specific compiler flags, otherwise the test for GLES fails!
CheckRPI()
CheckX11()
CheckDirectFB()
CheckOpenGLX11()
CheckOpenGLESX11()
CheckWayland()
CheckVivante()
[Buildsystem] Only build KMSDRM support if EGL+OpenGL is detected by CMake.
2020-12-23 00:17:30 +08:00
# Need to check EGL before checking KMSDRM because KMSDRM depends on it.
CheckEGLKMSDRM()
Fixed bug 3690 - SDL2 KMS/DRM render context support Manuel The attached patch adds support for KMS/DRM context graphics. It builds with no problem on X86_64 GNU/Linux systems, provided the needed libraries are present, and on ARM GNU/Linux systems that have KMS/DRM support and a GLES2 implementation. Tested on Raspberry Pi: KMS/DRM is what the Raspberry Pi will use as default in the near future, once the propietary DispmanX API by Broadcom is overtaken by open graphics stack, it's possible to boot current Raspbian system in KMS mode by adding "dtoverlay=vc4-kms-v3d" to config.txt on Raspbian's boot partition. X86 systems use KMS right away in every current GNU/Linux system. Simple build instructions: $./autogen.sh $./configure --enable-video-kmsdrm $make
2017-08-03 01:22:48 +08:00
CheckKMSDRM()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
cmake: add core/unix sources to the build. Fixes Bugzilla #3757.
2017-08-18 03:44:47 +08:00
if(UNIX)
file(GLOB CORE_UNIX_SOURCES ${SDL2_SOURCE_DIR}/src/core/unix/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${CORE_UNIX_SOURCES})
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(LINUX)
check_c_source_compiles("
#include <linux/input.h>
#ifndef EVIOCGNAME
#error EVIOCGNAME() ioctl not available
#endif
int main(int argc, char** argv) {}" HAVE_INPUT_EVENTS)
check_c_source_compiles("
#include <linux/kd.h>
#include <linux/keyboard.h>
Added debug postfix to install command and fixed library path
2018-04-10 23:03:54 +08:00
int main(int argc, char **argv)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
{
struct kbentry kbe;
kbe.kb_table = KG_CTRL;
ioctl(0, KDGKBENT, &kbe);
}" HAVE_INPUT_KD)
if(HAVE_INPUT_EVENTS)
set(SDL_INPUT_LINUXEV 1)
endif()
if(SDL_HAPTIC AND HAVE_INPUT_EVENTS)
set(SDL_HAPTIC_LINUX 1)
file(GLOB HAPTIC_SOURCES ${SDL2_SOURCE_DIR}/src/haptic/linux/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${HAPTIC_SOURCES})
set(HAVE_SDL_HAPTIC TRUE)
endif()
if(HAVE_INPUT_KD)
set(SDL_INPUT_LINUXKD 1)
endif()
check_include_file("libudev.h" HAVE_LIBUDEV_H)
fix bug #5395: handle old systems where inotify_init1 is not available
2020-12-13 20:32:24 +08:00
check_include_file("sys/inotify.h" HAVE_SYS_INOTIFY_H)
check_symbol_exists(inotify_init "sys/inotify.h" HAVE_INOTIFY_INIT)
check_symbol_exists(inotify_init1 "sys/inotify.h" HAVE_INOTIFY_INIT1)
if(HAVE_SYS_INOTIFY_H AND HAVE_INOTIFY_INIT)
set(HAVE_INOTIFY 1)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
Fixed bug 3301 - IBus support on Linux with CMake joe.gsoc16 I recently looked into Unicode support in SDL2 and realized that SDL_TEXTEDITING doesn't get triggered at all (Japanese IME). According to others on IRC it works fine on Windows/Mac but not for me on (arch)Linux. When compiling SDL with autotools, IBus support is enabled by default but not so with CMake. I never used CMake before but got it working and also included that pkg-config determines flags for dbus (FIXME in CMakeLists).
2016-10-02 03:48:08 +08:00
if(PKG_CONFIG_FOUND)
pkg_search_module(DBUS dbus-1 dbus)
if(DBUS_FOUND)
set(HAVE_DBUS_DBUS_H TRUE)
include_directories(${DBUS_INCLUDE_DIRS})
list(APPEND EXTRA_LIBS ${DBUS_LIBRARIES})
Fixed bug 5103 - Port fcitx support to both fcitx 4 & 5 wengxt Due to the new major fcitx version is coming close, the existing code need to be ported to use new Fcitx dbus interface. The new dbus interface is supported by both fcitx 4 and 5, and has a good side effect, which is that it will work with flatpak for free. Also the patch remove the dependency on fcitx header. Instead, it just hardcodes a few enum value in the code so need to handle the different header for fcitx4 or 5.
2020-05-12 05:31:04 +08:00
# Fcitx need only dbus.
set(HAVE_FCITX TRUE)
Fixed bug 3301 - IBus support on Linux with CMake joe.gsoc16 I recently looked into Unicode support in SDL2 and realized that SDL_TEXTEDITING doesn't get triggered at all (Japanese IME). According to others on IRC it works fine on Windows/Mac but not for me on (arch)Linux. When compiling SDL with autotools, IBus support is enabled by default but not so with CMake. I never used CMake before but got it working and also included that pkg-config determines flags for dbus (FIXME in CMakeLists).
2016-10-02 03:48:08 +08:00
endif()
pkg_search_module(IBUS ibus-1.0 ibus)
if(IBUS_FOUND)
set(HAVE_IBUS_IBUS_H TRUE)
include_directories(${IBUS_INCLUDE_DIRS})
list(APPEND EXTRA_LIBS ${IBUS_LIBRARIES})
Fixed bug 5103 - Port fcitx support to both fcitx 4 & 5 wengxt Due to the new major fcitx version is coming close, the existing code need to be ported to use new Fcitx dbus interface. The new dbus interface is supported by both fcitx 4 and 5, and has a good side effect, which is that it will work with flatpak for free. Also the patch remove the dependency on fcitx header. Instead, it just hardcodes a few enum value in the code so need to handle the different header for fcitx4 or 5.
2020-05-12 05:31:04 +08:00
endif()
if (HAVE_IBUS_IBUS_H OR HAVE_FCITX)
cmake: Build Linux-specific files like the configure script does. Otherwise we unconditionally compile things we shouldn't. Fixes Bugzilla #5175.
2020-06-27 05:42:30 +08:00
set(SDL_USE_IME TRUE)
add_definitions(-DSDL_USE_IME) # !!! FIXME: why isn't this a definition and not in SDL_config.h.cmake?
Fixed bug 3301 - IBus support on Linux with CMake joe.gsoc16 I recently looked into Unicode support in SDL2 and realized that SDL_TEXTEDITING doesn't get triggered at all (Japanese IME). According to others on IRC it works fine on Windows/Mac but not for me on (arch)Linux. When compiling SDL with autotools, IBus support is enabled by default but not so with CMake. I never used CMake before but got it working and also included that pkg-config determines flags for dbus (FIXME in CMakeLists).
2016-10-02 03:48:08 +08:00
endif()
commit 8f4dedc039190f5e734c47dcc4fc021b5793b659 Author: Micha? Janiszewski <janisozaur+signed@gmail.com> Date: Fri Sep 28 20:38:04 2018 +0200 CMake: fix building tests on Linux In case where libunwind.h has been found, it will be used by compiler, but linker wasn't updated to reflect use of this new library.
2018-09-29 12:03:39 +08:00
if(HAVE_LIBUNWIND_H)
# We've already found the header, so REQUIRE the lib to be present
pkg_search_module(UNWIND REQUIRED libunwind)
pkg_search_module(UNWIND_GENERIC REQUIRED libunwind-generic)
list(APPEND EXTRA_LIBS ${UNWIND_LIBRARIES} ${UNWIND_GENERIC_LIBRARIES})
endif()
Fixed bug 3301 - IBus support on Linux with CMake joe.gsoc16 I recently looked into Unicode support in SDL2 and realized that SDL_TEXTEDITING doesn't get triggered at all (Japanese IME). According to others on IRC it works fine on Windows/Mac but not for me on (arch)Linux. When compiling SDL with autotools, IBus support is enabled by default but not so with CMake. I never used CMake before but got it working and also included that pkg-config determines flags for dbus (FIXME in CMakeLists).
2016-10-02 03:48:08 +08:00
endif()
cmake: Build Linux-specific files like the configure script does. Otherwise we unconditionally compile things we shouldn't. Fixes Bugzilla #5175.
2020-06-27 05:42:30 +08:00
if(HAVE_DBUS_DBUS_H)
set(SOURCE_FILES ${SOURCE_FILES} "${SDL2_SOURCE_DIR}/src/core/linux/SDL_dbus.c")
endif()
if(SDL_USE_IME)
set(SOURCE_FILES ${SOURCE_FILES} "${SDL2_SOURCE_DIR}/src/core/linux/SDL_ime.c")
endif()
if(HAVE_IBUS_IBUS_H)
set(SOURCE_FILES ${SOURCE_FILES} "${SDL2_SOURCE_DIR}/src/core/linux/SDL_ibus.c")
endif()
if(HAVE_FCITX)
set(SOURCE_FILES ${SOURCE_FILES} "${SDL2_SOURCE_DIR}/src/core/linux/SDL_fcitx.c")
endif()
if(HAVE_LIBUDEV_H)
set(SOURCE_FILES ${SOURCE_FILES} "${SDL2_SOURCE_DIR}/src/core/linux/SDL_udev.c")
endif()
if(HAVE_INPUT_EVENTS)
set(SOURCE_FILES ${SOURCE_FILES} "${SDL2_SOURCE_DIR}/src/core/linux/SDL_evdev.c")
set(SOURCE_FILES ${SOURCE_FILES} "${SDL2_SOURCE_DIR}/src/core/linux/SDL_evdev_kbd.c")
endif()
# Always compiled for Linux, unconditionally:
udev: Factor out SDL_EVDEV_GuessDeviceClass This works on capability bitfields that can either come from udev or from ioctls, so it is equally applicable to both udev and non-udev input device detection. Signed-off-by: Simon McVittie <smcv@collabora.com>
2020-11-12 11:14:34 +08:00
set(SOURCE_FILES ${SOURCE_FILES} "${SDL2_SOURCE_DIR}/src/core/linux/SDL_evdev_capabilities.c")
cmake: Build Linux-specific files like the configure script does. Otherwise we unconditionally compile things we shouldn't. Fixes Bugzilla #5175.
2020-06-27 05:42:30 +08:00
set(SOURCE_FILES ${SOURCE_FILES} "${SDL2_SOURCE_DIR}/src/core/linux/SDL_threadprio.c")
# src/core/unix/*.c is included in a generic if(UNIX) section, elsewhere.
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
if(SDL_JOYSTICK)
cmake: Fix locating libusb header at configure-time Removing the CheckUSBHID call is necessary to avoid caching the failed header check result before we find libusb via pkg-config.
2019-08-01 00:34:00 +08:00
if(FREEBSD OR NETBSD OR OPENBSD OR BSDI)
CheckUSBHID()
endif()
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
CheckHIDAPI()
CMake: don't try to use the Linux joystick code on Android (thanks, Martin!). Fixes Bugzilla #3201.
2015-12-29 13:24:37 +08:00
if(LINUX AND NOT ANDROID)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SDL_JOYSTICK_LINUX 1)
Added stub Steam Controller sources to Android and iOS command line builds
2017-09-23 07:33:34 +08:00
file(GLOB JOYSTICK_SOURCES ${SDL2_SOURCE_DIR}/src/joystick/linux/*.c ${SDL2_SOURCE_DIR}/src/joystick/steam/*.c)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SOURCE_FILES ${SOURCE_FILES} ${JOYSTICK_SOURCES})
set(HAVE_SDL_JOYSTICK TRUE)
endif()
endif()
CheckPTHREAD()
if(CLOCK_GETTIME)
check_library_exists(rt clock_gettime "" FOUND_CLOCK_GETTIME)
if(FOUND_CLOCK_GETTIME)
list(APPEND EXTRA_LIBS rt)
set(HAVE_CLOCK_GETTIME 1)
else()
check_library_exists(c clock_gettime "" FOUND_CLOCK_GETTIME)
if(FOUND_CLOCK_GETTIME)
set(HAVE_CLOCK_GETTIME 1)
endif()
endif()
endif()
check_include_file(linux/version.h HAVE_LINUX_VERSION_H)
if(HAVE_LINUX_VERSION_H)
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -DHAVE_LINUX_VERSION_H")
endif()
Removed SDL_AndroidOpenURL, added SDL_OpenURL. Still needs to be wired into Xcode and Visual Studio projects.
2020-10-05 23:30:33 +08:00
file(GLOB MISC_SOURCES ${SDL2_SOURCE_DIR}/src/misc/unix/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${MISC_SOURCES})
set(HAVE_SDL_MISC TRUE)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_POWER)
if(LINUX)
set(SDL_POWER_LINUX 1)
file(GLOB POWER_SOURCES ${SDL2_SOURCE_DIR}/src/power/linux/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${POWER_SOURCES})
set(HAVE_SDL_POWER TRUE)
endif()
endif()
locale: Implemented SDL_GetPreferredLocales(). This was something I proposed a long time ago, Sylvain Becker did additional work on it, then back to me. Fixes Bugzilla #2131.
2020-05-04 14:27:29 +08:00
if(SDL_LOCALE)
file(GLOB LOCALE_SOURCES ${SDL2_SOURCE_DIR}/src/locale/unix/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${LOCALE_SOURCES})
set(HAVE_SDL_LOCALE TRUE)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_FILESYSTEM)
set(SDL_FILESYSTEM_UNIX 1)
file(GLOB FILESYSTEM_SOURCES ${SDL2_SOURCE_DIR}/src/filesystem/unix/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${FILESYSTEM_SOURCES})
set(HAVE_SDL_FILESYSTEM TRUE)
endif()
if(SDL_TIMERS)
set(SDL_TIMER_UNIX 1)
file(GLOB TIMER_SOURCES ${SDL2_SOURCE_DIR}/src/timer/unix/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${TIMER_SOURCES})
set(HAVE_SDL_TIMERS TRUE)
endif()
build: Merge pkg-config Libs.private into Libs for static-only builds A project being built entirely statically will call pkg-config with --static, which utilises the Libs.private field. Conversely it will not use --static when not being built entirely statically, even if there is only a static build of SDL available. This will most likely cause the build to fail due to underlinking unless we merge the Libs fields. This is what the Meson build system does when it generates pkg-config files. This also also follows the behaviour of sdl2-config. At the same time, the runtime linker flags are not applicable to static-only builds so only add them for shared builds.
2020-04-12 06:38:34 +08:00
set(SDL_RLD_FLAGS "")
if(RPATH AND SDL_SHARED)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(BSDI OR FREEBSD OR LINUX OR NETBSD)
Use --enable-new-dtags to set RUNPATH rather than RPATH so that LD_LIBRARY_PATH is not overridden by the application.
2016-01-08 08:42:30 +08:00
set(CMAKE_REQUIRED_FLAGS "-Wl,--enable-new-dtags")
check_c_compiler_flag("" HAVE_ENABLE_NEW_DTAGS)
set(CMAKE_REQUIRED_FLAGS ${ORIG_CMAKE_REQUIRED_FLAGS})
if(HAVE_ENABLE_NEW_DTAGS)
Fixed --enable-new-dtags check with cmake
2016-01-08 09:21:50 +08:00
set(SDL_RLD_FLAGS "-Wl,-rpath,\${libdir} -Wl,--enable-new-dtags")
else()
set(SDL_RLD_FLAGS "-Wl,-rpath,\${libdir}")
Use --enable-new-dtags to set RUNPATH rather than RPATH so that LD_LIBRARY_PATH is not overridden by the application.
2016-01-08 08:42:30 +08:00
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
elseif(SOLARIS)
set(SDL_RLD_FLAGS "-R\${libdir}")
endif()
set(CMAKE_BUILD_WITH_INSTALL_RPATH TRUE)
set(HAVE_RPATH TRUE)
endif()
elseif(WINDOWS)
find_program(WINDRES windres)
check_c_source_compiles("
#include <windows.h>
int main(int argc, char **argv) { }" HAVE_WIN32_CC)
file(GLOB CORE_SOURCES ${SDL2_SOURCE_DIR}/src/core/windows/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${CORE_SOURCES})
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
if(WINDOWS_STORE)
file(GLOB WINRT_SOURCE_FILES ${SDL2_SOURCE_DIR}/src/core/winrt/*.c ${SDL2_SOURCE_DIR}/src/core/winrt/*.cpp)
list(APPEND SOURCE_FILES ${WINRT_SOURCE_FILES})
endif()
CMake: Fixed building for Windows with VS2015 (bug #3080). - Don't try to link with the Visual C runtime. - Avoid code generation that would use functions from the VC runtime.
2016-01-01 03:26:40 +08:00
if(MSVC)
# Prevent codegen that would use the VC runtime libraries.
Fixed bug 3651 - CMake build does not install CMake package configuration tschwinger@elitemail.org Most ironically, although autoconf/automake-based builds install (pretty half-assed) CMake package configuration files, they're missing in installations resulting from CMake-based builds entirely. A proper configuration file typically also loads target exports (implemented in patch 3572, also fixing this issue - see my comment on that issue for details). I believe it would be best to let the dinosaurs go extinct and redirect all build efforts to the CMake end for two reasons: 1. It potentially provides the best user experience, but you'd have to give it some love and ship with less quirky buildfiles. 2. It would force distros to build SDL via CMake and thus would ensure target exports are actually available everywhere. Various CMake patches I submitted today in summary (directly converted from the HG commits and `am`d onto a fork of a git mirror that happened to be on `tip`). https://github.com/tschw/SDL/commits/patched Fixing #2576 #3572, #3613, and this fresh ticket, which is almost entirely advertisement ;). These already do to make SDL much less of a quirky fella to have in your dependency tree...
2017-08-10 10:03:10 +08:00
set_property(DIRECTORY . APPEND PROPERTY COMPILE_OPTIONS "/GS-")
CMake: Fixed building for Windows with VS2015 (bug #3080). - Don't try to link with the Visual C runtime. - Avoid code generation that would use functions from the VC runtime.
2016-01-01 03:26:40 +08:00
if(NOT ARCH_64)
Fixed bug 3651 - CMake build does not install CMake package configuration tschwinger@elitemail.org Most ironically, although autoconf/automake-based builds install (pretty half-assed) CMake package configuration files, they're missing in installations resulting from CMake-based builds entirely. A proper configuration file typically also loads target exports (implemented in patch 3572, also fixing this issue - see my comment on that issue for details). I believe it would be best to let the dinosaurs go extinct and redirect all build efforts to the CMake end for two reasons: 1. It potentially provides the best user experience, but you'd have to give it some love and ship with less quirky buildfiles. 2. It would force distros to build SDL via CMake and thus would ensure target exports are actually available everywhere. Various CMake patches I submitted today in summary (directly converted from the HG commits and `am`d onto a fork of a git mirror that happened to be on `tip`). https://github.com/tschw/SDL/commits/patched Fixing #2576 #3572, #3613, and this fresh ticket, which is almost entirely advertisement ;). These already do to make SDL much less of a quirky fella to have in your dependency tree...
2017-08-10 10:03:10 +08:00
set_property(DIRECTORY . APPEND PROPERTY COMPILE_OPTIONS "/arch:SSE")
CMake: Fixed building for Windows with VS2015 (bug #3080). - Don't try to link with the Visual C runtime. - Avoid code generation that would use functions from the VC runtime.
2016-01-01 03:26:40 +08:00
endif()
endif()
Removed SDL_AndroidOpenURL, added SDL_OpenURL. Still needs to be wired into Xcode and Visual Studio projects.
2020-10-05 23:30:33 +08:00
file(GLOB MISC_SOURCES ${SDL2_SOURCE_DIR}/src/misc/windows/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${MISC_SOURCES})
set(HAVE_SDL_MISC TRUE)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# Check for DirectX
if(DIRECTX)
CMake: Fixed to work on Windows 8 SDK (thanks, Martin!). Fixes Bugzilla #3207.
2015-12-29 13:02:43 +08:00
if(DEFINED MSVC_VERSION AND NOT ${MSVC_VERSION} LESS 1700)
set(USE_WINSDK_DIRECTX TRUE)
endif()
if(NOT CMAKE_COMPILER_IS_MINGW AND NOT USE_WINSDK_DIRECTX)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if("$ENV{DXSDK_DIR}" STREQUAL "")
message_error("DIRECTX requires the \$DXSDK_DIR environment variable to be set")
endif()
set(CMAKE_REQUIRED_FLAGS "/I\"$ENV{DXSDK_DIR}\\Include\"")
endif()
Fixed bug 3098: CMake project never enables xinput support on Windows. Thanks to EntranceJew for debugging the issue.
2015-08-29 06:05:24 +08:00
if(HAVE_WIN32_CC)
# xinput.h may need windows.h, but doesn't include it itself.
check_c_source_compiles("
#include <windows.h>
#include <xinput.h>
int main(int argc, char **argv) { }" HAVE_XINPUT_H)
Fixed bug 3609 - Windows build fails due to conflicting types for 'XINPUT_GAMEPAD_EX' Ozkan Sezer (In reply to Ryan C. Gordon from comment #9) > I've put this patch in as https://hg.libsdl.org/SDL/rev/7213ae46e870 ...can > you verify this works on the latest MinGW? > > Thanks, > --ryan. This patch is wrong: the structure in question has nothing to do with any gcc version in use. I suggest reverting this adding a conigury check for it, instead. Something like the following should do it: (configure needs regenerating.)
2017-07-11 09:31:28 +08:00
check_c_source_compiles("
#include <windows.h>
#include <xinput.h>
XINPUT_GAMEPAD_EX x1;
int main(int argc, char **argv) { }" HAVE_XINPUT_GAMEPAD_EX)
check_c_source_compiles("
#include <windows.h>
#include <xinput.h>
XINPUT_STATE_EX s1;
int main(int argc, char **argv) { }" HAVE_XINPUT_STATE_EX)
Fixed bug 3098: CMake project never enables xinput support on Windows. Thanks to EntranceJew for debugging the issue.
2015-08-29 06:05:24 +08:00
else()
check_include_file(xinput.h HAVE_XINPUT_H)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
check_include_file(d3d9.h HAVE_D3D_H)
check_include_file(d3d11_1.h HAVE_D3D11_H)
check_include_file(ddraw.h HAVE_DDRAW_H)
check_include_file(dsound.h HAVE_DSOUND_H)
check_include_file(dinput.h HAVE_DINPUT_H)
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
if(WINDOWS_STORE OR VCPKG_TARGET_TRIPLET MATCHES "arm-windows")
set(HAVE_DINPUT_H 0)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
check_include_file(dxgi.h HAVE_DXGI_H)
audio: Port WASAPI to WinRT, remove XAudio2 backend. XAudio2 doesn't have capture support, so WASAPI was to replace it; the holdout was WinRT, which still needed it as its primary audio target until the WASAPI code code be made to work. The support matrix now looks like: WinXP: directsound by default, winmm as a fallback for buggy drivers. Vista+: WASAPI (directsound and winmm as fallbacks for debugging). WinRT: WASAPI
2017-12-07 01:24:32 +08:00
if(HAVE_D3D_H OR HAVE_D3D11_H OR HAVE_DDRAW_H OR HAVE_DSOUND_H OR HAVE_DINPUT_H)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(HAVE_DIRECTX TRUE)
CMake: Fixed to work on Windows 8 SDK (thanks, Martin!). Fixes Bugzilla #3207.
2015-12-29 13:02:43 +08:00
if(NOT CMAKE_COMPILER_IS_MINGW AND NOT USE_WINSDK_DIRECTX)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# TODO: change $ENV{DXSDL_DIR} to get the path from the include checks
link_directories($ENV{DXSDK_DIR}\\lib\\${PROCESSOR_ARCH})
include_directories($ENV{DXSDK_DIR}\\Include)
endif()
endif()
set(CMAKE_REQUIRED_FLAGS ${ORIG_CMAKE_REQUIRED_FLAGS})
endif()
endpointvolume.h checks not needed since changeset 13078:8ab094a9df6b .
2019-10-01 19:00:02 +08:00
# headers needed elsewhere
configury, cmake: add check for endpointvolume.h : add HAVE_ENDPOINTVOLUME_H, HAVE_MMDEVICEAPI_H and HAVE_AUDIOCLIENT_H in SDL_config.h.in, SDL_config.h.cmake, SDL_config_windows.h, and in SDL_config_winrt.h.
2018-09-28 18:41:04 +08:00
check_include_file(mmdeviceapi.h HAVE_MMDEVICEAPI_H)
check_include_file(audioclient.h HAVE_AUDIOCLIENT_H)
build: fix / update sensors (windows) configuration - SDL_config.h.in: add missing defines SDL_SENSOR_COREMOTION and SDL_SENSOR_WINDOWS (configure did set SDL_SENSOR_WINDOWS but it never went in SDL_config.h or Makefile.) - SDL_config.h.cmake: remove duplicated SDL_SENSOR_XXX cmake defines. - autofoo, cmake: check for sensorsapi.h header before enabling windows sensors.
2020-10-12 06:02:28 +08:00
check_include_file(sensorsapi.h HAVE_SENSORSAPI_H)
configury, cmake: add check for endpointvolume.h : add HAVE_ENDPOINTVOLUME_H, HAVE_MMDEVICEAPI_H and HAVE_AUDIOCLIENT_H in SDL_config.h.in, SDL_config.h.cmake, SDL_config_windows.h, and in SDL_config_winrt.h.
2018-09-28 18:41:04 +08:00
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_AUDIO)
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
if(NOT WINDOWS_STORE)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SDL_AUDIO_DRIVER_WINMM 1)
file(GLOB WINMM_AUDIO_SOURCES ${SDL2_SOURCE_DIR}/src/audio/winmm/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${WINMM_AUDIO_SOURCES})
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(HAVE_SDL_AUDIO TRUE)
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
if(HAVE_DSOUND_H AND NOT WINDOWS_STORE)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SDL_AUDIO_DRIVER_DSOUND 1)
file(GLOB DSOUND_AUDIO_SOURCES ${SDL2_SOURCE_DIR}/src/audio/directsound/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${DSOUND_AUDIO_SOURCES})
endif()
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
if(WASAPI AND HAVE_AUDIOCLIENT_H AND HAVE_MMDEVICEAPI_H AND NOT WINDOWS_STORE)
cmake: add WASAPI audio target to the build (thanks, Martin!). Fixes Bugzilla #3588.
2017-02-24 11:38:04 +08:00
set(SDL_AUDIO_DRIVER_WASAPI 1)
file(GLOB WASAPI_AUDIO_SOURCES ${SDL2_SOURCE_DIR}/src/audio/wasapi/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${WASAPI_AUDIO_SOURCES})
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
if(SDL_VIDEO)
# requires SDL_LOADSO on Windows (IME, DX, etc.)
if(NOT SDL_LOADSO)
message_error("SDL_VIDEO requires SDL_LOADSO, which is not enabled")
endif()
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
if(WINDOWS_STORE)
set(SDL_VIDEO_DRIVER_WINRT 1)
file(GLOB WIN_VIDEO_SOURCES
${SDL2_SOURCE_DIR}/src/video/winrt/*.c
${SDL2_SOURCE_DIR}/src/video/winrt/*.cpp
${SDL2_SOURCE_DIR}/src/render/direct3d11/*.cpp
)
else()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SDL_VIDEO_DRIVER_WINDOWS 1)
file(GLOB WIN_VIDEO_SOURCES ${SDL2_SOURCE_DIR}/src/video/windows/*.c)
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SOURCE_FILES ${SOURCE_FILES} ${WIN_VIDEO_SOURCES})
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
if(RENDER_D3D AND HAVE_D3D_H AND NOT WINDOWS_STORE)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SDL_VIDEO_RENDER_D3D 1)
set(HAVE_RENDER_D3D TRUE)
endif()
if(RENDER_D3D AND HAVE_D3D11_H)
set(SDL_VIDEO_RENDER_D3D11 1)
set(HAVE_RENDER_D3D TRUE)
endif()
set(HAVE_SDL_VIDEO TRUE)
endif()
if(SDL_THREADS)
set(SDL_THREAD_WINDOWS 1)
set(SOURCE_FILES ${SOURCE_FILES}
${SDL2_SOURCE_DIR}/src/thread/windows/SDL_sysmutex.c
${SDL2_SOURCE_DIR}/src/thread/windows/SDL_syssem.c
${SDL2_SOURCE_DIR}/src/thread/windows/SDL_systhread.c
${SDL2_SOURCE_DIR}/src/thread/windows/SDL_systls.c
${SDL2_SOURCE_DIR}/src/thread/generic/SDL_syscond.c)
set(HAVE_SDL_THREADS TRUE)
endif()
build: fix / update sensors (windows) configuration - SDL_config.h.in: add missing defines SDL_SENSOR_COREMOTION and SDL_SENSOR_WINDOWS (configure did set SDL_SENSOR_WINDOWS but it never went in SDL_config.h or Makefile.) - SDL_config.h.cmake: remove duplicated SDL_SENSOR_XXX cmake defines. - autofoo, cmake: check for sensorsapi.h header before enabling windows sensors.
2020-10-12 06:02:28 +08:00
if(SDL_SENSOR AND HAVE_SENSORSAPI_H)
First pass at Windows sensor implementation
2020-04-08 23:34:27 +08:00
set(SDL_SENSOR_WINDOWS 1)
set(HAVE_SDL_SENSORS TRUE)
file(GLOB WINDOWS_SENSOR_SOURCES ${SDL2_SOURCE_DIR}/src/sensor/windows/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${WINDOWS_SENSOR_SOURCES})
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_POWER)
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
if(WINDOWS_STORE)
set(SDL_POWER_WINRT 1)
set(SOURCE_FILES ${SOURCE_FILES} ${SDL2_SOURCE_DIR}/src/power/winrt/SDL_syspower.cpp)
else()
set(SDL_POWER_WINDOWS 1)
set(SOURCE_FILES ${SOURCE_FILES} ${SDL2_SOURCE_DIR}/src/power/windows/SDL_syspower.c)
set(HAVE_SDL_POWER TRUE)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
locale: Implemented SDL_GetPreferredLocales(). This was something I proposed a long time ago, Sylvain Becker did additional work on it, then back to me. Fixes Bugzilla #2131.
2020-05-04 14:27:29 +08:00
if(SDL_LOCALE)
file(GLOB LOCALE_SOURCES ${SDL2_SOURCE_DIR}/src/locale/windows/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${LOCALE_SOURCES})
set(HAVE_SDL_LOCALE TRUE)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_FILESYSTEM)
set(SDL_FILESYSTEM_WINDOWS 1)
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
if(WINDOWS_STORE)
file(GLOB FILESYSTEM_SOURCES ${SDL2_SOURCE_DIR}/src/filesystem/winrt/*.cpp)
else()
file(GLOB FILESYSTEM_SOURCES ${SDL2_SOURCE_DIR}/src/filesystem/windows/*.c)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SOURCE_FILES ${SOURCE_FILES} ${FILESYSTEM_SOURCES})
set(HAVE_SDL_FILESYSTEM TRUE)
endif()
# Libraries for Win32 native and MinGW
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
if(NOT WINDOWS_STORE)
list(APPEND EXTRA_LIBS user32 gdi32 winmm imm32 ole32 oleaut32 version uuid advapi32 setupapi shell32)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
configure.in: Rename to configure.ac to fix an 'aclocal' warning Also rename references in related files.
2019-03-25 20:01:32 +08:00
# TODO: in configure.ac the check for timers is set on
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# cygwin | mingw32* - does this include mingw32CE?
if(SDL_TIMERS)
set(SDL_TIMER_WINDOWS 1)
file(GLOB TIMER_SOURCES ${SDL2_SOURCE_DIR}/src/timer/windows/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${TIMER_SOURCES})
set(HAVE_SDL_TIMERS TRUE)
endif()
if(SDL_LOADSO)
set(SDL_LOADSO_WINDOWS 1)
file(GLOB LOADSO_SOURCES ${SDL2_SOURCE_DIR}/src/loadso/windows/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${LOADSO_SOURCES})
set(HAVE_SDL_LOADSO TRUE)
endif()
file(GLOB CORE_SOURCES ${SDL2_SOURCE_DIR}/src/core/windows/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${CORE_SOURCES})
if(SDL_VIDEO)
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
if(VIDEO_OPENGL AND NOT WINDOWS_STORE)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SDL_VIDEO_OPENGL 1)
set(SDL_VIDEO_OPENGL_WGL 1)
set(SDL_VIDEO_RENDER_OGL 1)
set(HAVE_VIDEO_OPENGL TRUE)
endif()
Fix bug 3081: CMake configuration auto-disables OpenGL ES support on Windows. Thanks to EntranceJew for the patch.
2015-08-29 06:10:46 +08:00
if(VIDEO_OPENGLES)
set(SDL_VIDEO_OPENGL_EGL 1)
set(SDL_VIDEO_OPENGL_ES2 1)
set(SDL_VIDEO_RENDER_OGL_ES2 1)
set(HAVE_VIDEO_OPENGLES TRUE)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
if(SDL_JOYSTICK)
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
CheckHIDAPI()
hidapi: Add SDL_hidapi.c, allows support for multiple hidapi backends. This is currently supported on Linux and macOS. iOS and Android are not supported at all, Windows support could be added with some changes to the libusb backend. The Visual Studio and Xcode projects do not use this feature. Based on Valve Software's hid.cpp, written in collaboration with Andrew Eikum.
2019-08-01 00:20:55 +08:00
# TODO: Remove this hid.c block when SDL_hidapi.c is supported on Windows!
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
if(HAVE_HIDAPI)
set(SOURCE_FILES ${SOURCE_FILES} ${SDL2_SOURCE_DIR}/src/hidapi/windows/hid.c)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
file(GLOB JOYSTICK_SOURCES ${SDL2_SOURCE_DIR}/src/joystick/windows/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${JOYSTICK_SOURCES})
if(HAVE_DINPUT_H)
set(SDL_JOYSTICK_DINPUT 1)
Windows CMake: don't try to link dxguid, it's unavailable when using Visual Studio's XP-compatible toolset.
2016-03-19 09:15:53 +08:00
list(APPEND EXTRA_LIBS dinput8)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
if(HAVE_XINPUT_H)
set(SDL_JOYSTICK_XINPUT 1)
endif()
if(NOT HAVE_DINPUT_H AND NOT HAVE_XINPUT_H)
set(SDL_JOYSTICK_WINMM 1)
endif()
set(HAVE_SDL_JOYSTICK TRUE)
if(SDL_HAPTIC)
if(HAVE_DINPUT_H OR HAVE_XINPUT_H)
file(GLOB HAPTIC_SOURCES ${SDL2_SOURCE_DIR}/src/haptic/windows/*.c)
if(HAVE_DINPUT_H)
set(SDL_HAPTIC_DINPUT 1)
endif()
if(HAVE_XINPUT_H)
set(SDL_HAPTIC_XINPUT 1)
endif()
else()
file(GLOB HAPTIC_SOURCES ${SDL2_SOURCE_DIR}/src/haptic/dummy/*.c)
set(SDL_HAPTIC_DUMMY 1)
endif()
set(SOURCE_FILES ${SOURCE_FILES} ${HAPTIC_SOURCES})
set(HAVE_SDL_HAPTIC TRUE)
endif()
endif()
file(GLOB VERSION_SOURCES ${SDL2_SOURCE_DIR}/src/main/windows/*.rc)
file(GLOB SDLMAIN_SOURCES ${SDL2_SOURCE_DIR}/src/main/windows/*.c)
if(MINGW OR CYGWIN)
list(APPEND EXTRA_LIBS mingw32)
list(APPEND EXTRA_LDFLAGS "-mwindows")
set(SDL_CFLAGS "${SDL_CFLAGS} -Dmain=SDL_main")
list(APPEND SDL_LIBS "-lmingw32" "-lSDL2main" "-mwindows")
endif()
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
elseif(APPLE)
cmake: some iOS fixes (don't link to Cocoa or Carbon frameworks). Fixes Bugzilla #3625.
2017-08-19 08:25:14 +08:00
# TODO: rework this all for proper MacOS X, iOS and Darwin support
# We always need these libs on macOS at the moment.
# !!! FIXME: we need Carbon for some very old API calls in
# !!! FIXME: src/video/cocoa/SDL_cocoakeyboard.c, but we should figure out
# !!! FIXME: how to dump those.
CMake: tvOS support/fixes To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=tvOS - CMAKE_OSX_SYSROOT=<SDK> (examples: appletvos, appletvsimulator, appletvos12.4, /full/path/to/AppleTVOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;x86_64")
2019-08-28 00:30:20 +08:00
if(DARWIN OR MACOSX)
cmake: some iOS fixes (don't link to Cocoa or Carbon frameworks). Fixes Bugzilla #3625.
2017-08-19 08:25:14 +08:00
set(SDL_FRAMEWORK_COCOA 1)
set(SDL_FRAMEWORK_CARBON 1)
endif()
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
set(SDL_FRAMEWORK_FOUNDATION 1)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# Requires the darwin file implementation
if(SDL_FILE)
CMake: Changes to get CMake project to work with Android (thanks, Martin!). Fixes Bugzilla #3194. (but note that Bugzilla #3200 still needs to be resolved to get this really going on Android, at a minimum.)
2015-12-29 13:57:24 +08:00
file(GLOB EXTRA_SOURCES ${SDL2_SOURCE_DIR}/src/file/cocoa/*.m)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SOURCE_FILES ${EXTRA_SOURCES} ${SOURCE_FILES})
cmake: added a FIXME for later. Have to figure out what cmake version fixed this and bump the minimum to that.
2017-08-19 08:00:29 +08:00
# !!! FIXME: modern CMake doesn't need "LANGUAGE C" for Objective-C.
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set_source_files_properties(${EXTRA_SOURCES} PROPERTIES LANGUAGE C)
set(HAVE_SDL_FILE TRUE)
cmake: some iOS fixes (don't link to Cocoa or Carbon frameworks). Fixes Bugzilla #3625.
2017-08-19 08:25:14 +08:00
# !!! FIXME: why is COREVIDEO inside this if() block?
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SDL_FRAMEWORK_COREVIDEO 1)
else()
message_error("SDL_FILE must be enabled to build on MacOS X")
endif()
Removed SDL_AndroidOpenURL, added SDL_OpenURL. Still needs to be wired into Xcode and Visual Studio projects.
2020-10-05 23:30:33 +08:00
file(GLOB MISC_SOURCES ${SDL2_SOURCE_DIR}/src/misc/macosx/*.m)
set(SOURCE_FILES ${SOURCE_FILES} ${MISC_SOURCES})
set(HAVE_SDL_MISC TRUE)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_AUDIO)
set(SDL_AUDIO_DRIVER_COREAUDIO 1)
Hopefully fixed the cmake build on Mac OS X
2016-09-15 23:46:14 +08:00
file(GLOB AUDIO_SOURCES ${SDL2_SOURCE_DIR}/src/audio/coreaudio/*.m)
Fixed bug 3882 - cmake fix for osx Ozkan Sezer In my cross-build environment with cmake-2.8.12.1, cmake does not add SDL_coreaudio.m to its makefiles and the result is a failure. The fix is simple: set the language to C for it as it is done at other places in CMakeLists.txt.
2017-10-16 12:07:01 +08:00
# !!! FIXME: modern CMake doesn't need "LANGUAGE C" for Objective-C.
set_source_files_properties(${AUDIO_SOURCES} PROPERTIES LANGUAGE C)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SOURCE_FILES ${SOURCE_FILES} ${AUDIO_SOURCES})
set(HAVE_SDL_AUDIO TRUE)
Fixed set of libraries needed to build on Mac OS X
2016-09-15 23:57:56 +08:00
set(SDL_FRAMEWORK_COREAUDIO 1)
set(SDL_FRAMEWORK_AUDIOTOOLBOX 1)
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
set(SDL_FRAMEWORK_AVFOUNDATION 1)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
if(SDL_JOYSTICK)
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
CheckHIDAPI()
if(HAVE_HIDAPI)
CMake: tvOS support/fixes To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=tvOS - CMAKE_OSX_SYSROOT=<SDK> (examples: appletvos, appletvsimulator, appletvos12.4, /full/path/to/AppleTVOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;x86_64")
2019-08-28 00:30:20 +08:00
if(IOS OR TVOS)
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
set(SOURCE_FILES ${SOURCE_FILES} ${SDL2_SOURCE_DIR}/src/hidapi/ios/hid.m)
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
set(SDL_FRAMEWORK_COREBLUETOOTH 1)
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
endif()
endif()
CMake: tvOS support/fixes To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=tvOS - CMAKE_OSX_SYSROOT=<SDK> (examples: appletvos, appletvsimulator, appletvos12.4, /full/path/to/AppleTVOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;x86_64")
2019-08-28 00:30:20 +08:00
if(IOS OR TVOS)
Added stub Steam Controller sources to Android and iOS command line builds
2017-09-23 07:33:34 +08:00
file(GLOB JOYSTICK_SOURCES ${SDL2_SOURCE_DIR}/src/joystick/iphoneos/*.m ${SDL2_SOURCE_DIR}/src/joystick/steam/*.c)
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
set(SDL_JOYSTICK_MFI 1)
CMake: tvOS support/fixes To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=tvOS - CMAKE_OSX_SYSROOT=<SDK> (examples: appletvos, appletvsimulator, appletvos12.4, /full/path/to/AppleTVOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;x86_64")
2019-08-28 00:30:20 +08:00
if(IOS)
set(SDL_FRAMEWORK_COREMOTION 1)
endif()
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
set(SDL_FRAMEWORK_GAMECONTROLLER 1)
set(HAVE_SDL_SENSORS 1)
Fixed bug 3268 - CMAKE: ios fixes Martin Gerhardy ios is not properly handled in the cmake file. this patch is a start to do so - it properly builds for me with it.
2016-10-02 03:54:48 +08:00
else()
cmake: fix building for mac (bug #5407.)
2020-12-21 02:55:02 +08:00
file(GLOB JOYSTICK_SOURCES ${SDL2_SOURCE_DIR}/src/joystick/darwin/*.c ${SDL2_SOURCE_DIR}/src/joystick/iphoneos/*.m)
CMakeLists.txt: add src/joystick/iphoneos/*.m to Darwin joystick sources .. so that there won't be missing symbols. TODO: add checks for SDL_JOYSTICK_MFI ???
2020-11-24 19:10:30 +08:00
# FIXME: add checks for SDL_JOYSTICK_MFI???
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
set(SDL_JOYSTICK_IOKIT 1)
set(SDL_FRAMEWORK_IOKIT 1)
set(SDL_FRAMEWORK_FF 1)
Fixed bug 3268 - CMAKE: ios fixes Martin Gerhardy ios is not properly handled in the cmake file. this patch is a start to do so - it properly builds for me with it.
2016-10-02 03:54:48 +08:00
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SOURCE_FILES ${SOURCE_FILES} ${JOYSTICK_SOURCES})
set(HAVE_SDL_JOYSTICK TRUE)
endif()
if(SDL_HAPTIC)
CMake: tvOS support/fixes To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=tvOS - CMAKE_OSX_SYSROOT=<SDK> (examples: appletvos, appletvsimulator, appletvos12.4, /full/path/to/AppleTVOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;x86_64")
2019-08-28 00:30:20 +08:00
if (IOS OR TVOS)
Fixed bug 3670 - CMake IOS haptic error
2017-06-13 07:35:34 +08:00
file(GLOB HAPTIC_SOURCES ${SDL2_SOURCE_DIR}/src/haptic/dummy/*.c)
Fixed bug 3268 - CMAKE: ios fixes Martin Gerhardy ios is not properly handled in the cmake file. this patch is a start to do so - it properly builds for me with it.
2016-10-02 03:54:48 +08:00
set(SDL_HAPTIC_DUMMY 1)
else()
file(GLOB HAPTIC_SOURCES ${SDL2_SOURCE_DIR}/src/haptic/darwin/*.c)
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
set(SDL_HAPTIC_IOKIT 1)
set(SDL_FRAMEWORK_IOKIT 1)
set(SDL_FRAMEWORK_FF 1)
Fixed bug 3268 - CMAKE: ios fixes Martin Gerhardy ios is not properly handled in the cmake file. this patch is a start to do so - it properly builds for me with it.
2016-10-02 03:54:48 +08:00
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SOURCE_FILES ${SOURCE_FILES} ${HAPTIC_SOURCES})
set(HAVE_SDL_HAPTIC TRUE)
if(NOT SDL_JOYSTICK)
message(FATAL_ERROR "SDL_HAPTIC requires SDL_JOYSTICK to be enabled")
endif()
endif()
if(SDL_POWER)
CMake: tvOS support/fixes To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=tvOS - CMAKE_OSX_SYSROOT=<SDK> (examples: appletvos, appletvsimulator, appletvos12.4, /full/path/to/AppleTVOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;x86_64")
2019-08-28 00:30:20 +08:00
if (IOS OR TVOS)
Fixed bug 3268 - CMAKE: ios fixes Martin Gerhardy ios is not properly handled in the cmake file. this patch is a start to do so - it properly builds for me with it.
2016-10-02 03:54:48 +08:00
file(GLOB POWER_SOURCES ${SDL2_SOURCE_DIR}/src/power/uikit/*.m)
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
set(SDL_POWER_UIKIT 1)
Fixed bug 3268 - CMAKE: ios fixes Martin Gerhardy ios is not properly handled in the cmake file. this patch is a start to do so - it properly builds for me with it.
2016-10-02 03:54:48 +08:00
else()
file(GLOB POWER_SOURCES ${SDL2_SOURCE_DIR}/src/power/macosx/*.c)
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
set(SDL_POWER_MACOSX 1)
CMake: bug-fix for tvOS support tvOS Device support wasn't working, at least not with the current-latest tvOS release (13.0), with CMake failing during its configuration stage.
2019-09-24 06:24:03 +08:00
set(SDL_FRAMEWORK_IOKIT 1)
Fixed bug 3268 - CMAKE: ios fixes Martin Gerhardy ios is not properly handled in the cmake file. this patch is a start to do so - it properly builds for me with it.
2016-10-02 03:54:48 +08:00
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(SOURCE_FILES ${SOURCE_FILES} ${POWER_SOURCES})
set(HAVE_SDL_POWER TRUE)
endif()
locale: Implemented SDL_GetPreferredLocales(). This was something I proposed a long time ago, Sylvain Becker did additional work on it, then back to me. Fixes Bugzilla #2131.
2020-05-04 14:27:29 +08:00
if(SDL_LOCALE)
file(GLOB LOCALE_SOURCES ${SDL2_SOURCE_DIR}/src/locale/macosx/*.m)
set(SOURCE_FILES ${SOURCE_FILES} ${LOCALE_SOURCES})
set(HAVE_SDL_LOCALE TRUE)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_TIMERS)
set(SDL_TIMER_UNIX 1)
file(GLOB TIMER_SOURCES ${SDL2_SOURCE_DIR}/src/timer/unix/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${TIMER_SOURCES})
set(HAVE_SDL_TIMERS TRUE)
endif(SDL_TIMERS)
if(SDL_FILESYSTEM)
set(SDL_FILESYSTEM_COCOA 1)
file(GLOB FILESYSTEM_SOURCES ${SDL2_SOURCE_DIR}/src/filesystem/cocoa/*.m)
cmake: added a FIXME for later. Have to figure out what cmake version fixed this and bump the minimum to that.
2017-08-19 08:00:29 +08:00
# !!! FIXME: modern CMake doesn't need "LANGUAGE C" for Objective-C.
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set_source_files_properties(${FILESYSTEM_SOURCES} PROPERTIES LANGUAGE C)
set(SOURCE_FILES ${SOURCE_FILES} ${FILESYSTEM_SOURCES})
set(HAVE_SDL_FILESYSTEM TRUE)
endif()
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
if(SDL_SENSOR)
if(IOS)
set(SDL_SENSOR_COREMOTION 1)
set(HAVE_SDL_SENSORS TRUE)
file(GLOB SENSOR_SOURCES ${SDL2_SOURCE_DIR}/src/sensor/coremotion/*.m)
set(SOURCE_FILES ${SOURCE_FILES} ${SENSOR_SOURCES})
endif()
endif()
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
# iOS hack needed - http://code.google.com/p/ios-cmake/ ?
if(SDL_VIDEO)
CMake: tvOS support/fixes To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=tvOS - CMAKE_OSX_SYSROOT=<SDK> (examples: appletvos, appletvsimulator, appletvos12.4, /full/path/to/AppleTVOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;x86_64")
2019-08-28 00:30:20 +08:00
if (IOS OR TVOS)
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
set(SDL_VIDEO_DRIVER_UIKIT 1)
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
set(SDL_FRAMEWORK_COREGRAPHICS 1)
set(SDL_FRAMEWORK_QUARTZCORE 1)
set(SDL_FRAMEWORK_UIKIT 1)
set(SDL_IPHONE_KEYBOARD 1)
set(SDL_IPHONE_LAUNCHSCREEN 1)
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
file(GLOB UIKITVIDEO_SOURCES ${SDL2_SOURCE_DIR}/src/video/uikit/*.m)
set(SOURCE_FILES ${SOURCE_FILES} ${UIKITVIDEO_SOURCES})
else()
CheckCOCOA()
if(VIDEO_OPENGL)
set(SDL_VIDEO_OPENGL 1)
set(SDL_VIDEO_OPENGL_CGL 1)
set(SDL_VIDEO_RENDER_OGL 1)
set(HAVE_VIDEO_OPENGL TRUE)
endif()
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
endif()
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
if(VIDEO_OPENGLES)
CMake: tvOS support/fixes To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=tvOS - CMAKE_OSX_SYSROOT=<SDK> (examples: appletvos, appletvsimulator, appletvos12.4, /full/path/to/AppleTVOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;x86_64")
2019-08-28 00:30:20 +08:00
if(IOS OR TVOS)
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
set(SDL_FRAMEWORK_OPENGLES 1)
set(SDL_VIDEO_OPENGL_ES 1)
set(SDL_VIDEO_RENDER_OGL_ES 1)
else()
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
set(SDL_VIDEO_OPENGL_EGL 1)
endif()
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
set(SDL_VIDEO_OPENGL_ES2 1)
set(SDL_VIDEO_RENDER_OGL_ES2 1)
set(HAVE_VIDEO_OPENGLES TRUE)
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
endif()
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
if(VIDEO_VULKAN OR VIDEO_METAL OR RENDER_METAL)
Fixed bug 5366 - cmake build doesn't detect Metal on macOS Tom Seddon 2nd time lucky, perhaps. patch 2 applies to current HEAD at time of writing - 4eb049c9bb1ca94efe3c40b57beda3169984d0cb from https://github.com/SDL-mirror/SDL. This basically goes back to what was there originally, but now manually adding "-x objective-c" to the clang command line rather than "-ObjC". clang is then invoked without the "-x c" that was causing the problem, the snippet builds, and Metal is detected. (I had a quick trawl through the cmake code, but I couldn't see where this is handled.) I was moved to try this after finding SDL's own CHECK_OBJC_SOURCE_COMPILES macro, and noting what it does: https://github.com/SDL-mirror/SDL/blob/4eb049c9bb1ca94efe3c40b57beda3169984d0cb/cmake/macros.cmake#L67 An alternative fix of course would be to use CHECK_OBJC_SOURCE_COMPILES instead of cmake's check_objc_source_compiles - but that had the same problem of getting confused by "return 0;". (Maybe that's because it's a macro? I'll defer to a cmake expert on this one.) I decided in the end to err on the side of leaving things looking basically the same as they were before my first patch.
2020-12-09 22:17:55 +08:00
set(ORIG_CMAKE_REQUIRED_FLAGS ${CMAKE_REQUIRED_FLAGS})
Backed out changes for 5366 - cmake build doesn't detect Metal on macOS These changes introduce regressions for other build environments, so I'm backing them out until we sort out the correct fix.
2020-12-09 23:32:10 +08:00
set(CMAKE_REQUIRED_FLAGS "${CMAKE_REQUIRED_FLAGS} -ObjC")
Fixed bug 5366 - cmake build doesn't detect Metal on macOS Tom Seddon 2nd time lucky, perhaps. patch 2 applies to current HEAD at time of writing - 4eb049c9bb1ca94efe3c40b57beda3169984d0cb from https://github.com/SDL-mirror/SDL. This basically goes back to what was there originally, but now manually adding "-x objective-c" to the clang command line rather than "-ObjC". clang is then invoked without the "-x c" that was causing the problem, the snippet builds, and Metal is detected. (I had a quick trawl through the cmake code, but I couldn't see where this is handled.) I was moved to try this after finding SDL's own CHECK_OBJC_SOURCE_COMPILES macro, and noting what it does: https://github.com/SDL-mirror/SDL/blob/4eb049c9bb1ca94efe3c40b57beda3169984d0cb/cmake/macros.cmake#L67 An alternative fix of course would be to use CHECK_OBJC_SOURCE_COMPILES instead of cmake's check_objc_source_compiles - but that had the same problem of getting confused by "return 0;". (Maybe that's because it's a macro? I'll defer to a cmake expert on this one.) I decided in the end to err on the side of leaving things looking basically the same as they were before my first patch.
2020-12-09 22:17:55 +08:00
check_c_source_compiles("
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
#include <AvailabilityMacros.h>
#import <Metal/Metal.h>
#import <QuartzCore/CAMetalLayer.h>
#if TARGET_OS_SIMULATOR || (!TARGET_CPU_X86_64 && !TARGET_CPU_ARM64)
#error Metal doesn't work on this configuration
#endif
int main()
{
Fixed bug 5366 - cmake build doesn't detect Metal on macOS Tom Seddon 2nd time lucky, perhaps. patch 2 applies to current HEAD at time of writing - 4eb049c9bb1ca94efe3c40b57beda3169984d0cb from https://github.com/SDL-mirror/SDL. This basically goes back to what was there originally, but now manually adding "-x objective-c" to the clang command line rather than "-ObjC". clang is then invoked without the "-x c" that was causing the problem, the snippet builds, and Metal is detected. (I had a quick trawl through the cmake code, but I couldn't see where this is handled.) I was moved to try this after finding SDL's own CHECK_OBJC_SOURCE_COMPILES macro, and noting what it does: https://github.com/SDL-mirror/SDL/blob/4eb049c9bb1ca94efe3c40b57beda3169984d0cb/cmake/macros.cmake#L67 An alternative fix of course would be to use CHECK_OBJC_SOURCE_COMPILES instead of cmake's check_objc_source_compiles - but that had the same problem of getting confused by "return 0;". (Maybe that's because it's a macro? I'll defer to a cmake expert on this one.) I decided in the end to err on the side of leaving things looking basically the same as they were before my first patch.
2020-12-09 22:17:55 +08:00
return 0;
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
}
" HAVE_FRAMEWORK_METAL)
Fixed bug 5366 - cmake build doesn't detect Metal on macOS Tom Seddon 2nd time lucky, perhaps. patch 2 applies to current HEAD at time of writing - 4eb049c9bb1ca94efe3c40b57beda3169984d0cb from https://github.com/SDL-mirror/SDL. This basically goes back to what was there originally, but now manually adding "-x objective-c" to the clang command line rather than "-ObjC". clang is then invoked without the "-x c" that was causing the problem, the snippet builds, and Metal is detected. (I had a quick trawl through the cmake code, but I couldn't see where this is handled.) I was moved to try this after finding SDL's own CHECK_OBJC_SOURCE_COMPILES macro, and noting what it does: https://github.com/SDL-mirror/SDL/blob/4eb049c9bb1ca94efe3c40b57beda3169984d0cb/cmake/macros.cmake#L67 An alternative fix of course would be to use CHECK_OBJC_SOURCE_COMPILES instead of cmake's check_objc_source_compiles - but that had the same problem of getting confused by "return 0;". (Maybe that's because it's a macro? I'll defer to a cmake expert on this one.) I decided in the end to err on the side of leaving things looking basically the same as they were before my first patch.
2020-12-09 22:17:55 +08:00
set(CMAKE_REQUIRED_FLAGS ${ORIG_CMAKE_REQUIRED_FLAGS})
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
if(HAVE_FRAMEWORK_METAL)
set(SDL_FRAMEWORK_METAL 1)
set(SDL_FRAMEWORK_QUARTZCORE 1)
else()
set(VIDEO_VULKAN 0)
set(VIDEO_METAL 0)
set(RENDER_METAL 0)
endif()
endif()
if(VIDEO_METAL)
set(SDL_VIDEO_METAL 1)
set(HAVE_VIDEO_METAL TRUE)
endif()
if(RENDER_METAL)
file(GLOB RENDER_METAL_SOURCES ${SDL2_SOURCE_DIR}/src/render/metal/*.m)
set(SOURCE_FILES ${SOURCE_FILES} ${RENDER_METAL_SOURCES})
set(SDL_VIDEO_RENDER_METAL 1)
set(HAVE_RENDER_METAL TRUE)
endif()
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# Actually load the frameworks at the end so we don't duplicate include.
if(SDL_FRAMEWORK_COREVIDEO)
find_library(COREVIDEO CoreVideo)
list(APPEND EXTRA_LIBS ${COREVIDEO})
endif()
if(SDL_FRAMEWORK_COCOA)
find_library(COCOA_LIBRARY Cocoa)
list(APPEND EXTRA_LIBS ${COCOA_LIBRARY})
endif()
if(SDL_FRAMEWORK_IOKIT)
find_library(IOKIT IOKit)
list(APPEND EXTRA_LIBS ${IOKIT})
endif()
if(SDL_FRAMEWORK_FF)
find_library(FORCEFEEDBACK ForceFeedback)
list(APPEND EXTRA_LIBS ${FORCEFEEDBACK})
endif()
if(SDL_FRAMEWORK_CARBON)
find_library(CARBON_LIBRARY Carbon)
list(APPEND EXTRA_LIBS ${CARBON_LIBRARY})
endif()
Fixed set of libraries needed to build on Mac OS X
2016-09-15 23:57:56 +08:00
if(SDL_FRAMEWORK_COREAUDIO)
find_library(COREAUDIO CoreAudio)
list(APPEND EXTRA_LIBS ${COREAUDIO})
endif()
if(SDL_FRAMEWORK_AUDIOTOOLBOX)
find_library(AUDIOTOOLBOX AudioToolbox)
list(APPEND EXTRA_LIBS ${AUDIOTOOLBOX})
mac: Patched to compile with CMake project.
2016-09-04 13:42:46 +08:00
endif()
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
if(SDL_FRAMEWORK_AVFOUNDATION)
find_library(AVFOUNDATION AVFoundation)
list(APPEND EXTRA_LIBS ${AVFOUNDATION})
endif()
if(SDL_FRAMEWORK_COREBLUETOOTH)
find_library(COREBLUETOOTH CoreBluetooth)
list(APPEND EXTRA_LIBS ${COREBLUETOOTH})
endif()
if(SDL_FRAMEWORK_COREGRAPHICS)
find_library(COREGRAPHICS CoreGraphics)
list(APPEND EXTRA_LIBS ${COREGRAPHICS})
endif()
if(SDL_FRAMEWORK_COREMOTION)
find_library(COREMOTION CoreMotion)
list(APPEND EXTRA_LIBS ${COREMOTION})
endif()
if(SDL_FRAMEWORK_FOUNDATION)
find_library(FOUNDATION Foundation)
list(APPEND EXTRA_LIBS ${FOUNDATION})
endif()
if(SDL_FRAMEWORK_GAMECONTROLLER)
find_library(GAMECONTROLLER GameController)
list(APPEND EXTRA_LIBS ${GAMECONTROLLER})
endif()
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
if(SDL_FRAMEWORK_METAL)
CMake: tvOS support/fixes To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=tvOS - CMAKE_OSX_SYSROOT=<SDK> (examples: appletvos, appletvsimulator, appletvos12.4, /full/path/to/AppleTVOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;x86_64")
2019-08-28 00:30:20 +08:00
if(IOS OR TVOS)
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
find_library(METAL Metal)
list(APPEND EXTRA_LIBS ${METAL})
Fixed bug 3268 - CMAKE: ios fixes Martin Gerhardy ios is not properly handled in the cmake file. this patch is a start to do so - it properly builds for me with it.
2016-10-02 03:54:48 +08:00
else()
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
list(APPEND EXTRA_LDFLAGS "-Wl,-weak_framework,Metal")
endif()
endif()
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
if(SDL_FRAMEWORK_OPENGLES)
find_library(OPENGLES OpenGLES)
list(APPEND EXTRA_LIBS ${OPENGLES})
endif()
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
if(SDL_FRAMEWORK_QUARTZCORE)
CMake: tvOS support/fixes To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=tvOS - CMAKE_OSX_SYSROOT=<SDK> (examples: appletvos, appletvsimulator, appletvos12.4, /full/path/to/AppleTVOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;x86_64")
2019-08-28 00:30:20 +08:00
if(IOS OR TVOS)
Add public APIs for creating a Metal view attached to an SDL window. Add SDL_metal.h.
2019-08-05 23:35:32 +08:00
find_library(QUARTZCORE QuartzCore)
list(APPEND EXTRA_LIBS ${QUARTZCORE})
else()
list(APPEND EXTRA_LDFLAGS "-Wl,-weak_framework,QuartzCore")
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
endif()
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
if(SDL_FRAMEWORK_UIKIT)
find_library(UIKIT UIKit)
list(APPEND EXTRA_LIBS ${UIKIT})
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
CheckPTHREAD()
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
elseif(HAIKU)
Removed SDL_AndroidOpenURL, added SDL_OpenURL. Still needs to be wired into Xcode and Visual Studio projects.
2020-10-05 23:30:33 +08:00
file(GLOB MISC_SOURCES ${SDL2_SOURCE_DIR}/src/misc/haiku/*.cc)
set(SOURCE_FILES ${SOURCE_FILES} ${MISC_SOURCES})
set(HAVE_SDL_MISC TRUE)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_VIDEO)
set(SDL_VIDEO_DRIVER_HAIKU 1)
file(GLOB HAIKUVIDEO_SOURCES ${SDL2_SOURCE_DIR}/src/video/haiku/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${HAIKUVIDEO_SOURCES})
set(HAVE_SDL_VIDEO TRUE)
if(VIDEO_OPENGL)
# TODO: Use FIND_PACKAGE(OpenGL) instead
set(SDL_VIDEO_OPENGL 1)
set(SDL_VIDEO_OPENGL_BGL 1)
set(SDL_VIDEO_RENDER_OGL 1)
list(APPEND EXTRA_LIBS GL)
set(HAVE_VIDEO_OPENGL TRUE)
endif()
endif()
cmake: various Haiku settings were accidentally under "if(SDL_VIDEO)"
2020-04-24 14:06:53 +08:00
set(SDL_FILESYSTEM_HAIKU 1)
file(GLOB FILESYSTEM_SOURCES ${SDL2_SOURCE_DIR}/src/filesystem/haiku/*.cc)
set(SOURCE_FILES ${SOURCE_FILES} ${FILESYSTEM_SOURCES})
set(HAVE_SDL_FILESYSTEM TRUE)
if(SDL_TIMERS)
set(SDL_TIMER_HAIKU 1)
file(GLOB TIMER_SOURCES ${SDL2_SOURCE_DIR}/src/timer/haiku/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${TIMER_SOURCES})
set(HAVE_SDL_TIMERS TRUE)
endif()
locale: Implemented SDL_GetPreferredLocales(). This was something I proposed a long time ago, Sylvain Becker did additional work on it, then back to me. Fixes Bugzilla #2131.
2020-05-04 14:27:29 +08:00
if(SDL_LOCALE)
file(GLOB LOCALE_SOURCES ${SDL2_SOURCE_DIR}/src/locale/haiku/*.cc)
set(SOURCE_FILES ${SOURCE_FILES} ${LOCALE_SOURCES})
set(HAVE_SDL_LOCALE TRUE)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
CheckPTHREAD()
Add basic support for compiling on RISC OS
2020-02-14 04:50:47 +08:00
elseif(RISCOS)
riscos: Implement SDL_OpenURL()
2020-10-12 00:32:32 +08:00
file(GLOB MISC_SOURCES ${SDL2_SOURCE_DIR}/src/misc/riscos/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${MISC_SOURCES})
set(HAVE_SDL_MISC TRUE)
Add basic support for compiling on RISC OS
2020-02-14 04:50:47 +08:00
if(SDL_TIMERS)
set(SDL_TIMER_UNIX 1)
file(GLOB TIMER_SOURCES ${SDL2_SOURCE_DIR}/src/timer/unix/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${TIMER_SOURCES})
set(HAVE_SDL_TIMERS TRUE)
if(CLOCK_GETTIME)
set(HAVE_CLOCK_GETTIME 1)
endif()
endif()
CheckPTHREAD()
if(SDL_AUDIO)
CheckOSS()
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
We removed QuartzCore link dependency in commit 5be6badaf7e1
2017-08-29 13:42:41 +08:00
if(VIDEO_VULKAN)
Be clear that disabling Vulkan surface support disables the entire SDL Vulkan integration
2017-08-28 15:11:38 +08:00
set(SDL_VIDEO_VULKAN 1)
cmake: correctly report Vulkan support at configure time (thanks, Tiago!). Fixes Bugzilla #4262.
2018-09-18 23:49:42 +08:00
set(HAVE_VIDEO_VULKAN TRUE)
vulkan: Initial Vulkan support! This work was done by Jacob Lifshay and Mark Callow; I'm just merging it into revision control.
2017-08-28 10:15:57 +08:00
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# Dummies
configure.in: Rename to configure.ac to fix an 'aclocal' warning Also rename references in related files.
2019-03-25 20:01:32 +08:00
# configure.ac does it differently:
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# if not have X
# if enable_X { SDL_X_DISABLED = 1 }
# [add dummy sources]
# so it always adds a dummy, without checking, if it was actually requested.
# This leads to missing internal references on building, since the
# src/X/*.c does not get included.
if(NOT HAVE_SDL_JOYSTICK)
Mark a subsystem as dummy, not disabled, if it was intended to be compiled in. From Tom Black: I'm having problems initializing the sensor module. I'm compiling with a standard ./configure && make && sudo make install, and the module says it's enabled, but SDL_Init(SDL_INIT_EVERYTHING) is failing with SDL_GetError() returning "SDL not built with sensor support".
2018-09-29 12:19:27 +08:00
set(SDL_JOYSTICK_DUMMY 1)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_JOYSTICK AND NOT APPLE) # results in unresolved symbols on OSX
file(GLOB JOYSTICK_SOURCES ${SDL2_SOURCE_DIR}/src/joystick/dummy/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${JOYSTICK_SOURCES})
endif()
endif()
if(NOT HAVE_SDL_HAPTIC)
Mark a subsystem as dummy, not disabled, if it was intended to be compiled in. From Tom Black: I'm having problems initializing the sensor module. I'm compiling with a standard ./configure && make && sudo make install, and the module says it's enabled, but SDL_Init(SDL_INIT_EVERYTHING) is failing with SDL_GetError() returning "SDL not built with sensor support".
2018-09-29 12:19:27 +08:00
set(SDL_HAPTIC_DUMMY 1)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
file(GLOB HAPTIC_SOURCES ${SDL2_SOURCE_DIR}/src/haptic/dummy/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${HAPTIC_SOURCES})
endif()
Mark a subsystem as dummy, not disabled, if it was intended to be compiled in. From Tom Black: I'm having problems initializing the sensor module. I'm compiling with a standard ./configure && make && sudo make install, and the module says it's enabled, but SDL_Init(SDL_INIT_EVERYTHING) is failing with SDL_GetError() returning "SDL not built with sensor support".
2018-09-29 12:19:27 +08:00
if(NOT HAVE_SDL_SENSORS)
set(SDL_SENSOR_DUMMY 1)
file(GLOB SENSORS_SOURCES ${SDL2_SOURCE_DIR}/src/sensor/dummy/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${SENSORS_SOURCES})
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(NOT HAVE_SDL_LOADSO)
set(SDL_LOADSO_DISABLED 1)
file(GLOB LOADSO_SOURCES ${SDL2_SOURCE_DIR}/src/loadso/dummy/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${LOADSO_SOURCES})
endif()
if(NOT HAVE_SDL_FILESYSTEM)
set(SDL_FILESYSTEM_DISABLED 1)
file(GLOB FILESYSTEM_SOURCES ${SDL2_SOURCE_DIR}/src/filesystem/dummy/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${FILESYSTEM_SOURCES})
endif()
locale: Implemented SDL_GetPreferredLocales(). This was something I proposed a long time ago, Sylvain Becker did additional work on it, then back to me. Fixes Bugzilla #2131.
2020-05-04 14:27:29 +08:00
if(NOT HAVE_SDL_LOCALE)
set(SDL_LOCALE_DISABLED 1)
file(GLOB LOCALE_SOURCES ${SDL2_SOURCE_DIR}/src/locale/dummy/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${LOCALE_SOURCES})
endif()
Removed SDL_AndroidOpenURL, added SDL_OpenURL. Still needs to be wired into Xcode and Visual Studio projects.
2020-10-05 23:30:33 +08:00
if(NOT HAVE_SDL_MISC)
set(SDL_MISC_DISABLED 1)
file(GLOB LOCALE_SOURCES ${SDL2_SOURCE_DIR}/src/misc/dummy/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${LOCALE_SOURCES})
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# We always need to have threads and timers around
if(NOT HAVE_SDL_THREADS)
set(SDL_THREADS_DISABLED 1)
file(GLOB THREADS_SOURCES ${SDL2_SOURCE_DIR}/src/thread/generic/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${THREADS_SOURCES})
endif()
if(NOT HAVE_SDL_TIMERS)
set(SDL_TIMERS_DISABLED 1)
file(GLOB TIMER_SOURCES ${SDL2_SOURCE_DIR}/src/timer/dummy/*.c)
set(SOURCE_FILES ${SOURCE_FILES} ${TIMER_SOURCES})
endif()
if(NOT SDLMAIN_SOURCES)
file(GLOB SDLMAIN_SOURCES ${SDL2_SOURCE_DIR}/src/main/dummy/*.c)
endif()
# Append the -MMD -MT flags
# if(DEPENDENCY_TRACKING)
# if(COMPILER_IS_GNUCC)
# set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -MMD -MT \$@")
# endif()
# endif()
configure_file("${SDL2_SOURCE_DIR}/include/SDL_config.h.cmake"
"${SDL2_BINARY_DIR}/include/SDL_config.h")
# Prepare the flags and remove duplicates
if(EXTRA_LDFLAGS)
list(REMOVE_DUPLICATES EXTRA_LDFLAGS)
endif()
if(EXTRA_LIBS)
list(REMOVE_DUPLICATES EXTRA_LIBS)
endif()
if(EXTRA_CFLAGS)
list(REMOVE_DUPLICATES EXTRA_CFLAGS)
endif()
listtostr(EXTRA_CFLAGS _EXTRA_CFLAGS)
set(EXTRA_CFLAGS ${_EXTRA_CFLAGS})
# Compat helpers for the configuration files
cmake: run updaterev.sh if(NOT CMAKE_HOST_WIN32) ... not if(NOT WINDOWS OR CYGWIN) This way, it generates SDL_revision.h in cross-build environments too.
2020-10-13 19:45:10 +08:00
if(NOT CMAKE_HOST_WIN32)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# TODO: we need a Windows script, too
Fixed bug 5250 - updaterev.sh failed using CMake Tools on VSCode Remote Sebastian Vargas Vargas Running CMake configure from a Windows Subsystem for Linux using Visual Studio Code Remote doesn't generate the header file with the current source revision, it throws "/home/sebva/SDL/build-scripts/updaterev.sh: 13: cannot create /mnt/c/Users/sebva/.vscode/extensions/ms-vscode-remote.remote-wsl-0.44.4/include/SDL_revision.h.new: Directory nonexistent".
2020-12-09 22:56:34 +08:00
execute_process(COMMAND sh ${SDL2_SOURCE_DIR}/build-scripts/updaterev.sh
WORKING_DIRECTORY ${SDL2_BINARY_DIR})
cmake: run updaterev.sh if(NOT CMAKE_HOST_WIN32) ... not if(NOT WINDOWS OR CYGWIN) This way, it generates SDL_revision.h in cross-build environments too.
2020-10-13 19:45:10 +08:00
endif()
if(NOT WINDOWS OR CYGWIN)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(prefix ${CMAKE_INSTALL_PREFIX})
set(exec_prefix "\${prefix}")
set(libdir "\${exec_prefix}/lib${LIB_SUFFIX}")
set(bindir "\${exec_prefix}/bin")
set(includedir "\${prefix}/include")
if(SDL_STATIC)
set(ENABLE_STATIC_TRUE "")
set(ENABLE_STATIC_FALSE "#")
else()
set(ENABLE_STATIC_TRUE "#")
set(ENABLE_STATIC_FALSE "")
endif()
if(SDL_SHARED)
build: Merge pkg-config Libs.private into Libs for static-only builds A project being built entirely statically will call pkg-config with --static, which utilises the Libs.private field. Conversely it will not use --static when not being built entirely statically, even if there is only a static build of SDL available. This will most likely cause the build to fail due to underlinking unless we merge the Libs fields. This is what the Meson build system does when it generates pkg-config files. This also also follows the behaviour of sdl2-config. At the same time, the runtime linker flags are not applicable to static-only builds so only add them for shared builds.
2020-04-12 06:38:34 +08:00
set(PKG_CONFIG_LIBS_PRIV "
Libs.private:")
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(ENABLE_SHARED_TRUE "")
set(ENABLE_SHARED_FALSE "#")
else()
build: Merge pkg-config Libs.private into Libs for static-only builds A project being built entirely statically will call pkg-config with --static, which utilises the Libs.private field. Conversely it will not use --static when not being built entirely statically, even if there is only a static build of SDL available. This will most likely cause the build to fail due to underlinking unless we merge the Libs fields. This is what the Meson build system does when it generates pkg-config files. This also also follows the behaviour of sdl2-config. At the same time, the runtime linker flags are not applicable to static-only builds so only add them for shared builds.
2020-04-12 06:38:34 +08:00
set(PKG_CONFIG_LIBS_PRIV "")
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(ENABLE_SHARED_TRUE "#")
set(ENABLE_SHARED_FALSE "")
endif()
# Clean up the different lists
listtostr(EXTRA_LIBS _EXTRA_LIBS "-l")
build: Don't duplicate Libs in Libs.private in pkg-config file pkg-config already prepends Libs to Libs.private when you specify --static so there's no need to duplicate them. Most other projects don't do this.
2020-04-12 20:24:36 +08:00
set(SDL_STATIC_LIBS ${EXTRA_LDFLAGS} ${_EXTRA_LIBS})
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
list(REMOVE_DUPLICATES SDL_STATIC_LIBS)
listtostr(SDL_STATIC_LIBS _SDL_STATIC_LIBS)
set(SDL_STATIC_LIBS ${_SDL_STATIC_LIBS})
listtostr(SDL_LIBS _SDL_LIBS)
set(SDL_LIBS ${_SDL_LIBS})
# MESSAGE(STATUS "SDL_LIBS: ${SDL_LIBS}")
# MESSAGE(STATUS "SDL_STATIC_LIBS: ${SDL_STATIC_LIBS}")
configure_file("${SDL2_SOURCE_DIR}/sdl2.pc.in"
"${SDL2_BINARY_DIR}/sdl2.pc" @ONLY)
configure_file("${SDL2_SOURCE_DIR}/sdl2-config.in"
"${SDL2_BINARY_DIR}/sdl2-config")
configure_file("${SDL2_SOURCE_DIR}/sdl2-config.in"
"${SDL2_BINARY_DIR}/sdl2-config" @ONLY)
configure_file("${SDL2_SOURCE_DIR}/SDL2.spec.in"
"${SDL2_BINARY_DIR}/SDL2.spec" @ONLY)
endif()
##### Info output #####
message(STATUS "")
message(STATUS "SDL2 was configured with the following options:")
message(STATUS "")
message(STATUS "Platform: ${CMAKE_SYSTEM}")
message(STATUS "64-bit: ${ARCH_64}")
message(STATUS "Compiler: ${CMAKE_C_COMPILER}")
message(STATUS "")
message(STATUS "Subsystems:")
foreach(_SUB ${SDL_SUBSYSTEMS})
string(TOUPPER ${_SUB} _OPT)
message_bool_option(${_SUB} SDL_${_OPT})
endforeach()
message(STATUS "")
message(STATUS "Options:")
list(SORT ALLOPTIONS)
foreach(_OPT ${ALLOPTIONS})
# Longest option is VIDEO_X11_XSCREENSAVER = 22 characters
# Get the padding
string(LENGTH ${_OPT} _OPTLEN)
math(EXPR _PADLEN "23 - ${_OPTLEN}")
string(RANDOM LENGTH ${_PADLEN} ALPHABET " " _PADDING)
message_tested_option(${_OPT} ${_PADDING})
endforeach()
message(STATUS "")
message(STATUS " CFLAGS: ${CMAKE_C_FLAGS}")
message(STATUS " EXTRA_CFLAGS: ${EXTRA_CFLAGS}")
message(STATUS " EXTRA_LDFLAGS: ${EXTRA_LDFLAGS}")
message(STATUS " EXTRA_LIBS: ${EXTRA_LIBS}")
message(STATUS "")
message(STATUS " Build Shared Library: ${SDL_SHARED}")
message(STATUS " Build Static Library: ${SDL_STATIC}")
CMake: Let static libraries optionally be built with Position Independent Code.
2016-01-05 14:26:32 +08:00
if(SDL_STATIC)
message(STATUS " Build Static Library with Position Independent Code: ${SDL_STATIC_PIC}")
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
message(STATUS "")
if(UNIX)
message(STATUS "If something was not detected, although the libraries")
message(STATUS "were installed, then make sure you have set the")
message(STATUS "CFLAGS and LDFLAGS environment variables correctly.")
message(STATUS "")
endif()
cmake: added support for enabling the ARM SIMD/NEON code.
2019-10-25 11:17:19 +08:00
if(WARN_ABOUT_ARM_SIMD_ASM_MIT)
message(STATUS "")
message(STATUS "SDL is being built with ARM SIMD optimizations, which")
message(STATUS "uses code licensed under the MIT license. If this is a")
message(STATUS "problem, please disable that code by rerunning CMake with:")
message(STATUS "")
message(STATUS " -DARMSIMD=OFF")
endif()
if(WARN_ABOUT_ARM_NEON_ASM_MIT)
message(STATUS "")
message(STATUS "SDL is being built with ARM NEON optimizations, which")
message(STATUS "uses code licensed under the MIT license. If this is a")
message(STATUS "problem, please disable that code by rerunning CMake with:")
message(STATUS "")
message(STATUS " -DARMNEON=OFF")
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
# Ensure that the extra cflags are used at compile time
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${EXTRA_CFLAGS}")
# Always build SDLmain
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
if(NOT WINDOWS_STORE)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
add_library(SDL2main STATIC ${SDLMAIN_SOURCES})
Further improvements for bug 4128 - CMAKE: Generated target import file contains incorrect include path Mohamed It would be useful to be able to do either `#include "SDL2/SDL.h"` or `#include "SDL.h"`. This patch allows that and adds compatibility with other build systems. It also allows differentiating between SDL1 and SDL2.
2020-02-16 16:08:36 +08:00
target_include_directories(SDL2main PUBLIC "$<BUILD_INTERFACE:${SDL2_SOURCE_DIR}/include>" $<INSTALL_INTERFACE:include> $<INSTALL_INTERFACE:include/SDL2>)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set(_INSTALL_LIBS "SDL2main")
cmake: Set debug library name suffix per target instead of setting it globally. Don't add a suffix on android
2018-02-26 01:22:47 +08:00
if (NOT ANDROID)
cmake: ${SDL_CMAKE_DEBUG_POSTFIX} needs to be wrapped in quotes. Otherwise, if it's an empty string, it causes problems. Fixes Bugzilla #4117.
2019-12-03 15:42:53 +08:00
set_target_properties(SDL2main PROPERTIES DEBUG_POSTFIX "${SDL_CMAKE_DEBUG_POSTFIX}")
cmake: Set debug library name suffix per target instead of setting it globally. Don't add a suffix on android
2018-02-26 01:22:47 +08:00
endif()
Accepted patch https://github.com/microsoft/vcpkg/blob/master/ports/sdl2/enable-winrt-cmake.patch
2020-12-09 01:13:08 +08:00
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
Fixed bug 3918 - HIDAPI, CMake support for android project
2019-08-27 17:38:43 +08:00
if (ANDROID AND HAVE_HIDAPI)
set(_INSTALL_LIBS ${_INSTALL_LIBS} "hidapi")
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_SHARED)
Fixed setting the version info in the shared library when using CMake Ray Molenkamp When building sdl as shared lib, the version info is lacking in the final binary. CMake gathers the right resource files into ${VERSION_SOURCES} but then doesn't do anything with them.
2016-10-03 18:42:10 +08:00
add_library(SDL2 SHARED ${SOURCE_FILES} ${VERSION_SOURCES})
cmake: Pacify warning about Policy CMP0042 not being set.
2017-08-10 10:43:16 +08:00
if(APPLE)
Correct the name of the SDL shared library in CMake for Mac OS
2018-08-19 05:23:40 +08:00
set_target_properties(SDL2 PROPERTIES
MACOSX_RPATH 1
OUTPUT_NAME "SDL2-${LT_RELEASE}")
cmake: Pacify warning about Policy CMP0042 not being set.
2017-08-10 10:43:16 +08:00
elseif(UNIX AND NOT ANDROID)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set_target_properties(SDL2 PROPERTIES
VERSION ${LT_VERSION}
SOVERSION ${LT_REVISION}
OUTPUT_NAME "SDL2-${LT_RELEASE}")
else()
fix bug #5384 -- define DLL_EXPORT in DLL builds and adjust begin_code.h
2020-12-09 17:01:10 +08:00
if(WINDOWS OR CYGWIN)
set_target_properties(SDL2 PROPERTIES
DEFINE_SYMBOL DLL_EXPORT)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set_target_properties(SDL2 PROPERTIES
VERSION ${SDL_VERSION}
SOVERSION ${LT_REVISION}
OUTPUT_NAME "SDL2")
endif()
cmake: don't use /NODEFAULTLIB if we are using the C runtime (thanks, Rob!). Fixes Bugzilla #3640.
2017-06-07 01:39:29 +08:00
if(MSVC AND NOT LIBC)
CMake: Fixed building for Windows with VS2015 (bug #3080). - Don't try to link with the Visual C runtime. - Avoid code generation that would use functions from the VC runtime.
2016-01-01 03:26:40 +08:00
# Don't try to link with the default set of libraries.
set_target_properties(SDL2 PROPERTIES LINK_FLAGS_RELEASE "/NODEFAULTLIB")
set_target_properties(SDL2 PROPERTIES LINK_FLAGS_DEBUG "/NODEFAULTLIB")
set_target_properties(SDL2 PROPERTIES STATIC_LIBRARY_FLAGS "/NODEFAULTLIB")
endif()
set(_INSTALL_LIBS "SDL2" ${_INSTALL_LIBS})
cmake: Bump minimum required CMake to 3.0.0 and mark link libraries PRIVATE. Fixes Bugzilla #2992.
2020-06-27 08:11:32 +08:00
target_link_libraries(SDL2 PRIVATE ${EXTRA_LIBS} ${EXTRA_LDFLAGS})
Further improvements for bug 4128 - CMAKE: Generated target import file contains incorrect include path Mohamed It would be useful to be able to do either `#include "SDL2/SDL.h"` or `#include "SDL.h"`. This patch allows that and adds compatibility with other build systems. It also allows differentiating between SDL1 and SDL2.
2020-02-16 16:08:36 +08:00
target_include_directories(SDL2 PUBLIC "$<BUILD_INTERFACE:${SDL2_SOURCE_DIR}/include>" $<INSTALL_INTERFACE:include> $<INSTALL_INTERFACE:include/SDL2>)
Fixed bug 5100 - compilation CMake Android armeabi-v7a (Thanks Steve Robinson!) fatal error: 'cpu-features.h' file not found on CMake Android armeabi-v7a
2020-05-16 03:12:23 +08:00
if(ANDROID)
target_include_directories(SDL2 PRIVATE ${ANDROID_NDK}/sources/android/cpufeatures)
else()
cmake: ${SDL_CMAKE_DEBUG_POSTFIX} needs to be wrapped in quotes. Otherwise, if it's an empty string, it causes problems. Fixes Bugzilla #4117.
2019-12-03 15:42:53 +08:00
set_target_properties(SDL2 PROPERTIES DEBUG_POSTFIX "${SDL_CMAKE_DEBUG_POSTFIX}")
cmake: Set debug library name suffix per target instead of setting it globally. Don't add a suffix on android
2018-02-26 01:22:47 +08:00
endif()
CMake: tvOS support/fixes To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=tvOS - CMAKE_OSX_SYSROOT=<SDK> (examples: appletvos, appletvsimulator, appletvos12.4, /full/path/to/AppleTVOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;x86_64")
2019-08-28 00:30:20 +08:00
if(IOS OR TVOS)
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
set_property(TARGET SDL2 APPEND_STRING PROPERTY COMPILE_FLAGS "-fobjc-arc")
target_compile_definitions(SDL2 PRIVATE IOS_DYLIB=1)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
Fixed bug 4639 - CMake build does not generate libhidapi.so for Android Manuel Sabogal I noticed that the current Android.mk builds a libhidapi.so library for Android but the CMake build hasn't been updated to do so. I'll attach a patch that fixes this issue.
2019-05-22 08:33:31 +08:00
if(ANDROID)
cmake: Added HIDAPI support.
2019-07-09 04:46:52 +08:00
if(HAVE_HIDAPI)
add_library(hidapi SHARED ${SDL2_SOURCE_DIR}/src/hidapi/android/hid.cpp)
endif()
Fixed bug 4639 - CMake build does not generate libhidapi.so for Android Manuel Sabogal I noticed that the current Android.mk builds a libhidapi.so library for Android but the CMake build hasn't been updated to do so. I'll attach a patch that fixes this issue.
2019-05-22 08:33:31 +08:00
if(MSVC AND NOT LIBC)
# Don't try to link with the default set of libraries.
set_target_properties(hidapi PROPERTIES LINK_FLAGS_RELEASE "/NODEFAULTLIB")
set_target_properties(hidapi PROPERTIES LINK_FLAGS_DEBUG "/NODEFAULTLIB")
set_target_properties(hidapi PROPERTIES STATIC_LIBRARY_FLAGS "/NODEFAULTLIB")
endif()
Fixed bug 3918 - CMake support for android project
2019-08-11 21:23:37 +08:00
if(HAVE_HIDAPI)
target_link_libraries(hidapi log)
endif()
Fixed bug 4639 - CMake build does not generate libhidapi.so for Android Manuel Sabogal I noticed that the current Android.mk builds a libhidapi.so library for Android but the CMake build hasn't been updated to do so. I'll attach a patch that fixes this issue.
2019-05-22 08:33:31 +08:00
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_STATIC)
set (BUILD_SHARED_LIBS FALSE)
add_library(SDL2-static STATIC ${SOURCE_FILES})
Fixed bug 3651 - CMake build does not install CMake package configuration tschwinger@elitemail.org Most ironically, although autoconf/automake-based builds install (pretty half-assed) CMake package configuration files, they're missing in installations resulting from CMake-based builds entirely. A proper configuration file typically also loads target exports (implemented in patch 3572, also fixing this issue - see my comment on that issue for details). I believe it would be best to let the dinosaurs go extinct and redirect all build efforts to the CMake end for two reasons: 1. It potentially provides the best user experience, but you'd have to give it some love and ship with less quirky buildfiles. 2. It would force distros to build SDL via CMake and thus would ensure target exports are actually available everywhere. Various CMake patches I submitted today in summary (directly converted from the HG commits and `am`d onto a fork of a git mirror that happened to be on `tip`). https://github.com/tschw/SDL/commits/patched Fixing #2576 #3572, #3613, and this fresh ticket, which is almost entirely advertisement ;). These already do to make SDL much less of a quirky fella to have in your dependency tree...
2017-08-10 10:03:10 +08:00
if (NOT SDL_SHARED OR NOT WIN32)
set_target_properties(SDL2-static PROPERTIES OUTPUT_NAME "SDL2")
# Note: Apparently, OUTPUT_NAME must really be unique; even when
# CMAKE_IMPORT_LIBRARY_SUFFIX or the like are given. Otherwise
# the static build may race with the import lib and one will get
# clobbered, when the suffix is realized via subsequent rename.
endif()
CMake: Let static libraries optionally be built with Position Independent Code.
2016-01-05 14:26:32 +08:00
set_target_properties(SDL2-static PROPERTIES POSITION_INDEPENDENT_CODE ${SDL_STATIC_PIC})
cmake: don't use /NODEFAULTLIB if we are using the C runtime (thanks, Rob!). Fixes Bugzilla #3640.
2017-06-07 01:39:29 +08:00
if(MSVC AND NOT LIBC)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
set_target_properties(SDL2-static PROPERTIES LINK_FLAGS_RELEASE "/NODEFAULTLIB")
set_target_properties(SDL2-static PROPERTIES LINK_FLAGS_DEBUG "/NODEFAULTLIB")
set_target_properties(SDL2-static PROPERTIES STATIC_LIBRARY_FLAGS "/NODEFAULTLIB")
endif()
# TODO: Win32 platforms keep the same suffix .lib for import and static
# libraries - do we need to consider this?
set(_INSTALL_LIBS "SDL2-static" ${_INSTALL_LIBS})
target_link_libraries(SDL2-static ${EXTRA_LIBS} ${EXTRA_LDFLAGS})
Further improvements for bug 4128 - CMAKE: Generated target import file contains incorrect include path Mohamed It would be useful to be able to do either `#include "SDL2/SDL.h"` or `#include "SDL.h"`. This patch allows that and adds compatibility with other build systems. It also allows differentiating between SDL1 and SDL2.
2020-02-16 16:08:36 +08:00
target_include_directories(SDL2-static PUBLIC "$<BUILD_INTERFACE:${SDL2_SOURCE_DIR}/include>" $<INSTALL_INTERFACE:include> $<INSTALL_INTERFACE:include/SDL2>)
Fixed bug 5100 - compilation CMake Android armeabi-v7a (Thanks Steve Robinson!) fatal error: 'cpu-features.h' file not found on CMake Android armeabi-v7a
2020-05-16 03:12:23 +08:00
if(ANDROID)
target_include_directories(SDL2-static PRIVATE ${ANDROID_NDK}/sources/android/cpufeatures)
else()
cmake: ${SDL_CMAKE_DEBUG_POSTFIX} needs to be wrapped in quotes. Otherwise, if it's an empty string, it causes problems. Fixes Bugzilla #4117.
2019-12-03 15:42:53 +08:00
set_target_properties(SDL2-static PROPERTIES DEBUG_POSTFIX "${SDL_CMAKE_DEBUG_POSTFIX}")
cmake: Set debug library name suffix per target instead of setting it globally. Don't add a suffix on android
2018-02-26 01:22:47 +08:00
endif()
CMake: tvOS support/fixes To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=tvOS - CMAKE_OSX_SYSROOT=<SDK> (examples: appletvos, appletvsimulator, appletvos12.4, /full/path/to/AppleTVOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;x86_64")
2019-08-28 00:30:20 +08:00
if(IOS OR TVOS)
CMake: iOS support added When using a recent version of CMake (3.14+), this should make it possible to: - build SDL for iOS, both static and dynamic - build SDL test apps (as iOS .app bundles) - generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis) To use, set the following CMake variables when running CMake's configuration stage: - CMAKE_SYSTEM_NAME=iOS - CMAKE_OSX_SYSROOT=<SDK> (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.) - CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s") Examples: - for Simulator, using the latest, installed SDK: cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64 - for Device, using the latest, installed SDK, 64-bit only cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64 - for Device, using the latest, installed SDK, mixed 32/64 bit cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s" - for Device, using a specific SDK revision (iOS 12.4, in this example): cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64 - for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles): cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 23:07:43 +08:00
set_property(TARGET SDL2-static APPEND_STRING PROPERTY COMPILE_FLAGS "-fobjc-arc")
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
Fixed bug 4092 - CMake support for building everything in the "test" directory Eric Wasylishen Patch to support building the tests with cmake. Disabled by default, use: "cmake .. -DSDL_TEST=YES" to enable the tests. Tested on macOS 10.13 with the ninja, makefile, and Xcode generators, and Windows 10 with the Visual Studio 2017 generator.
2018-02-25 00:59:58 +08:00
##### Tests #####
if(SDL_TEST)
file(GLOB TEST_SOURCES ${SDL2_SOURCE_DIR}/src/test/*.c)
add_library(SDL2_test STATIC ${TEST_SOURCES})
Added debug postfix to install command and fixed library path
2018-04-10 23:03:54 +08:00
Fixed bug 4092 - CMake support for building everything in the "test" directory Eric Wasylishen Patch to support building the tests with cmake. Disabled by default, use: "cmake .. -DSDL_TEST=YES" to enable the tests. Tested on macOS 10.13 with the ninja, makefile, and Xcode generators, and Windows 10 with the Visual Studio 2017 generator.
2018-02-25 00:59:58 +08:00
add_subdirectory(test)
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
##### Installation targets #####
Fixed bug 3651 - CMake build does not install CMake package configuration tschwinger@elitemail.org Most ironically, although autoconf/automake-based builds install (pretty half-assed) CMake package configuration files, they're missing in installations resulting from CMake-based builds entirely. A proper configuration file typically also loads target exports (implemented in patch 3572, also fixing this issue - see my comment on that issue for details). I believe it would be best to let the dinosaurs go extinct and redirect all build efforts to the CMake end for two reasons: 1. It potentially provides the best user experience, but you'd have to give it some love and ship with less quirky buildfiles. 2. It would force distros to build SDL via CMake and thus would ensure target exports are actually available everywhere. Various CMake patches I submitted today in summary (directly converted from the HG commits and `am`d onto a fork of a git mirror that happened to be on `tip`). https://github.com/tschw/SDL/commits/patched Fixing #2576 #3572, #3613, and this fresh ticket, which is almost entirely advertisement ;). These already do to make SDL much less of a quirky fella to have in your dependency tree...
2017-08-10 10:03:10 +08:00
install(TARGETS ${_INSTALL_LIBS} EXPORT SDL2Targets
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
LIBRARY DESTINATION "lib${LIB_SUFFIX}"
ARCHIVE DESTINATION "lib${LIB_SUFFIX}"
RUNTIME DESTINATION bin)
Fixed bug 3651 - CMake build does not install CMake package configuration tschwinger@elitemail.org Most ironically, although autoconf/automake-based builds install (pretty half-assed) CMake package configuration files, they're missing in installations resulting from CMake-based builds entirely. A proper configuration file typically also loads target exports (implemented in patch 3572, also fixing this issue - see my comment on that issue for details). I believe it would be best to let the dinosaurs go extinct and redirect all build efforts to the CMake end for two reasons: 1. It potentially provides the best user experience, but you'd have to give it some love and ship with less quirky buildfiles. 2. It would force distros to build SDL via CMake and thus would ensure target exports are actually available everywhere. Various CMake patches I submitted today in summary (directly converted from the HG commits and `am`d onto a fork of a git mirror that happened to be on `tip`). https://github.com/tschw/SDL/commits/patched Fixing #2576 #3572, #3613, and this fresh ticket, which is almost entirely advertisement ;). These already do to make SDL much less of a quirky fella to have in your dependency tree...
2017-08-10 10:03:10 +08:00
##### Export files #####
Fixed bug 4594 - Fix install location of CMake targets on Apple platforms tschwinger Followup to #3651 As already noted by Ryan, no framework is being built, so we better install to lib/cmake. That code was originally part of a patch submitted by David Demelier, whose credit BTW got lost (I combined his patch for #3572 with fixes for #2576 and #3613 resulting in #3651 because things started to depend on another). I tested that the configuration files are found correctly in the new location on MacOS X based on a hint to the root (see https://cmake.org/cmake/help/latest/command/find_package.html#search-procedure).
2019-06-09 05:34:38 +08:00
if (WINDOWS)
Fixed bug 3651 - CMake build does not install CMake package configuration tschwinger@elitemail.org Most ironically, although autoconf/automake-based builds install (pretty half-assed) CMake package configuration files, they're missing in installations resulting from CMake-based builds entirely. A proper configuration file typically also loads target exports (implemented in patch 3572, also fixing this issue - see my comment on that issue for details). I believe it would be best to let the dinosaurs go extinct and redirect all build efforts to the CMake end for two reasons: 1. It potentially provides the best user experience, but you'd have to give it some love and ship with less quirky buildfiles. 2. It would force distros to build SDL via CMake and thus would ensure target exports are actually available everywhere. Various CMake patches I submitted today in summary (directly converted from the HG commits and `am`d onto a fork of a git mirror that happened to be on `tip`). https://github.com/tschw/SDL/commits/patched Fixing #2576 #3572, #3613, and this fresh ticket, which is almost entirely advertisement ;). These already do to make SDL much less of a quirky fella to have in your dependency tree...
2017-08-10 10:03:10 +08:00
set(PKG_PREFIX "cmake")
else ()
Fixed bug 4928 - CMakeLists.txt: put cmake package helpers in proper libdir Tiago O. Use LIB_SUFFIX variable to determine the correct path.
2020-01-06 23:06:58 +08:00
set(PKG_PREFIX "lib${LIB_SUFFIX}/cmake/SDL2")
Fixed bug 3651 - CMake build does not install CMake package configuration tschwinger@elitemail.org Most ironically, although autoconf/automake-based builds install (pretty half-assed) CMake package configuration files, they're missing in installations resulting from CMake-based builds entirely. A proper configuration file typically also loads target exports (implemented in patch 3572, also fixing this issue - see my comment on that issue for details). I believe it would be best to let the dinosaurs go extinct and redirect all build efforts to the CMake end for two reasons: 1. It potentially provides the best user experience, but you'd have to give it some love and ship with less quirky buildfiles. 2. It would force distros to build SDL via CMake and thus would ensure target exports are actually available everywhere. Various CMake patches I submitted today in summary (directly converted from the HG commits and `am`d onto a fork of a git mirror that happened to be on `tip`). https://github.com/tschw/SDL/commits/patched Fixing #2576 #3572, #3613, and this fresh ticket, which is almost entirely advertisement ;). These already do to make SDL much less of a quirky fella to have in your dependency tree...
2017-08-10 10:03:10 +08:00
endif ()
include(CMakePackageConfigHelpers)
write_basic_package_version_file("${CMAKE_BINARY_DIR}/SDL2ConfigVersion.cmake"
VERSION ${SDL_VERSION}
COMPATIBILITY AnyNewerVersion
)
install(EXPORT SDL2Targets
FILE SDL2Targets.cmake
NAMESPACE SDL2::
DESTINATION ${PKG_PREFIX}
)
install(
FILES
Fixed bug 3862 - Install is broken when adding SDL2 to an existing CMake project Steve Robinson In my existing CMake project, I use add_subdirectory to add the source for SDL2. This worked fine in 2.0.5, but now in 2.0.6 when I build the INSTALL CMake target, I get this error: file INSTALL cannot find "D:/path/to/SDL2Config.cmake". Call Stack (most recent call first): 3rdparty/SDL2/cmake_install.cmake:32 (include) 3rdparty/cmake_install.cmake:36 (include) cmake_install.cmake:32 (include) To fix this, I changed line 1770 from this: ${CMAKE_SOURCE_DIR}/SDL2Config.cmake To this: ${CMAKE_CURRENT_SOURCE_DIR}/SDL2Config.cmake
2017-10-07 07:42:43 +08:00
${CMAKE_CURRENT_SOURCE_DIR}/SDL2Config.cmake
Fixed bug 3651 - CMake build does not install CMake package configuration tschwinger@elitemail.org Most ironically, although autoconf/automake-based builds install (pretty half-assed) CMake package configuration files, they're missing in installations resulting from CMake-based builds entirely. A proper configuration file typically also loads target exports (implemented in patch 3572, also fixing this issue - see my comment on that issue for details). I believe it would be best to let the dinosaurs go extinct and redirect all build efforts to the CMake end for two reasons: 1. It potentially provides the best user experience, but you'd have to give it some love and ship with less quirky buildfiles. 2. It would force distros to build SDL via CMake and thus would ensure target exports are actually available everywhere. Various CMake patches I submitted today in summary (directly converted from the HG commits and `am`d onto a fork of a git mirror that happened to be on `tip`). https://github.com/tschw/SDL/commits/patched Fixing #2576 #3572, #3613, and this fresh ticket, which is almost entirely advertisement ;). These already do to make SDL much less of a quirky fella to have in your dependency tree...
2017-08-10 10:03:10 +08:00
${CMAKE_BINARY_DIR}/SDL2ConfigVersion.cmake
DESTINATION ${PKG_PREFIX}
COMPONENT Devel
)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
file(GLOB INCLUDE_FILES ${SDL2_SOURCE_DIR}/include/*.h)
file(GLOB BIN_INCLUDE_FILES ${SDL2_BINARY_DIR}/include/*.h)
foreach(_FNAME ${BIN_INCLUDE_FILES})
get_filename_component(_INCNAME ${_FNAME} NAME)
list(REMOVE_ITEM INCLUDE_FILES ${SDL2_SOURCE_DIR}/include/${_INCNAME})
endforeach()
list(APPEND INCLUDE_FILES ${BIN_INCLUDE_FILES})
install(FILES ${INCLUDE_FILES} DESTINATION include/SDL2)
Fixed bug 4135 - Broken symlink libSDL2.so since rev11940 Tiago O. Symlink points to the wrong folder, and target will always have debug postfix, so it'll be broken for other build types.
2018-04-16 00:37:51 +08:00
string(TOUPPER "${CMAKE_BUILD_TYPE}" UPPER_BUILD_TYPE)
if (UPPER_BUILD_TYPE MATCHES DEBUG)
set(SOPOSTFIX "${SDL_CMAKE_DEBUG_POSTFIX}")
else()
set(SOPOSTFIX "")
endif()
CMake: Don't make a libSDL2.so symlink on Mac OS X (do .dylib instead).
2016-01-04 09:52:44 +08:00
if(NOT (WINDOWS OR CYGWIN))
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
if(SDL_SHARED)
Correct the name of the SDL shared library in CMake for Mac OS
2018-08-19 05:23:40 +08:00
set(SOEXT ${CMAKE_SHARED_LIBRARY_SUFFIX}) # ".so", ".dylib", etc.
get_target_property(SONAME SDL2 OUTPUT_NAME)
Description: fix installation of shared library for Android Author: Boris Pek <tehnick-8@yandex.ru> Last-Update: 2018-01-10
2018-02-08 06:03:24 +08:00
if(NOT ANDROID)
install(CODE "
execute_process(COMMAND ${CMAKE_COMMAND} -E create_symlink
Fix WORKING_DIR parameter
2018-12-04 02:14:35 +08:00
\"lib${SONAME}${SOPOSTFIX}${SOEXT}\" \"libSDL2${SOPOSTFIX}${SOEXT}\"
WORKING_DIRECTORY \"${SDL2_BINARY_DIR}\")")
Correct the name of the SDL shared library in CMake for Mac OS
2018-08-19 05:23:40 +08:00
install(FILES ${SDL2_BINARY_DIR}/libSDL2${SOPOSTFIX}${SOEXT} DESTINATION "lib${LIB_SUFFIX}")
Description: fix installation of shared library for Android Author: Boris Pek <tehnick-8@yandex.ru> Last-Update: 2018-01-10
2018-02-08 06:03:24 +08:00
endif()
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
if(FREEBSD)
# FreeBSD uses ${PREFIX}/libdata/pkgconfig
install(FILES ${SDL2_BINARY_DIR}/sdl2.pc DESTINATION "libdata/pkgconfig")
else()
install(FILES ${SDL2_BINARY_DIR}/sdl2.pc
DESTINATION "lib${LIB_SUFFIX}/pkgconfig")
endif()
install(PROGRAMS ${SDL2_BINARY_DIR}/sdl2-config DESTINATION bin)
# TODO: what about the .spec file? Is it only needed for RPM creation?
Fixed bug 3078 - cmake: fix sdl2.m4 install location on cross hosts Timo Gurr On cross hosts running autotools for SDL2_gfx-1.0.1 fails to find sdl2.m4: eautoreconf: running in /var/tmp/paludis/build/media-libs-SDL2_gfx-1.0.1/work/SDL2_gfx-1.0.1 ... aclocal aclocal-1.13: warning: autoconf input should be named 'configure.ac', not 'configure.in' configure.in:128: warning: macro 'AM_PATH_SDL2' not found in library libtoolize --copy --force --automake aclocal aclocal-1.13: warning: autoconf input should be named 'configure.ac', not 'configure.in' configure.in:128: warning: macro 'AM_PATH_SDL2' not found in library autoconf configure.in:128: error: possibly undefined macro: AM_PATH_SDL2 If this token and others are legitimate, please use m4_pattern_allow. See the Autoconf documentation. * Failed Running autoconf ! SDL2 installs it to /usr/x86_64-pc-linux-gnu/share/aclocal on cross hosts, attached patch makes use of CMAKE_INSTALL_FULL_DATAROOTDIR to support correctly installing to the arch independent location /usr/share/aclocal.
2016-10-02 06:11:18 +08:00
install(FILES "${SDL2_SOURCE_DIR}/sdl2.m4" DESTINATION "${CMAKE_INSTALL_FULL_DATAROOTDIR}/aclocal")
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 23:33:46 +08:00
endif()
##### Uninstall target #####
Fixed bug 3866 - CMake error when trying to make the 'uninstall' target when it already exists Steve Robinson In my project, the 'uninstall' target is already created by the glew library. I get this error when SDL2 tries to create it: CMake Error at _build/3rdparty/SDL2/SDL2-2.0.6/CMakeLists.txt:1816 (add_custom_target): add_custom_target cannot create target "uninstall" because another target with the same name already exists. The existing target is a custom target created in source directory "D:/Code/sdl2-tutorial/_build/3rdparty/glew/glew-2.1.0/build/cmake". See documentation for policy CMP0002 for more details. To fix it, go to the bottom of the SDL2 CMakeLists.txt file. Add an if statement to check for the existence of the target before creating it. The end result looks like this: if(NOT TARGET uninstall) configure_file( "${CMAKE_CURRENT_SOURCE_DIR}/cmake_uninstall.cmake.in" "${CMAKE_CURRENT_BINARY_DIR}/cmake_uninstall.cmake" IMMEDIATE @ONLY) add_custom_target(uninstall COMMAND ${CMAKE_COMMAND} -P ${CMAKE_CURRENT_BINARY_DIR}/cmake_uninstall.cmake) endif() This is how the glew library deals with this possibility in their CMakeLists.txt file.
2017-10-12 23:47:02 +08:00
if(NOT TARGET uninstall)
configure_file(
"${CMAKE_CURRENT_SOURCE_DIR}/cmake_uninstall.cmake.in"
"${CMAKE_CURRENT_BINARY_DIR}/cmake_uninstall.cmake"
IMMEDIATE @ONLY)
add_custom_target(uninstall
COMMAND ${CMAKE_COMMAND} -P ${CMAKE_CURRENT_BINARY_DIR}/cmake_uninstall.cmake)
endif()
Reference in New Issue Copy Permalink