mesa.git
14 years agonv50: bswap32 the polygon stipple pattern
Christoph Bumiller [Thu, 26 Nov 2009 15:59:39 +0000 (16:59 +0100)]
nv50: bswap32 the polygon stipple pattern

The hardware wants the pattern the same way it is
passed to glPolygonStipple.

14 years agogallium/util: added util_bswap32()
Christoph Bumiller [Thu, 26 Nov 2009 15:58:59 +0000 (16:58 +0100)]
gallium/util: added util_bswap32()

14 years agosvga: fix for not using texture width/height/depth arrays
Roland Scheidegger [Fri, 27 Nov 2009 16:41:42 +0000 (17:41 +0100)]
svga: fix for not using texture width/height/depth arrays

14 years agor300g,llvmpipe: fix some more merge problems
Roland Scheidegger [Fri, 27 Nov 2009 16:40:24 +0000 (17:40 +0100)]
r300g,llvmpipe: fix some more merge problems

14 years agoMerge branch 'width0'
Roland Scheidegger [Fri, 27 Nov 2009 16:15:47 +0000 (17:15 +0100)]
Merge branch 'width0'

Conflicts:
src/gallium/drivers/r300/r300_texture.c
src/gallium/state_trackers/xorg/xorg_exa.c
src/mesa/state_tracker/st_cb_texture.c

14 years agogallium: fix more statetrackers/drivers for not using texture width/height/depth...
Roland Scheidegger [Thu, 26 Nov 2009 21:49:58 +0000 (22:49 +0100)]
gallium: fix more statetrackers/drivers for not using texture width/height/depth arrays

14 years agotgsi/exec: Fix orientation of DDY.
Michal Krol [Thu, 26 Nov 2009 19:38:43 +0000 (20:38 +0100)]
tgsi/exec: Fix orientation of DDY.

14 years agotgsi/exec: Force return from a subroutine at ENDSUB.
Michal Krol [Thu, 26 Nov 2009 19:30:04 +0000 (20:30 +0100)]
tgsi/exec: Force return from a subroutine at ENDSUB.

14 years agodraw: Fix max_index check.
Keith Whitwell [Thu, 26 Nov 2009 18:24:57 +0000 (19:24 +0100)]
draw: Fix max_index check.

We want to fallback to draw splitting when vertex element indices
might be too high for atomic draw path (currently limited to 4095).

14 years agotgsi/ureg: Add forgotten goto in ureg_DECL_constant().
Michal Krol [Thu, 26 Nov 2009 18:21:55 +0000 (19:21 +0100)]
tgsi/ureg: Add forgotten goto in ureg_DECL_constant().

14 years agollvmpipe: Fake missing SSSE3 when simulation less capabable machines.
José Fonseca [Thu, 26 Nov 2009 11:16:19 +0000 (11:16 +0000)]
llvmpipe: Fake missing SSSE3 when simulation less capabable machines.

SSE3 != SSSE3 and so far we only use the later.

14 years agollvmpipe: Update/correct CPU requirements.
José Fonseca [Thu, 26 Nov 2009 11:15:08 +0000 (11:15 +0000)]
llvmpipe: Update/correct CPU requirements.

There are no hard requirements at the moment.

We don't really emit any sse3 yet. Just some ssse3.

Thanks to Roland for spotting these incorrections.

14 years agoMerge commit 'origin/st-shader-varients'
Roland Scheidegger [Thu, 26 Nov 2009 00:15:25 +0000 (01:15 +0100)]
Merge commit 'origin/st-shader-varients'

Conflicts:
src/mesa/state_tracker/st_atom_shader.c
src/mesa/state_tracker/st_program.c

14 years agoscons: Make it work with MinGW build of LLVM 2.6.
José Fonseca [Wed, 25 Nov 2009 18:06:12 +0000 (18:06 +0000)]
scons: Make it work with MinGW build of LLVM 2.6.

LLVM 2.5 is no longer supported on windows.

14 years agor600: add ARB_texture_non_power_of_two support.
Dave Airlie [Wed, 25 Nov 2009 05:45:31 +0000 (15:45 +1000)]
r600: add ARB_texture_non_power_of_two support.

This makes the miptree rounds up to the near POT for each level for
all radeons, however since mipmaps aren't support with NPOT on previous
radeons this calculation shouldn't cause any problems. If it does
we can just make it r600 only.

I tested a few mipmap demos on r500 and they all seem to work.

Signed-off-by: Dave Airlie <airlied@redhat.com>
14 years agor600 : reset stack flag with one channel only.
Richard Li [Tue, 24 Nov 2009 21:00:25 +0000 (16:00 -0500)]
r600 : reset stack flag with one channel only.

14 years agor600 : fix stack depth setting bug.
Richard Li [Tue, 24 Nov 2009 17:16:39 +0000 (12:16 -0500)]
r600 : fix stack depth setting bug.

14 years agollvmpipe: Update instructions.
José Fonseca [Tue, 24 Nov 2009 16:01:01 +0000 (16:01 +0000)]
llvmpipe: Update instructions.

14 years agosvga: Handle comment tokens when dumping.
José Fonseca [Sun, 15 Nov 2009 20:14:03 +0000 (12:14 -0800)]
svga: Handle comment tokens when dumping.

14 years agosvga: Use consistent names for public symbol names of shader dumping facilities.
José Fonseca [Tue, 24 Nov 2009 14:43:30 +0000 (14:43 +0000)]
svga: Use consistent names for public symbol names of shader dumping facilities.

14 years agosvga: Use consistent file names for dumping facilities.
José Fonseca [Tue, 24 Nov 2009 14:37:45 +0000 (14:37 +0000)]
svga: Use consistent file names for dumping facilities.

14 years agoutil: Describe a few more formats.
José Fonseca [Tue, 24 Nov 2009 14:24:38 +0000 (14:24 +0000)]
util: Describe a few more formats.

14 years agollvmpipe: Use assert instead of abort. Only verify functions on debug builds.
José Fonseca [Mon, 23 Nov 2009 11:21:11 +0000 (11:21 +0000)]
llvmpipe: Use assert instead of abort. Only verify functions on debug builds.

14 years agollvmpipe: Fix memory leak.
José Fonseca [Sun, 15 Nov 2009 14:46:48 +0000 (06:46 -0800)]
llvmpipe: Fix memory leak.

14 years agollvmpipe: Use the generic conversion routine for depths.
José Fonseca [Wed, 11 Nov 2009 01:52:53 +0000 (17:52 -0800)]
llvmpipe: Use the generic conversion routine for depths.

This allows for z32f depth format to work correctly.

14 years agollvmpipe: Be more conservative with the supported formats.
José Fonseca [Wed, 11 Nov 2009 01:51:06 +0000 (17:51 -0800)]
llvmpipe: Be more conservative with the supported formats.

We'll likely support much more formats, but doing this allows to run
more testsuites without immediately hit assertion failures.

14 years agowgl: Fix copy'n'paste typo in comment.
José Fonseca [Wed, 11 Nov 2009 00:56:43 +0000 (16:56 -0800)]
wgl: Fix copy'n'paste typo in comment.

14 years agollvmpipe: Fix typo in comparison operator.
José Fonseca [Wed, 11 Nov 2009 00:55:44 +0000 (16:55 -0800)]
llvmpipe: Fix typo in comparison operator.

14 years agotgsi: Document Declaration Semantic token and FACE semantic name.
Michal Krol [Tue, 24 Nov 2009 10:58:01 +0000 (11:58 +0100)]
tgsi: Document Declaration Semantic token and FACE semantic name.

14 years agoslang: Fix allocation size.
Michal Krol [Tue, 24 Nov 2009 10:22:03 +0000 (11:22 +0100)]
slang: Fix allocation size.

We don't need 16K+ to store a single pointer.

14 years agomesa: Fix pointer arithmetic.
Michal Krol [Tue, 24 Nov 2009 10:17:16 +0000 (11:17 +0100)]
mesa: Fix pointer arithmetic.

14 years agotgsi: Implement predicated instructions in exec.
Michal Krol [Tue, 24 Nov 2009 08:03:41 +0000 (09:03 +0100)]
tgsi: Implement predicated instructions in exec.

14 years agotgsi: Add ureg_DECL_loop().
Michal Krol [Tue, 24 Nov 2009 08:02:29 +0000 (09:02 +0100)]
tgsi: Add ureg_DECL_loop().

14 years agotgsi: Account for gallium shader token representation changes.
Michal Krol [Tue, 24 Nov 2009 08:01:48 +0000 (09:01 +0100)]
tgsi: Account for gallium shader token representation changes.

14 years agogallium: Refactor the instruction predicate TGSI token.
Michal Krol [Tue, 24 Nov 2009 07:55:40 +0000 (08:55 +0100)]
gallium: Refactor the instruction predicate TGSI token.

Rename it to tgsi_instruction_predicate -- it's no longer an extended
token. Its presence is indicated by a new flag in tgsi_instruction that
indicates whether an instruction is predicated.

Also, change predicate index representation to match the other
tokens that specify register indices.

14 years agor300: fix swtcl bo leak problem.
Dave Airlie [Tue, 24 Nov 2009 01:56:45 +0000 (11:56 +1000)]
r300: fix swtcl bo leak problem.

We can get a lot of swtcl bo allocations - need to probably abstract
this a bit further.

Signed-off-by: Dave Airlie <airlied@redhat.com>
14 years agoradeon/r200/r300/r600: make bo mapping be explicit
Dave Airlie [Tue, 24 Nov 2009 00:49:34 +0000 (10:49 +1000)]
radeon/r200/r300/r600: make bo mapping be explicit

This moves the bo mapping outside the DMA layer and makes it explicit,
this should in theory make it simpler to split the clean up the dma/cmdbuf
linkage that I created before that is broken.

Tested on: r600, rv380 (tcl/no-tcl), rv200 (tcl/no-tcl)

Signed-off-by: Dave Airlie <airlied@redhat.com>
14 years agomesa: use gcc __builtin_popcount()
Brian Paul [Tue, 24 Nov 2009 01:09:46 +0000 (18:09 -0700)]
mesa: use gcc __builtin_popcount()

14 years agor300g: use util_bitcount()
Brian Paul [Tue, 24 Nov 2009 01:06:19 +0000 (18:06 -0700)]
r300g: use util_bitcount()

14 years agoegl: use util_bitcount()
Brian Paul [Tue, 24 Nov 2009 01:04:47 +0000 (18:04 -0700)]
egl: use util_bitcount()

14 years agogallium/util: added util_bitcount()
Brian Paul [Tue, 24 Nov 2009 01:04:22 +0000 (18:04 -0700)]
gallium/util: added util_bitcount()

14 years agoMerge commit 'origin/mesa_7_7_branch'
Maciej Cencora [Mon, 23 Nov 2009 21:14:48 +0000 (22:14 +0100)]
Merge commit 'origin/mesa_7_7_branch'

14 years agoradeon: fix errors in miptree related function
Maciej Cencora [Mon, 23 Nov 2009 20:59:08 +0000 (21:59 +0100)]
radeon: fix errors in miptree related function

- typo
- memory leak
- off by one (spotted by airlied)

14 years agoMerge branch 'mesa_7_7_branch' of http://anongit.freedesktop.org/git/mesa/mesa into...
Maciej Cencora [Mon, 23 Nov 2009 20:28:22 +0000 (21:28 +0100)]
Merge branch 'mesa_7_7_branch' of anongit.freedesktop.org/git/mesa/mesa into mesa_7_7_branch

14 years agosvga: Scrub Makefiles a bit
Jakob Bornecrantz [Mon, 23 Nov 2009 18:59:02 +0000 (19:59 +0100)]
svga: Scrub Makefiles a bit

Remove x86 specific hacks. Not that they will ever be used on
none x86 arches, but they are built by default. And the way the
flags where added was a hack.

14 years agoslang: Check return value from emit_instruction().
Michal Krol [Mon, 23 Nov 2009 12:22:04 +0000 (13:22 +0100)]
slang: Check return value from emit_instruction().

14 years agotgsi: Clamp the source argument in micro_exp2() to avoid Inf.
Michal Krol [Mon, 23 Nov 2009 10:32:58 +0000 (11:32 +0100)]
tgsi: Clamp the source argument in micro_exp2() to avoid Inf.

14 years agotgsi: Remove code that actually had no effect.
Michal Krol [Mon, 23 Nov 2009 10:29:29 +0000 (11:29 +0100)]
tgsi: Remove code that actually had no effect.

14 years agosoftpipe: Initialise TGSI machine's Face.
Michal Krol [Mon, 23 Nov 2009 09:51:07 +0000 (10:51 +0100)]
softpipe: Initialise TGSI machine's Face.

14 years agotgsi: Fix POSITION and FACE fragment shader inputs.
Michal Krol [Mon, 23 Nov 2009 09:49:41 +0000 (10:49 +0100)]
tgsi: Fix POSITION and FACE fragment shader inputs.

14 years agotgsi: Bring BGNFOR/ENDFOR implementation up to spec.
Michal Krol [Mon, 23 Nov 2009 09:40:05 +0000 (10:40 +0100)]
tgsi: Bring BGNFOR/ENDFOR implementation up to spec.

14 years agotgsi: Fake TXD implementation in exec.
Michal Krol [Thu, 19 Nov 2009 12:05:58 +0000 (13:05 +0100)]
tgsi: Fake TXD implementation in exec.

14 years agotgsi: Add execution debugging facilities to exec.
Michal Krol [Thu, 19 Nov 2009 12:01:08 +0000 (13:01 +0100)]
tgsi: Add execution debugging facilities to exec.

14 years agoslang: Be more robust with memory in concat_shaders().
Michal Krol [Thu, 19 Nov 2009 11:52:58 +0000 (12:52 +0100)]
slang: Be more robust with memory in concat_shaders().

14 years agomesa: Initialize variable in _mesa_get_texel_fetch_func.
Vinson Lee [Mon, 23 Nov 2009 06:52:59 +0000 (01:52 -0500)]
mesa: Initialize variable in _mesa_get_texel_fetch_func.

14 years agoglx: Prevent potential null pointer deference in driCreateContext.
Vinson Lee [Mon, 23 Nov 2009 06:30:32 +0000 (01:30 -0500)]
glx: Prevent potential null pointer deference in driCreateContext.

14 years agoglu/sgi: Fix memory leak in bitmapBuild2DMipmaps.
Vinson Lee [Mon, 23 Nov 2009 06:09:06 +0000 (01:09 -0500)]
glu/sgi: Fix memory leak in bitmapBuild2DMipmaps.

14 years agoglu/sgi: Fix memory leak in gluScaleImage3D.
Vinson Lee [Mon, 23 Nov 2009 05:57:37 +0000 (00:57 -0500)]
glu/sgi: Fix memory leak in gluScaleImage3D.

14 years agor600 : add support for shader instruction trunc and discard.
Richard Li [Mon, 23 Nov 2009 02:31:46 +0000 (21:31 -0500)]
r600 : add support for shader instruction trunc and discard.

14 years agor600: hopefully fix segfault.
Dave Airlie [Sun, 22 Nov 2009 20:44:29 +0000 (06:44 +1000)]
r600: hopefully fix segfault.

14 years agor600: fix inline issues
Dave Airlie [Sun, 22 Nov 2009 20:31:29 +0000 (06:31 +1000)]
r600: fix inline issues

14 years agor600 : add stack depth calculation, enable CF pop.
Richard Li [Sun, 22 Nov 2009 17:58:28 +0000 (12:58 -0500)]
r600 : add stack depth calculation, enable CF pop.

14 years agor600 : use cf for all pop now, left optimization for future.
Richard Li [Fri, 20 Nov 2009 21:36:55 +0000 (16:36 -0500)]
r600 : use cf for all pop now, left optimization for future.

14 years agor300: fix VP source conflict resolution on 64-bit machines
Maciej Cencora [Sun, 22 Nov 2009 14:08:46 +0000 (15:08 +0100)]
r300: fix VP source conflict resolution on 64-bit machines

On 32bit machines we were lucky because the sizeof(reg) == sizeof(rc_src_register).
On 64bit machines pointers are 8 bytes long, so we were overwriting other data.

14 years agor300: fix SIN/COS/SCS instructions for R300 fp
Maciej Cencora [Sun, 22 Nov 2009 13:13:18 +0000 (14:13 +0100)]
r300: fix SIN/COS/SCS instructions for R300 fp

14 years agoglu/sgi: Fix memory leak in gluScaleImage.
Vinson Lee [Sun, 22 Nov 2009 06:57:35 +0000 (01:57 -0500)]
glu/sgi: Fix memory leak in gluScaleImage.

14 years agoglu: Fix memory leak in __gl_meshMakeEdge.
Vinson Lee [Sun, 22 Nov 2009 06:45:53 +0000 (01:45 -0500)]
glu: Fix memory leak in __gl_meshMakeEdge.

14 years agosvga: Prevent potential null pointer deference in vmw_surface.c.
Vinson Lee [Sun, 22 Nov 2009 06:26:32 +0000 (01:26 -0500)]
svga: Prevent potential null pointer deference in vmw_surface.c.

14 years agosvga: Fix memory leak in vmw_screen_ioctl.c
Vinson Lee [Sun, 22 Nov 2009 06:20:07 +0000 (01:20 -0500)]
svga: Fix memory leak in vmw_screen_ioctl.c

14 years agost/xorg: Prevent potential null pointer deference in xorg_xv.c.
Vinson Lee [Sun, 22 Nov 2009 06:10:38 +0000 (01:10 -0500)]
st/xorg: Prevent potential null pointer deference in xorg_xv.c.

14 years agost/xorg: Reorder cases in switch statement.
Vinson Lee [Sun, 22 Nov 2009 05:55:52 +0000 (00:55 -0500)]
st/xorg: Reorder cases in switch statement.

Silences missing break statement warning.

14 years agor300g: Build fix.
Corbin Simpson [Sun, 22 Nov 2009 03:13:26 +0000 (19:13 -0800)]
r300g: Build fix.

Oops.

14 years agor300g: do not reallocate r300_vertex_info and r300_rs_block all the time
Marek Olšák [Fri, 20 Nov 2009 03:52:49 +0000 (04:52 +0100)]
r300g: do not reallocate r300_vertex_info and r300_rs_block all the time

14 years agor300g: clean up vs/fs tabs
Marek Olšák [Sat, 21 Nov 2009 04:51:13 +0000 (05:51 +0100)]
r300g: clean up vs/fs tabs

Instead of vs_tab, we use vs_output_tab and it's local now. fs_tab hasn't
been used anywhere, so I removed it and r300_update_fs_tab too.

14 years agoradeon: fix compressed mipmapped textures
Maciej Cencora [Sat, 21 Nov 2009 14:56:23 +0000 (15:56 +0100)]
radeon: fix compressed mipmapped textures

Tested on r300 only, other cards may require adjusting texture_compressed_row_align.

14 years agoradeon: fix glCompressedTexSubImage
Maciej Cencora [Sat, 21 Nov 2009 14:56:02 +0000 (15:56 +0100)]
radeon: fix glCompressedTexSubImage

14 years agoradeon: fix compressed mipmapped textures
Maciej Cencora [Sat, 21 Nov 2009 14:56:23 +0000 (15:56 +0100)]
radeon: fix compressed mipmapped textures

Tested on r300 only, other cards may require adjusting texture_compressed_row_align.

14 years agoradeon: fix glCompressedTexSubImage
Maciej Cencora [Sat, 21 Nov 2009 14:56:02 +0000 (15:56 +0100)]
radeon: fix glCompressedTexSubImage

14 years agointel: make CopyTex[Sub]Image fallback debug messages more consistent
Roland Scheidegger [Sat, 21 Nov 2009 12:39:30 +0000 (04:39 -0800)]
intel: make CopyTex[Sub]Image fallback debug messages more consistent

14 years agomesa: handle different RowStride in _mesa_get_compressed_teximage
Roland Scheidegger [Sat, 21 Nov 2009 12:22:50 +0000 (04:22 -0800)]
mesa: handle different RowStride in _mesa_get_compressed_teximage

drivers storing compressed textures with non-native stride but relying on
_mesa_get_compressed_teximage for GetCompressedTexImage otherwise won't
work correctly (for instance i965 compressed mipmaps).

14 years agoi965: Fix several memory leaks on exit.
Eric Anholt [Sat, 21 Nov 2009 12:33:21 +0000 (13:33 +0100)]
i965: Fix several memory leaks on exit.

Bug #25194.

14 years agor300g: silence warnings
Marek Olšák [Thu, 19 Nov 2009 21:40:11 +0000 (22:40 +0100)]
r300g: silence warnings

14 years agor300g: fix typo in r300_reg.h to prevent the RS unit from doing random things
Marek Olšák [Thu, 19 Nov 2009 19:41:19 +0000 (20:41 +0100)]
r300g: fix typo in r300_reg.h to prevent the RS unit from doing random things

And reorder fragment shader inputs so that the colors are before texcoords,
as is allocated by the shader compiler. This commit makes VS->FS attribute
routing work on R500.

14 years agor300g: fix updating a vertex format
Marek Olšák [Sun, 15 Nov 2009 15:41:25 +0000 (16:41 +0100)]
r300g: fix updating a vertex format

We must update PSC when we change the vertex format, e.g. vertex colors
from RGBA to BGRA.

14 years agor300g: add texture lod clamping
Marek Olšák [Sat, 14 Nov 2009 22:27:20 +0000 (23:27 +0100)]
r300g: add texture lod clamping

These now work:
piglit/lodclamp
piglit/levelclamp

14 years agor300g: Begin separating HW TCL and SW TCL state and setup.
Corbin Simpson [Sun, 15 Nov 2009 04:25:15 +0000 (05:25 +0100)]
r300g: Begin separating HW TCL and SW TCL state and setup.

This patch removes draw_context entirely from the HW TCL path and cleans up
a few other things along the way. Hopefully, nothing got broken.

Thanks to Marek Olšák for testing, review, and pointing out my bugs. :3

14 years agost/egl: Fix memory leak in egl_tracker.c.
Vinson Lee [Sat, 21 Nov 2009 00:33:25 +0000 (16:33 -0800)]
st/egl: Fix memory leak in egl_tracker.c.

14 years agodraw: Initialize variable in draw_pt.c.
Vinson Lee [Sat, 21 Nov 2009 00:03:48 +0000 (16:03 -0800)]
draw: Initialize variable in draw_pt.c.

14 years agoprogs/xdemos: Fix memory leak in pbinfo.c.
Vinson Lee [Fri, 20 Nov 2009 23:52:53 +0000 (15:52 -0800)]
progs/xdemos: Fix memory leak in pbinfo.c.

14 years agoprogs/xdemos: Fix memory leak in pbdemo.c.
Vinson Lee [Fri, 20 Nov 2009 23:48:29 +0000 (15:48 -0800)]
progs/xdemos: Fix memory leak in pbdemo.c.

14 years agoprogs/xdemos: Fix memory leak in glxpbdemo.c.
Vinson Lee [Fri, 20 Nov 2009 23:18:19 +0000 (15:18 -0800)]
progs/xdemos: Fix memory leak in glxpbdemo.c.

14 years agor300g: emit R300_TEX_ENABLE to indicate there are no textures bound
Marek Olšák [Fri, 20 Nov 2009 04:17:00 +0000 (05:17 +0100)]
r300g: emit R300_TEX_ENABLE to indicate there are no textures bound

Previously, this reg wasn't emitted at all if texture_count == 0.

14 years agor300g: remove variant states from emit_state_invariant
Marek Olšák [Thu, 19 Nov 2009 20:07:20 +0000 (21:07 +0100)]
r300g: remove variant states from emit_state_invariant

14 years agoprogs/xdemos: Remove duplicate code in glxinfo.c.
Vinson Lee [Fri, 20 Nov 2009 23:06:56 +0000 (15:06 -0800)]
progs/xdemos: Remove duplicate code in glxinfo.c.

14 years agoprogs/xdemos: Add assert in corender.c.
Vinson Lee [Fri, 20 Nov 2009 23:03:16 +0000 (15:03 -0800)]
progs/xdemos: Add assert in corender.c.

14 years agor300g: Texrect factor cleanup.
Corbin Simpson [Fri, 20 Nov 2009 22:55:22 +0000 (14:55 -0800)]
r300g: Texrect factor cleanup.

(0, 0, 0, 1) is a much saner default value, and texrect factors only need
to be (1/s, 1/t, 0, 1).

14 years agor300g: fix rectangle textures on r3xx
Marek Olšák [Sat, 14 Nov 2009 21:14:42 +0000 (22:14 +0100)]
r300g: fix rectangle textures on r3xx

Adapted from Maciej Cencora's patch.

14 years agoi915g: Add missing break statement in i915_debug.c.
Vinson Lee [Fri, 20 Nov 2009 22:31:42 +0000 (14:31 -0800)]
i915g: Add missing break statement in i915_debug.c.

14 years agogallium/util: Initialize variables in u_pack_color.h.
Vinson Lee [Fri, 20 Nov 2009 22:20:15 +0000 (14:20 -0800)]
gallium/util: Initialize variables in u_pack_color.h.

14 years agor300g: Clean up bitcounting function.
Corbin Simpson [Fri, 20 Nov 2009 22:10:59 +0000 (14:10 -0800)]
r300g: Clean up bitcounting function.

I didn't see this in u_math; surely somebody else has this wheel
reinvented elsewhere.