Generate the C protocol files from the protocol XML files installed by
wayland-protocols, and use them to implement support for relative pointer
motions and pointer locking.
Note that at the time, the protocol is unstable and may change in the future.
Any future breaking changes will, however, fail gracefully and result in no
regressions compared to before this patch.
Since we are loading shared objects dynamically, build our own version of the
core protocol symbols, so that we in the future can include protocol
extensions.
Jonas Kulla
The configure script didn't differentiate between Linux and Android, unconditionally compiling in the unix implementation of SDL_sysfilesystem.c.
I'm probably one of the very few people building SDL for android using classic configure + standalone toolchain, so this has gone undetected all along.
With this commit, you can compile SDL2 with Emscripten
( http://emscripten.org/ ), and make your SDL-based C/C++ program
into a web app.
This port was due to the efforts of several people, including: Charlie Birks,
Sathyanarayanan Gunasekaran, Jukka Jylänki, Alon Zakai, Edward Rudd,
Bruce Mitchener, and Martin Gerhardy. (Thanks, everyone!)
Chris Beck
When creating a homebrew recipe for wesnoth, I discovered that the SDL image configuration routine does not detect libpng properly -- if you have multiple instances of libpng on your system, and you use environment variables to select an instance which is not in your system directory, the build can be broken, because it will run configuration tests against the system installed version, but deduce that it should use the filename of the system-installed version. In a vanilla build of wesnoth using homebrew, this results in segfaults at runtime, because you end up linking against two different versions of libpng, which is also needed independently of SDL.
The problem is essentially in the "find_lib" routine in the configure file:
find_lib()
{
gcc_bin_path=[`$CC -print-search-dirs 2>/dev/null | fgrep programs: | sed 's/[^=]*=\(.*\)/\1/' | sed 's/:/ /g'`]
gcc_lib_path=[`$CC -print-search-dirs 2>/dev/null | fgrep libraries: | sed 's/[^=]*=\(.*\)/\1/' | sed 's/:/ /g'`]
env_lib_path=[`echo $LIBS $LDFLAGS | sed 's/-L[ ]*//g'`]
for path in $gcc_bin_path $gcc_lib_path $env_lib_path /usr/lib /usr/local/lib; do
lib=[`ls -- $path/$1 2>/dev/null | sort -r | sed 's/.*\/\(.*\)/\1/; q'`]
if test x$lib != x; then
echo $lib
return
fi
done
}
Because the for loop goes over the system directories before the environment directories, any system-installed lib will shadow the lib selected via environment variables. This is contrary to the behavior of the configuration tests earlier in the script, which prefers the environment variable libs over the system-installed libs. The 'for' loop should instead be:
for path in $env_lib_path $gcc_bin_path $gcc_lib_path /usr/lib /usr/local/lib; do
You can see the full discussion on the Homebrew / linuxbrew issue tracker here: https://github.com/Homebrew/linuxbrew/issues/172
I have checked that this bug also affects SDL 1.2.15, SDL_mixer and SDL_ttf 1.2, which all use this same "find_lib" routine. I have not determined if the bug affects SDL 2.0, which seems not to use this exact routine.
Alex Szpakowski
SDL's Cocoa backend uses the CGDisplayMode API to get refresh rate information about a display mode, but CGDisplayModeGetRefreshRate will return 0 on most non-CRT monitors.
The only way I know of to get correct refresh rate information in OS X is via the CoreVideo DisplayLink API.
I have attached a patch which tries to use the CVDisplayLinkGetNominalOutputVideoRefreshPeriod function if CGDisplayModeGetRefreshRate fails, which fixes display mode refresh rate information on the monitors I tested.
The CVDisplayLink API requires linking with the CoreVideo framework, and the patch updates the various build files to do so.
Alex Szpakowski
Now that SDL on iOS requires CoreMotion to be linked, some of the Xcode projects included with the SDL source (such as the iOS tests and the iOS app template) as well as the premake and automake scripts need to be updated.
I've attached a patch which does so. It also fixes the SDL Xcode project to build for 64-bit ARM as well as armv7 by default (or whatever the default ARM targets are for the Xcode version used), which is what the iOS app template expects.
If the EGL extension EGL_KHR_create_context is available, we can use it to
set the core/compatability profile and the minimum OpenGL version.
Use this if it is available to get the context requested by the GL attributes.
I added -Wshadow and then turned it off again because of massive variable shadowing in the blit macros.
Feel free to go through that code and fix these if you want. Just uncomment CheckWarnShadow in configure.in if you want to try this.
Add V=1 to the make command line will show the full commands but by default
we just show the tool-type and the output file. This is generally much easier
on the eye and makes warnings and errors more clearly visible.
The D3D11 renderer is now slightly faster than D3D9 on my Windows 8 machine (testsprite2 runs at 3400 FPS vs 3100 FPS)
This will need tweaking to fix the Windows RT build.
configure wraps this C code in a main() function, so you can't declare a
function body in there. Besides, I'm not sure why we declared a function
that's part of the multitouch API anyhow.
Now we just reference a type that only exists if the headers have multitouch
support and call it a day.
Brian Minton
When building static or shared libraries on cygwin 1.7.25 on Windows 7 (32-bit), I get the following link errors:
build/.libs/SDL_windowskeyboard.o: In function `UIElementSink_QueryInterface@12':
/home/c-bminton/src/SDL/src/video/windows/SDL_windowskeyboard.c:995: undefined reference to `IID_IUnknown'
build/.libs/SDL_windowskeyboard.o: In function `IPPASink_QueryInterface@12':
/home/c-bminton/src/SDL/src/video/windows/SDL_windowskeyboard.c:1101: undefined reference to `IID_IUnknown'
collect2: error: ld returned 1 exit status
Makefile:126: recipe for target 'build/libSDL2.la' failed
make: *** [build/libSDL2.la] Error 1
--
The libuuid from e2fsprogs is completely unrelated to the w32api UUID.DLL implib. FWIW, any *NIX software that I've seen obviously wants the e2fsprogs version, but if libuuid-devel is installed (in /usr/lib) then you can't link against the w32api implib with a simple -luuid.
We'll flip the default back to enabled right after 2.0.2 is finalized, and
try to declare them stable and ready by 2.0.3.
Those that have an interest in supporting them in 2.0.2 can manually enable
them in the configure script with --enable-video-wayland and/or
--enable-video-mir.
Based on the original port to Wayland by: Joel Teichroeb, Benjamin Franzke, Scott Moreau, et al.
Additional changes in this commit, done by me:
* Wayland uses the common EGL framework
* EGL can now create a desktop OpenGL context
* testgl2 loads GL functions dynamically, no need to link to libGL anymore
* Assorted fixes to the Wayland backend
Tested on the Weston Compositor (v1.0.5) that ships with Ubuntu 13.10,
running Weston under X. Tests ran: testrendercopyex (all backends), testgl2, testgles2,testintersections
Haiku uses most of the standard pthread API, with a few #ifdefs where we
still need to fallback onto the old BeOS APIs.
BeOS, however, does not support pthreads (or maybe doesn't support it well),
so I'm unplugging support for the platform with this changeset. Be Inc went
out of business in 2001.
Joseph Carter
There's a whole set of configure tests for BSD's libusbhid, and they only matter on BSD. However, if you have the library on Linux, it gets pulled in as library bloat. And it's bloat of the highest order since not a single function call to the library is ever made unless you're on a *BSD.
Andreas
With the patch applied, make is not able to find the rule for Makefile.in anymore. Removing the patch resolves the issue.
The path is in fact correct (in my case: /c/external/SDL64/SDL). But it seems the windows build of GNU Make doesn't work well with pathnames in rules. Both the dependencies in "$(srcdir)/configure: $(srcdir)/configure.in" and "Makefile: $(srcdir)/Makefile.in" will cause rules not to be found when srcdir is defined.
The same problem occurs if the patch is removed and I supply configure with a srcdir manually.
The configure script was still overwriting SIZEOF_VOIDP so both ended up as either 4 or 8 depending on the arch. This simply removes the check from configure.in
Unexplicable computer sciences phenomenon: Instead of returning an empty set,
*.c in an folder with no .c files produces the "*.c" string to be added as
a source. I'm sorry future generations, we are doing the best we can :)
leighmanthegreat@hotmail.com
I downloaded SDL2 from hg source.
I built to a ./build directory.
I downloaded SDL_image 2 hg and attempted to build.
When it cannot find a sdl2-config the SDL_image configure correctly suggests setting SDL_CONFIG variable.
Setting this the configure still fails with 'Permission denied' on the call to sdl2-config.
Setting execute permission solves the problem.
If possible, sdl2-config should have executable bit set when it is created.
tomaszewski.p
Recent changes in SDL_sysrenderer.h and SDL_sysvideo.h had no impact on directfb backend.
Attached patch:
- updates interface,
- resolves uninitialized variable reading,
- changes logging tio use SDL_Log API,
- updates configure to use DIRECTFBCONFIG variable instead direct call to directfb-config.
The OpenBSD case is puzzling - it's complaining about symbols in the C library. Do we need to explicitly link with -lc for shared libraries on OpenBSD?
The BeOS case is legitimate, it's calling back into the application code to initialize BeApp if it isn't already initialized.
Brad Smith 2012-07-19 11:39:09 PDT
I noticed this error from the OpenBSD/amd64 buildbot log..
../configure[15018]: test: -O2: unexpected operator/operand
The attached patch fixes the issue.
It was very confusing to have configure generate an SDL_config.h and then not have it be used when building on Mac OS X or Windows. I'll just have to remember to use SDL_config_windows.h when building official releases that are supposed to be ABI compatible with Visual Studio.
Scott Percival 2011-07-03 06:41:51 PDT
This submission is aimed at making life easier for OpenGL ES capable devices
running a X11 stack (e.g. Maemo, Meego, TrimSlice, other ARM SoC boards not
running Android). SDL's Pandora support already has the neccesary GLES-to-X11
glue code, however it's all ghetto'd off in Makefile.pandora and not very
flexible.
The patch:
- adds an awesome --enable-video-opengles option to configure
- re-modifies the opengles and opengles2 SDL_renderers to use function pointers
- no idea why this was removed?
- for SDL_Renderers, links in libGLESv1_CM, libGLES_CM (for PowerVR fans) or
libGLESv2 at runtime
- links in libEGL.so at runtime - the old code made an assumption that
eglFunctions could be pulled from the active GLES library, PowerVR for one
doesn't let you do that with their libGLESv2
- allows you to pick which of GLES v1 or v2 to load via
SDL_GL_CONTEXT_MAJOR_VERSION
So far I've tested this on a Nokia N900 (OMAP 3430/SGX 530 running Maemo 5) and
a Toshiba AC100 (Tegra 2 running Ubuntu 10.10). I haven't tested it on... well,
everything that isn't those two, such as a Pandora, iOS or Android device. The
Pandora specific code should be kept intact (fingers crossed), and nothing
painfully drastic has been added to the SDL_renderers. The library loading
sequence in SDL_x11opengles has been updated to accomodate both NVIDIA's
propensity to let developers get away with murder and PowerVR's alternative of
punishing every missed step.
The test apps work okay with GLES or GLES2 as the renderer. For some reason
alpha blending doesn't seem to work on the Tegra 2; last week NVIDIA pushed out
a new set of X11 GLES drivers, so I'll try and investigate once I upgrade
those. Also, this patch adds things to configure.in, include/SDL_config.h.in
and test/configure.in. I didn't know what the policy was re. committing
generated spaghetti from autotools, so ./autogen.sh has to be run again. Sorry.
I think that's about everything, let me know if there's anything I've
overlooked.
t.grundner@goto3d.de 2011-09-01 03:59:17 PDT
I figured out what is going on. GCC 4.5.2 assumes the stack is 16 byte aligned
by default. Therefore there are no AND alignment corrections necessary if we
wish to align a stack variable to a 16 byte boundary. That is bad if your OS
ABI is not 16 byte aligned. Windows 32 bit stacks are 4 byte aligned. This
results in the above mentioned SIGSEGV. This is also no problem if I compile
both SDL.dll and my app with MingW because MinGW/GCC inserts a
andl $-16, %esp
instruction right in the beginning of the main function. So at least the stack
of the thread calling the main function is 16 byte aligned. But as soon as I
start to use the SDL.dll from an application not compiled by MinGW there is no
ANDL safing my app.
However there is a GCC option that can change the default stack alignment:
-mpreferred-stack-boundary=num
Setting num=2 assumes a the stack is aligned to a 4 byte boundary. This results
in GCC inserting the necessary
andl $-16, %esp
into SDL_FillRect. Rebuilding SDL with
./configure "CFLAGS=-mpreferred-stack-boundary=2 -g -O3"
solved the problem.
IMHO this should also be a problem on Solaris.
The following links contain further information:
http://gcc.gnu.org/onlinedocs/gcc-4.5.2/gcc/i386-and-x86_002d64-Options.html#i386-and-x86_002d64-Optionshttp://www.agner.org/optimize/calling_conventions.pdf
