While the news was delivered on the 6th of September, XDS Cambridge was going on at that point, and it wasn't until the 17th of September that the RadeonHD Linux driver was published, so it's hard to call a precise birth-date for this strategy. It could also be considered the point at which SUSE wrote a letter to AMD with this open-source idea.
It's been a bumpy road over the past four years, but AMD's open-source strategy remains in full-swing. While the RadeonHD driver is no more, the goals as proposed by SUSE in a letter to AMD back in early 2007 have largely been realized.
a. Provide an open source driver with full support for the latest generations of ATI hardware
including new chipset versions of the same generation.
b. Improve the quality of the open source driver for older hardware generations.
c. Add missing support for already released series of ATI hardware to the open source driver including support for video playback and 3D (with programmable shader support).
d. Provide specification and programming documentation for graphics chipsets for the development of an open source driver to the open source community including specification updates for new chips of the same series including erratas.
e. Continue to implement open source drivers and publication of specifications for future generation chipsets.
There's support for all of AMD's new hardwarem, including Gallium3D support, but the churn of documentation drops have slowed down a bit. Fortunately, things seem to be working out and we're beginning to see a quicker turn-around time for new hardware enablement following major Radeon product launches.
Last month I talked about AMD's open-source progress over four years and in two weeks will be the results from the largest-ever Radeon Linux driver comparison with more than two dozen Radeon GPUs on both the open and closed-source drivers.
Cheers to all of the developers involved. Next week at XDC2011 Chicago there shall be celebrating.