* Fix D3D12 debug crash due to validation layers SDK bug
* Change LONG to FARPROC to be technically correct
* Make requested changes:
Move windows version checking to function in bgfx.cpp
Change windowsVersionIs to also use this technique
Use bgfx code style
Specify dwOSVersionInfoSize prior to calling rtlGetVersion
* Merge both functions into one
Use bx::dlsym instead of GetProcAddress directly
Fix major & minor comparison logic
* Fix win10 version comment
* Revert rendererCreate check back to Win 0x0602
* Clean up logic as per suggestion
* Add 'frameNumber' to Frame struct
Previously, the frame number returned from bgfx::frame() was tracked separately in the Context. Now,
we store that information in the Frame. This will allow us to attach the frame number to ViewStats.
* Add frame number to ViewStats
When ViewStats are enabled, we tag each timer query with the current frame number, then include
that information in the final results. In this way, clients can correlate specific work that they
submitted to specific GPU timing information.
NOTE: Some backends not implemented, yet. They will always have 0 for frame number.
The ones which are implemented are:
* OpenGL
* Vulkan
* D3D 9,11,12
* Noop
* Add indirect draw with indirect count (BGFX_CAPS_DRAW_INDIRECT_COUNT)
* Update bindings
* VK: Add support for BGFX_CAPS_DRAW_INDIRECT_COUNT
* D3D12: Add support for BGFX_CAPS_DRAW_INDIRECT_COUNT
* GL: Add support for BGFX_CAPS_DRAW_INDIRECT_COUNT
* 48-drawindirect: Use BGFX_CAPS_DRAW_INDIRECT_COUNT if available
* 48-drawindirect: Update shaders
Compute kSortKeyViewNumBits automatically via log2(BGFX_CONFIG_MAX_VIEWS), so that we could change BGFX_CONFIG_MAX_VIEWS to be above 2^10 = 1024 without changing the `bgfx` source code.
One possible problem with this approach is running out of bits, so maybe we could assert for that, I am not sure where to do it and how many bits we can use though.
Using `bx::uint32_cntlz` because `bx::log2` is not constexpr.
Tested on linux opengl with BGFX_CONFIG_MAX_VIEWS=4096, seems to work.
