Currently each time a screen resolution or setting change occurs, fbdev_destroy, fbdev_init and fbdev_set_mode are called in sequence, trying to destroy context and surface (though context pointer seems to remain unchanged) and create again both of them.
However it seems something is wrong with egl_destroy, as after the call fps drops from 60 to 33fps in GUI with huge performance impact, at least with libmali.
Philosophy is changed with this commit to avoid destroying and creating context and surface each time (creation only occurs in fbdev_init and egl_destroy is called only if retroarch is shutting down).
As a minor modification, framebuffer is resetted to 0 on retroarch shutdown to avoid any chance of freezed screen effect.
-In case yoffset is not set=0 by the driver (which seems to occur randomly on certain configurations), on next set_video_mode call screen freezes and cannot be updated. Condition to set yoffset to 0 if current value is >0 is added as fix.
-If setterm is not available, sh error occurs in the shell on gfx_ctx_mali_fbdev_destroy call. Condition based on setterm availability is added to avoid this.
-If pixclock from driver is not available (=0), divide by 0 occurs when trying to calculate refresh rate. Condition is added as a fix, if pixclock=0 refresh rate is set as 60.
- move shader flags from the video drivers to the context drivers
- rework config_load_shader_preset() from configuration.c into retroarch_load_shader_preset() in retroarch.c with proper compatibility check
- implicitly call retroarch_load_shader_preset() in retroarch_get_shader_preset() once per core/content life cycle
- use video_shader_is_supported() instead of video_driver_test_all_flags() where appropriate
- remove GFX_CTX_FLAGS_NONE because it is meaningless
Added a platform parameter to egl_init_context. If the caller provides a
platform other than EGL_NONE, then it will try to use eglGetPlatformDisplay or
eglGetPlatformDisplayEXT instead of eglGetDisplay.
If neither eglGetPlatformDisplay or eglGetPlatformDisplayEXT is supported, then
it will still fall back to calling eglGetDisplay.
Updated the Wayland, X11, and DRM callers to use the correct platform enum.
Those are the callers that don't just pass EGL_DEFAULT_DISPLAY as the native
display handle.
Calling eglGetDisplay with any value other than EGL_DEFAULT_DISPLAY is
inherently unreliable, because it requires the EGL implementation to guess a
platform type based on a (void *) pointer. Some implementations might not
identify a particular platform, or worse, might guess wrong.
Fixes https://github.com/libretro/RetroArch/issues/4790
Old kronos headers used mali_native_window, ARM's headers used fbdev_window, and
latest mali-fbdev package on the odroid with newer khronos headers, the struct seems missing
