Always turn on the TAM_DEBUG3=0x6 workaround on real r200s. It appears that
the current cases for turning it on were insufficient (Bugzilla #1519, 729, 814)
and it has no significant performance impact. Performance tested with quake3
in GL_LINEAR mode both with and without anisotropy, with the workaround always
on or always off.
- Color buffer clear is accelerated, but flickers (possibly caused by a
recent DDX or Mesa change or bad merge)
- Everything else uses software fallback rendering
- There should be no clipping-related artifacts with the
sw-clipspan-fixes.patch against Mesa (posted on dri-devel)
- Multiple clients should be rock solid with a DDX patch that is soon to
come (soon = within the next hour or so)
Add support for optimized versions of the code underlying ReadPixels
(and DrawPixels). The R200, R128, and Unichrome drivers get support
in this commit. Other drivers would be easy enough to add for people
that have the cards.
The DRI (CVS) build will need to be updated to account for the new
source files.
Initial support for PowerPC specific code in Mesa and DRI drivers. DRI
drivers built on PowerPC systems should now show things like "PowerPC" or
"PowerPC/Altivec" in the GL_RENDERER string.
The VMX moniker is used for Altivec/Velocity Engine/VMX SIMD additions. I
chose this not because I work for IBM but because it's a LOT shorter to
type. :)
Add Roland Scheidegger's S3TC patch. This patch does not implement the
(patented) S3TC/DXTC algorithms, but adds an option to dlopen a library module
providing functions to do so. Because it uses dlopen, it is only enabled if
USE_EXTERNAL_DXTN_LIB=1 is defined (which is only in linux-dri config, so far).
It adds support for S3TC to several DRI drivers, and adds a DRI config option to
force enabling S3TC even if the software compression/decompression is
unavailable. This may allow people to use apps that require S3TC even though
they don't have a license to implement the patented material themselves, if
those apps use precompressed textures.
Ideally we would get permission from the current holder of the patents to
implement the algorithm in Mesa, at which point the dlopen mess could go away.
Until then, this allows some to run applications they couldn't otherwise, and
hopefully will provide us with more push to get the final step of getting that
permission done.