To add a new GL extension to Mesa you have to do the following.
If glext.h doesn't define the extension, edit include/GL/gl.h and add: - new enum tokens - new API function entry points - #define GL_EXT_the_extension_name 1 If adding a new API function (call it glNewFunctionEXT): - insert glNewFunctionEXT()into src/apiext.h - edit src/types.h and add NewFunction to the gl_api_table struct - implement gl_NewFunction() in the appropriate src file - hook gl_NewFunction() into pointers.c - add display list support in dlist.c for save_NewFunction() - add glNewFunctionEXT to gl_GetProcAddress() in extensions.c or in the device driver's GetProcAddress() function if appropriate
If adding new GL state be sure to update get.c and enable.c
In general, look for an extension similar to the new one that's already implemented in Mesa and follow it by example.
Mesa's code style has changed over the years. Here's the latest.
Comment your code! It's extremely important that open-source code be well documented. Also, strive to write clean, easily understandable code.
3-space indentation
If you use tabs, set them to 8 columns
Brace example:
if (condition) {
foo;
}
else {
bar;
}
Here's the GNU indent command which will best approximate my preferred style:
indent -br -i3 -npcs infile.c -o outfile.c
Local variable name example: localVarName (no underscores)
Constants and macros are ALL_UPPERCASE, with _ between words
Global vars not allowed.
Function name examples:
glFooBar() - a public GL entry point (in dispatch.c) _mesa_FooBar() - the internal immediate mode function save_FooBar() - retained mode (display list) function in dlist.c foo_bar() - a static (private) function _mesa_foo_bar() - an internal non-static Mesa function
XXX to do
These are the instructions for making a new Mesa release.
Prerequisites (later versions may work):
Be sure to do a "cvs update -d ." in the Mesa directory to get all the latest files.
Update the version strings in src/get.c and src/X/fakeglx.c to return the new Mesa version number.
Create/edit the docs/RELNOTES-X-Y file to document what's new in the release. Edit the docs/VERSIONS file too. Update the docs/IAFA-PACKAGE file.
Edit Make-config and change the MESA_MAJOR and/or MESA_MINOR versions.
Edit the GNU configure stuff to change versions numbers as needed: Update the version string (second argument) in the line "AM_INIT_AUTOMAKE(Mesa, 3.3)" in the configure.in file.
Remove the leading `dnl' from the line "dnl AM_MAINTAINER_MODE".
Verify the version numbers near the top of configure.in
Run "fixam -f" to disable automatic dependency tracking.
Run the bootstrap script to generate the configure script.
Edit Makefile.X11 and verify DIRECTORY is set correctly. The Mesa sources must be in that directory (or there must be a symbolic link).
Edit Makefile.X11 and verify that LIB_NAME and DEMO_NAME are correct. If it's a beta release, be sure the bump up the beta release number.
cp Makefile.X11 to Makefile so that the old-style Mesa makefiles still work. ./configure will overwrite it if that's what the user runs.
Make a symbolic link from $(DIRECTORY) to Mesa. For example, ln -s Mesa Mesa-3.3 This is needed in order to make a correct tar file in the next step.
Make the distribution files. From inside the Mesa directory:
make -f Makefile.X11 lib_tar make -f Makefile.X11 demo_tar make -f Makefile.X11 lib_zip make -f Makefile.X11 demo_zip
Copy the distribution files to a temporary directory, unpack them, compile everything, and run some demos to be sure everything works.
Upload the *.tar.gz and *.zip files to ftp.mesa3d.org
Update the web site.
Make an announcement on the mailing lists: mesa3d-dev@lists.sf.net, mesa3d-users@lists.sf.net and mesa3d-announce@lists.sf.net
In order to run the bootstrap script you'll need:
autoconf 2.50 automake 1.4-p5 libtool 1.4