Some Early Bits Of The "Soft FP64" Infrastructure Will Be Mainlined Soon In Mesa

David Airlie has announced his plans to begin mainlining some early infrastructure work on the "soft" FP64 code into Mesa Git. This doesn't yet allow for soft FP64 on older GPUs lacking the hardware capability to do this otherwise, but will help in another area and can make for easier mainlining of the actual soft FP64 support in the future.

Due to some of his int/double conversion code not passing the OpenGL Conformance Test Suite or Piglit tests, David Airlie has decided to start bringing in some of the soft FP64 code into Mesa for addressing the problem. This doesn't yet enable soft FP64 but is a step in that direction and could make it easier to merge the actual support when the time comes. Elie Tournier at Collabora has been working on most of the soft FP64 code itself for emulating this capability with GLSL shaders.

Airlie announced his plans on the Mesa-dev list. The soft FP64 code will eventually allow for ARB_gpu_shader_fp64 on older GPUs lacking native double-precision floating-point support. Right now the R600g driver is limited to OpenGL 3.3 for all but the HD 5800/6900 series hardware due to the absence of native FP64 elsewhere, but once this code is in place, the Radeon HD 5000/6000 series in full will be able to have OpenGL 4.x.