[g3dvl] move zscan into shaders

[g3dvl] remove quantizer_scale from picture structure

[g3dvl] remove dc_dct_pred from picture structure

[g3dvl] divide mpg12 width height by 16

[g3dvl] remove the dubble zscan from the mpg12 decoder

vdpau: make state tracker far less noisy

vdpau: remove empty color.c file

[g3dvl] remove the unused bitstream parser components

vdpau: remove the unused and incomplete mpeg2 parser

[g3dvl] Implement MPEG2 VLD

Based uppon xine's slice_xvmc.c.
This gets VDPAU up and running.

[g3dvl] Name of pkg-config file is libva.

Signed-off-by: Christian König <deathsimple@vodafone.de>

[g3dvl] Fix compile error.

Signed-off-by: Christian König <deathsimple@vodafone.de>

vdpau: start to implement MPEG2 decoder part

[g3dvl] make clear color configureable

[g3dvl] make reset_dirty_area a public interface

[g3dvl] implement clearing of dirty destination surface areas

[g3dvl] move compositor src and dst normalisation into layer setting

[g3dvl] fix setting width, height and chroma format in video buffer

xvmc: recreate drawable surface only when dst area changes

[g3dvl] fix a couple of bugs around paletted subpictures

r600g: revert commit 68cc6bc5d8b6986acc7f5780d705f4ae9be2a446

Revert commit "remove the unneeded bo from COLOR[0-9]_INFO
Also implement a working alternative.

vdpau: implement OutputSurfaceDestroy

vdpau: implement PresentationQueueDestroy

vdpau: implement VideoMixerDestroy

vdpau: add stups for the missing functions

vdpau: implement and cleanup PresentationQueueTarget destruction

xvmc: futher optimize ycbcr upload

[g3dvl] merge idct stage 2 and mc ycbcr stage into a single draw

[g3dvl] rework fence handling and add r600g workaround

[g3dvl] start supporting different render target formats

Let's start with NV12, but anything else shouldn't be much of a problem any more.

[g3dvl] remove resource_format workaround

r600g: implement clamp_fragment_color handling

The first small step to get arb_color_buffer_float working.

r600g: remove the unneeded bo from COLOR[0-9]_INFO

The COLOR[0-7]_INFO registers doesn't neeed a bo.

Merge remote branch 'origin/master' into pipe-video

Conflicts:
configs/linux-dri
src/gallium/drivers/r600/r600_pipe.c
src/gallium/drivers/r600/r600_state.c
src/gallium/include/pipe/p_format.h
src/gallium/tests/graw/fragment-shader/frag-abs.sh
src/gallium/tests/graw/fragment-shader/frag-add.sh
src/gallium/tests/graw/fragment-shader/frag-cb-1d.sh
src/gallium/tests/graw/fragment-shader/frag-cb-2d.sh
src/gallium/tests/graw/fragment-shader/frag-dp3.sh
src/gallium/tests/graw/fragment-shader/frag-dp4.sh
src/gallium/tests/graw/fragment-shader/frag-dst.sh
src/gallium/tests/graw/fragment-shader/frag-ex2.sh
src/gallium/tests/graw/fragment-shader/frag-face.sh
src/gallium/tests/graw/fragment-shader/frag-flr.sh
src/gallium/tests/graw/fragment-shader/frag-frc.sh
src/gallium/tests/graw/fragment-shader/frag-kil.sh
src/gallium/tests/graw/fragment-shader/frag-lg2.sh
src/gallium/tests/graw/fragment-shader/frag-lit.sh
src/gallium/tests/graw/fragment-shader/frag-lrp.sh
src/gallium/tests/graw/fragment-shader/frag-mad-immx.sh
src/gallium/tests/graw/fragment-shader/frag-mad.sh
src/gallium/tests/graw/fragment-shader/frag-max.sh
src/gallium/tests/graw/fragment-shader/frag-min.sh
src/gallium/tests/graw/fragment-shader/frag-mov.sh
src/gallium/tests/graw/fragment-shader/frag-mul.sh
src/gallium/tests/graw/fragment-shader/frag-rcp.sh
src/gallium/tests/graw/fragment-shader/frag-rsq.sh
src/gallium/tests/graw/fragment-shader/frag-sge.sh
src/gallium/tests/graw/fragment-shader/frag-slt.sh
src/gallium/tests/graw/fragment-shader/frag-srcmod-abs.sh
src/gallium/tests/graw/fragment-shader/frag-srcmod-absneg.sh
src/gallium/tests/graw/fragment-shader/frag-srcmod-neg.sh
src/gallium/tests/graw/fragment-shader/frag-srcmod-swz.sh
src/gallium/tests/graw/fragment-shader/frag-sub.sh
src/gallium/tests/graw/fragment-shader/frag-tempx.sh
src/gallium/tests/graw/fragment-shader/frag-xpd.sh
src/gallium/tests/graw/vertex-shader/vert-abs.sh
src/gallium/tests/graw/vertex-shader/vert-add.sh
src/gallium/tests/graw/vertex-shader/vert-arl.sh
src/gallium/tests/graw/vertex-shader/vert-arr.sh
src/gallium/tests/graw/vertex-shader/vert-cb-1d.sh
src/gallium/tests/graw/vertex-shader/vert-cb-2d.sh
src/gallium/tests/graw/vertex-shader/vert-dp3.sh
src/gallium/tests/graw/vertex-shader/vert-dp4.sh
src/gallium/tests/graw/vertex-shader/vert-dst.sh
src/gallium/tests/graw/vertex-shader/vert-ex2.sh
src/gallium/tests/graw/vertex-shader/vert-flr.sh
src/gallium/tests/graw/vertex-shader/vert-frc.sh
src/gallium/tests/graw/vertex-shader/vert-lg2.sh
src/gallium/tests/graw/vertex-shader/vert-lit.sh
src/gallium/tests/graw/vertex-shader/vert-lrp.sh
src/gallium/tests/graw/vertex-shader/vert-mad.sh
src/gallium/tests/graw/vertex-shader/vert-max.sh
src/gallium/tests/graw/vertex-shader/vert-min.sh
src/gallium/tests/graw/vertex-shader/vert-mov.sh
src/gallium/tests/graw/vertex-shader/vert-mul.sh
src/gallium/tests/graw/vertex-shader/vert-rcp.sh
src/gallium/tests/graw/vertex-shader/vert-rsq.sh
src/gallium/tests/graw/vertex-shader/vert-sge.sh
src/gallium/tests/graw/vertex-shader/vert-slt.sh
src/gallium/tests/graw/vertex-shader/vert-srcmod-abs.sh
src/gallium/tests/graw/vertex-shader/vert-srcmod-absneg.sh
src/gallium/tests/graw/vertex-shader/vert-srcmod-neg.sh
src/gallium/tests/graw/vertex-shader/vert-srcmod-swz.sh
src/gallium/tests/graw/vertex-shader/vert-sub.sh
src/gallium/tests/graw/vertex-shader/vert-xpd.sh
src/gallium/tools/trace/dump.py
src/gallium/tools/trace/format.py
src/gallium/tools/trace/model.py
src/gallium/tools/trace/parse.py

[g3dvl] cleanup error handling

scons: Build classic mesa gdi driver.

Build as

  scons platform=windows mesagdi

[g3dvl] make ycbcr stream and block data a public interface

osmesa: Fix Mingw build.

Build as

  scons platform=windows osmesa

scons: Build osmesa.

Just type

  scons osmesa

mapi/glapi: Avoid warnings about visibility being unsupported on MinGW.

mesa/vf: Remove.

Unused.  Probably replaced by translate module.

graw: Replace dead symlinks to delete python statetracker shaders with their contents.

[g3dvl] start implementing zscan and quantification

Not 100% complete, but at least a good start.

draw: fix point/line/tri flushing bug in vbuf code

Need to reset the point/line/tri functions to point to the "first"
versions whenever we flush vertices.  Fixes unfilled polygon rendering
errors seen in demos/samples/logo.c.  See comments for more info.

NOTE: This is a candidate for the 7.10 branch.

r300g: do not advertise color_buffer_float on r300 and r400

It seems they can't do unclamped vertex colors. Tested on RV350.

r300/compiler: fix up error message

egl/wayland: Follow buffer.attach removal

egl/wayland: Update to buffer.damage addition

configure.ac: fix up the help text for --enable-gallium-r300

r300g: disallow A16F,L16F,LA16F,I16F texture formats on DRM<2.8.0

st/egl: Update to wayland protocol,egl changes

egl_dri2: Create wl_buffers for pixmap surfaces

Needed since wayland-egl doesnt do this anymore.

egl/wayland: Add new version parameter for post_drm

egl/wayland: Fix prefixes for interface names

swrast: fix readpix clamping

Broken with e5c6a92a12b5cd7db205d72039f58d302b0be9d5. (ARB_color_buffer_float)

Clamping should occur if type != float, otherwise the MSBs of the resulting
pixels are killed off. For example, reading back LUMINANCE = R+G+B can be
greater than 0xff, but the result is naturally masked by 0xff
for UNSIGNED_BYTE, leading to bogus results.

The following bug report seems to want clamping to occur if type == half_float
too. Not sure what's correct.

Bug: [bisected pineview] oglc case pxconv-read failed
https://bugs.freedesktop.org/show_bug.cgi?id=35852

Tested by: Fang Xun <xunx.fang@intel.com>
Reviewed-and-tested-by: Ian Romanick <ian.d.romanick@intel.com>

WL_bind_wayland_display: clarify that context must be NULL

i915: Gut all remaining bits of hardware fog

None of this ever gets used.  Fog is always calculated by a fragment
program.  Even though the fixed-function fog unit is never used, state
updates are still sent to the hardware.  Removing those spurious state
updates can't hurt performance.

Reviewed-by: Eric Anholt <eric@anholt.net>
Acked-by: Corbin Simpson <MostAwesomeDude@gmail.com>
Acked-by: Alex Deucher <alexdeucher@gmail.com>

i915: i915_context::vertex_fog is always I915_FOG_NONE, so kill it

Reviewed-by: Eric Anholt <eric@anholt.net>
Acked-by: Corbin Simpson <MostAwesomeDude@gmail.com>
Acked-by: Alex Deucher <alexdeucher@gmail.com>

i915: There's always a fragment program

Fragment programs are generated by core Mesa for fixed-function.
Because of this, there's no reason to handle cases where there is no
fragment program for fog.

Reviewed-by: Eric Anholt <eric@anholt.net>
Acked-by: Corbin Simpson <MostAwesomeDude@gmail.com>
Acked-by: Alex Deucher <alexdeucher@gmail.com>

i915: Delete disabled try_pixel_fog paths

Reviewed-by: Eric Anholt <eric@anholt.net>
Acked-by: Corbin Simpson <MostAwesomeDude@gmail.com>
Acked-by: Alex Deucher <alexdeucher@gmail.com>

mesa: Kill gl_fragment_program::FogOption with fire

All drivers expect this to always be GL_NONE.  Don't let there be any
opportunity for a bad value to leak out and infect some unsuspecting
driver.  If any driver for hardware that had fixed-function
per-fragment fog (i915 and perhaps some r300-ish) was ever going to
add support, it would have done it by now.

Reviewed-by: Eric Anholt <eric@anholt.net>
Acked-by: Corbin Simpson <MostAwesomeDude@gmail.com>
Acked-by: Alex Deucher <alexdeucher@gmail.com>

i915: gl_fragment_program::FogOption is always GL_NONE so don't check it

Reviewed-by: Eric Anholt <eric@anholt.net>
Acked-by: Corbin Simpson <MostAwesomeDude@gmail.com>
Acked-by: Alex Deucher <alexdeucher@gmail.com>

i965: gl_fragment_program::FogOption is always GL_NONE so don't check it

Reviewed-by: Eric Anholt <eric@anholt.net>
Acked-by: Corbin Simpson <MostAwesomeDude@gmail.com>
Acked-by: Alex Deucher <alexdeucher@gmail.com>

mesa: gl_fragment_program::FogOption is always GL_NONE so don't check it

Reviewed-by: Eric Anholt <eric@anholt.net>
Acked-by: Corbin Simpson <MostAwesomeDude@gmail.com>
Acked-by: Alex Deucher <alexdeucher@gmail.com>

mesa: Fix bugs in ff fragment shader fog handling

This patch fixes two bugs related to fog in the fixed-function
fragment shader generation code.

Fog was only lowered to instructions if MRTs were used.  The fragment
shader assembler always lowers "fog option" code to instructions, and
many drivers (e.g., r300) expect this.

When fog lowering did happen, it was after the instruction count was
checked against implementation limits.  Since fog lowering may add up
to 5 instructions, a program that was below the limits before lowering
may exceed the limits after lowering.

NOTE: This is a candidate for the stable branches.

Reviewed-by: Eric Anholt <eric@anholt.net>
Acked-by: Corbin Simpson <MostAwesomeDude@gmail.com>
Acked-by: Alex Deucher <alexdeucher@gmail.com>

st/mesa: check image size before copy_image_data_to_texture()

We should only copy images into the dest texture if the size is correct.
This fixes a failed assertion when finalizing a texture with mis-defined
mipmap levels such as:
level 0: 32x32
level 1: 8x8

Also, fix incorrect mipmap level used in assertion at the top of
copy_image_data_to_texture().

NOTE: This is a candidate for the 7.10 branch.

galahad,util: warn on resource target mismatch in copy_region

Reviewed-by: Jakob Bornecrantz <jakob@vmware.com>

st/mesa: implement CopyBufferSubData using resource_copy_region

Reviewed-by: Jakob Bornecrantz <jakob@vmware.com>

gallium: add fallback for copying buffers to all drivers

Just to keep drivers working.

Reviewed-by: Jakob Bornecrantz <jakob@vmware.com>

util: add a simple memcpy path for copying buffers in util_resource_copy_region

Reviewed-by: Jakob Bornecrantz <jakob@vmware.com>

glsl/ast_to_hir: Only use the local 'type' variable as a temporary.

Lots of code (deleted by this patch) tried to make type == result->type,
but not all cases did.  Don't pretend; just use result->type.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>

i965: Remove dead vertex buffer structs.

We do this OUT_BATCH-style in brw_draw_upload.c.

wayland-egl: Update to wayland-egl.h changes, drop struct wl_egl_display

mesa: fix void pointer arithmetic warning

docs: Mention that GLSL 1.30 is started.

Things definitely remaining todo: switch statements, clip distances.
On 965, we also need real integers in the VS, and implementations of
some things like isinf/isnan.

Reviewed-by: Brian Paul <brianp@vmware.com>

docs: Update for Intel ARB_texture_float and ARB_color_buffer_float.

Reviewed-by: Brian Paul <brianp@vmware.com>

intel: Add support for ARB_color_buffer_float.

Reviewed-by: Brian Paul <brianp@vmware.com>

meta: Add support for ARB_color_buffer_float to _mesa_meta_Clear().

Tested with piglit arb_color_buffer_float-clear.

Reviewed-by: Brian Paul <brianp@vmware.com>

meta: Add support for ARB_color_buffer_float to _mesa_meta_DrawPixels.

Tested with piglit arb_color_buffer_float-drawpixels.

Reviewed-by: Brian Paul <brianp@vmware.com>

intel: Add support for ARB_texture_float.

For 1 and 2-channel formats the hardware only supports rendering to R
and RG.  To do I and L render targets we just call them R and
everything works out.  For A, we would need to rewrite the CC to do
the alpha channel's blending on color instead, and send the fragment
alpha down the red channel.  For LA, there doesn't seem to be any
hope, because we can't do independent color/alpha blending while
treating the LA surface as RG.

Reviewed-by: Brian Paul <brianp@vmware.com>

intel: Add support for blit copies of >32bpp formats.

The blitter only does up 32bpp at a time, so we handle it by mangling
coordinates and calling the surface 32bpp.

Fixes ARB_texture_rg/fbo-generatemipmap-formats-float with ARB_texture_float.

Reviewed-by: Brian Paul <brianp@vmware.com>

mesa: Add renderbuffer accessors for A, I, L, FLOAT32.

Of these, intel will be using I and L initially, and A once we rewrite
fragment shaders and the CC for rendering to it as R.

Reviewed-by: Brian Paul <brianp@vmware.com>

mesa: Add renderbuffer accessors for MESA_FORMAT_R_FLOAT32, RG_FLOAT32.

Reviewed-by: Brian Paul <brianp@vmware.com>

mesa: Add ChooseTexFormat fallbacks for floating point textures.

This covers X_FLOAT16 -> X_FLOAT32, and X -> RGBA_FLOAT32.

Reviewed-by: Brian Paul <brianp@vmware.com>

mesa: Add renderbuffer accessors for MESA_FORMAT_RGBA_FLOAT32.

Reviewed-by: Brian Paul <brianp@vmware.com>

swrast: Add LUMINANCE, INTENSITY, LUMINANCE_ALPHA to span asserts.

Fixes ARB_texture_float/fbo-alphatest-formats.

Reviewed-by: Brian Paul <brianp@vmware.com>

r600g: fix userspace fences again

reinstate b7617346dcff50a66a10c61b95c33682cf629c9e after the
rework in 6067a2a67f9a7aab2aee051469bea8af03747a95.

Signed-off-by: Alex Deucher <alexdeucher@gmail.com>

[g3dvl] make motion vector buffers a public interface

r600g: don't flush the dest caches on every draw

Keep track of when the caches are dirty, and only flush them when
the framebuffer state is set and when the context is flushed.

Signed-off-by: Dave Airlie <airlied@redhat.com>

translate: disable clamping of instanced array indexes

This fixes piglit's draw-instanced-divisor test for softpipe on both
the generic and SSE paths.  This is temporary until we have the
correct per-array max_index information.

[g3dvl] revert commit 310eea52ca1e997295c84163066cc5d0fd4f8cf6

Using a seperate vertex buffer for mc and ycbcr handling is still better.

r600g: add evergreen+ big endian support

Based on Cédric's r6xx/r7xx patch.

Signed-off-by: Alex Deucher <alexdeucher@gmail.com>

r600g: add big endian support for r6xx/r7xx

Signed-off-by: Cedric Cano <ccano@interfaceconcept.com>
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>

r600c: add evergreen big endian support

Based on Cedric's r6xx/r7xx patch.

Signed-off-by: Alex Deucher <alexdeucher@gmail.com>

r600c: add big endian support for r6xx/r7xx

Signed-off-by: Cedric Cano <ccano@interfaceconcept.com>
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>

Makefile: add missing Scons files

Makefile: add missing .cpp, .ll, .yy files

Signed-off-by: Brian Paul <brianp@vmware.com>

r600g: add cb support for snorm formats.

Check for signed type and enable SNORM.

Signed-off-by: Dave Airlie <airlied@redhat.com>

configure.ac: rename --enable-gallium-radeon to --enable-gallium-r300

Also fix up the help string for both r300 and r600.

r200: enable some extensions

Such as:
- GL_ARB_half_float_pixel
- GL_ARB_vertex_array_object
- GL_APPLE_vertex_array_object
- GL_EXT_gpu_program_parameters