This is nasty because there's no way in GL to output data to the stencil
buffer directly, so we have to do a dance to wrap the depth/stencil buffer
in an ARGB renderbuffer.
Improves performance of several oglconform testcases by better than a factor
of 2.
If we shift bytes into the texel word (or use the PACK_COLOR_8888 macro),
we don't have to worry about big vs. little endian. See comments about
texel formats in texformat.h.
Remove an unneeded/incorrect else-if clause that produced wrong results
on big-endian systems.
If the tnl output attributes matches the swsetup input attributes we still
need to check if the desired vertex color type (float vs. chan) has changed
so that we use the right emit functions.
Fixes a conformance failure found with logicop test at pathlevel 3.
This lets us short-circuit when we're leaving the same cliprects in place,
which becomes quite common with metaops clears, and may be useful for some of
our FBO paths.
Without this, we would reject programs which sampled multiple times from
registers defined in the same phase (block of instructions with the same
texture indirection count), as each sample would count as a new phase
beginning. Instead, keep track of which phases registers were written in,
and only bump phase when we're reading from one generated in this phase.
On the other hand, we failed to count oC or oD texture samples as being new
phases.
Bug #17865.
Move GL_ARB_texture_non_power_of_two and GL_ATI_separate_stencil
from the generic extension list to the 965-specific list. Neither
extension is supported on i830-class hardware, and
GL_ATI_separate_stencil is not supported on i915-class hardare.
GL_ARB_texture_non_power_of_two is supported on i915-class hardare and
is already in the i915-specific list.
The ARB extension is a superset of the older SGIX extension. Any
hardware that can support the SGIX version can also support the ARB
version. In Mesa, any driver that supports one also supports the
other. This unification just simplifies some bits of code.
- Sort extensions by ARB, then EXT, then vendor by name
- Remove redundant (only one of GL_{ARB,EXT,NV}_texture_rectangle) or
duplicate extension strings
The GL_OBJECT_TYPE_ARB query is handled directly in
_mesa_GetObjectParamterivARB because it is only supported in the
extension version of the shanding language API. glGetProgramiv and
glGetShaderiv should not accept this enum.
The GL_ARB_shader_objects spec says that glAttachShaderARB is supposed
to return GL_INVALID_OPERATION if a shader is attached to a program
where it is already attached. _mesa_attach_shader perviously returned
without error in this case.
Added an x86-64 CPUID function and use it to detect 3Dnow! If 3Dnow!
is available, use _mesa_3dnow_transform_points4_3d_no_rot,
_mesa_3dnow_transform_points4_perspective,
_mesa_3dnow_transform_points4_2d_no_rot, and _mesa_3dnow_transform_points4_2d.
This fixes long standing bug #8724.
The i965 driver previously had it's own set of code to convert
fixed-function TNL state to a vertex program. Core Mesa has code to
do this, so there is no reason to duplicate that effort in the driver.
In fact, this duplication leads to bugs when other aspects of the Mesa
infrastructure change.
