Age | Commit message (Collapse) | Author |
|
TLS requires kernel >= 2.4.20. Per glapi.
|
|
It should be u_thread_self, not _glthread_GetID.
|
|
The preprocessor magic in mapi was nothing but obfuscation. Rewrite
mapi_abi.py to generate real C code.
This commit removes the hack added in
43121f20866bb89e8dac92bd92ec85a943704b7e.
|
|
Right fix is probably stop C-preprocessor abuse and stick 100% with
scripted code generation.
|
|
Fixes nasty bug where some parts of the code didn't define WIN32_THREADS
and were using the integer mutex implementation, causing even confusion
to the debuggers.
And there is little interest of other thread implemenation on Win32
besides Win32 threads.
|
|
Do not use "void *" in arithmetics.
|
|
mapi.h is included by vgapi and st/vega. On win32, the macro expands to
dllexport and dllimport respectively.
|
|
That is, replace the old _glapi_* names by new names that start with
u_current_. When MAPI_GLAPI_CURRENT is defined, u_current.h defines
rename macros to restore the old names. That is done for ABI
compatibility.
|
|
glapi defines an interface that is used by DRI drivers. It must not be
changed in an ABI incompatible way. This commit moves all
functions/variables belong to the interface to glapi.h. Instead of
including u_current.h from glapi.h, u_current.h now includes glapi.h.
|
|
|
|
Define PUBLIC to __declspec(dllexport) when _MVC_VER is defined.
|
|
|
|
|
|
|
|
|
|
The idea is to have an API generate a header using mapi_abi.py. The API
can then use the header to build a dispatcher.
|
|
Specifically, move all or most of
glapi/glapi.c to mapi/u_current.c,
glapi/glapi_execmem.c to mapi/u_execmem.c,
glapi/glthread.[ch] to mapi/u_thread.[ch]
and remove their dependencies on core Mesa headers.
|