platform/upstream/mesa.git
14 years agodraw: fix typo in comment
Brian Paul [Wed, 20 Oct 2010 23:18:40 +0000 (17:18 -0600)]
draw: fix typo in comment

14 years agoi965: Correct scratch space allocation.
Eric Anholt [Thu, 21 Oct 2010 22:07:45 +0000 (15:07 -0700)]
i965: Correct scratch space allocation.

One, it was allocating increments of 1kb, but per thread scratch space
is a power of two.  Two, the new FS wasn't getting total_scratch set
at all, so everyone thought they had 1kb and writes beyond 1kb would
go stomping on a neighbor thread.

With this plus the previous register spilling for the new FS,
glsl-fs-convolution-1 passes.

14 years agoi965: Don't emit register spill offsets directly into g0.
Eric Anholt [Thu, 21 Oct 2010 21:40:49 +0000 (14:40 -0700)]
i965: Don't emit register spill offsets directly into g0.

g0 is used by others, and is expected to be left exactly as it was
dispatched to us.  So manually move g0 into our message reg when
spilling/unspilling and update the offset in the MRF.  Fixes failures
in texture sampling after having spilled a register.

14 years agoi965: Add support for register spilling.
Eric Anholt [Tue, 19 Oct 2010 16:25:51 +0000 (09:25 -0700)]
i965: Add support for register spilling.

It can be tested with if (0) replaced with if (1) to force spilling for all
virtual GRFs.  Some simple tests work, but large texturing tests fail.

14 years agoi965: Fix gl_FrontFacing emit on pre-gen6.
Eric Anholt [Wed, 20 Oct 2010 22:21:53 +0000 (15:21 -0700)]
i965: Fix gl_FrontFacing emit on pre-gen6.

It's amazing this code worked.  Basically, we would get lucky in
register allocation and the tests using frontfacing would happen to
allocate gl_FrontFacing storage and the instructions generating
gl_FrontFacing but pointing at another register to the same hardware
register.  Noticed during register spilling debug, when suddenly they
didn't get allocatd the same storage.

14 years agoi965: Split register allocation out of the ever-growing brw_fs.cpp.
Eric Anholt [Wed, 20 Oct 2010 17:26:29 +0000 (10:26 -0700)]
i965: Split register allocation out of the ever-growing brw_fs.cpp.

14 years agoRefresh autogenerated file builtin_function.cpp.
Kenneth Graunke [Thu, 21 Oct 2010 18:43:39 +0000 (11:43 -0700)]
Refresh autogenerated file builtin_function.cpp.

Since this is just generated by python, it's questionable whether this
should continue to live in the repository - Mesa already has other
things generated from python as part of the build process.

14 years agogenerate_builtins.py: Output large strings as arrays of characters.
Kenneth Graunke [Thu, 21 Oct 2010 18:42:45 +0000 (11:42 -0700)]
generate_builtins.py: Output large strings as arrays of characters.

This works around MSVC's 65535 byte limit, unfortunately at the expense
of any semblance of readability and much larger file size.  Hopefully I
can implement a better solution later, but for now this fixes the build.

14 years agogallivm: Silence uninitialized variable warning.
Vinson Lee [Thu, 21 Oct 2010 18:27:35 +0000 (11:27 -0700)]
gallivm: Silence uninitialized variable warning.

Fixes this GCC warning.
gallivm/lp_bld_tgsi_aos.c: In function 'lp_build_tgsi_aos':
gallivm/lp_bld_tgsi_aos.c:516: warning: 'dst0' may be used uninitialized in this function
gallivm/lp_bld_tgsi_aos.c:516: note: 'dst0' was declared here

14 years agogallivm: Silence uninitialized variable warnings.
Vinson Lee [Thu, 21 Oct 2010 18:21:03 +0000 (11:21 -0700)]
gallivm: Silence uninitialized variable warnings.

Fixes these GCC warnings.
gallivm/lp_bld_sample_aos.c: In function 'lp_build_sample_image_nearest':
gallivm/lp_bld_sample_aos.c:271: warning: 't_ipart' may be used uninitialized in this function
gallivm/lp_bld_sample_aos.c:271: warning: 'r_ipart' may be used uninitialized in this function

14 years agogallivm: Silence uninitialized variable warnings.
Vinson Lee [Thu, 21 Oct 2010 18:10:15 +0000 (11:10 -0700)]
gallivm: Silence uninitialized variable warnings.

Fixes these GCC warnings.
gallivm/lp_bld_sample_aos.c: In function 'lp_build_sample_image_linear':
gallivm/lp_bld_sample_aos.c:439: warning: 'r_ipart' may be used uninitialized in this function
gallivm/lp_bld_sample_aos.c:438: warning: 't_ipart' may be used uninitialized in this function
gallivm/lp_bld_sample_aos.c:438: warning: 't_fpart' may be used uninitialized in this function
gallivm/lp_bld_sample_aos.c:439: warning: 'r_fpart' may be used uninitialized in this function
gallivm/lp_bld_sample_aos.c:438: warning: 't_fpart_hi' may be used uninitialized in this function
gallivm/lp_bld_sample_aos.c:438: warning: 't_fpart_lo' may be used uninitialized in this function
gallivm/lp_bld_sample_aos.c:439: warning: 'r_fpart_hi' may be used uninitialized in this function
gallivm/lp_bld_sample_aos.c:439: warning: 'r_fpart_lo' may be used uninitialized in this function

14 years agomesa: Remove unused vtxfmt_tmp.h.
Chia-I Wu [Fri, 15 Oct 2010 02:09:37 +0000 (10:09 +0800)]
mesa: Remove unused vtxfmt_tmp.h.

It was used by the "neutral" tnl module that was dropped in
81ccb3e2ce708619f4c23537a237d61bdffdd35f.

14 years agor600g: start splitting out common code from eg/r600.
Dave Airlie [Thu, 21 Oct 2010 09:11:23 +0000 (19:11 +1000)]
r600g: start splitting out common code from eg/r600.

no point duplicating code that doesn't touch hw, also make it easier
to spot mistakes

14 years agor600g: initial translate state support
Dave Airlie [Thu, 21 Oct 2010 07:34:25 +0000 (17:34 +1000)]
r600g: initial translate state support

14 years agodraw: Remove unnecessary header.
Vinson Lee [Thu, 21 Oct 2010 08:47:52 +0000 (01:47 -0700)]
draw: Remove unnecessary header.

14 years agollvmpipe: Remove unnecessary header.
Vinson Lee [Thu, 21 Oct 2010 08:44:48 +0000 (01:44 -0700)]
llvmpipe: Remove unnecessary header.

14 years agoglsl: Refresh autogenerated file builtin_function.cpp.
Kenneth Graunke [Tue, 19 Oct 2010 18:57:15 +0000 (11:57 -0700)]
glsl: Refresh autogenerated file builtin_function.cpp.

14 years agoglsl: Add support for GLSL 1.30's modf built-in.
Kenneth Graunke [Tue, 19 Oct 2010 18:56:47 +0000 (11:56 -0700)]
glsl: Add support for GLSL 1.30's modf built-in.

14 years agoglcpp: Refresh autogenerated lexer file.
Kenneth Graunke [Thu, 21 Oct 2010 04:53:36 +0000 (21:53 -0700)]
glcpp: Refresh autogenerated lexer file.

14 years agoglcpp: Return NEWLINE token for newlines inside multi-line comments.
Kenneth Graunke [Thu, 21 Oct 2010 04:51:03 +0000 (21:51 -0700)]
glcpp: Return NEWLINE token for newlines inside multi-line comments.

This is necessary for the main compiler to get correct line numbers.

14 years agor600g: add texture tiling enable under a debug option.
Dave Airlie [Thu, 21 Oct 2010 03:40:45 +0000 (13:40 +1000)]
r600g: add texture tiling enable under a debug option.

At the moment you need kernel patches to have texture tiling work
with the kernel CS checker, so once they are upstream and the drm version
is bumped we can make this enable flip the other way most likely.

14 years agor600g: add texture tiling alignment support.
Dave Airlie [Thu, 21 Oct 2010 03:37:54 +0000 (13:37 +1000)]
r600g: add texture tiling alignment support.

this sets things up to align stride/height with tile sizes,
it also adds support for the 2D/1D array mode cross over point.

14 years agor600g: introduce a per-driver resource flag for transfers.
Dave Airlie [Thu, 21 Oct 2010 03:36:01 +0000 (13:36 +1000)]
r600g: introduce a per-driver resource flag for transfers.

this is to be used to decide not to tile a surface being used for transfers.

14 years agor600g: add r600 surface to store the aligned height.
Dave Airlie [Thu, 21 Oct 2010 03:31:27 +0000 (13:31 +1000)]
r600g: add r600 surface to store the aligned height.

we need to know the aligned height when binding the surface to cb/zb,
not the gallium surface height.

14 years agor600g: start adding hooks for aligning width/height for tiles.
Dave Airlie [Thu, 21 Oct 2010 03:27:07 +0000 (13:27 +1000)]
r600g: start adding hooks for aligning width/height for tiles.

14 years agor600g: move to per-miplevel array mode.
Dave Airlie [Thu, 21 Oct 2010 03:26:04 +0000 (13:26 +1000)]
r600g: move to per-miplevel array mode.

Since the hw transitions from 2D->1D sampling below the 2D macrotile
size we need to keep track of the array mode per level so we can
render to it using the CB.

14 years agor600g: all non-0 mipmap levels need to be w/h aligned to POT.
Dave Airlie [Thu, 21 Oct 2010 03:20:14 +0000 (13:20 +1000)]
r600g: all non-0 mipmap levels need to be w/h aligned to POT.

this adds a new minify function to the driver to ensure this.

14 years agoswrast: Print out format on unexpected failure in _swrast_DrawPixels.
Vinson Lee [Wed, 20 Oct 2010 22:27:48 +0000 (15:27 -0700)]
swrast: Print out format on unexpected failure in _swrast_DrawPixels.

14 years agomesa: Remove FEATURE_ARB_shading_language_120 macro.
Kenneth Graunke [Wed, 20 Oct 2010 20:52:16 +0000 (13:52 -0700)]
mesa: Remove FEATURE_ARB_shading_language_120 macro.

Everything should be able to support 1.20 at this point.

14 years agoglsl: Remove useless ir_shader enumeration value.
Kenneth Graunke [Wed, 20 Oct 2010 21:59:40 +0000 (14:59 -0700)]
glsl: Remove useless ir_shader enumeration value.

14 years agoglsl: Add assert for unhandled ir_shader case.
Vinson Lee [Wed, 20 Oct 2010 21:10:26 +0000 (14:10 -0700)]
glsl: Add assert for unhandled ir_shader case.

Silences this GCC warning.
ast_to_hir.cpp: In function 'void apply_type_qualifier_to_variable(const
ast_type_qualifier*, ir_variable*, _mesa_glsl_parse_state*, YYLTYPE*)'
ast_to_hir.cpp:1768: warning: enumeration value 'ir_shader' not handled
in switch

14 years agodraw: use float version of LLVM Mul/Add instructions
Brian Paul [Wed, 20 Oct 2010 20:54:38 +0000 (14:54 -0600)]
draw: use float version of LLVM Mul/Add instructions

LLVM 2.8 is pickier about int vs float instructions and operands.

14 years agollvmpipe/draw: always enable LLVMAddInstructionCombiningPass()
Brian Paul [Wed, 20 Oct 2010 20:47:32 +0000 (14:47 -0600)]
llvmpipe/draw: always enable LLVMAddInstructionCombiningPass()

We were working around an LLVM 2.5 bug but we're using LLVM 2.6 or later now.
This basically reverts commit baddcbc5225e12052b3bc8c07a8b65243d76574d.
This fixes the piglit bug/tri-tex-crash.c failure.

14 years agoosmesa: link against libtalloc
Orion Poplawski [Wed, 20 Oct 2010 19:54:57 +0000 (15:54 -0400)]
osmesa: link against libtalloc

Otherwise consumers have to, and that's lame.

Signed-off-by: Adam Jackson <ajax@redhat.com>
14 years agor600g: Ensure r600_src is initialized in tgsi_exp function.
Vinson Lee [Wed, 20 Oct 2010 19:44:08 +0000 (12:44 -0700)]
r600g: Ensure r600_src is initialized in tgsi_exp function.

Silences these GCC warnings.
r600_shader.c: In function 'tgsi_exp':
r600_shader.c:2339: warning: 'r600_src[0].rel' is used uninitialized in this function
r600_shader.c:2339: warning: 'r600_src[0].abs' is used uninitialized in this function
r600_shader.c:2339: warning: 'r600_src[0].neg' is used uninitialized in this function
r600_shader.c:2339: warning: 'r600_src[0].chan' is used uninitialized in this function
r600_shader.c:2339: warning: 'r600_src[0].sel' is used uninitialized in this function

14 years agodraw: Move loop variable declaration outside for loop.
Vinson Lee [Wed, 20 Oct 2010 06:48:59 +0000 (23:48 -0700)]
draw: Move loop variable declaration outside for loop.

Fixes MSVC build.

14 years agodraw: make sure viewport gets updated in draw llvm shader
Keith Whitwell [Wed, 20 Oct 2010 05:11:49 +0000 (22:11 -0700)]
draw: make sure viewport gets updated in draw llvm shader

The viewport state was being baked in at compile time (oops...)

14 years agoMerge branch 'llvm-cliptest-viewport'
Keith Whitwell [Wed, 20 Oct 2010 04:41:28 +0000 (21:41 -0700)]
Merge branch 'llvm-cliptest-viewport'

14 years agodraw: corrections to allow for different cliptest cases
Hui Qi Tay [Mon, 18 Oct 2010 17:48:18 +0000 (18:48 +0100)]
draw: corrections to allow for different cliptest cases

14 years agoi965: Use the new style of IF statement with embedded comparison on gen6.
Eric Anholt [Tue, 19 Oct 2010 19:32:55 +0000 (12:32 -0700)]
i965: Use the new style of IF statement with embedded comparison on gen6.

"Everyone else" does it this way, so follow suit.  It's fewer
instructions, anyway.

14 years agoi965: Set the source operand types for gen6 if/else/endif to integer.
Eric Anholt [Tue, 19 Oct 2010 21:50:42 +0000 (14:50 -0700)]
i965: Set the source operand types for gen6 if/else/endif to integer.

I don't think this should matter, but I'm not sure, and it's
recommended by a kernel checker in fulsim.

14 years agoi965: Add EU emit support for gen6's new IF instruction with comparison.
Eric Anholt [Tue, 19 Oct 2010 19:55:04 +0000 (12:55 -0700)]
i965: Add EU emit support for gen6's new IF instruction with comparison.

14 years agolinker: Improve handling of unread/unwritten shader inputs/outputs
Ian Romanick [Wed, 20 Oct 2010 00:59:10 +0000 (17:59 -0700)]
linker: Improve handling of unread/unwritten shader inputs/outputs

Previously some shader input or outputs that hadn't received location
assignments could slip through.  This could happen when a shader
contained user-defined varyings and was used with either
fixed-function or assembly shaders.

See the piglit tests glsl-[fv]s-user-varying-ff and
sso-user-varying-0[12].

NOTE: this is a candidate for the 7.9 branch.

14 years agoglsl: Commit generated file glsl_lexer.cpp
Chad Versace [Fri, 15 Oct 2010 21:46:00 +0000 (14:46 -0700)]
glsl: Commit generated file glsl_lexer.cpp

Changes are due to commit "glsl: Fix lexer rule for ^=".

14 years agoglsl: Fix lexer rule for ^=
Chad Versace [Fri, 15 Oct 2010 21:44:28 +0000 (14:44 -0700)]
glsl: Fix lexer rule for ^=

The caret is a special character, and needs to be quoted or escaped.

14 years agoglsl: Implement ast-to-hir for bit-logic ops
Chad Versace [Fri, 15 Oct 2010 19:08:28 +0000 (12:08 -0700)]
glsl: Implement ast-to-hir for bit-logic ops

Implement by adding to ast_expression::hir() the following cases:
    - ast_and_assign
    - ast_or_assign
    - ast_xor_assign

14 years agoglsl: Define bit_logic_result_type() in ast_to_hir.cpp
Chad Versace [Fri, 15 Oct 2010 18:28:05 +0000 (11:28 -0700)]
glsl: Define bit_logic_result_type() in ast_to_hir.cpp

This function type checks the operands of and returns the result type of
bit-logic operations.  It replaces the type checking performed in the
following cases of ast_expression::hir() :
    - ast_bit_and
    - ast_bit_or
    - ast_bit_xor

14 years agoglsl: Implement ast-to-hir for bit-shift-assignment
Chad Versace [Fri, 15 Oct 2010 17:05:50 +0000 (10:05 -0700)]
glsl: Implement ast-to-hir for bit-shift-assignment

Implement by adding to ast_expression::hir() these cases:
    - ast_ls_assign
    - ast_rs_assign

14 years agoglsl: Define shift_result_type() in ast_to_hir.cpp
Chad Versace [Fri, 15 Oct 2010 16:49:46 +0000 (09:49 -0700)]
glsl: Define shift_result_type() in ast_to_hir.cpp

This function type checks the operands of and returns the result type of
bit-shift operations. It replaces the type checking performed in the following
cases of ast_expression::hir() :
    - ast_lshift
    - ast_rshift

14 years agoi965: Disable thread dispatch when the FS doesn't do any work.
Eric Anholt [Mon, 18 Oct 2010 22:43:20 +0000 (15:43 -0700)]
i965: Disable thread dispatch when the FS doesn't do any work.

This should reduce the cost of generating shadow maps, for example.
No performance difference measured in nexuiz, though it does trigger
this path.

14 years agoi965: Remove the gen6 emit_mi_flushes I sprinkled around the driver.
Eric Anholt [Tue, 19 Oct 2010 16:50:44 +0000 (09:50 -0700)]
i965: Remove the gen6 emit_mi_flushes I sprinkled around the driver.

These were for debugging in bringup.  Now that relatively complicated
apps are working, they haven't helped debug anything in quite a while.

14 years agoi965: Tell the shader compiler when we expect depth writes for gen6.
Eric Anholt [Tue, 19 Oct 2010 16:44:20 +0000 (09:44 -0700)]
i965: Tell the shader compiler when we expect depth writes for gen6.

This fixes hangs in some Z-writes-in-shaders tests, though other
pieces don't come out correctly.

Bug #30392: hang in fbo-fblit-d24s8. (still fails with bad color drawn
to some targets)

14 years agollvmpipe: Initialize variable.
Vinson Lee [Tue, 19 Oct 2010 17:14:11 +0000 (10:14 -0700)]
llvmpipe: Initialize variable.

14 years agollvmpipe: Initialize state variable in debug_bin function.
Vinson Lee [Tue, 19 Oct 2010 17:02:28 +0000 (10:02 -0700)]
llvmpipe: Initialize state variable in debug_bin function.

14 years agost/xorg: Fix memory leak on error path.
Vinson Lee [Tue, 19 Oct 2010 16:49:15 +0000 (09:49 -0700)]
st/xorg: Fix memory leak on error path.

14 years agogallivm: fix incorrect type for zero vector in emit_kilp()
Brian Paul [Tue, 19 Oct 2010 15:14:17 +0000 (09:14 -0600)]
gallivm: fix incorrect type for zero vector in emit_kilp()

http://bugs.freedesktop.org/show_bug.cgi?id=30974

14 years agomesa: fix mesa version string construction
Brian Paul [Mon, 18 Oct 2010 15:07:54 +0000 (09:07 -0600)]
mesa: fix mesa version string construction

Now that MESA_MINOR=10, we no longer need the extra '0' in the
version string.

14 years agomesa: Make sure we have the talloc cflags when using the talloc headers
Thomas Hellstrom [Thu, 14 Oct 2010 20:08:03 +0000 (22:08 +0200)]
mesa: Make sure we have the talloc cflags when using the talloc headers

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
14 years agost/xorg: Fix compilation for Xservers >= 1.10
Thomas Hellstrom [Tue, 19 Oct 2010 09:44:08 +0000 (11:44 +0200)]
st/xorg: Fix compilation for Xservers >= 1.10

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
14 years agoxorg/vmwgfx: Don't use deprecated x*alloc / xfree functions
Thomas Hellstrom [Tue, 19 Oct 2010 09:25:15 +0000 (11:25 +0200)]
xorg/vmwgfx: Don't use deprecated x*alloc / xfree functions

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
14 years agost/xorg: Don't use deprecated x*alloc / xfree functions
Thomas Hellstrom [Tue, 19 Oct 2010 08:39:24 +0000 (10:39 +0200)]
st/xorg: Don't use deprecated x*alloc / xfree functions

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
14 years agost/xorg: Fix compilation errors for Xservers compiled without Composite
Thomas Hellstrom [Tue, 19 Oct 2010 07:35:16 +0000 (09:35 +0200)]
st/xorg: Fix compilation errors for Xservers compiled without Composite

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
14 years agost/xorg, xorg/vmwgfx: Be a bit more frendly towards cross-compiling environments
Thomas Hellstrom [Thu, 14 Oct 2010 20:18:38 +0000 (22:18 +0200)]
st/xorg, xorg/vmwgfx: Be a bit more frendly towards cross-compiling environments

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
14 years agor300/compiler: Remove unused variable.
Vinson Lee [Tue, 19 Oct 2010 07:07:22 +0000 (00:07 -0700)]
r300/compiler: Remove unused variable.

14 years agor300g: Add new debug option for logging vertex/fragment program stats
Tom Stellard [Mon, 18 Oct 2010 06:17:01 +0000 (23:17 -0700)]
r300g: Add new debug option for logging vertex/fragment program stats

14 years agor300/compiler: Add a new function for more efficient dataflow analysis
Tom Stellard [Sun, 10 Oct 2010 19:39:00 +0000 (12:39 -0700)]
r300/compiler: Add a new function for more efficient dataflow analysis

rc_get_readers_normal() supplies a list of readers for a given
instruction.  This function is now being used by the copy propagate
optimization and will eventually be used by most other optimization
passes as well.

14 years agor300/compiler: Clear empty registers after constant folding
Tom Stellard [Thu, 14 Oct 2010 06:45:12 +0000 (23:45 -0700)]
r300/compiler: Clear empty registers after constant folding

14 years agor300/compiler: Fix incorrect assumption
Tom Stellard [Wed, 6 Oct 2010 05:53:40 +0000 (22:53 -0700)]
r300/compiler: Fix incorrect assumption

It is possible for a single pair instruction arg to select from both an
RGB and an Alpha source.

14 years agor300/compiler: Create a helper function for merging presubtract sources
Tom Stellard [Fri, 1 Oct 2010 06:36:57 +0000 (23:36 -0700)]
r300/compiler: Create a helper function for merging presubtract sources

14 years agoi965: Remove unused variable.
Kenneth Graunke [Tue, 19 Oct 2010 01:14:33 +0000 (18:14 -0700)]
i965: Remove unused variable.

14 years agoglsl: Regenerate parser files.
Kenneth Graunke [Tue, 19 Oct 2010 00:30:16 +0000 (17:30 -0700)]
glsl: Regenerate parser files.

14 years agoglsl: Fix copy and paste error in ast_bit_and node creation.
Kenneth Graunke [Tue, 19 Oct 2010 00:25:52 +0000 (17:25 -0700)]
glsl: Fix copy and paste error in ast_bit_and node creation.

All & operations were incorrectly being generated as ast_bit_or.

14 years agoi965: Avoid blits in BufferCopySubdata on gen6.
Eric Anholt [Mon, 18 Oct 2010 21:12:12 +0000 (14:12 -0700)]
i965: Avoid blits in BufferCopySubdata on gen6.

Fixes glean/bufferObject.

14 years agoi965: Fix scissor-offscreen on gen6 like we did pre-gen6.
Eric Anholt [Mon, 18 Oct 2010 20:11:29 +0000 (13:11 -0700)]
i965: Fix scissor-offscreen on gen6 like we did pre-gen6.

14 years agoi965: Assert out on gen6 VS constant buffer reads that hang the GPU for now.
Eric Anholt [Mon, 18 Oct 2010 19:41:10 +0000 (12:41 -0700)]
i965: Assert out on gen6 VS constant buffer reads that hang the GPU for now.

14 years agoi965: Fix assertion failure on gen6 BufferSubData to busy BO.
Eric Anholt [Mon, 18 Oct 2010 18:32:19 +0000 (11:32 -0700)]
i965: Fix assertion failure on gen6 BufferSubData to busy BO.

Fixes fbo-blit and probably several other tests.

14 years agoi965: Fix a weirdness in NOT handling.
Eric Anholt [Mon, 18 Oct 2010 18:13:26 +0000 (11:13 -0700)]
i965: Fix a weirdness in NOT handling.

XOR makes much more sense.  Note that the previous code would have
failed for not(not(x)), but that gets optimized out.

14 years agoi965: Disable the debug printf I added for FS disasm.
Eric Anholt [Mon, 18 Oct 2010 17:59:46 +0000 (10:59 -0700)]
i965: Disable the debug printf I added for FS disasm.

14 years agoi965: Add missing "break" statement.
Kenneth Graunke [Mon, 18 Oct 2010 19:21:20 +0000 (12:21 -0700)]
i965: Add missing "break" statement.

Otherwise, it would try to handle arrays as structures, use
uninitialized memory, and crash.

14 years agollvmpipe: Don't test rounding of x.5 numbers.
José Fonseca [Mon, 18 Oct 2010 16:35:21 +0000 (09:35 -0700)]
llvmpipe: Don't test rounding of x.5 numbers.

SSE4.1 has different rules, and so far this doesn't seem to cause any
problems with conformance test suites.

14 years agogallivm: Add a note about SSE4.1's nearest mode rounding.
José Fonseca [Mon, 18 Oct 2010 16:32:35 +0000 (09:32 -0700)]
gallivm: Add a note about SSE4.1's nearest mode rounding.

14 years agomesa: Add missing else in do_row_3D
Brian Rogers [Sun, 17 Oct 2010 21:21:06 +0000 (14:21 -0700)]
mesa: Add missing else in do_row_3D

This fixes erroneous "bad format in do_row()" messages

Signed-off-by: Brian Paul <brianp@vmware.com>
14 years agollvmpipe: remove lp_setup_coef*.c files from Makefile
Brian Paul [Mon, 18 Oct 2010 13:59:02 +0000 (07:59 -0600)]
llvmpipe: remove lp_setup_coef*.c files from Makefile

14 years agoegl/i965: include inline_wrapper_sw_helper.h
Victor Tseng [Mon, 18 Oct 2010 13:55:37 +0000 (07:55 -0600)]
egl/i965: include inline_wrapper_sw_helper.h

Signed-off-by: Brian Paul <brianp@vmware.com>
14 years agoglsl: Don't return NULL IR for erroneous bit-shift operators.
Kenneth Graunke [Mon, 18 Oct 2010 07:24:46 +0000 (00:24 -0700)]
glsl: Don't return NULL IR for erroneous bit-shift operators.

Existing code relies on IR being generated (possibly with error type)
rather than returning NULL.  So, don't break - go ahead and generate the
operation.  As long as an error is flagged, things will work out.

Fixes fd.o bug #30914.

14 years agor600g: retrieve tiling info from kernel for shared buffers.
Dave Airlie [Sun, 17 Oct 2010 23:45:58 +0000 (09:45 +1000)]
r600g: retrieve tiling info from kernel for shared buffers.

we need to know if the back is tiled so we can blit from it properly.

14 years agor600g: fix transfer function for tiling.
Dave Airlie [Wed, 13 Oct 2010 00:44:46 +0000 (10:44 +1000)]
r600g: fix transfer function for tiling.

this makes readback with tiled back work better.

14 years agor600g: attempt to cleanup depth blit
Dave Airlie [Mon, 18 Oct 2010 02:05:27 +0000 (12:05 +1000)]
r600g: attempt to cleanup depth blit

cleanup what I'm nearly sure is unnecessary work in the depth blit code.

14 years agor600g: depth needs to bound to ds
Dave Airlie [Mon, 18 Oct 2010 03:23:34 +0000 (13:23 +1000)]
r600g: depth needs to bound to ds

14 years agor600g: fix typo in tiling setup cb code.
Dave Airlie [Mon, 18 Oct 2010 02:05:38 +0000 (12:05 +1000)]
r600g: fix typo in tiling setup cb code.

14 years agor600g: do proper tracking of views/samplers.
Dave Airlie [Mon, 18 Oct 2010 02:04:57 +0000 (12:04 +1000)]
r600g: do proper tracking of views/samplers.

we need to do pretty much what r300g does in for this, this fixes some
issues seen while working on tiling.

14 years agollvmpipe: remove unused arg from jit_setup_tri function
Keith Whitwell [Mon, 18 Oct 2010 02:23:40 +0000 (19:23 -0700)]
llvmpipe: remove unused arg from jit_setup_tri function

14 years agollvmpipe: remove unused file
Keith Whitwell [Mon, 18 Oct 2010 02:11:47 +0000 (19:11 -0700)]
llvmpipe: remove unused file

14 years agoMerge remote branch 'origin/master' into lp-setup-llvm
Keith Whitwell [Mon, 18 Oct 2010 02:03:42 +0000 (19:03 -0700)]
Merge remote branch 'origin/master' into lp-setup-llvm

Conflicts:
src/gallium/drivers/llvmpipe/lp_setup_coef.c
src/gallium/drivers/llvmpipe/lp_setup_coef.h
src/gallium/drivers/llvmpipe/lp_setup_coef_intrin.c
src/gallium/drivers/llvmpipe/lp_setup_point.c
src/gallium/drivers/llvmpipe/lp_setup_tri.c
src/gallium/drivers/llvmpipe/lp_state_derived.c
src/gallium/drivers/llvmpipe/lp_state_fs.h

14 years agollvmpipe: fail cleanly on malloc failure in lp_setup_alloc_triangle
Keith Whitwell [Mon, 18 Oct 2010 01:48:11 +0000 (18:48 -0700)]
llvmpipe: fail cleanly on malloc failure in lp_setup_alloc_triangle

14 years agollvmpipe: remove setup fallback path
Keith Whitwell [Mon, 18 Oct 2010 01:29:28 +0000 (18:29 -0700)]
llvmpipe: remove setup fallback path

14 years agogallivm: Comment lp_build_insert_new_block().
José Fonseca [Sun, 17 Oct 2010 20:41:45 +0000 (13:41 -0700)]
gallivm: Comment lp_build_insert_new_block().

14 years agollvmpipe: clean up fields in draw_llvm_variant_key
Keith Whitwell [Mon, 18 Oct 2010 00:53:29 +0000 (17:53 -0700)]
llvmpipe: clean up fields in draw_llvm_variant_key

14 years agor600g: set tiling bits in hw state
Dave Airlie [Fri, 8 Oct 2010 01:57:04 +0000 (11:57 +1000)]
r600g: set tiling bits in hw state

14 years agor600g: get tiling info from kernel
Dave Airlie [Fri, 8 Oct 2010 01:56:43 +0000 (11:56 +1000)]
r600g: get tiling info from kernel