OSDN Git Service

android-x86/external-mesa.git
12 years agor600g: always clear query memory
Alex Deucher [Tue, 7 Jun 2011 20:44:40 +0000 (16:44 -0400)]
r600g: always clear query memory

According to the hw documentation, the driver needs to:
- allocate 128 bits for each possible DB
- clear the 128 bits for each possible DB
- write 1 to bits 127 and 63 for upper DBs that don't
exist on a particular asic

Previously we were only doing these steps if the
asic had less than the max possible DBs.

Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
12 years agor300g: consolidate deducing chipset info
Marek Olšák [Tue, 7 Jun 2011 18:59:56 +0000 (20:59 +0200)]
r300g: consolidate deducing chipset info

Use the new PCI ID table, make it simpler.

12 years agor300_pci_ids: set families to match r300g
Marek Olšák [Tue, 7 Jun 2011 18:50:32 +0000 (20:50 +0200)]
r300_pci_ids: set families to match r300g

12 years agor600g: add missing r300 families
Marek Olšák [Tue, 7 Jun 2011 18:51:03 +0000 (20:51 +0200)]
r600g: add missing r300 families

Wondering why r600g needs to include r300_pci_ids.h

12 years agointel: Update intel-decode.c from intel-gpu-tools.
Eric Anholt [Mon, 6 Jun 2011 06:05:20 +0000 (23:05 -0700)]
intel: Update intel-decode.c from intel-gpu-tools.

12 years agointel: Implement glFinish() correctly by waiting on all previous rendering.
Eric Anholt [Tue, 31 May 2011 19:32:06 +0000 (12:32 -0700)]
intel: Implement glFinish() correctly by waiting on all previous rendering.

Before, we were waiting for (most of) the current framebuffer to be
done, which is not quite the same thing.

12 years agodarwin: Fix VG_LIB_GLOB to also match the unversioned symlink
Jeremy Huddleston [Tue, 7 Jun 2011 17:07:26 +0000 (13:07 -0400)]
darwin: Fix VG_LIB_GLOB to also match the unversioned symlink

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agodarwin: Don't link against libGL when building libOSMesa
Jeremy Huddleston [Tue, 7 Jun 2011 17:06:44 +0000 (13:06 -0400)]
darwin: Don't link against libGL when building libOSMesa

Everything should be resolved through glapi.

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agor300_pci_ids: add missing 4B48
Marek Olšák [Tue, 7 Jun 2011 16:45:23 +0000 (18:45 +0200)]
r300_pci_ids: add missing 4B48

Found in r300_chipset.h.

12 years agor300g: Remove is_r3xx
Benjamin Franzke [Mon, 6 Jun 2011 16:31:12 +0000 (18:31 +0200)]
r300g: Remove is_r3xx

Use r300_pci_ids.h instead.

Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
12 years agor600g: Use radeon pciid list for the family lookup table
Benjamin Franzke [Mon, 6 Jun 2011 11:13:01 +0000 (13:13 +0200)]
r600g: Use radeon pciid list for the family lookup table

Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
12 years agoradeon: Use pciid list to generate PCI_CHIP_<FAMILY>_<ID> defines
Benjamin Franzke [Mon, 6 Jun 2011 11:12:26 +0000 (13:12 +0200)]
radeon: Use pciid list to generate PCI_CHIP_<FAMILY>_<ID> defines

Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
12 years agotargets/egl: Support driver name lookup using pci lists
Benjamin Franzke [Mon, 30 May 2011 07:55:14 +0000 (09:55 +0200)]
targets/egl: Support driver name lookup using pci lists

Make use of this in drm and wayland st/egl backends.

Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
12 years agoegl_dri2: Use external driver pci list
Benjamin Franzke [Mon, 30 May 2011 08:49:55 +0000 (10:49 +0200)]
egl_dri2: Use external driver pci list

Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
12 years agoAdd radeon pci id lists
Benjamin Franzke [Thu, 26 May 2011 12:54:38 +0000 (14:54 +0200)]
Add radeon pci id lists

Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
12 years agoAdd intel pci id lists
Benjamin Franzke [Mon, 30 May 2011 09:58:32 +0000 (11:58 +0200)]
Add intel pci id lists

Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
12 years agotgsi: s/varient/variant/
Brian Paul [Tue, 7 Jun 2011 13:57:04 +0000 (07:57 -0600)]
tgsi: s/varient/variant/

12 years agodraw: rename draw_vs_varient.c to draw_vs_variant.c
Brian Paul [Tue, 7 Jun 2011 13:55:06 +0000 (07:55 -0600)]
draw: rename draw_vs_varient.c to draw_vs_variant.c

12 years agodraw/llvm: whitespace, formatting fixes
Brian Paul [Tue, 7 Jun 2011 13:31:34 +0000 (07:31 -0600)]
draw/llvm: whitespace, formatting fixes

12 years agodraw: s/1/TRUE/
Brian Paul [Mon, 6 Jun 2011 20:33:05 +0000 (14:33 -0600)]
draw: s/1/TRUE/

12 years agoi965: Fix flipped GT1 vs GT2 URB VS entry count limits.
Eric Anholt [Mon, 6 Jun 2011 06:20:57 +0000 (23:20 -0700)]
i965: Fix flipped GT1 vs GT2 URB VS entry count limits.

12 years agoMerge branch 'master' of ssh://git.freedesktop.org/git/mesa/mesa
Stéphane Marchesin [Tue, 7 Jun 2011 03:42:30 +0000 (20:42 -0700)]
Merge branch 'master' of ssh://git.freedesktop.org/git/mesa/mesa

12 years agoi915g: Do generic remapping.
Stéphane Marchesin [Tue, 7 Jun 2011 03:40:16 +0000 (20:40 -0700)]
i915g: Do generic remapping.

With complex shaders there are often "holes" in the fs inputs, and we only
have 8 tex coorsd to map those to. To fix this, we remap fs inputs to [0..8].
This lets us to run many more GLSL programs.

12 years agodarwin: Set VG_LIB_{NAME,GLOB} to fix make install
Jeremy Huddleston [Mon, 6 Jun 2011 23:54:55 +0000 (19:54 -0400)]
darwin: Set VG_LIB_{NAME,GLOB} to fix make install

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agor600g: optimise the draw emission packets for r600/eg
Dave Airlie [Mon, 6 Jun 2011 10:03:13 +0000 (20:03 +1000)]
r600g: optimise the draw emission packets for r600/eg

This just reduces code size a bit for this chunk.

Signed-off-by: Dave Airlie <airlied@redhat.com>
12 years agor600g: use an enabled list to track enabled blocks.
Dave Airlie [Thu, 2 Jun 2011 23:59:12 +0000 (09:59 +1000)]
r600g: use an enabled list to track enabled blocks.

At the end of flushing we were scanning over 450 blocks
with generally about 50 enabled. This reduces the scanning
to just the list of enabled blocks.

Signed-off-by: Dave Airlie <airlied@redhat.com>
12 years agor600g/winsys: overhaul resource range/blocks.
Dave Airlie [Mon, 6 Jun 2011 08:00:36 +0000 (18:00 +1000)]
r600g/winsys: overhaul resource range/blocks.

There isn't much point taking the overhead of range/block lookups on resources
we aren't going to be getting resource registers at wierd offsets.

Signed-off-by: Dave Airlie <airlied@redhat.com>
12 years agor600g: split out block init
Dave Airlie [Mon, 6 Jun 2011 04:33:23 +0000 (14:33 +1000)]
r600g: split out block init

This just splits this function up as pre-cursor to reusing
the internals of it.

Signed-off-by: Dave Airlie <airlied@redhat.com>
12 years agor600g: move resource setting to its own structures.
Dave Airlie [Fri, 3 Jun 2011 05:34:31 +0000 (15:34 +1000)]
r600g: move resource setting to its own structures.

resource setting could be a fair bit more lightweight,
this patch just separates the resource structs from the standard
reg tracking structs in the driver, later patches will improve
the winsys.

Signed-off-by: Dave Airlie <airlied@redhat.com>
12 years agor600g: only call bo loop if nbos is positive.
Dave Airlie [Mon, 6 Jun 2011 01:15:10 +0000 (11:15 +1000)]
r600g: only call bo loop if nbos is positive.

we don't need to loop over all the registers unless we have
some bos in the block, also avoid setting the ctx flags,
and move the optional stuff down below this chunk.

Signed-off-by: Dave Airlie <airlied@redhat.com>
12 years agor600g: don't need to call the packet dirty function if not dirty.
Dave Airlie [Fri, 3 Jun 2011 05:24:55 +0000 (15:24 +1000)]
r600g: don't need to call the packet dirty function if not dirty.

also fix a unneeded dirty check and add a dirty check speedup.

Signed-off-by: Dave Airlie <airlied@redhat.com>
12 years agoi915g: implement more opcodes.
Stéphane Marchesin [Mon, 6 Jun 2011 19:32:17 +0000 (12:32 -0700)]
i915g: implement more opcodes.

12 years agoi915g: implement TGSI_OPCODE_SEQ.
Stéphane Marchesin [Sat, 4 Jun 2011 01:57:16 +0000 (18:57 -0700)]
i915g: implement TGSI_OPCODE_SEQ.

12 years agoi915g: handle varyings properly.
Stéphane Marchesin [Sat, 4 Jun 2011 00:06:59 +0000 (17:06 -0700)]
i915g: handle varyings properly.

12 years agoi915g: add missing and unsupported PIPE_CAPs to avoid debug messages.
Stéphane Marchesin [Sat, 4 Jun 2011 00:03:37 +0000 (17:03 -0700)]
i915g: add missing and unsupported PIPE_CAPs to avoid debug messages.

12 years agodri2: protect dri2FlushFrontBuffer against NULL buffers.
Stéphane Marchesin [Sat, 4 Jun 2011 00:03:09 +0000 (17:03 -0700)]
dri2: protect dri2FlushFrontBuffer against NULL buffers.

12 years agoGallium: fix indentation in u_blitter.c
Stéphane Marchesin [Thu, 2 Jun 2011 03:51:59 +0000 (20:51 -0700)]
Gallium: fix indentation in u_blitter.c

12 years agoi915g: don't destroy a texture buffer if it's NULL.
Stéphane Marchesin [Wed, 1 Jun 2011 22:32:39 +0000 (15:32 -0700)]
i915g: don't destroy a texture buffer if it's NULL.

12 years agollvmpipe: use $(CXX) instead of g++ for linking.
Stéphane Marchesin [Wed, 1 Jun 2011 17:37:13 +0000 (10:37 -0700)]
llvmpipe: use $(CXX) instead of g++ for linking.

This allows setting the path to the C++ compiler.

12 years agoi965: Update SURFACE_STATE dumping for Ivybridge.
Kenneth Graunke [Tue, 17 May 2011 21:37:18 +0000 (14:37 -0700)]
i965: Update SURFACE_STATE dumping for Ivybridge.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
12 years agoi965: Update SAMPLER_STATE dumping for Ivybridge.
Kenneth Graunke [Tue, 17 May 2011 21:22:16 +0000 (14:22 -0700)]
i965: Update SAMPLER_STATE dumping for Ivybridge.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
12 years agoi965: Update SF_CLIP_VIEWPORT state dumping for Ivybridge.
Kenneth Graunke [Tue, 17 May 2011 21:28:20 +0000 (14:28 -0700)]
i965: Update SF_CLIP_VIEWPORT state dumping for Ivybridge.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
12 years agoapple: Package applegl source into MesaLib tarball
Jeremy Huddleston [Mon, 6 Jun 2011 16:48:08 +0000 (12:48 -0400)]
apple: Package applegl source into MesaLib tarball

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agodarwin: Define GALLIUM_DRIVERS_DIRS in darwin config
Jeremy Huddleston [Mon, 6 Jun 2011 15:38:05 +0000 (11:38 -0400)]
darwin: Define GALLIUM_DRIVERS_DIRS in darwin config

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agoapple: Finish build fixes for applegl_glx.c
Jeremy Huddleston [Mon, 6 Jun 2011 15:27:19 +0000 (11:27 -0400)]
apple: Finish build fixes for applegl_glx.c

libGL.dylib now *builds* on darwin.

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agoapple: More fixes for compilation failures
Jeremy Huddleston [Mon, 6 Jun 2011 15:15:32 +0000 (11:15 -0400)]
apple: More fixes for compilation failures

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agosoftpipe: Anisotropic filtering extension.
Andreas Faenger [Mon, 6 Jun 2011 07:13:16 +0000 (07:13 +0000)]
softpipe: Anisotropic filtering extension.

Reference implementation which produces high quality renderings.
Based on Higher Quality Elliptical Weighted Avarage Filter (EWA).

Signed-off-by: Brian Paul <brianp@vmware.com>
12 years agost/mesa: fix renderbuffer internal format for accum buffer
Brian Paul [Mon, 6 Jun 2011 14:06:34 +0000 (08:06 -0600)]
st/mesa: fix renderbuffer internal format for accum buffer

Fixes an accum buffer regression since switching to the table-based
format selection code.

12 years agoutil: add \n to debug_checkpoint_full
Marcin Slusarz [Mon, 16 May 2011 19:53:06 +0000 (21:53 +0200)]
util: add \n to debug_checkpoint_full

Signed-off-by: José Fonseca <jfonseca@vmware.com>
12 years agonv50: fix nv50_sampler_state_delete array overflow
Marcin Slusarz [Sun, 5 Jun 2011 19:00:28 +0000 (21:00 +0200)]
nv50: fix nv50_sampler_state_delete array overflow

num_samplers is array of 3 elements, not 5

12 years agoapple: Fix multiple build failures in applegl_glx.c
Jeremy Huddleston [Mon, 6 Jun 2011 01:19:40 +0000 (21:19 -0400)]
apple: Fix multiple build failures in applegl_glx.c

It is still not building, but this gets us many steps closer

See https://bugs.freedesktop.org/show_bug.cgi?id=29162

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agoapple: Build darwin using applegl rather than indirect
Jeremy Huddleston [Sun, 5 Jun 2011 22:50:55 +0000 (18:50 -0400)]
apple: Build darwin using applegl rather than indirect

This reverts portions of 6849916170c0275c13510251a7b217c20f2b993e that caused
the darwin config to fail to build due to missing implementations in that
commit.

See https://bugs.freedesktop.org/show_bug.cgi?id=29162

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agoglx: Dead code removal
Jeremy Huddleston [Sun, 5 Jun 2011 23:55:51 +0000 (19:55 -0400)]
glx: Dead code removal

Remove a redundant ifndef GLX_USE_APPLEGL

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agoapple: ifdef out come glapi-foo on darwin
Jeremy Huddleston [Sun, 5 Jun 2011 23:57:52 +0000 (19:57 -0400)]
apple: ifdef out come glapi-foo on darwin

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agoapple: Change from XExtDisplayInfo to struct glx_display
Jeremy Huddleston [Sun, 5 Jun 2011 23:26:19 +0000 (19:26 -0400)]
apple: Change from XExtDisplayInfo to struct glx_display

Fixes regression introduced by: ab434f6b7641a64d30725a9ac24929240362d466 and
                                c356f5867f2c1fad7155df538b9affa8dbdcf869

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agodri/nouveau: fix gnome-shell segfault
Ben Skeggs [Mon, 6 Jun 2011 00:17:59 +0000 (10:17 +1000)]
dri/nouveau: fix gnome-shell segfault

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
12 years agoapple: Rename GLXcontext
Jeremy Huddleston [Sun, 5 Jun 2011 22:22:47 +0000 (18:22 -0400)]
apple: Rename GLXcontext

Fixes regression introduced by: c356f5867f2c1fad7155df538b9affa8dbdcf869

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agoapple: Rename _gl_context_modes_find_visual to glx_config_find_visual
Jeremy Huddleston [Sun, 5 Jun 2011 22:19:59 +0000 (18:19 -0400)]
apple: Rename _gl_context_modes_find_visual to glx_config_find_visual

Fixes regression introduced by: 6ddf66e9230ee862ac341c4767cf6b3b2dd2552b

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agoapple: Re-add driContext and do_destroy
Jeremy Huddleston [Sun, 5 Jun 2011 22:02:44 +0000 (18:02 -0400)]
apple: Re-add driContext and do_destroy

Fixes regression introduced by: c491e585e43d48a2aeec96ccc4008da6c443fb42

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agoapple: Rename GLXcontext
Jeremy Huddleston [Sun, 5 Jun 2011 21:22:56 +0000 (17:22 -0400)]
apple: Rename GLXcontext

Fixes regression introduced by: c356f5867f2c1fad7155df538b9affa8dbdcf869

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agoapple: Rename __GLcontextModes to struct glx_config
Jeremy Huddleston [Sun, 5 Jun 2011 21:02:33 +0000 (17:02 -0400)]
apple: Rename __GLcontextModes to struct glx_config

Fixes regression introduced by: 6ddf66e9230ee862ac341c4767cf6b3b2dd2552b

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agoapple: Rename glcontextmodes.[ch] to glxconfig.[ch]
Jeremy Huddleston [Sun, 5 Jun 2011 20:56:01 +0000 (16:56 -0400)]
apple: Rename glcontextmodes.[ch] to glxconfig.[ch]

Fixes regression introduced by: 65d98e25770487456eb3d7eb8ec3ec8272f170b1

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agoapple: Update GL specs
Jeremy Huddleston [Sun, 5 Jun 2011 21:14:04 +0000 (17:14 -0400)]
apple: Update GL specs

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
12 years agometa: Don't do sRGB encode for framebuffer blits on sRGB-enabled framebuffers.
Eric Anholt [Wed, 1 Jun 2011 20:15:28 +0000 (13:15 -0700)]
meta: Don't do sRGB encode for framebuffer blits on sRGB-enabled framebuffers.

Fixes fbo-srgb-blit.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=35373
Reviewed-by: Brian Paul <brianp@vmware.com>
12 years agometa: Don't do srgb to linear decode when blitting srgb textures.
Eric Anholt [Wed, 1 Jun 2011 20:00:14 +0000 (13:00 -0700)]
meta: Don't do srgb to linear decode when blitting srgb textures.

Fixes the GL_SRGB8_ALPHA8 -> GL_RGBA8 blits in fbo-srgb-blit.c

Reviewed-by: Brian Paul <brianp@vmware.com>
12 years agoglx: Fix use-before-null-check in dri2InvalidateBuffers().
Eric Anholt [Wed, 1 Jun 2011 20:39:39 +0000 (13:39 -0700)]
glx: Fix use-before-null-check in dri2InvalidateBuffers().

The compiler used our dereference here to skip the NULL check below.
Fixes window resize in "jconsole -J-Dsun.java2d.opengl=True" under
OpenJDK 6.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=37766
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
12 years agoi965: Drop remaining strict conformance fallback for GL_POINT_SMOOTH.
Eric Anholt [Wed, 25 May 2011 19:59:25 +0000 (12:59 -0700)]
i965: Drop remaining strict conformance fallback for GL_POINT_SMOOTH.

We actually could do this in hardware in the fragment shader using
gl_PointCoord and the point's size.

12 years agoi965: Drop strict conformance fallback for GL_LINE_STIPPLE.
Eric Anholt [Wed, 25 May 2011 19:58:21 +0000 (12:58 -0700)]
i965: Drop strict conformance fallback for GL_LINE_STIPPLE.

We implement line stipples, just not *quite* correctly.  We have a
piglit testcase to use when we want to fix it, if we do.  Until then,
don't lie to our test suites.

12 years agoi965: Drop strict conformance fallback for GL_LINE_SMOOTH.
Eric Anholt [Wed, 25 May 2011 19:54:49 +0000 (12:54 -0700)]
i965: Drop strict conformance fallback for GL_LINE_SMOOTH.

We do have hardware antialised lines.  If we care, we should actually
fix them to be conformant (or as close as possible) instead of using
this knob to fool testcases using swrast.

For some interesting reading on the state of GL_*_SMOOTH across
several drivers, see:
http://homepage.mac.com/arekkusu/bugs/invariance/HWAA.html

12 years agoi965: Drop strict conformance fallback for GL_POLYGON_SMOOTH.
Eric Anholt [Wed, 25 May 2011 19:51:47 +0000 (12:51 -0700)]
i965: Drop strict conformance fallback for GL_POLYGON_SMOOTH.

From my reading of the GL 2.1 spec, no antialiasing is strictly
conformant for polygon smoothing.  Yes, it's absurd, but then,
hardware doesn't support this so maybe it's not so absurd.

12 years agoi965: Drop INTEL_CONFORMANCE=2 fallback code.
Eric Anholt [Wed, 25 May 2011 19:51:05 +0000 (12:51 -0700)]
i965: Drop INTEL_CONFORMANCE=2 fallback code.

This was just a duplicate of no_rast=true driconf option, which is
relatively standard across drivers.

12 years agoglsl: fixed printing of structure constants.
Paul Berry [Fri, 3 Jun 2011 18:23:31 +0000 (11:23 -0700)]
glsl: fixed printing of structure constants.

ir_print_visitor::visit(ir_constant *) was failing to index properly
into ir->type->fields.structure, so the first field name was being
reprinted for every field in the structure.

Signed-off-by: Brian Paul <brianp@vmware.com>
12 years agoAST dump: fixed printing of conditionals.
Paul Berry [Fri, 3 Jun 2011 17:02:32 +0000 (10:02 -0700)]
AST dump: fixed printing of conditionals.

ast_expression::print() had an incorrect index into the subexpressions
array, so (a ? b : c) was being incorrectly rendered as (a ? b : b).

Signed-off-by: Brian Paul <brianp@vmware.com>
12 years agor600g: move spi update to only when states change.
Dave Airlie [Fri, 3 Jun 2011 05:18:59 +0000 (15:18 +1000)]
r600g: move spi update to only when states change.

This updates the spi state after ps/vs binding or rasteriser state
change.

Signed-off-by: Dave Airlie <airlied@redhat.com>
12 years agocso: move cso hashes to a more table driven scheme
Dave Airlie [Fri, 3 Jun 2011 01:05:47 +0000 (11:05 +1000)]
cso: move cso hashes to a more table driven scheme

this removes a bad branch pain in the hash table lookup fn.

Signed-off-by: Dave Airlie <airlied@redhat.com>
12 years agou_prim: convert u_trim_pipe_prim to table driven.
Dave Airlie [Fri, 3 Jun 2011 00:36:48 +0000 (10:36 +1000)]
u_prim: convert u_trim_pipe_prim to table driven.

This makes this function not be an always miss for the branch predictor.

Noticed using cachegrind, makes a minor difference to gears numbers on r600g.

Signed-off-by: Dave Airlie <airlied@redhat.com>
12 years agor600g: make conv pipe prim table driven.
Dave Airlie [Fri, 3 Jun 2011 00:10:01 +0000 (10:10 +1000)]
r600g: make conv pipe prim table driven.

This is a lot more branch predictor friendly, it actually
showed up in cachegrind profiles.

Signed-off-by: Dave Airlie <airlied@redhat.com>
12 years agomesa: fix typo (s/GLGL/GLSL/)
Brian Paul [Fri, 3 Jun 2011 03:49:03 +0000 (21:49 -0600)]
mesa: fix typo (s/GLGL/GLSL/)

12 years agomesa: 80-column wrapping
Brian Paul [Fri, 3 Jun 2011 02:56:23 +0000 (20:56 -0600)]
mesa: 80-column wrapping

12 years agomesa: consolidate glGetUniform code
Brian Paul [Fri, 3 Jun 2011 02:56:23 +0000 (20:56 -0600)]
mesa: consolidate glGetUniform code

12 years agomesa: refactor/consolidate uniform lookup code
Brian Paul [Fri, 3 Jun 2011 02:56:23 +0000 (20:56 -0600)]
mesa: refactor/consolidate uniform lookup code

12 years agomesa: add some minor fixes for geometry shaders
Brian Paul [Fri, 3 Jun 2011 02:56:23 +0000 (20:56 -0600)]
mesa: add some minor fixes for geometry shaders

12 years agodri: add missing files from 873379a8818eed9ab16c24728b7091a3a3705c5b
Brian Paul [Fri, 3 Jun 2011 00:27:10 +0000 (18:27 -0600)]
dri: add missing files from 873379a8818eed9ab16c24728b7091a3a3705c5b

12 years agomesa: add implementation of glDrawElementsInstancedBaseVertex
Pierre-Eric Pelloux-Prayer [Tue, 31 May 2011 11:33:54 +0000 (13:33 +0200)]
mesa: add implementation of glDrawElementsInstancedBaseVertex

Signed-off-by: Brian Paul <brianp@vmware.com>
12 years agoglapi: regenerated files
Pierre-Eric Pelloux-Prayer [Tue, 31 May 2011 11:34:34 +0000 (13:34 +0200)]
glapi: regenerated files

Signed-off-by: Brian Paul <brianp@vmware.com>
12 years agoglapi: add glDrawElementsInstancedBaseVertex() function to xml
Brian Paul [Thu, 2 Jun 2011 23:54:40 +0000 (17:54 -0600)]
glapi: add glDrawElementsInstancedBaseVertex() function to xml

Signed-off-by: Brian Paul <brianp@vmware.com>
12 years agodri/nouveau: Fix build with --enable-shared-dricore.
Johannes Obermayr [Thu, 2 Jun 2011 23:15:44 +0000 (17:15 -0600)]
dri/nouveau: Fix build with --enable-shared-dricore.

- Based on the work of Себастьян Gliţa Κατινα <cglita@yahoo.com>
- Split Makefile.template into Makefile.defines and Makefile.targets
- Adapt other drivers to new situation
- Fixes https://bugs.freedesktop.org/show_bug.cgi?id=35441

Signed-off-by: Brian Paul <brianp@vmware.com>
12 years agor300: remove MIN3 macro, already defined in macros.h
Brian Paul [Thu, 2 Jun 2011 23:04:30 +0000 (17:04 -0600)]
r300: remove MIN3 macro, already defined in macros.h

12 years agor600g: sampler and texture state doesn't need a range/block.
Dave Airlie [Thu, 2 Jun 2011 22:50:58 +0000 (08:50 +1000)]
r600g: sampler and texture state doesn't need a range/block.

These are handled separately in the winsys, so don't need the calculations
done at this point. this manifested as a crash in point-sprite,

Thanks to XoD on #radeon for pointing it out.

Signed-off-by: Dave Airlie <airlied@redhat.com>
13 years agoglx: Remove (unused, broken) fastImageUnpack fast path
Adam Jackson [Wed, 1 Jun 2011 17:00:57 +0000 (13:00 -0400)]
glx: Remove (unused, broken) fastImageUnpack fast path

Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
13 years agoglx: Fix another case of using req outside of the display lock
Adam Jackson [Wed, 1 Jun 2011 15:33:48 +0000 (11:33 -0400)]
glx: Fix another case of using req outside of the display lock

Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
13 years agosoftpipe: add a better fake implementation of fences
Marek Olšák [Wed, 1 Jun 2011 22:50:45 +0000 (00:50 +0200)]
softpipe: add a better fake implementation of fences

The flush function, when asked for, should not return a NULL fence.

NULL can only be returned if fences are not implemented, and st/mesa
doesn't call any of the fence functions if it receives a NULL fence
(because some drivers don't even set the fence hooks).

ARB_sync is exposed if fence_finish is set.

13 years agoutil: faster logbase2
Brian Paul [Thu, 2 Jun 2011 14:43:07 +0000 (08:43 -0600)]
util: faster logbase2

13 years agomesa: faster logbase2
Benjamin Bellec [Thu, 2 Jun 2011 14:31:16 +0000 (08:31 -0600)]
mesa: faster logbase2

With minor clean-ups by Brian Paul.

Signed-off-by: Brian Paul <brianp@vmware.com>
13 years agost/mesa: add GL_R11F_G11F_B10F to format table
Brian Paul [Thu, 2 Jun 2011 14:01:26 +0000 (08:01 -0600)]
st/mesa: add GL_R11F_G11F_B10F to format table

Fixes http://bugs.freedesktop.org/show_bug.cgi?id=37839

13 years agod3d1x/sm4: don't reset 1st index of multi-dimensional operands to 0
Christoph Bumiller [Thu, 2 Jun 2011 12:04:05 +0000 (14:04 +0200)]
d3d1x/sm4: don't reset 1st index of multi-dimensional operands to 0

13 years agod3d1x/sm4: fix swizzle for 1 component operands
Christoph Bumiller [Thu, 2 Jun 2011 12:01:24 +0000 (14:01 +0200)]
d3d1x/sm4: fix swizzle for 1 component operands

For example, "mov o0.w, l(1)" would use imm_values[3], which is
not valid.

13 years agoi965: Raise const.MaxTextureLevels to 14 (8192)
Chris Wilson [Thu, 2 Jun 2011 07:27:09 +0000 (08:27 +0100)]
i965: Raise const.MaxTextureLevels to 14 (8192)

Mesa now limits, by default, the max number of texture levels to 15 so we
can now support the architectural maximum for gen4-6 of 14.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agor600g: add spi state and move spi/vgt to modify register
Dave Airlie [Thu, 2 Jun 2011 05:08:24 +0000 (15:08 +1000)]
r600g: add spi state and move spi/vgt to modify register

This modifies the VGT state and move the SPI setup to its own discrete state.

It then just sets the SPI state up and the VGT state up once and modifies
them thereafter.

Signed-off-by: Dave Airlie <airlied@redhat.com>
13 years agor600g: decrease CPU time on set buffer resources
Dave Airlie [Thu, 2 Jun 2011 05:03:52 +0000 (15:03 +1000)]
r600g: decrease CPU time on set buffer resources

This splits the initialisation and the setting of values in the resource
buffers. We only should end up initialising once and updateing with new values
when needed.

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