As was just talked about in announcing the open-source 2D and 3D support for ATI Evergreen GPUs
, the R600g driver has been gaining lots of momentum in the past few weeks. Ever since this open-source Gallium3D driver that aims to provide hardware-acceleration for ATI Radeon HD 2000/3000/4000 (and potentially Radeon HD 5000 series) hardware took a shader compiler shortcut
a few weeks back, it seems almost every time our RSS feed for the Mesa Git change-log refreshes there is a new R600g driver change.
In late July we were able to run the infamous glxgears with the R600g driver
and shortly thereafter it picked up OpenGL texture support
and mip-map and face culling support
Since our last R600g status report, some of the changes to this driver that will eventually replace the R600 classic Mesa driver include support for new TGSI opcode instructions, segmentation fault fixes, OpenGL occlusion query support, fixed pitch alignment, user-clip plane support, an improved texture format checker, point/sprite rendering support, and various other technical changes. Some of the new instructions supported include POW, COS, SIN, SSG, SEQ, SGT, SNE, FRC, FLR, DDX, DDY, SGE, SLE, TXB, and many more. You get the point.
Progress on the ATI R600 Gallium3D driver can be followed via this CGit page
. While the R300g driver has been a greater focus for developers than the effectively-complete R300 classic driver, the R600g driver too is becoming a greater focus for Mesa developers than the classic alternative driver. Most of the changes as of late to the R600g driver have been by Red Hat's David Airlie.
Of course, as mentioned in the open-source Evergreen article today, this creates an interesting dilemma of whether to continue focusing on the R600g driver and ignore the classic Evergreen 3D driver, get the ATI Radeon HD 5000 series support merged into the R600g driver, or attempt to divert some resources to getting this Evergreen driver that's written against Mesa's classic architecture up and running.