Compare commits

..

451 Commits

Author SHA1 Message Date
Ian Romanick
7e9d853665 docs: Add change log to 7.9.2 release notes 2011-03-02 14:18:05 -08:00
Ian Romanick
ca5cf9bf66 mesa: set version string to 7.9.2 (final) 2011-03-02 13:48:22 -08:00
Ian Romanick
dfae3ffa0f intel: Remove driver date and related bits from renderer string
Not only did this contain lies, it contained lies that wouldn't be
useful even if true.
(cherry picked from commit b0a7492aeb)
2011-03-02 13:47:41 -08:00
Ian Romanick
2b41289691 docs: Clean up bug fixes list
All the unnumbered bugs are first.  These are followed by numbered
bugs sorted by bug number.  This is analogous to 8aabb1bc on the 7.10
branch.
2011-03-01 13:22:02 -08:00
Ian Romanick
42c3e00bdb docs: Update 7.9.2 release notes for recent cherry picks 2011-03-01 13:16:53 -08:00
Cyril Brulebois
0ebbfa4335 Point to bugs.freedesktop.org rather than bugzilla.freedesktop.org
Suggested by a freedesktop.org admin.

Signed-off-by: Cyril Brulebois <kibi@debian.org>
(cherry picked from commit d252db7af1)
2011-03-01 12:57:23 -08:00
Brian Paul
fc5889a528 docs: updated environment variable list
(cherry picked from commit 1bf9954bb4)

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2011-03-01 12:57:07 -08:00
Sam Hocevar
a5b6d8ec4d docs: fix glsl_compiler name
(cherry picked from commit fde4943688)
2011-03-01 12:54:21 -08:00
Sam Hocevar
6eb1f39b55 docs: add glsl info
(cherry picked from commit 3e8fb54fb8)
2011-03-01 12:54:05 -08:00
Chad Versace
5e987d0849 tnl: Add support for datatype GL_FIXED in vertex arrays
Before populating the vertex buffer attribute pointer (VB->AttribPtr[]),
convert vertex data in GL_FIXED format to GL_FLOAT.

Fixes bug: http://bugs.freedesktop.org/show_bug.cgi?id=34047

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit a231ac23f4)
2011-03-01 11:53:26 -08:00
Ian Romanick
9fe29f34f3 i915: Force lowering of all types of indirect array accesses in the FS
NOTE: This is a candidate for the 7.9 and 7.10 branches.

(cherry picked from commit 4c1dc1c4d7)
2011-03-01 11:53:06 -08:00
Ian Romanick
24a9d95ec7 i915: Calculate partial result to temp register first
Previously the SNE and SEQ instructions would calculate the partial
result to the destination register.  This would cause problems if the
destination register was also one of the source registers.

Fixes piglit tests glsl-fs-any, glsl-fs-struct-equal,
glsl-fs-struct-notequal, glsl-fs-vec4-operator-equal,
glsl-fs-vec4-operator-notequal.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit 53b8b68843)
2011-03-01 10:49:56 -08:00
Ian Romanick
38c93ba921 i915: Only mark a register as available if all components are written
Previously a register would be marked as available if any component
was written.  This caused shaders such as this:

  0: TEX TEMP[0].xyz, INPUT[14].xyyy, texture[0], 2D;
  1: MUL TEMP[1], UNIFORM[0], TEMP[0].xxxx;
  2: MAD TEMP[2], UNIFORM[1], TEMP[0].yyyy, TEMP[1];
  3: MAD TEMP[1], UNIFORM[2], TEMP[0].zzzz, TEMP[2];
  4: ADD TEMP[0].xyz, TEMP[1].xyzx, UNIFORM[3].xyzx;
  5: TEX TEMP[1].w, INPUT[14].xyyy, texture[0], 2D;
  6: MOV TEMP[0].w, TEMP[1].wwww;
  7: MOV OUTPUT[2], TEMP[0];
  8: END

to produce incorrect code such as this:

  BEGIN
  DCL S[0]
  DCL T_TEX0
  R[0] = MOV T_TEX0.xyyy
  U[0] = TEXLD S[0],R[0]
  R[0].xyz = MOV U[0]
  R[1] = MUL CONST[0], R[0].xxxx
  R[2] = MAD CONST[1], R[0].yyyy, R[1]
  R[1] = MAD CONST[2], R[0].zzzz, R[2]
  R[0].xyz = ADD R[1].xyzx, CONST[3].xyzx
  R[0] = MOV T_TEX0.xyyy
  U[0] = TEXLD S[0],R[0]
  R[1].w = MOV U[0]
  R[0].w = MOV R[1].wwww
  oC = MOV R[0]
  END

Note that T_TEX0 is copied to R[0], but the xyz components of R[0] are
still expected to hold a calculated value.

Fixes piglit tests draw-elements-vs-inputs, fp-kill, and
glsl-fs-color-matrix.  It also fixes Meego bugzilla #13005.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit a04582739e)
2011-03-01 10:49:49 -08:00
Kenneth Graunke
1156006e4a glsl: Use reralloc instead of plain realloc.
Plugs a memory leak when compiling shaders with user defined structures.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit eb639349e2)
2011-02-28 16:33:09 -08:00
Ian Romanick
7be5744b5b linker: Fix off-by-one error implicit array sizing
Arrays are zero based.  If the highest element accessed is 6, the
array needs to have 7 elements.

Fixes piglit test glsl-fs-implicit-array-size-03 and bugzilla #34198.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit 25b36e8ff8)
2011-02-28 16:01:08 -08:00
nobled
c73bedd191 glx: Put null check before use
'dpy' was being checked for null *after* it was already used once.

Also add a null check for psc, and drop gc's redundant initialization.
(cherry picked from commit b5dc40710d)
2011-02-28 16:01:08 -08:00
Ian Romanick
bf66888a81 glsl: Regenerate compiler files from cherry picks 2011-02-28 16:01:08 -08:00
Ian Romanick
808ba96f3e glsl: Finish out the reduce/reduce error fixes
Track variables, functions, and types during parsing.  Use this
information in the lexer to return the currect "type" for identifiers.

Change the handling of structure constructors.  They will now show up
in the AST as constructors (instead of plain function calls).

Fixes piglit tests constructor-18.vert, constructor-19.vert, and
constructor-20.vert.  Also fixes bugzilla #29926.

NOTE: This is a candidate for the 7.9 and 7.10 branches.

(cherry picked from commit 8842158944)
2011-02-28 16:01:08 -08:00
Keith Packard
cf2076f89d glsl: Eliminate reduce/reduce conflicts in glsl grammar
This requires lexical disambiguation between variable and type
identifiers (as most C compilers do).

Signed-off-by: Keith Packard <keithp@keithp.com>

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit f4b812e1a6)
2011-02-28 16:01:08 -08:00
Chad Versace
e1e829689d glsl: Reinstate constant-folding for division by zero
Fixes regression: https://bugs.freedesktop.org/show_bug.cgi?id=34160

Commit e7c1f058d1 disabled constant-folding
when division-by-zero occured. This was a mistake, because the spec does
allow division by zero. (From section 5.9 of the GLSL 1.20 spec: Dividing
by zero does not cause an exception but does result in an unspecified
value.)

For floating-point division, the original pre-e7c1f05 behavior is
reinstated.

For integer division, constant-fold 1/0 to 0.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>

(cherry picked from commit 62c8c77333)
2011-02-28 15:27:41 -08:00
Chad Versace
3df97ee70c glsl: Set operators '%' and '%=' to be reserved when GLSL < 1.30
From section 5.9 of the GLSL 1.20 spec:
   The operator modulus (%) is reserved for future use.

From section 5.8 of the GLSL 1.20 spec:
   The assignments modulus into (%=), left shift by (<<=), right shift by
   (>>=), inclusive or into ( |=), and exclusive or into ( ^=). These
   operators are reserved for future use.

The GLSL ES 1.00 spec and GLSL 1.10 spec have similiar language.

Fixes bug:
https://bugs.freedesktop.org//show_bug.cgi?id=33916

Fixes Piglit tests:
spec/glsl-1.00/compiler/arithmetic-operators/modulus-00.frag
spec/glsl-1.00/compiler/assignment-operators/modulus-assign-00.frag
spec/glsl-1.10/compiler/arithmetic-operators/modulus-00.frag
spec/glsl-1.10/compiler/assignment-operators/modulus-assign-00.frag
spec/glsl-1.20/compiler/arithmetic-operators/modulus-00.frag
spec/glsl-1.20/compiler/assignment-operators/modulus-assign-00.frag
(cherry picked from commit 82f994f386)
2011-02-28 15:27:40 -08:00
Ian Romanick
7d6d3d6133 glcpp: Regenerate files from recent cherry picks 2011-02-28 15:27:40 -08:00
Chad Versace
a068f7a2df glcpp: Raise error when modulus is zero
For example, this now raises an error:
   #define XXX 1 / 0

Fixes bug: https://bugs.freedesktop.org//show_bug.cgi?id=33507
Fixes Piglit test: spec/glsl-1.10/preprocessor/modulus-by-zero.vert

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit fd1252ab67)
2011-02-28 14:47:33 -08:00
Ian Romanick
d72c16a0d3 mesa: Initial size for secondary color array is 3
See table 6.7 on page 347 of the OpenGL 3.0 specification.
(cherry picked from commit 09e15ac76a)
2011-02-28 14:47:05 -08:00
Kenneth Graunke
2eb4816939 glsl: Fix use of uninitialized values in _mesa_glsl_parse_state ctor.
This has probably existed since e5e34ab18e or so.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit dfdb9fda82)
2011-02-28 14:33:59 -08:00
Ian Romanick
8cf57be878 glsl: Reject shader versions not supported by the implementation
Previously we'd happily compile GLSL 1.30 shaders on any driver.  We'd
also happily compile GLSL 1.10 and 1.20 shaders in an ES2 context.
This has been a long standing FINISHME in the compiler.

NOTE: This is a candidate for the 7.9 and 7.10 branches
(cherry picked from commit 14880a510a)
2011-02-28 14:33:31 -08:00
Ian Romanick
bf50184f42 glsl: Ensure that all GLSL versions are supported in the stand-alone compiler
NOTE: This is a candidate for the 7.9 and 7.10 branches
(cherry picked from commit e5e34ab18e)
2011-02-28 14:33:22 -08:00
Ian Romanick
4d2a725196 intel: Fix typeos from 3d028024 and 790ff232
...and remove egg from face.
(cherry picked from commit 2fb0aebd4a)
2011-02-28 14:27:36 -08:00
Ian Romanick
bcfa054b29 i915: Set correct values for range/precision of fragment shader types
(cherry picked from commit 790ff232e2)
2011-02-28 14:27:36 -08:00
Ian Romanick
499936406b i965: Set correct values for range/precision of fragment shader types
(cherry picked from commit 3d028024e5)
2011-02-28 14:27:36 -08:00
Ian Romanick
56c63dba48 mesa: Connect glGetShaderPrecisionFormat into the dispatch table
(cherry picked from commit dde3270c19)
2011-02-28 14:27:35 -08:00
Brian Paul
9c8d677a16 mesa: implement glGetShaderPrecisionFormat()
Drivers should override the default range/precision info as needed.
No drivers do this yet.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>

(cherry picked from commit 3ee60a3558)
2011-02-28 14:27:35 -08:00
Chia-I Wu
f8d1b4fd26 mesa: Add glDepthRangef and glClearDepthf to APIspec.xml.
Core mesa has gained support for GL_ARB_ES2_compatibility.  Make GLES
generated dispatch table use them.
(cherry picked from commit a4a5a9a5ce)
2011-02-28 14:27:35 -08:00
Eric Anholt
3eda10aebf mesa: Add getter for GL_SHADER_COMPILER with ARB_ES2_compatibility.
Fixes piglit arb_es2_compatibility-shadercompiler
(cherry picked from commit 4620de7eea)
2011-02-28 14:27:35 -08:00
Eric Anholt
215ec57e69 mesa: Add getters for ARB_ES2_compatibility MAX_*_VECTORS.
Fixes piglit arb_es2_compatibility-maxvectors.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>

(cherry picked from commit 8395f206a8)
2011-02-28 14:27:35 -08:00
Eric Anholt
33aee8177e mesa: Add support for glDepthRangef and glClearDepthf.
These are ARB_ES2_compatibility float variants of the core double
entrypoints.  Fixes arb_es2_compatibility-depthrangef.
(cherry picked from commit e12c4faf7e)
2011-02-28 14:27:35 -08:00
Eric Anholt
c9d5d6ebc6 mesa: Add actual support for glReleaseShaderCompiler from ES2.
Fixes no-op dispatch warning in piglit
arb_es2_compatibility-releaseshadercompiler.c.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>

(cherry picked from commit 7b987578a9)
2011-02-28 14:27:34 -08:00
Eric Anholt
e63f8638cd mesa: Add extension enable bit for GL_ARB_ES2_compatibility.
(cherry picked from commit 9c6954fc9d)

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2011-02-28 14:27:34 -08:00
Ian Romanick
ae775f3f37 glapi: Regenerate for GL_ARB_ES2_compatibility.
This is not a cherry pick, but it matches 841ad6bf (and ac06d610).
2011-02-28 14:27:34 -08:00
Eric Anholt
bd0457346f glapi: Add entrypoints and enums for GL_ARB_ES2_compatibility.
(cherry picked from commit 8560cb939b)
2011-02-28 13:56:06 -08:00
Ian Romanick
161e36c0f7 mesa: Remove unsupported OES extensions
Based commit 19418e921 from master.  Chad's extension rework series,
of which 19418e921 is part, wasn't cherry picked to 7.9 because there
were too many conflicts.  Most of the conflicts were caused by
extensions that were removed in 7.10.
2011-02-25 17:34:02 -08:00
Paulo Zanoni
4996e4a845 dri_util: fail driCreateNewScreen if InitScreen is NULL
Without this, X doesn't start with UMS on r300g.

NOTE: This is a candidate for the 7.9 and 7.10 branches.

Signed-off-by: Paulo Zanoni <pzanoni@mandriva.com>
Signed-off-by: Brian Paul <brianp@vmware.com>
2011-02-22 15:20:35 -05:00
Dave Airlie
b093c9939e radeon: avoid segfault on 3D textures.
This is a candidate for 7.9 and 7.10
2011-02-22 15:11:09 -05:00
Dave Airlie
87a02b952b radeon/r200: fix fbo-clearmipmap + gen-teximage
sw clears were being used and not getting the correct offsets in the span
code.

also not emitting correct offsets for CB draws to texture levels.

(I've no idea why I'm playing with r100).

This is a candidate for 7.9 and 7.10
2011-02-22 15:07:45 -05:00
Brian Paul
1bac4815a8 glsl2: fix signed/unsigned comparison warning
(cherry picked from commit 3633e1f538)
2011-02-22 08:17:27 -07:00
Brian Paul
ffd9033227 glsl: add ir_shader case in switch stmt to silence warning 2011-02-22 08:17:21 -07:00
Brian Paul
439c298595 mesa: include compiler.h for ASSERT macro
This was already done in the 7.10 branch and master.
2011-02-22 07:20:12 -07:00
Marek Olšák
cd42dda3a5 st/dri: Track drawable context bindings
Needs to track this ourself since because we get into a race condition with
the dri_util.c code on make current when rendering to the front buffer.

This is what happens:
Old context is rendering to the front buffer.

App calls MakeCurrent with a new context. dri_util.c sets
drawable->driContextPriv to the new context and then calls the driver make
current. st/dri make current flushes the old context, which calls back into
st/dri via the flush frontbuffer hook. st/dri calls dri loader flush
frontbuffer, which calls invalidate buffer on the drawable into st/dri.

This is where things gets wrong. st/dri grabs the context from the dri
drawable (which now points to the new context) and calls invalidate
framebuffer to the new context which has not yet set the new drawable as its
framebuffers since we have not called make current yet, it asserts.
(cherry picked from commit 94ccc31ba4)

Conflicts:

	src/gallium/state_trackers/dri/common/dri_context.c
2011-02-20 17:08:20 +01:00
Brian Paul
c2b76d966e docs: add links to 7.9.1 and 7.9.2 release notes 2011-02-21 18:05:18 -07:00
Brian Paul
a28050de4f docs: update 7.9.2 release notes with Brian's cherry-picks 2011-02-21 18:05:01 -07:00
Brian Paul
9c122ce963 st/mesa: need to translate clear color according to surface's base format
When clearing a GL_LUMINANCE_ALPHA buffer, for example, we need to convert
the clear color (R,G,B,A) to (R,R,R,A).  We were doing this for texture border
colors but not renderbuffers.  Move the translation function to st_format.c
and share it.

This fixes the piglit fbo-clear-formats test.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit e2d108ec82)

Conflicts:

	src/mesa/state_tracker/st_atom_sampler.c
2011-02-21 17:40:06 -07:00
Brian Paul
f0eff3061c st/mesa: fix the default case in st_format_datatype()
Part of the fix for piglit fbo-clear-formats

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit c966c6980c)
2011-02-21 17:39:03 -07:00
Brian Paul
5a79a6dc04 st/mesa: set renderbuffer _BaseFormat in a few places
NOTE: This is a candidate for the 7.9 and 7.10 branches
(cherry picked from commit 633c9fcf78)
2011-02-21 17:38:32 -07:00
Brian Paul
652ceece0e cso: fix loop bound in cso_set_vertex_samplers()
Before we were looping to nr_samplers, which is the number of fragment
samplers, not vertex samplers.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit d087cfaabf)
2011-02-21 17:31:48 -07:00
Julien Cristau
e39aa1966b glx: fix length of GLXGetFBConfigsSGIX
The extra length is the size of the request *minus* the size of the
VendorPrivate header, not the addition.

NOTE: This is a candidate for the 7.9 and 7.10 branches

Signed-off-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit cbe9fc12a6)
2011-02-21 17:30:44 -07:00
Julien Cristau
ba764ae5b1 glx: fix GLXChangeDrawableAttributesSGIX request
xGLXChangeDrawableAttributesSGIXReq follows the GLXVendorPrivate header
with a drawable, number of attributes, and list of (type, value)
attribute pairs.  Don't forget to put the number of attributes in there.
I don't think this can ever have worked.

NOTE: This is a candidate for the 7.9 and 7.10 branches

Signed-off-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit e27913f805)
2011-02-21 17:30:32 -07:00
Dimitry Andric
3959f9f7c8 glapi: add @GOTPCREL relocation type
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=33440
This replaces commit 731ec60da3

NOTE: This is a candidate for the 7.9 and 7.10 branches

Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit cfb9aae3ec)
2011-02-21 17:30:01 -07:00
Brian Paul
fa5c4b988a softpipe: fix off-by-one error in setup_fragcoord_coeff()
If we invert Y, need to subtract one from the surface height.

Fixes https://bugs.freedesktop.org/show_bug.cgi?id=26795
for softpipe.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit 40ac24e631)
2011-02-21 17:29:18 -07:00
Brian Paul
0a218b00e7 st/mesa: fix incorrect fragcoord.x translation
emit_adjusted_wpos() needs separate x,y translation values.  If we
invert Y, we don't want to effect X.

Part of the fix for http://bugs.freedesktop.org/show_bug.cgi?id=26795

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit bb56631f0c)
2011-02-21 17:28:53 -07:00
Dimitry Andric
8198e26969 glapi: adding @ char before type specifier in glapi_x86.S
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=33433
NOTE: This is a candidate for the 7.9 and 7.10 branches.

Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 37bffe8d12)
2011-02-21 17:28:36 -07:00
Julien Cristau
1b641f1b32 glx: fix request lengths
We were sending too long requests for GLXChangeDrawableAttributes,
GLXGetDrawableAttributes, GLXDestroyPixmap and GLXDestroyWindow.

NOTE: This is a candidate for the 7.9 and 7.10 branches

Signed-off-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 4324d6fdfb)
2011-02-21 17:26:10 -07:00
Dimitry Andric
8a9e6a57d0 mesa: s/movzxw/movzwl/ in read_rgba_span_x86.S
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=33386
NOTE: This is a candidate for the 7.9 and 7.10 branches

Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 811ee32a9e)
2011-02-21 17:25:17 -07:00
Dimitry Andric
279c03b98f mesa: s/movzx/movzbl/
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=33388
NOTE: This is a candidate for the 7.9 and 7.10 branches.

Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 3fda80246f)
2011-02-21 17:23:58 -07:00
Brian Paul
edfe056747 llvmpipe: make sure binning is active when we begin/end a query
This fixes a potential failure when a begin/end_query is the first
thing to happen after flushing the scene.

NOTE: This is a candidate for the 7.10 and 7.9 branches.
(cherry picked from commit 42dbc2530b)
2011-02-21 17:23:13 -07:00
Brian Paul
8ad9670866 mesa: check for dummy renderbuffer in _mesa_FramebufferRenderbufferEXT()
Fixes a failed assertion when a renderbuffer ID that was gen'd but not
previously bound was passed to glFramebufferRenderbuffer().  Generate
the same error that NVIDIA does.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit dd973cd9e8)
2011-02-21 17:21:59 -07:00
Brian Paul
09cb5815ae mesa: don't assert in GetIntegerIndexed, etc
We were getting an assertion upon invalid pname.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit 67722ae403)
2011-02-21 17:21:34 -07:00
Brian Paul
d1424f0185 mesa: fix num_draw_buffers==0 in fixed-function fragment program generation
This fixes a problem when glDrawBuffers(GL_NONE).  The fragment program
was writing to color output[0] but OutputsWritten was 0.  That led to a
failed assertion in the Mesa->TGSI translation code.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit 2fa6012f6a)
2011-02-21 17:20:54 -07:00
Brian Paul
7e43495053 mesa: fix a few format table mistakes, assertions
The BaseFormat field was incorrect for a few R and RG formats.
Fix a couple assertions too.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit 0073f50cd4)
2011-02-21 17:19:47 -07:00
Ian Romanick
246716c8f6 docs: Update 7.9.2 release notes
Add recent cherry picks
2011-02-21 14:47:43 -08:00
Ian Romanick
8159bef77f mesa: Fix error checks in GetVertexAttrib functions
Querying index zero is not an error in OpenGL ES 2.0.

Querying an index larger than the value returned by
GL_MAX_VERTEX_ATTRIBS is an error in all APIs.

Fixes bugzilla #32375.
(cherry picked from commit 5c3f1cdbbe)
2011-02-21 14:47:38 -08:00
Ian Romanick
d8392a2ca8 linker: Generate link errors when ES shaders are missing stages
ES requires that a vertex shader and a fragment shader be present.

Fixes bugzilla #32214.
(cherry picked from commit ce9171f9d8)
2011-02-21 13:50:35 -08:00
Ian Romanick
ac270e43e6 mesa: glGetUniform only returns a single element of an array
Also return it as the correct type.  Previously the whole array would
be returned and each element would be expanded to a vec4.

Fixes piglit test getuniform-01 and bugzilla #29823.

(cherry picked from commit 20d278a7ff)
2011-02-21 13:50:03 -08:00
Eric Anholt
fc09445c87 intel: Make renderbuffer tiling choice match texture tiling choice.
There really shouldn't be any difference between the two for us.
Fixes a bug where Z16 renderbuffers would be untiled on gen6, likely
leading to hangs.
(cherry picked from commit 29c4f95cbc)
(cherry picked from commit 4e0d6cf7ba)
2011-02-21 13:48:46 -08:00
Ian Romanick
0f3a9643c6 Use C-style system headers in C++ code to avoid issues with std:: namespace
Based on commit 497baf4e4a from master.
2011-02-21 13:19:05 -08:00
Eric Anholt
59422a1716 glsl: Fix the lowering of variable array indexing to not lose write_masks.
Fixes glsl-complex-subscript on 965.
(cherry picked from commit c00bc13564)
2011-02-21 13:19:05 -08:00
Marek Olšák
231e72cdbd mesa: fix texture3D mipmap generation for UNSIGNED_BYTE_3_3_2
(cherry picked from commit d5a1325f81)
2011-02-16 20:53:39 +01:00
Bryce Harrington
da623c9c44 r300g: Null pointer check for buffer deref in gallium winsys
radeon_drm_bufmgr_create_buffer_from_handle() can return NULL buffers
sometimes (seen when alt-tabbing in compiz).  Avoid dereferencing the
buffer pointer in this case.

Ref.: https://bugs.launchpad.net/ubuntu/+source/mesa/+bug/691653
Also: https://bugzilla.redhat.com/show_bug.cgi?id=660143

Signed-off-by: Bryce Harrington <bryce@canonical.com>
2011-02-11 02:32:48 +01:00
Vinson Lee
175f3c9674 ralloc: Add missing va_end following va_copy.
(cherry picked from commit cde443e0b9)
2011-02-07 15:03:10 -08:00
Kenneth Graunke
4e4a9d4050 Remove the talloc sources from the Mesa repository.
(cherry picked from commit 1568b19e3b)
2011-02-07 15:02:30 -08:00
Kenneth Graunke
f7993b66a6 Remove talloc from the SCons build system.
(cherry picked from commit 1f4fbc56e52ff2215f7eac7fc84ef8d8a8ec3c2f)

Conflicts:
	src/gallium/targets/libgl-gdi/SConscript
	src/mesa/SConscript
2011-02-07 15:02:30 -08:00
Kenneth Graunke
7f1a7a26e0 Remove talloc from the make and automake build systems.
(cherry picked from commit d610a57c8c69a0197ac7c3d0b22be1b8f0dc446f)
2011-02-07 15:02:30 -08:00
Kenneth Graunke
cdcc5da9fe ralloc: a new MIT-licensed recursive memory allocator.
(cherry picked from commit 42fd9c2ebb)
2011-02-07 15:02:30 -08:00
Kenneth Graunke
5afd1d6ac2 Convert everything from the talloc API to the ralloc API.
(cherry-picked from 2ac83dba30dfe6f61b2c9639100acbb329ac4543)

Conflicts:
	src/glsl/ast_function.cpp
	src/glsl/glcpp/glcpp.c
	src/glsl/glsl_parser_extras.cpp
	src/glsl/ir.cpp
	src/glsl/ir.h
	src/glsl/ir_explog_to_explog2.cpp
	src/glsl/ir_mat_op_to_vec.cpp
	src/glsl/ir_sub_to_add_neg.cpp
	src/glsl/lower_texture_projection.cpp
	src/glsl/lower_variable_index_to_cond_assign.cpp
	src/glsl/lower_vector.cpp
	src/mesa/drivers/dri/i965/brw_cubemap_normalize.cpp
	src/mesa/drivers/dri/i965/brw_fs.cpp
	src/mesa/drivers/dri/i965/brw_fs.h
	src/mesa/drivers/dri/i965/brw_fs_reg_allocate.cpp
	src/mesa/main/shaderapi.c
	src/mesa/main/shaderobj.h
	src/mesa/program/ir_to_mesa.cpp
	src/mesa/program/register_allocate.c
	src/mesa/program/sampler.cpp
2011-02-07 15:02:30 -08:00
Kenneth Graunke
c9954dab04 ralloc: Add a fake implementation of ralloc based on talloc.
(cherry picked from commit dc55254f5b)

Conflicts:
	src/glsl/SConscript
2011-02-07 15:02:30 -08:00
Kenneth Graunke
e33d9b48c4 glcpp: Remove use of talloc reference counting.
We almost always want to simply steal; we only need to copy when copying
a token list (in which case we're already cloning stuff anyway).

(cherry picked from commit 6ecee54a9a)
2011-02-07 15:02:30 -08:00
Kenneth Graunke
f5bef74066 glsl, i965: Remove unnecessary talloc includes.
These are already picked up by ir.h or glsl_types.h.

(cherry picked from commit e256e4743c)
2011-02-07 15:02:30 -08:00
Kenneth Graunke
5e8523895a glsl: Don't bother unsetting a destructor that was never set.
This was totally copied and pasted from glsl_symbol_table.

(cherry picked from commit 21031b4e88)
2011-02-07 15:02:29 -08:00
Ian Romanick
ccc11aacbb docs: Update 7.9.2 release notes
Add recent cherry picks for precision qualifers, linker bugs, and other issues.
2011-02-04 16:53:25 -08:00
Eric Anholt
9acbde11f5 i965/vs: When MOVing to produce ABS, strip negate of the operand.
We were returning the negative absolute value, instead of the absolute
value.  Fixes glsl-vs-abs-neg.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>

(cherry picked from commit 9351ef7a44)
2011-02-04 16:53:25 -08:00
Eric Anholt
6c6894eac7 docs: Add a relnote for the Civ IV on i965. 2011-02-04 16:53:24 -08:00
Eric Anholt
41d4af17a1 i965: Avoid double-negation of immediate values in the VS.
In general, we have to negate in immediate values we pass in because
the src1 negate field in the register description is in the bits3 slot
that the 32-bit value is loaded into, so it's ignored by the hardware.
However, the src0 negate field is in bits1, so after we'd negated the
immediate value loaded in, it would also get negated through the
register description.  This broke this VP instruction in the position
calculation in civ4:

MAD TEMP[1], TEMP[1], CONST[256].zzzz, CONST[256].-y-y-y-y;

Bug #30156
(cherry picked from commit 1d1ad6306d)
2011-02-04 16:53:24 -08:00
Chris Wilson
bf3932aad6 intel: Check for unsupported texture when finishing using as a render target
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=32541
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
(cherry picked from commit 8b9570e685)
2011-02-04 18:17:46 -06:00
Ian Romanick
2b1150170a linker: Propagate max_array_access while linking functions
Update the max_array_access of a global as functions that use that
global are pulled into the linked shader.

Fixes piglit test glsl-fs-implicit-array-size-01 and bugzilla #33219.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit 0f4b2a0a23)
2011-02-04 15:18:20 -08:00
Ian Romanick
a46bc59b27 linker: Set sizes for non-global arrays as well
Previously only global arrays with implicit sizes would be patched.
This causes all arrays that are actually accessed to be sized.

Fixes piglit test glsl-fs-implicit-array-size-02.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit c87e9ef4d2)
2011-02-04 15:18:15 -08:00
Ian Romanick
1f34edbbde glsl: Don't assert when the value returned by a function has no rvalue
The rvalue of the returned value can be NULL if the shader says
'return foo();' and foo() is a function that returns void.

Existing GLSL specs do *NOT* say that this is an error.  The type of
the return value is void.  If the return type of the function is also
void, then this should compile without error.  I expect that future
versions of the GLSL spec will fix this (wink, wink, nudge, nudge).

Fixes piglit test glsl-1.10/compiler/expressions/return-01.vert and
bugzilla #33308.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit 2db46fe5f0)
2011-02-04 15:18:06 -08:00
Kenneth Graunke
427d574920 glsl: Expose a public glsl_type::void_type const pointer.
This is analogous to glsl_type::int_type and all the others.
(cherry picked from commit 5c229e5fbd)
2011-02-04 15:18:00 -08:00
Chad Versace
13711c8a5a glsl: Mark 'in' variables at global scope as read-only
Fixes Piglit tests:
spec/glsl-1.30/compiler/storage-qualifiers/static-write-centroid-in-01.frag
spec/glsl-1.30/compiler/storage-qualifiers/static-write-in-01.frag
spec/glsl-1.30/compiler/storage-qualifiers/static-write-in-02.frag

(cherry picked from commit 01a584d093)

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2011-02-04 15:16:56 -08:00
Chad Versace
3ddf406530 glsl: Fix segfault due to missing printf argument
Fixes the following Piglit tests:
glslparsertest/shaders/array2.frag
glslparsertest/shaders/dataType6.frag

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit 46f7105df4)
2011-02-04 14:37:06 -08:00
Chad Versace
a246791c20 glsl: Fix semantic checks on precision qualifiers
The check for
   Precision qualifiers only apply to floating point and integer types.
was incomplete. It rejected only type 'bool' and structures.
(cherry picked from commit 45e8e6c6b1)
2011-02-04 14:36:23 -08:00
Ian Romanick
b481b8059d glsl: Refresh autogenerated lexer and parser files.
For the previous few commits.
2011-02-04 14:35:59 -08:00
Chad Versace
6fc2c15dbd glsl: Remove redundant semantic check in parser
The removed semantic check also exists in ast_type_specifier::hir(), which
is a more natural location for it.

The check verified that precision statements are applied only to types
float and int.
(cherry picked from commit a9bf8c12ee)
2011-02-04 14:33:50 -08:00
Chad Versace
f5819379ed glsl: Add support for default precision statements
* Add new field ast_type_specifier::is_precision_statement.
* Add semantic checks in ast_type_specifier::hir().
* Alter parser rules accordingly.
(cherry picked from commit 08a286c9cc)
2011-02-04 14:33:37 -08:00
Chad Versace
dbc19fe94c glsl: Add semantic checks for precision qualifiers
* Check that precision qualifiers only appear in language versions 1.00,
  1.30, and later.
* Check that precision qualifiers do not apply to bools and structs.

Fixes the following Piglit tests:
* spec/glsl-1.30/precision-qualifiers/precision-bool-01.frag
* spec/glsl-1.30/precision-qualifiers/precision-struct-01.frag
* spec/glsl-1.30/precision-qualifiers/precision-struct-02.frag

(cherry picked from commit 889e1a5b6c)

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2011-02-04 14:33:12 -08:00
Chad Versace
6292060193 glsl: Change default value of ast_type_specifier::precision
Change default value to ast_precision_none, which denotes the absence of
a precision of a qualifier.

Previously, the default value was ast_precision_high. This made it
impossible to detect if a precision qualifier was present or not.
(cherry picked from commit aaa31bf8f4)
2011-02-04 14:31:36 -08:00
Chad Versace
378910ab97 glsl: Fix parser rule for type_specifier
Do not assign a value to ast_type_specifier::precision when no precision
qualifier is present.
(cherry picked from commit 33279cd2d3)
2011-02-04 14:31:30 -08:00
Ian Romanick
2925a5a1f5 i965: Make OPCODE_KIL_NV do its work in a temp, not the null reg!
This is similar to commit 0efea25c in 7.10.  The change was made
directly in the 7.9 branch because the affected code does not exist in
Mesa 7.10 or later.  It was removed with the change to the new
fragment shader back-end.

Fixes the GPU hang in piglit test glsl-fs-discard-03.
2011-02-04 10:05:56 -08:00
Ian Romanick
b632543289 docs: Update 7.9.2 release notes
Replace "Fix an error in uniform arrays in row calculating" with the
actual bugzilla that was fixed.
2011-01-26 10:16:19 -08:00
Ian Romanick
9e15c878bb mesa: bump version to 7.9.2-devel 2011-01-25 16:27:22 -08:00
Ian Romanick
ab6c7bbf4e docs: Initial set of release notes for 7.9.2 2011-01-25 16:22:26 -08:00
Jian Zhao
e4c1bcacef mesa: fix an error in uniform arrays in row calculating.
Fix the error in uniform row calculating, it may alloc one line
more which may cause out of range on memory usage, sometimes program
aborted when free the memory.

NOTE: This is a candidate for 7.9 and 7.10 branches.

Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 2a7380e9c3)
2011-01-19 10:58:06 -08:00
Ian Romanick
179edc673b glsl: Allow 'in' and 'out' when 'layout' is also available
All of the extensions that add the 'layout' keyword also enable (and
required) the use of 'in' and 'out' with shader globals.

This is related to (piglit) bugzilla #31804.

NOTE: This is a candidate for the 7.9 and 7.10 branches.

(cherry picked from commit 82c4b4f88a)
2011-01-19 10:50:35 -08:00
Ian Romanick
48e81e3b9e glsl: Track variable usage, use that to enforce semantics
In particular, variables cannot be redeclared invariant after being
used.

Fixes piglit test invariant-05.vert and bugzilla #29164.

NOTE: This is a candidate for the 7.9 and 7.10 branches.

(cherry picked from commit bd33055ef4)
2011-01-19 10:48:39 -08:00
Ian Romanick
1785904c5c glsl: Disallow 'in' and 'out' on globals in GLSL 1.20
Fixes piglit tests glsl-1.20/compiler/qualifiers/in-01.vert and
glsl-1.20/compiler/qualifiers/out-01.vert and bugzilla #32910.

NOTE: This is a candidate for the 7.9 and 7.10 branches.  This patch
also depends on the previous two commits.

(cherry picked from commit 469ea695bb)
2011-01-19 10:47:32 -08:00
Ian Romanick
28d47fed61 glsl & glcpp: Refresh autogenerated lexer and parser files.
For the previous few commits.
2011-01-19 10:46:15 -08:00
Ian Romanick
df033be3f7 glsl: Add version_string containing properly formatted GLSL version
(cherry picked from commit eebdfdfbcf)
2011-01-19 10:43:03 -08:00
Ian Romanick
ef2f4b5469 glcpp: Generate an error for division by zero
When GCC encounters a division by zero in a preprocessor directive, it
generates an error.  Since the GLSL spec says that the GLSL
preprocessor behaves like the C preprocessor, we should generate that
same error.

It's worth noting that I cannot find any text in the C99 spec that
says this should be an error.  The only text that I can find is line 5
on page 82 (section 6.5.5 Multiplicative Opertors), which says,

    "The result of the / operator is the quotient from the division of
    the first operand by the second; the result of the % operator is
    the remainder. In both operations, if the value of the second
    operand is zero, the behavior is undefined."

Fixes 093-divide-by-zero.c test and bugzilla #32831.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit 9ca5300b6e)
2011-01-19 10:42:55 -08:00
Chad Versace
8d1e0de5f7 glcpp: Fix segfault when validating macro redefinitions
In _token_list_equal_ignoring_space(token_list_t*, token_list_t*), add
a guard that prevents dereferncing a null token list.

This fixes test src/glsl/glcpp/tests/092-redefine-macro-error-2.c and
Bugzilla #32695.
(cherry picked from commit 4fff52f1c9)
2011-01-19 10:42:45 -08:00
Chad Versace
5f9c50053c glsl: At link-time, check that globals have matching centroid qualifiers
Fixes bug 31923: http://bugs.freedesktop.org/show_bug.cgi?id=31923
(cherry picked from commit 61428dd2ab)
2011-01-19 10:42:38 -08:00
Ian Romanick
7757980139 glsl: Support the 'invariant(all)' pragma
Previously the 'STDGL invariant(all)' pragma added in GLSL 1.20 was
simply ignored by the compiler.  This adds support for setting all
variable invariant.

In GLSL 1.10 and GLSL ES 1.00 the pragma is ignored, per the specs,
but a warning is generated.

Fixes piglit test glsl-invariant-pragma and bugzilla #31925.

NOTE: This is a candidate for the 7.9 and 7.10 branches.

(cherry-picked from commit 86b4398cd1)
2011-01-19 10:42:00 -08:00
Brian Paul
dab59f70e6 draw: Fix an off-by-one bug in a vsplit assertion.
When use_spoken is true, istart (the first vertex of this segment) is
replaced by i0 (the spoken vertex of the fan).  There are still icount
vertices.

Thanks to Brian Paul for spotting this.
(cherry picked from commit abbb1c8f08)
2011-01-13 11:35:24 -07:00
Alberto Milone
4200c6bae4 r600c: add evergreen ARL support.
Signed-off-by: Alberto Milone <alberto.milone@canonical.com>
2011-01-11 14:49:30 -05:00
Marek Olšák
67aeab0b77 docs: fix messed up names with special characters in relnotes-7.9.1 2011-01-08 02:54:51 +01:00
Ian Romanick
f912245ed1 docs: Add 7.9.1 md5sums 2011-01-07 13:31:30 -08:00
Ian Romanick
42a9472dc6 docs: Add change log to 7.9.1 release notes 2011-01-07 13:28:22 -08:00
Ian Romanick
234144b48d mesa: set version string to 7.9.1 2011-01-07 12:16:29 -08:00
Ian Romanick
3d90628138 docs: Initial set of release notes for 7.9.1 2011-01-07 12:12:55 -08:00
Marek Olšák
d0f0a6920b r300/compiler: disable the rename_regs pass for loops
This workaround fixes rendering of kwin thumbnails.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit 8543902bfb)

Conflicts:

	src/mesa/drivers/dri/r300/compiler/radeon_rename_regs.c
2011-01-07 07:25:12 +01:00
Chia-I Wu
110c49fddd egl: Check extensions.
Do not call into the driver if the extension for the called function is
not enabled.
(cherry picked from commit 6b2f1561ad)
2011-01-07 12:43:38 +08:00
Chia-I Wu
c438a09b4c egl_glx: Fix borken driver.
The driver was broken since 6eda3f311b.
All configs fail to pass _eglValidateConfig.  This fix might not be
complete, but at least EGL demos run.

This fixes bug #32825.
2011-01-05 18:54:44 +08:00
Brian Paul
0ab166d556 mesa: Bump the number of bits in the register index.
More than 1023 temporaries were being used for a Cinebench shader before
doing temporary optimization, causing the index value to wrap around to
-1024.
(cherry picked from commit a9fa0f3a2f)
2011-01-04 11:10:19 -07:00
Kenneth Graunke
b50658c804 i965: Internally enable GL_NV_blend_square on ES2.
Hopefully should fix bug #32520.
(cherry picked from commit 6bb1e4541e)
2011-01-04 09:50:14 -08:00
Kenneth Graunke
3c8dd6a8de Fix build on systems where "python" is python 3.
First, it changes autoconf to use a "python2" binary when available,
rather than plain "python" (which is ambiguous).  Secondly, it changes
the Makefiles to use $(PYTHON) $(PYTHON_FLAGS) rather than calling
python directly.

Signed-off-by: Xavier Chantry <chantry.xavier@gmail.com>
Signed-off-by: Matthew William Cox <matt@mattcox.ca>
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
(cherry picked from commit 3acc826520)
2011-01-04 09:50:14 -08:00
Marek Olšák
801e26a02a mesa: fix texel store functions for some float formats
These are copy-paste errors obviously.
(cherry picked from commit bf7b6f60ae)
2011-01-04 09:50:14 -08:00
Ian Romanick
247774085c Refresh autogenerated file builtin_function.cpp.
See also a954dbeb.
2011-01-04 09:50:14 -08:00
Kenneth Graunke
d5d573b592 glsl/builtins: Compute the correct value for smoothstep(vec, vec, vec).
These mistakenly computed 't' instead of t * t * (3.0 - 2.0 * t).

Also, properly vectorize the smoothstep(float, float, vec) variants.

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit d7423a6531)
2011-01-04 09:50:14 -08:00
Brian Paul
54c2846f23 st/mesa: fix renderbuffer pointer check in st_Clear()
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=30694

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit efbd33aff9)
2011-01-04 08:00:25 -07:00
Brian Paul
8941450010 tdfx: s/Format/_BaseFormat/
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=31560
(cherry picked from commit d18df9e336)
2011-01-04 07:53:44 -07:00
Brian Paul
0ff45dd5d7 mesa: no-op glBufferSubData() on size==0
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=31439

NOTE: this is a candidate for the 7.9 branch
(cherry picked from commit 6e2e136428)
2011-01-03 17:05:24 -08:00
Kenneth Graunke
4e7cdca57a Remove OES_compressed_paletted_texture from the ES2 extension list.
We don't support it.
(cherry picked from commit d0f8eea9a0)
2011-01-03 17:05:24 -08:00
Brian Paul
69cc356832 glsl: new glsl_strtod() wrapper to fix decimal point interpretation
We always want to use '.' as the decimal point.

See http://bugs.freedesktop.org/show_bug.cgi?id=24531

NOTE: this is a candidate for the 7.10 branch.

(cherry picked from commit bb10e081c8)
2011-01-03 17:05:24 -08:00
Ian Romanick
0d98ceb4bf ir_to_mesa: Don't generate swizzles for record derefs of non-scalar/vectors
This is the same as what the array dereference handler does.

Fixes piglit test glsl-link-struct-array (bugzilla #31648).

NOTE: This is a candidate for the 7.9 and 7.10 branches.
(cherry picked from commit 2d577ee730)
2011-01-03 17:05:24 -08:00
Ian Romanick
40390b80e9 linker: Allow built-in arrays to have different sizes between shader stages
Fixes pitlit test glsl-link-varying-TexCoord (bugzilla #31650).
(cherry picked from commit cb2b547a47)
2011-01-03 17:05:24 -08:00
Henri Verbeet
7ffeced927 st/mesa: Handle wrapped depth buffers in st_copy_texsubimage().
(cherry picked from commit 59051ad443)
2010-12-31 07:51:53 +01:00
Henri Verbeet
b8c16ee7b1 r600: Evergreen has two extra frac_bits for the sampler LOD state.
Note: this is a candidate for the 7.9 branch.
(cherry picked from commit 228d0d1153)
2010-12-31 07:51:49 +01:00
Brian Paul
f5435ec3a9 st/mesa: fix mipmap generation bug
In st_finalize_texture() we were looking at the st_texture_object::
lastLevel field instead of the pipe_resource::last_level field to
determine which resource to store the mipmap in.

Then, in st_generate_mipmap() we need to call st_finalize_texture() to
make sure the destination resource is properly allocated.

These changes fix the broken piglit fbo-generatemipmap-formats test.
(cherry picked from commit cae2bb76c1)
2010-12-30 17:40:07 +01:00
richard
a3740ba31e r600c : inline vertex format is not updated in an app, switch to use vfetch constants. For the 7.9 and 7.10 branches as well. 2010-12-29 10:54:44 -05:00
Alex Deucher
166da97889 r600c: fix some opcodes on evergreen
There were a few places where we were using the wrong opcodes
on evergreen.  arl still needs to be fixed on evergreen; see
r600g for reference.

NOTE: This is a candidate for the 7.9 branch.

Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
2010-12-29 10:54:14 -05:00
Luca Barbieri
2fb170c90f glsl: Unroll loops with conditional breaks anywhere (not just the end)
Currently we only unroll loops with conditional breaks at the end, which is
the form that lower_jumps generates.

However, if breaks are not lowered, they tend to appear at the beginning, so
add support for a conditional break anywhere.

Signed-off-by: Luca Barbieri <luca@luca-barbieri.com>
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
(cherry picked from commit 0e50c21e24)
2010-12-27 13:12:20 -08:00
Kenneth Graunke
4f2f3440d0 glsl: Consider the "else" branch when looking for loop breaks.
Found this bug by code inspection.  Based off the comments just before
this code, the intent is to find whether the break exists in the "then"
branch or the "else" branch.  However, the code actually looked at the
last instruction in the "then" branch twice.
(cherry picked from commit 13c45c590b)
2010-12-27 13:12:12 -08:00
Kenneth Graunke
760288bec1 glsl: Clean up code by adding a new is_break() function.
(cherry picked from commit 528fa8ce32)
2010-12-27 13:12:04 -08:00
Eric Anholt
84e7dd0114 glsl: Correct the marking of InputsRead/OutputsWritten on in/out matrices.
If you used a constant array index to access the matrix, we'd flag a
bunch of wrong inputs/outputs as being used because the index was
multiplied by matrix columns and the actual used index was left out.

Fixes glsl-mat-attribute.
(cherry picked from commit b13a2e640f)
2010-12-27 13:11:54 -08:00
Fabian Bieler
9d1c6dd79b glsl: fix lowering conditional returns in subroutines
this fix applies to the lower_sub_return 'branch' of the lower_jumps pass

Fixes piglit tests glsl-functions-5 and glsl-functions-6.
(cherry picked from commit ef534f3838)
2010-12-27 13:11:26 -08:00
Chia-I Wu
47552b2c66 mesa: Do not advertise GL_OES_texture_3D.
GL_OES_texture_3D has a GLSL counterpart.  Since it is not implemented,
GL_OES_texture_3D should not be advertised.
(cherry picked from commit d2028ba339)
2010-12-27 13:11:26 -08:00
Ian Romanick
edfe057f94 linker: Fix regressions caused by previous commit
That's what I get for not running piglit before pushing.

Don't try to patch types of unsized arrays when linking fails.

Don't try to patch types of unsized arrays that are shared between
shader stages.
(cherry picked from commit 002cd2c8d4)
2010-12-27 13:11:26 -08:00
Ian Romanick
a4cda571f0 linker: Ensure that unsized arrays have a size after linking
Fixes piglit test case glsl-vec-array (bugzilla #31908).

NOTE: This bug does not affect 7.9, but I think this patch is a
candiate for the 7.9 branch anyway.
(cherry picked from commit 6f53921c4b)
2010-12-27 13:09:26 -08:00
Eric Anholt
ecb68d51aa i965: Correct the dp_read message descriptor setup on g4x.
It's mostly like gen4 message descriptor setup, except that the sizes
of type/control changed to be like gen5.  Fixes 21 piglit cases on
gm45, including the regressions in bug #32311 from increased VS
constant buffer usage.
(cherry picked from commit 5dc53444c8)
2010-12-23 14:27:11 -08:00
Marek Olšák
be1f4cfd96 r300g: finally fix the texture corruption on r3xx-r4xx
Even though a bound texture stays bound when calling set_fragment_sampler_views,
it must be assigned a new cache region depending on the occupancy of other
texture units.

This fixes:
https://bugs.freedesktop.org/show_bug.cgi?id=28800

Thanks to Álmos <aaalmosss@gmail.com> for finding the bug in the code.

NOTE: This is a candidate for both the 7.9 and 7.10 branches.
(cherry picked from commit daffaca53e)

Conflicts:

	src/gallium/drivers/r300/r300_state.c
2010-12-18 10:17:52 +01:00
Ian Romanick
7b6b66d617 glsl: Inherrit type of declared variable from initializer after processing assignment
do_assignment may apply implicit conversions to coerce the base type
of initializer to the base type of the variable being declared.  Fixes
piglit test glsl-implicit-conversion-02 (bugzilla #32287).  This
probably also fixes bugzilla #32273.

NOTE: This is a candidate for the 7.9 branch and the 7.10 branch.
(cherry picked from commit d7f27e2e76)

Conflicts:

	src/glsl/ast_to_hir.cpp
2010-12-17 13:51:46 -08:00
Ian Romanick
2b4277dcf3 glsl: Inherrit type of declared variable from initializer
Types of declared variables and their initializer must match excatly
except for unsized arrays.  Previously the type inherritance for
unsized arrays happened implicitly in the emitted assignment.
However, this assignment is never emitted for uniforms.  Now that type
is explicitly copied unconditionally.

Fixes piglit test array-compare-04.vert (bugzilla #32035) and
glsl-array-uniform-length (bugzilla #31985).

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit b0fc5103cb)
2010-12-17 13:51:46 -08:00
Ian Romanick
585fa612c9 glcpp: Commit changes in generated files cause by previous commit 2010-12-17 13:51:45 -08:00
Kenneth Graunke
b8e73354d9 glcpp: Don't emit SPACE tokens in conditional_tokens production.
Fixes glslparsertest defined-01.vert.

Reported-by: José Fonseca <jfonseca@vmware.com>
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Acked-by: Carl Worth <cworth@cworth.org>
(cherry picked from commit 800eed6765)
2010-12-17 13:14:25 -08:00
Eric Anholt
154d0a3e7e mesa: Add getters for ARB_copy_buffer's attachment points.
Fixes more complaints by oglconform.
(cherry picked from commit 43491adc44)
2010-12-17 12:58:15 -08:00
Eric Anholt
7c0ff4a5df mesa: Add getters for the rest of the supported draw buffers.
MAX_DRAW_BUFFERS is 8, so allow all 8 GL_DRAW_BUFFER# to be retrieved.
Fixes complaints by oglconform.
(cherry picked from commit 7cba339375)
2010-12-17 12:57:57 -08:00
Eric Anholt
d192179605 glsl: Fix flipped return of has_value() for array constants.
Fixes glsl-array-uniform.
(cherry picked from commit b381eff141)
2010-12-17 12:57:35 -08:00
Chad Versace
b2cd388b79 glsl: Fix linker bug in cross_validate_globals()
Cause linking to fail if a global has mismatching invariant qualifiers.

See https://bugs.freedesktop.org/show_bug.cgi?id=30261
(cherry picked from commit 7528f143df)
2010-12-17 12:56:40 -08:00
Aras Pranckevicius
bb2cd98e51 glsl: fix matrix type check in ir_algebraic
Fixes glsl-mat-mul-1.
(cherry picked from commit 4ce084c707)
2010-12-16 17:11:50 -08:00
Ian Romanick
d9d4f22e65 glsl: Ensure that equality comparisons don't return a NULL IR tree
This fixes bugzilla #32035 and piglit test case array-compare-01 and
array-compare-02.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 6d36be508f)
2010-12-16 17:11:50 -08:00
Eric Anholt
d347b90b94 glsl: Mark the array access for whole-array comparisons.
By not doing so, the uniform contents of
glsl-uniform-non-uniform-array-compare.shader_test was getting thrown
out since nobody was recorded as dereferencing the array.
(cherry picked from commit b4f585665c)
2010-12-16 17:11:50 -08:00
Eric Anholt
23a868930c glsl: Quiet unreachable no-return-from-function warning.
(cherry picked from commit d56c97413e)
2010-12-16 16:56:51 -08:00
Eric Anholt
1e88b3461a glsl: Fix structure and array comparisions.
We were trying to emit a single ir_expression to compare the whole
thing.  The backends (ir_to_mesa.cpp and brw_fs.cpp so far) expected
ir_binop_any_nequal or ir_binop_all_equal to apply to at most a vector
(with matrices broken down by the lowering pass).  Break them down to
a bunch of ORed or ANDed any_nequals/all_equals.

Fixes:
glsl-array-compare
glsl-array-compare-02
glsl-fs-struct-equal
glsl-fs-struct-notequal
Bug #31909
(cherry picked from commit ff79633d9f)
2010-12-16 16:56:51 -08:00
Eric Anholt
573c78803d glsl: Add a helper constructor for expressions that works out result type.
This doesn't cover all expressions or all operand types, but it will
complain if you overreach and it allows for much greater slack on the
programmer's part.
(cherry picked from commit 6b937465d4)

The new constructors required some changes because the
ir_quadop_vector changes were not cherry picked over from master.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2010-12-16 16:56:51 -08:00
Ian Romanick
5650cc462d glsl: Add unary ir_expression constructor
(cherry picked from commit 13f57d42b6)
2010-12-16 16:56:51 -08:00
Vinson Lee
54a596e34a glsl: Fix 'control reaches end of non-void function' warning.
Fix this GCC warning.
ir.cpp: In static member function
'static unsigned int ir_expression::get_num_operands(ir_expression_operation)':
ir.cpp:199: warning: control reaches end of non-void function
(cherry picked from commit 855c66bde7)
2010-12-16 16:56:51 -08:00
Kenneth Graunke
2127692542 glsl: Refactor get_num_operands.
This adds sentinel values to the ir_expression_operation enum type:
ir_last_unop, ir_last_binop, and ir_last_opcode.  They are set to the
previous one so they don't trigger "unhandled case in switch statement"
warnings, but should never be handled directly.

This allows us to remove the huge array of 1s and 2s in
ir_expression::get_num_operands().
(cherry picked from commit 007f488150)
2010-12-16 16:56:40 -08:00
Ian Romanick
0795cb71fc i915: Correctly generate unconditional KIL instructions
Fixes piglit test glsl-fs-discard-03.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit c92550be64)
2010-12-16 10:34:59 -08:00
Kenneth Graunke
cefccebd89 glsl: Remove "discard" support from lower_jumps.
The new lower_discard and opt_discard_simplification passes should
handle all the necessary transformations, so lower_jumps doesn't need to
support it.

Also, lower_jumps incorrectly handled conditional discards - it would
unconditionally truncate all code after the discard.  Rather than fixing
the bug, simply remove the code.

NOTE: This is a candidate for the 7.9 branch.
2010-12-16 10:34:59 -08:00
Ian Romanick
927c752dd6 glsl: Add a lowering pass to move discards out of if-statements.
This should allow lower_if_to_cond_assign to work in the presence of
discards, fixing bug #31690 and likely #31983.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 940df10100)
2010-12-16 10:34:59 -08:00
Marek Olšák
0b4287a687 ir_to_mesa: Add support for conditional discards.
NOTE: This is a candidate for the 7.9 branch.

Signed-off-by: Marek Olšák <maraeo@gmail.com>
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
(cherry picked from commit ead2ea89f4)
2010-12-16 10:34:59 -08:00
Ian Romanick
69d19cb84b glsl: Add an optimization pass to simplify discards.
NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 9a1d063c6d)
2010-12-16 10:34:59 -08:00
Ian Romanick
0ab2bb8270 glsl/linker: Free any IR discarded by optimization passes.
Previously, IR for a linked shader was allocated directly out of the
gl_shader object - meaning all of it lived as long as the shader.

Now, IR is allocated out of a temporary context, and any -live- IR is
reparented/stolen to (effectively) the gl_shader.  Any remaining IR can
be freed.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 2da02e75b1)
2010-12-16 10:34:59 -08:00
Kenneth Graunke
1e2e387a15 glsl: Add a virtual as_discard() method.
NOTE: This is candidate for the 7.9 branch.
(cherry picked from commit 1eb7a81f2e)
2010-12-16 10:34:59 -08:00
Marek Olšák
dcc7a8a80a r300g: fix rendering with a vertex attrib having a zero stride
The hardware apparently does support a zero stride, so let's use it.

This fixes missing objects in ETQW, but might also fix a ton of other
similar-looking bugs.

NOTE: This is a candidate for both the 7.9 and 7.10 branches.
(cherry picked from commit 2af8a19831)
2010-12-16 16:55:59 +01:00
Marek Olšák
0a8ad737e7 r300/compiler: fix swizzle lowering with a presubtract source operand
If a source operand has a non-native swizzle (e.g. the KIL instruction
cannot have a swizzle other than .xyzw), the lowering pass uses one or more
MOV instructions to move the operand to an intermediate temporary with
native swizzles.

This commit fixes that the presubtract information was lost during
the lowering.

NOTE: This is a candidate for both the 7.9 and 7.10 branches.
(cherry picked from commit d0990db6bd)
2010-12-16 16:55:16 +01:00
Dave Airlie
4b19941399 r300g: fixup rs690 tiling stride alignment calculations.
The RS690 memory controller prefers things to be on a different
boundary than the discrete GPUs, we had an attempt to fix this,
but it still failed, this consolidates the stride calculation
into one place and removes the really special case check.

This fixes gnome-shell and 16 piglit tests on my rs690 system.

NOTE: This is a candidate for both the 7.9 and 7.10 branches.

Signed-off-by: Dave Airlie <airlied@redhat.com>
(cherry picked from commit 32218e4cc8)
2010-12-16 18:15:25 +10:00
Ian Romanick
1cfd517b15 i915: Disallow alpha, red, RG, and sRGB as render targets
Fixes bugzilla #31832

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 78a340fd48)
2010-12-15 15:11:01 -08:00
Eric Anholt
a0d5da96b5 i965: Add support for rendering to SARGB8 FBOs.
Tested with fbo-generatemipmap-formats GL_EXT_texture_srgb.  The test
still fails on SLA8, though.
(cherry picked from commit 1946b81e70)
2010-12-15 14:39:58 -08:00
Kenneth Graunke
50ec3c74e3 glsl: Don't inline function prototypes.
Currently, the standalone compiler tries to do function inlining before
linking shaders (including linking against the built-in functions).
This resulted in the built-in function _prototypes_ being inlined rather
than the actual function definition.

This is only known to fix a bug in the standalone compiler; most
programs should be unaffected.  Still, it seems like a good idea.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit e8a24c65bc)
2010-12-15 14:38:37 -08:00
Kenneth Graunke
d595657e5a glsl: Use do_common_optimization in the standalone compiler.
NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit a82592de92)
2010-12-15 14:38:37 -08:00
Ian Romanick
c055239bae mesa: Fix glGet of ES2's GL_MAX_*_VECTORS properties.
Previously, the get table listed all three as having custom locations,
yet find_custom_value did not have cases to handle them.

MAX_VARYING_VECTORS does not need a custom location since MaxVaryings is
already stored as float[4] (or vec4).  MaxUniformComponents is stored as
the number of floats, however, so a custom implementation that divides
by 4 is necessary.

Fixes bugs.freedesktop.org #31495.
(cherry picked from commit 1197393faa)

Conflicts:

	src/mesa/main/get.c

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2010-12-15 14:38:37 -08:00
Kristian Høgsberg
6b96b002c8 i965: Don't write mrf assignment for pointsize output
https://bugs.freedesktop.org/show_bug.cgi?id=31894
(cherry picked from commit a889f9ee5c)
2010-12-15 14:38:37 -08:00
Ian Romanick
ce8f4f266e glsl: better handling of linker failures
Upon link error, exit translation loop, free program instructions.
Check for null pointers in calling code.
(cherry picked from commit 6162773ea4)
2010-12-15 14:25:42 -08:00
Brian Paul
e0dbd44a00 glsl: fix off by one in register index assertion
(cherry picked from commit 50fd99d172)
2010-12-15 14:24:09 -08:00
Kenneth Graunke
67b7a3844a glsl: Fix constant expression handling for <, >, <=, >= on vectors.
ir_binop_less, ir_binop_greater, ir_binop_lequal, and ir_binop_gequal
are defined to work on vectors as well as scalars, as long as the two
operands have the same type.

This is evident from both ir_validate.cpp and our use of these opcodes
in the GLSL lessThan, greaterThan, lessThanEqual, greaterThanEqual
built-in functions.

Found by code inspection.  Not known to fix any bugs.  Presumably, our
tests for the built-in comparison functions must pass because C.E.
handling is done on the ir_call of "greaterThan" rather than the inlined
opcode.  The C.E. handling of the built-in function calls is correct.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit e16c9d5d03)
2010-12-15 14:19:16 -08:00
Chad Versace
69cc62b907 glsl: Fix erroneous cast in ast_jump_statement::hir()
Return values were erroneously cast from (ir_rvalue*) to (ir_expression*).

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit b4cdba687c)
2010-12-15 14:19:07 -08:00
Kenneth Graunke
8c8c1b2c99 glsl: Unconditionally define GL_FRAGMENT_PRECISION_HIGH in ES2 shaders.
This is really supposed to be defined only if the driver supports highp
in the fragment shader - but all of our current ES2 implementations do.
So, just define it.  In the future, we'll need to add a flag to
gl_context and only define the macro if the flag is set.

"Fixes" freedesktop.org bug #31673.
(cherry picked from commit d719bf8fb4)
2010-12-15 14:18:01 -08:00
Kenneth Graunke
9b8e56a502 glcpp: Define GL_FRAGMENT_PRECISION_HIGH if GLSL version >= 1.30.
Per section 4.5.4 of the GLSL 1.30 specification.
(cherry picked from commit 3fb83038a0)
2010-12-15 14:17:51 -08:00
Robert Hooker
5a8d1d5b72 egl_dri2: Add missing intel chip ids.
Signed-off-by: Robert Hooker <robert.hooker@canonical.com>
(cherry picked from commit 778917069c)
2010-12-15 14:17:29 -08:00
Robert Hooker
029252a448 intel: Add a new B43 pci id.
Signed-off-by: Robert Hooker <robert.hooker@canonical.com>
(cherry picked from commit e8b2d36723)
2010-12-15 14:17:11 -08:00
Peter Clifton
4b30da86a7 intel: Fix emit_linear_blit to use DWORD aligned width blits
The width of the 2D blits used to copy the data is defined as a 16-bit
signed integer, but the pitch must be DWORD aligned. Limit to an integral
number of DWORDs, (1 << 15 - 4) rather than (1 << 15 -1).

Fixes corruption to data uploaded with glBufferSubData.

Signed-off-by: Peter Clifton <pcjc2@cam.ac.uk>
(cherry picked from commit 10b9e018ca)
2010-12-15 12:39:53 -08:00
Eric Anholt
da5d6c42c0 i965: Allow OPCODE_SWZ to put immediates in the first arg.
Fixes assertion failure with texture swizzling in the GLSL path when
it's triggered (such as gen6 FF or ARB_fp shadow comparisons).

Fixes:
texdepth
texSwizzle
fp1-DST test
fp1-LIT test 3
(cherry picked from commit f00929cbdd)
2010-12-15 12:36:01 -08:00
Eric Anholt
0c56721cfc glsl: Free the loop state context when we free the loop state.
Since this was talloced off of NULL instead of the compile state, it
was a real leak over the course of the program.  Noticed with
valgrind --leak-check=full --show-reachable=yes.  We should really
change these passes to generally get the compile context as an argument
so simple mistakes like this stop mattering.
(cherry picked from commit 6929cdd14b)
2010-12-15 12:26:59 -08:00
Aras Pranckevicius
6333f1d224 glsl: fix crash in loop analysis when some controls can't be determined
Fixes loop-07.frag.
(cherry picked from commit d67df5dd9d)
2010-12-15 12:26:34 -08:00
Ian Romanick
639e65e054 mesa: Allow query of MAX_SAMPLES with EXT_framebuffer_multisample
Previously queries of MAX_SAMPLES were only allowed with
ARB_framebuffer_object, but EXT_framebuffer_multisample also enables
this query.  This seems to only effect the i915.  All other drivers
support both extensions or neither extension.

This patch is based on a patch that Kenneth sent along with the report.

NOTE: this is a candidate for the 7.9 branch.

Reported-by: Kenneth Waters <kwaters@chromium.org>
(cherry picked from commit bcef51c3b8)
2010-12-15 12:26:19 -08:00
Ian Romanick
8454f931d6 glsl: Fix incorrect gl_type of sampler2DArray and sampler1DArrayShadow
NOTE: this is a candidate for the 7.9 branch.
(cherry picked from commit 956ae44dcf)
2010-12-15 12:24:45 -08:00
Chad Versace
0da1c7b90f glsl: Fix ir_expression::constant_expression_value()
When the type of the ir_expression is error_type, return NULL.
This fixes bug 31371.
(cherry picked from commit b62c1c4595)
2010-12-15 12:24:25 -08:00
Kenneth Graunke
6e5907c390 ir_dead_functions: Actually free dead functions and signatures.
This makes linked shaders use around 36k less memory since the
built-in prototypes are now freed.
(cherry picked from commit a457ca7844)
2010-12-15 12:24:15 -08:00
Francisco Jerez
4c1db8c7a4 meta: Don't try to disable cube maps if the driver doesn't expose the extension.
Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 8eaa97592a)
2010-12-15 12:20:20 -08:00
Francisco Jerez
edd9b9350e meta: Fix incorrect rendering of the bitmap alpha component.
Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 7831994868)
2010-12-15 12:19:02 -08:00
Francisco Jerez
5c71690096 meta: Don't leak alpha function/reference value changes.
Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit d846362389)
2010-12-15 12:18:44 -08:00
Ian Romanick
44df2b0844 mesa: Make metaops use program refcounts instead of names.
Fixes failure on restoring state when the program was active but
deleted, and the name no longer exists.

Bug #31194
(cherry picked from commit a974949f3b)

Conflicts:

	src/mesa/drivers/common/meta.c
	src/mesa/main/shaderapi.c
	src/mesa/main/shaderapi.h

This required substantial edits because the 7.9 branch doesn't have
the EXT_separate_shader_objects changes.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2010-12-15 12:17:00 -08:00
Brian Paul
5b28db24ff mesa, st/mesa: disable GL_ARB_geometry_shader4
The new GLSL compiler doesn't support geom shaders yet so disable the
GL_ARB_geometry_shader4 extension.  Undo this when geom shaders work again.

NOTE: This is a candidate for the 7.10 branch.

(cherry picked from commit bb7c2691d2)
2010-12-14 16:29:52 -07:00
Brian Paul
2f4af9e37b tnl: a better way to initialize the gl_program_machine memory
This improves commit ef3f7e61b3

NOTE: This is a candidate for the 7.9 and 7.10 branches.

(cherry picked from commit 6577f753b2)
2010-12-14 16:29:39 -07:00
Brian Paul
34e0da8e73 tnl: Initialize gl_program_machine memory in run_vp.
Fixes piglit valgrind glsl-array-bounds-04 failure (FDO bug 29946).

NOTE:
This is a candidate for the 7.10 branch.
This is a candidate for the 7.9 branch.

(cherry picked from commit ef3f7e61b3)
2010-12-14 16:29:06 -07:00
Brian Paul
1ee289960f mesa: test for cube map completeness in glGenerateMipmap()
The texture is not cube complete if the base level images aren't of
the same size and format.

NOTE: This is a candidate for the 7.9 branch.

(cherry picked from commit ecb7cc3319)
2010-12-14 16:28:14 -07:00
Brian Paul
22ba37aae2 mesa: make glGet*(GL_NONE) generate GL_INVALID_ENUM
In find_value() check if we've hit the 0th/invalid entry before checking
if the pname matches.

Fixes http://bugs.freedesktop.org/show_bug.cgi?id=31987

NOTE: This is a candidate for the 7.9 branch.

(cherry picked from commit 503983b09e)
2010-12-14 16:27:35 -07:00
Brian Paul
3b200dd3c5 mesa: return GL_FRAMEBUFFER_DEFAULT as FBO attachment type
If querying the default/window-system FBO's attachment type, return
GL_FRAMEBUFFER_DEFAULT (per the GL_ARB_framebuffer_object spec).

See http://bugs.freedesktop.org/show_bug.cgi?id=31947

NOTE: This is a candidate for the 7.9 branch.

(cherry picked from commit 6d13ec7dc0)
2010-12-14 16:26:30 -07:00
Brian Paul
5f05ade503 mesa: fix GL_FRAMEBUFFER_ATTACHMENT_OBJECT_NAME query
Return 0 instead of generating an error.

See http://bugs.freedesktop.org/show_bug.cgi?id=30993

Note that piglit fbo-getframebufferattachmentparameter-01 still does
not pass.  But Mesa behaves the same as the NVIDIA driver in this case.
Perhaps the test is incorrect.

NOTE: This is a candidate for the 7.9 branch.

(cherry picked from commit 20cf1851d8)
2010-12-14 16:26:14 -07:00
Ian Romanick
9ef05d27a6 glsl: Commit lexer files changed by previous cherry picking 2010-12-14 13:10:14 -08:00
Kenneth Graunke
1d63a77177 glsl: Fix constant component count in vector constructor emitting.
Fixes freedesktop.org bug #31101 as well as piglit test cases
assignment-type-mismatch.vert and constructor-28.vert.
(cherry picked from commit ba2382f50d)
2010-12-14 11:55:55 -08:00
Chad Versace
554dbaa8a0 glsl: Fix ast-to-hir for ARB_fragment_coord_conventions
Function ast_declarator_list::hir(), when processing keywords added by
extension ARB_fragment_coord_conventions, made the mistake of checking only if
the extension was __supported by the driver__. The correct behavior is to check
if the extensi0n is __enabled in the parse state__.

NOTE: this is a candidate for the 7.9 branch.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit 6e00627384)
2010-12-14 11:55:55 -08:00
Kenneth Graunke
1c089271fd 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.
(cherry picked from commit 2cacaf6e7b)
2010-12-14 11:55:54 -08:00
Kenneth Graunke
2c57c020a3 glcpp: Return NEWLINE token for newlines inside multi-line comments.
This is necessary for the main compiler to get correct line numbers.
(cherry picked from commit bd55ba568b)
2010-12-14 11:55:54 -08:00
Ian Romanick
2f398485f2 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.

(cherry picked from commit cc90e62d70)
2010-12-14 11:55:54 -08:00
Ian Romanick
e5ad6c3366 glsl: Slightly change the semantic of _LinkedShaders
Previously _LinkedShaders was a compact array of the linked shaders
for each shader stage.  Now it is arranged such that each slot,
indexed by the MESA_SHADER_* defines, refers to a specific shader
stage.  As a result, some slots will be NULL.  This makes things a
little more complex in the linker, but it simplifies things in other
places.

As a side effect _NumLinkedShaders is removed.

NOTE: This may be a candidate for the 7.9 branch.  If there are other
patches that get backported to 7.9 that use _LinkedShader, this patch
should be cherry picked also.

(cherry picked from commit 3322fbaf3b)
2010-12-14 11:55:54 -08:00
Chad Versace
1c15fb4414 glsl: Fix lexer rule for ^=
The caret is a special character, and needs to be quoted or escaped.
(cherry picked from commit cba9062d58)
2010-12-14 11:55:54 -08:00
Ian Romanick
a138a59966 mesa: Refactor validation of shader targets
Actually validate that the implementation supports the particular
shader target as well.  Previously if a driver only supported vertex
shaders, for example, glCreateShaderObjectARB would gladly create a
fragment shader.

NOTE: this is a candidate for the 7.9 branch.
(cherry picked from commit 5cb24c4a75)
2010-12-14 11:55:54 -08:00
Ian Romanick
6fc544dd50 linker: Reject shaders that have unresolved function calls
This really amounts to just using the return value from
link_function_calls.  All the work was being done, but the result was
being ignored.

Fixes piglit test link-unresolved-funciton.

NOTE: this is a candidate for the 7.9 branch.
(cherry picked from commit 4a45595cf3)
2010-12-14 11:55:54 -08:00
Eric Anholt
42e9750975 intel: Allow CopyTexSubImage to InternalFormat 3/4 textures, like RGB/RGBA.
The format selection of the CopyTexSubImage is pretty bogus still, but
this at least avoids software fallbacks in nexuiz, bringing
performance from 7.5fps to 12.8fps on my machine.
(cherry picked from commit 3316a54205)
2010-12-14 11:55:54 -08:00
Ian Romanick
72a3bddb24 mesa: Validate assembly shaders when GLSL shaders are used
If an GLSL shader is used that does not provide all stages and
assembly shaders are provided for the missing stages, validate the
assembly shaders.

Fixes bugzilla #30787 and piglit tests glsl-invalid-asm0[12].

NOTE: this is a candidate for the 7.9 branch.
(cherry picked from commit b2b9b22c10)
2010-12-14 11:55:54 -08:00
Tom Stellard
e8bc8d6a5c r300/compiler: Don't allow presubtract sources to be remapped twice
https://bugs.freedesktop.org/show_bug.cgi?id=31193

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 4265c2f819)
2010-12-10 11:07:24 -08:00
Alex Deucher
dc4956922d gallium/egl: fix r300 vs r600 loading
Should fix:
https://bugs.freedesktop.org/show_bug.cgi?id=31841
2010-12-07 10:59:20 +10:00
Marek Olšák
b486197013 st/mesa: initialize key in st_vp_varient
This fixes endless vertex shader recompilations in find_translated_vp
if the shader contains an edge flag output.

NOTE: This is a candidate for the 7.9 branch.

Signed-off-by Brian Paul <brianp@vmware.com>
(cherry picked from commit c0c929cdac)
2010-12-06 22:44:04 +01:00
Marek Olšák
2cb8522523 mesa, st/mesa: fix gl_FragCoord with FBOs in Gallium
gl_FragCoord.y needs to be flipped upside down if a FBO is bound.

This fixes:
- piglit/fbo-fragcoord
- https://bugs.freedesktop.org/show_bug.cgi?id=29420

Here I add a new program state STATE_FB_WPOS_Y_TRANSFORM, which is set based
on whether a FBO is bound. The state contains a pair of transformations.
It can be either (XY=identity, ZW=transformY) if a FBO is bound,
or (XY=transformY, ZW=identity) otherwise, where identity = (1, 0),
transformY = (-1, height-1).

A classic driver (or st/mesa) may, based on some other state, choose whether
to use XY or ZW, thus negate the conditional "if (is a FBO bound) ...".
The reason for this is that a Gallium driver is allowed to only support WPOS
relative to either the lower left or the upper left corner, so we must flip
the Y axis accordingly again. (the "invert" parameter in emit_wpos_inversion)

NOTE: This is a candidate for the 7.9 branch.

Signed-off-by: Marek Olšák <maraeo@gmail.com>
Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit d531f9c2f5)
2010-12-06 22:43:15 +01:00
Brian Paul
564a6a15fe llvmpipe: fix broken stencil writemask
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=32070
(cherry picked from commit dbf996f308)

Conflicts:

	src/gallium/drivers/llvmpipe/lp_bld_depth.c
2010-12-03 09:47:23 -07:00
Marek Olšák
7d280a3bbe r300g: disable ARB_texture_swizzle if S3TC is enabled on r3xx-only
r3xx cannot swizzle compressed textures. r4xx+ is unaffected.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit fcf6b353bf)
2010-12-01 23:56:56 +01:00
Marek Olšák
988a35c7f7 r300g: fix texture swizzling with compressed textures on r400-r500
This fixes all S3TC piglit/texwrap tests.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 6478a4de14)
2010-12-01 23:56:56 +01:00
Marek Olšák
4e7548aafc r300g: fix texture border color once again
I made the texwrap test be more thorough and realized that this driver code
had not been quite right. This commit fixes the border color for depth
textures, compressed textures, and 16-bits-per-channel textures
with up to 2 channels (R16, RG16).

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 1f1375d4d8)
2010-12-01 23:56:55 +01:00
Marek Olšák
277655d7fa r300/compiler: fix rc_rewrite_depth_out for it to work with any instruction
It looks like the function was originally written for ARB_fragment_program.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit ae3e58d973)
2010-12-01 23:56:55 +01:00
Marek Olšák
d42db5b738 r300g: fix rendering with no vertex elements
Fixes glsl-vs-point-size, although I meant to fix glsl-novertexdata.
Since swrast fails glsl-novertexdata too, I guess it's a core issue.
(cherry picked from commit ffb732d8bd)
2010-12-01 23:56:51 +01:00
Marek Olšák
62f2b7e759 r300g: fix texture border color for all texture formats
This fixes 8 texwrap format tests.
The code should handle arbitrary formats now and is cleaner.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 7e2256688a)
2010-12-01 23:38:52 +01:00
Marek Olšák
44ba61c518 r300g: add a default channel ordering of texture border for unhandled formats
It should fix the texture border for compressed textures.
Broken since 8449a4772a.
(cherry picked from commit 676c3f08bd)

Conflicts:

	src/gallium/drivers/r300/r300_state_derived.c
2010-12-01 23:38:36 +01:00
Marek Olšák
1c746ba9d4 r300g: fix texture border for 16-bits-per-channel formats
This is kinda hacky, but it's hard to come up with a generic solution for
all formats when only a few are used in practice (I mostly get B8G8R8*8).
(cherry picked from commit 8449a4772a)
2010-12-01 23:34:59 +01:00
Marek Olšák
c9661d695b r300g: fix microtiling for 16-bits-per-channel formats
(cherry picked from commit d0408cf55d)
2010-12-01 23:33:33 +01:00
Brian Paul
aa7bf2d88e mesa: check for posix_memalign() errors
Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit b1097607db)
2010-11-29 09:21:26 -07:00
Brian Paul
9d644e7b95 st/mesa: fix mapping of zero-sized buffer objects
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=31934

(cherry picked from commit 97ae4dad1c)
2010-11-26 13:53:33 -07:00
Brian Paul
54047fcd8d meta: Mask Stencil.Clear against stencilMax in _mesa_meta_Clear
This fixes incorrect behaviour when the stencil clear value exceeds
the size of the stencil buffer, for example, when set with:

glClearStencil (~1); /* Set a bit pattern of 111...11111110 */
glClear (GL_STENCIL_BUFFER_BIT);

The clear value needs to be masked by the value 2^m - 1, where m is the
number of bits in the stencil buffer. Previously, we passed the value
masked with 0x7fffffff to _mesa_StencilFuncSeparate which then clamps,
NOT masks the value to the range 0 to 2^m - 1.

The result would be clearing the stencil buffer to 0xff, rather than 0xfe.

Signed-off-by: Peter Clifton <pcjc2@cam.ac.uk>
Signed-off-by: Brian Paul <brianp@vmware.com>

(cherry picked from commit ee88727df8)
2010-11-24 12:16:44 -07:00
Thomas Hellstrom
bcdd93e3a7 gallium/targets: Trivial crosscompiling fix
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-11-23 11:43:52 +01:00
Thomas Hellstrom
75065b0b55 svga/drm: Optionally resolve calls to powf during link-time
When linked with certain builds of libstdc++, it appears like powf is resolved
by a symbol in that library. Other builds of libstdc++ doesn't contain that
symbol resulting in a linker / loader error. Optionally
resolve that symbol and replace it with calls to logf and expf.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-11-23 11:43:45 +01:00
Thomas Hellstrom
8ae23176d6 st/egl: Fix build for include files in nonstandard places
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-11-23 11:43:34 +01:00
Thomas Hellstrom
7d0c0d32b5 mesa: Add talloc includes for gles
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-11-23 11:43:26 +01:00
Thomas Hellstrom
e5c78c2003 egl: Add an include for size_t
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-11-23 11:43:17 +01:00
Tom Stellard
930c168df6 r300/compiler: Use correct swizzles for all presubtract sources 2010-11-22 12:00:10 -08:00
Tom Stellard
5551faf649 r300/compiler: Ignore alpha dest register when replicating the result
When the result of the alpha instruction is being replicated to the RGB
destination register, we do not need to use alpha's destination register.
This fixes an invalid "Too many hardware temporaries used" error in
the case where a transcendent operation writes to a temporary register
greater than max_temp_regs.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 23f577dbd4)
2010-11-22 11:21:53 -08:00
Tom Stellard
80ff25b628 r300/compiler: Use zero as the register index for unused sources
This fixes an invalid "Too many hardware temporaries used" error in the
case where a source reads from a temporary register with an index greater
than max_temp_regs and then the source is marked as unused before the
register allocation pass.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit d668659003)
2010-11-22 11:21:53 -08:00
Tom Stellard
7c8715ef0a r300/compiler: Fix instruction scheduling within IF blocks
Reads of registers that where not written to within the same block were
not being tracked.  So in a situations like this:
0: IF
1: ADD t0, t1, t2
2: MOV t2, t1

Instruction 2 didn't know that instruction 1 read from t2, so
in some cases instruction 2 was being scheduled before instruction 1.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 3e5f9789d6)
2010-11-22 11:21:48 -08:00
Tom Stellard
06fa5d81da r300/compiler: Fix register allocator's handling of loops
NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit e2301b45c2)
2010-11-22 11:15:33 -08:00
Tom Stellard
b86bf31b05 r300/compiler: Make sure presubtract sources use supported swizzles
NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 412803b5cd)
2010-11-22 11:04:07 -08:00
Brian Paul
68afe8f685 swrast: init alpha value to 1.0 in opt_sample_rgb_2d()
(cherry picked from commit 6a0255122a)
2010-11-22 09:05:31 -07:00
Chia-I Wu
0d060d9933 st/vega: Do not wait NULL fences.
(cherry picked from commit cc5c908d7d)
2010-11-20 15:59:58 +08:00
Chia-I Wu
aa5574e78f egl_dri2: Fix __DRI_DRI2 version 1 support.
Correctly set __DRI_API_OPENGL flag.
(cherry picked from commit aa139a14ba)
2010-11-20 15:59:58 +08:00
Chia-I Wu
5105e3d415 st/mesa: Unreference the sampler view in st_bind_surface.
Without this, update_textures may not pick up the new pipe_resource.

It is actually update_textures that should check
stObj->sampler_view->texture != stObj->pt, but let's follow st_TexImage
and others for now.
(cherry picked from commit cfc81d93f7)
2010-11-20 15:59:58 +08:00
Chia-I Wu
feacfc4a93 egl: Fix a false negative check in _eglCheckMakeCurrent.
This call sequence

  eglMakeCurrent(dpy, surf, surf, ctx1);
  eglMakeCurrent(dpy, surf, surf, ctx2);

should be valid if ctx1 and ctx2 have the same client API and are not
current in another thread.
(cherry picked from commit 0d43cbed2f)
2010-11-20 15:59:57 +08:00
Julien Cristau
b6ae3d7b23 Makefile: don't include the same files twice in the tarball
src/mesa/drivers/dri/*/*/*.[chS] is a superset of
src/mesa/drivers/dri/*/server/*.[ch] and
src/mesa/drivers/dri/common/xmlpool/*.[ch].
include/GL/internal/glcore.h is already in MAIN_FILES, no need for it in
DRI_FILES too.  src/glx/Makefile was listed twice.

Signed-off-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Brian Paul <brianp@vmware.com>
2010-11-15 08:53:12 -07:00
Brian Paul
3db8f85446 mesa: fix glDeleteBuffers() regression
This fixes a regression (failed assertion) from commit
c552f273f5 which was hit if glDeleteBuffers()
was called on a buffer that was never bound.

NOTE: this is a candidate for the 7.9 branch.
(cherry picked from commit 78587ea012)
2010-11-15 08:53:12 -07:00
Brian Paul
3e4ed38791 mesa: make glIsBuffer() return false for never bound buffers
Use a dummy buffer object as we do for frame/renderbuffer objects.
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=31514

Note: this is a candidate for the 7.9 branch.
(cherry picked from commit c552f273f5)
2010-11-15 08:53:12 -07:00
Brian Paul
acee46d9a1 mesa: handle more pixel types in mipmap generation code
NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit b3b6476695)
2010-11-15 08:53:11 -07:00
Brian Paul
97805cfedd mesa: add missing formats in _mesa_format_to_type_and_comps()
NOTE: this is a candidate for the 7.9 branch
(cherry picked from commit 79c65410c1)

Conflicts:

	src/mesa/main/formats.c
2010-11-15 08:53:04 -07:00
Daniel Lichtenberger
9b1ac4d127 radeon: fix potential segfault in renderbuffer update
Fixes:
https://bugs.freedesktop.org/show_bug.cgi?id=31617

Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
2010-11-15 01:36:02 -05:00
Roland Scheidegger
78ccca5a69 r200: fix r200 large points
DD_POINT_SIZE got never set for some time now (as it was set only in ifdefed
out code), which caused the r200 driver to use the point primitive mistakenly
in some cases which can only do size 1 instead of point sprite. Since the
logic to use point instead of point sprite prim is flaky at best anyway (can't
work correctly for per-vertex point size), just drop this and always emit point
sprites (except for AA points) - reasons why the driver tried to use points for
size 1.0 are unknown though it is possible they are faster or more conformant.
Note that we can't emit point sprites without point sprite cntl as that might
result in undefined point sizes, hence need drm version check (which was
unnecessary before as it should always have selected points). An
alternative would be to rely on the RE point size clamp controls which could
clamp the size to 1.0 min/max even if the SE point size is undefined, but currently
always use 0 for min clamp. (As a side note, this also means the driver does
not honor the gl spec which mandates points, but not point sprites, with zero size
to be rendered as size 1.)
This should fix recent reports of https://bugs.freedesktop.org/show_bug.cgi?id=702.
This is a candidate for the mesa 7.9 branch.
2010-11-11 01:14:10 +01:00
Mario Kleiner
2fe82c281d mesa/r300classic: Fix dri2Invalidate/radeon_prepare_render for page flipping.
A call to radeon_prepare_render() at the beginning of draw
operations was placed too deep in the call chain,
inside r300RunRenderPrimitive(), instead of
r300DrawPrims() where it belongs. This leads to
emission of stale target color renderbuffer into the cs if
bufferswaps via page-flipping are used, and thereby causes
massive rendering corruption due to unsynchronized
rendering into the active frontbuffer.

This patch fixes such problems for use with the
upcoming radeon page-flipping patches.

Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
2010-11-08 13:54:10 -05:00
Alex Deucher
773e8fadc8 r600c: properly align mipmaps to group size
fixes:
https://bugs.freedesktop.org/show_bug.cgi?id=31400
2010-11-08 12:11:14 -05:00
Alex Deucher
7f2d1280e7 r600c: add missing radeon_prepare_render() call on evergreen 2010-10-27 14:32:02 -04:00
Alex Deucher
075640e37e r100: revalidate after radeon_update_renderbuffers
This is a port of 603741a86d
to r100.

Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
2010-10-27 13:58:54 -04:00
Thomas Hellstrom
fb7a095785 mesa: Make sure we have the talloc cflags when using the talloc headers
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-10-19 14:21:13 +02:00
Thomas Hellstrom
f245bb4c9c st/xorg: Fix compilation for Xservers >= 1.10
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-10-19 12:11:44 +02:00
Thomas Hellstrom
cc71f76be2 xorg/vmwgfx: Don't use deprecated x*alloc / xfree functions
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-10-19 12:11:31 +02:00
Thomas Hellstrom
47887b1efe st/xorg: Don't use deprecated x*alloc / xfree functions
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-10-19 12:11:21 +02:00
Thomas Hellstrom
764f9974e5 st/xorg: Fix compilation errors for Xservers compiled without Composite
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-10-19 12:11:13 +02:00
Thomas Hellstrom
aa2b751abb st/xorg, xorg/vmwgfx: Be a bit more frendly towards cross-compiling environments
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-10-19 12:11:02 +02:00
Brian Paul
6f5c6e5cc5 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>
(cherry picked from commit 746b602fbd)
2010-10-18 08:11:45 -06:00
Brian Paul
e5b2a8d350 egl/i965: include inline_wrapper_sw_helper.h
Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit e19187e1be)
2010-10-18 08:02:42 -06:00
Chia-I Wu
da580c734f st/egl: Do not finish a fence that is NULL.
i915g would dereference the NULL pointer.
(cherry picked from commit d6de1f44a0)
2010-10-15 11:20:39 +08:00
Chia-I Wu
db7de707a3 st/vega: Fix version check in context creation.
This fixes a regression since 4531356817.
(cherry picked from commit b2c0ef8b51)
2010-10-15 11:20:39 +08:00
Chia-I Wu
bce3917192 targets/egl: Fix linking with libdrm.
(cherry picked from commit da495ee870)
2010-10-15 11:20:39 +08:00
Thomas Hellstrom
2eac388bfe st/xorg: Fix typo
Pointed out by Jakob Bornecrantz.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-10-12 18:26:56 +02:00
Thomas Hellstrom
6730630804 st/xorg: Don't try to remove invalid fbs
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-10-12 14:20:27 +02:00
Thomas Hellstrom
d5618c55ec xorg/vmwgfx: Don't hide HW cursors when updating them
Gets rid of annoying cursor flicker

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-10-12 14:12:36 +02:00
Thomas Hellstrom
ee6417f97f st/xorg: Add a customizer option to get rid of annoying cursor update flicker
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-10-12 14:09:09 +02:00
Thomas Hellstrom
14bae5cdf2 xorg/vmwgfx: Make vmwarectrl work also on 64-bit servers
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-10-12 11:08:39 +02:00
Thomas Hellstrom
799e6d4d15 st/xorg: Don't try to use option values before processing options
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
2010-10-12 10:28:57 +02:00
Daniel Vetter
6936fbc039 r200: revalidate after radeon_update_renderbuffers
By calling radeon_draw_buffers (which sets the necessary flags
in radeon->NewGLState) and revalidating if NewGLState is non-zero
in r200TclPrimitive. This fixes an assert in libdrm (the color-/
depthbuffer was changed but not yet validated) and and stops the
kernel cs checker from complaining about them (when they're too
small).

Thanks to Mario Kleiner for the hint to call radeon_draw_buffer
(instead of my half-broken hack).

v2: Also fix the swtcl r200 path.

Cc: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2010-10-11 15:17:44 -04:00
Brian Paul
0a060c8fb0 st/mesa: replace assertion w/ conditional in framebuffer invalidation
https://bugs.freedesktop.org/show_bug.cgi?id=30632

NOTE: this is a candidate for the 7.9 branch.
(cherry picked from commit 3d6eec0a87)
2010-10-06 08:41:44 -06:00
Brian Paul
34c518be52 swrast: fix choose_depth_texture_level() to respect mipmap filtering state
NOTE: this is a candidate for the 7.9 branch.
(cherry picked from commit fb5e6f88fc)
2010-10-05 14:34:58 -06:00
Ian Romanick
2b4a05e74a docs: added news item for 7.9 release
Also fix link to release notes in 7.9-rc1 news item.
2010-10-05 10:08:28 -07:00
Ian Romanick
7afce798ee docs: Update mailing lines from sf.net to freedesktop.org
(cherry picked from commit c19bc5de96)
2010-10-05 10:08:21 -07:00
Ian Romanick
ab502d0d6b docs: download.html does not need to be updated for each release
(cherry picked from commit 41e371e351)
2010-10-05 10:08:14 -07:00
Ian Romanick
548a50caae docs: Import 7.8.x release notes from 7.8 branch. 2010-10-05 10:08:06 -07:00
Ian Romanick
11e290daa4 Add 7.9 md5sums 2010-10-04 18:59:34 -07:00
Ian Romanick
889cb2539b docs: Add 7.9 release date 2010-10-04 18:56:21 -07:00
Ian Romanick
242facde45 mesa: set version string to 7.9 (final) 2010-10-04 18:55:57 -07:00
Alex Deucher
73be0dfe03 Add evergreen support to release notes 2010-10-04 21:46:57 -04:00
Ian Romanick
0b300a4e3d docs: Update 7.9 release notes with one more bug fix 2010-10-04 17:43:52 -07:00
Eric Anholt
586797d7a8 i965: Fix glean/texSwizzle regression in previous commit.
Easy enough patch, who needs a full test run.  Oh, that's right.  Me.
(cherry picked from commit 14bf92ba19)
2010-10-04 17:41:56 -07:00
Eric Anholt
9b646b1fd3 i965: Set up swizzling of shadow compare results for GL_DEPTH_TEXTURE_MODE.
The brw_wm_surface_state.c handling of GL_DEPTH_TEXTURE_MODE doesn't
apply to shadow compares, which always return an intensity value.  The
texture swizzles can do the job for us.

Fixes:
glsl1-shadow2D(): 1
glsl1-shadow2D(): 3
(cherry picked from commit a7fa00dfc5)
2010-10-04 17:41:45 -07:00
Kenneth Graunke
dc8e22e4f0 i965: Fix incorrect batchbuffer size in gen6 clip state command.
FORCE_ZERO_RTAINDEX should be in the fourth (and final) dword.
(cherry picked from commit f317713432)
2010-10-04 17:41:33 -07:00
Ian Romanick
21078eef10 docs: Add list of bugs fixed in 7.9
(cherry picked from commit 7f440b3f9d76fa3a02877074d9701ab499d5de34)
2010-10-04 17:38:54 -07:00
Marek Olšák
968a9ec76e update release notes for Gallium
I am trying to be exhaustive, but still I might have missed tons of other
changes to Gallium.
2010-10-05 02:24:50 +02:00
Vinson Lee
2cdfda8851 tgsi: Remove duplicate case value.
Cherry-picked from master commit 2491258436
2010-10-04 17:31:33 +02:00
Jakob Bornecrantz
09d89dc95b tgsi: Actually care what check_soa_dependencies says
Thanks to José for the more complete list of supported opcodes.

Cherry-picked from master commit d21301675c
Also includes fix commit 4bb42a4f7e
2010-10-04 17:30:34 +02:00
José Fonseca
15ad34f012 tgsi: Don't ignore indirect registers in tgsi_check_soa_dependencies
Cherry-picked from master commit c66f0c4629
2010-10-04 17:30:00 +02:00
Alex Deucher
094fe2aa23 r600c: fix segfault in evergreen stencil code
Fixes:
https://bugs.freedesktop.org/show_bug.cgi?id=30551
2010-10-01 20:15:17 -04:00
Tom Fogal
ba889a2b3e Implement x86_64 atomics for compilers w/o intrinsics.
Really old gcc's (3.3, at least) don't have support for the
intrinsics we need.  This implements a fallback for that case.
2010-10-01 17:56:02 -06:00
richard
71cc467f4c evergreen : fix z format setting, enable stencil. 2010-10-01 16:18:16 -04:00
Alex Deucher
d6e2b707ba r600c: pull over 6xx/7xx vertex fixes for evergreen 2010-10-01 00:52:36 -04:00
Alex Deucher
0ab9d8d94b r600c: add reloc for CB_COLOR0_ATTRIB
We'll need a reloc for tiling eventually,
so add it now.
2010-09-30 20:57:38 -04:00
Adam Jackson
759e9fb737 i965: Update renderer strings for sandybridge
Signed-off-by: Adam Jackson <ajax@redhat.com>
(cherry picked from commit 0c86e1f294)
2010-09-30 13:32:42 -07:00
Zhenyu Wang
4af00f1a5e i965: always set tiling for fbo depth buffer on sandybridge
Sandybridge requires depth buffer must be tiling.

Fix 'fbo_firecube' demo.
(cherry picked from commit 72b368ae69)
2010-09-30 13:23:38 -07:00
Zhenyu Wang
1c8795075f i965: fallback lineloop on sandybridge for now
Until we fixed GS hang issue.
(cherry picked from commit 73dab75b41)
2010-09-30 13:23:37 -07:00
Kenneth Graunke
5d43c78d03 glsl: "Copyright", not "Constantright"
Clearly this started out as ir_copy_propagation.cpp, but the search and
replace was a bit overzealous.
(cherry picked from commit 565ff67688)
2010-09-30 13:23:37 -07:00
Marek Olšák
9f076a0848 r300g: fix conditional rendering in non-wait path
NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 83278d384e)
2010-09-30 22:17:15 +02:00
Chia-I Wu
9bab8ca4f8 st/egl: Drop context argument from egl_g3d_get_egl_image.
Fix a regression since 17eace581d.
(cherry picked from commit e2b51b7c5b)
2010-09-30 23:48:34 +08:00
Chia-I Wu
f6eb5d991c mapi: Fix compiler warnings.
Do not use "void *" in arithmetics.
(cherry picked from commit ebeb4a7e8a)
2010-09-30 17:12:30 +08:00
Tom Stellard
f282d72678 r300/compiler: Don't merge instructions that write output regs and ALU result
https://bugs.freedesktop.org/show_bug.cgi?id=30415

(cherry picked from commit b27a809266)
2010-09-29 11:05:01 -07:00
Tom Stellard
e1918fd5d5 r300/compiler: Don't use rc_error() unless the error is unrecoverable
https://bugs.freedesktop.org/show_bug.cgi?id=30416

(cherry picked from commit 1b76dde0cd)
2010-09-29 11:04:12 -07:00
Tom Stellard
fcf5819945 r300/compiler: Fix segfault in error path
https://bugs.freedesktop.org/show_bug.cgi?id=30415

(cherry picked from commit d40ff5510c)
2010-09-29 11:03:02 -07:00
Brian Paul
c584318a21 Makefile: ensure Gallium's Makefile.xorg and SConscript.dri are in the tarball
Signed-off-by: Brian Paul <brianp@vmware.com>
(cherry picked from commit 698893889a)
2010-09-29 09:53:02 -06:00
Kristian Høgsberg
b24238c49a glx: Only remove drawables from the hash when we actually delete them
https://bugs.freedesktop.org/show_bug.cgi?id=30457
2010-09-29 09:10:05 -04:00
Ian Romanick
5d3a4317e8 mesa: set version string to 7.9-rc2 2010-09-28 19:22:55 -07:00
Eric Anholt
836b44b2c8 glsl: Also update implicit sizes of varyings at link time.
Otherwise, we'll often end up with gl_TexCoord being 0 length, for
example.  With ir_to_mesa, things ended up working out anyway, as long
as multiple implicitly-sized arrays weren't involved.
(cherry picked from commit 586b4b500f)
2010-09-28 18:41:05 -07:00
Eric Anholt
83854a9c84 i965: Fix up part of my Sandybridge attributes support patch.
I confused the array sizing for number of files for the number of regs
in a file.
(cherry picked from commit ef8e002c75)
2010-09-28 18:40:32 -07:00
Eric Anholt
b4824096b0 i965: Fix all non-snb regression in the snb attribute interpolation commit.
This apparently had never been tested elsewhere before being merged to
master.
(cherry picked from commit f1dba03056)
2010-09-28 18:40:24 -07:00
Zhenyu Wang
71bdc82c19 i965: fallback bitmap operation on sandybridge
Need to bring back correct fb write with header to set pixel
write mask. Fallback for now.

Note: Conflicts in intel_pixel_bitmap.c due to the removal of
intel_texture_bitmap from master.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2010-09-28 18:37:20 -07:00
Zhenyu Wang
aaf7f78dfa i965: fix occlusion query on sandybridge
Fix pipe control command for depth stall and PS_DEPTH_COUNT write.
(cherry picked from commit 3074b61f64)
2010-09-28 18:36:05 -07:00
Zhenyu Wang
5c7a7d0b09 i965: fix point sprite on sandybridge
Need to set point sprite function in fixed SF state now on sandybridge.
(cherry picked from commit ec99833e92)
2010-09-28 18:36:01 -07:00
Zhenyu Wang
f736d85a5a i965: fix scissor state on sandybridge
Fix incorrect scissor rect struct and missed scissor state pointer
setting for sandybridge.
(cherry picked from commit 4b6b0bf24a)
2010-09-28 18:35:57 -07:00
Zhenyu Wang
40dc275431 i965: enable polygon offset on sandybridge
Depth offset function is moved to SF stage on sandybridge.
(cherry picked from commit 3f3059fcc0)
2010-09-28 18:35:41 -07:00
Zhenyu Wang
3dd52d51ce i965: fix pixel w interpolation on sandybridge
(cherry picked from commit 15a8e7ec90)
2010-09-28 18:35:35 -07:00
Zhenyu Wang
e00bb8732c i965: don't do calculation for delta_xy on sandybridge
Sandybridge doesn't have Xstart/Ystart in payload header.
(cherry picked from commit 85fa900b93)
2010-09-28 18:35:30 -07:00
Zhenyu Wang
11215cd535 i965: only allow SIMD8 kernel on sandybridge now
Until we fixed SIMD16 kernel, force to SIMD8 on sandybridge now.
(cherry picked from commit c58bf2cee5)
2010-09-28 18:35:26 -07:00
Zhenyu Wang
6831cf67d1 i965: sandybridge pipe control workaround before write cache flush
Must issue a pipe control with any non-zero post sync op before
write cache flush = 1 pipe control.
(cherry picked from commit 18c3b754f9)
2010-09-28 18:35:21 -07:00
Zhenyu Wang
ece465219f i965: Add all device ids for sandybridge
(cherry picked from commit c8033f1b1e)
2010-09-28 18:35:16 -07:00
Zhenyu Wang
0dc83cc437 i965: fix const register count for sandybridge
Sandybridge's PS constant buffer payload size is decided from
push const buffer command, incorrect size would cause wrong data
in payload for position and vertex attributes. This fixes coefficients
for tex2d/tex3d.
(cherry picked from commit 81aae67e58)
2010-09-28 18:35:11 -07:00
Zhenyu Wang
922d7e1025 i965: Fix sampler on sandybridge
Sandybridge has not much change on texture sampler with Ironlake.
(cherry picked from commit 956f866030)
2010-09-28 18:35:04 -07:00
Zhenyu Wang
4afb1c8451 i965: fix jump count on sandybridge
Jump count is for 64bit long each, so one instruction requires 2
like on Ironlake.
(cherry picked from commit c5a3b25bb9)
2010-09-28 18:34:58 -07:00
Zhenyu Wang
ad52c2b773 i965: VS use SPF mode on sandybridge for now
Until conditional instructions were fixed, use SPF mode instead for now.
(cherry picked from commit 9c39a9fcb2)
2010-09-28 18:34:53 -07:00
Zhenyu Wang
364b6e0709 i965: add sandybridge viewport state bo into validation list
(cherry picked from commit 7401a98e29)
2010-09-28 18:34:47 -07:00
Zhenyu Wang
32c1dc76af i965: ignore quads for GS kernel on sandybridge
Sandybridge's VF would convert quads to polygon which not required
for GS then. Current GS state still would cause hang on lineloop.
(cherry picked from commit a0b1d7b2b8)
2010-09-28 18:34:43 -07:00
Zhenyu Wang
f9ad343cf7 i965: ff sync message change for sandybridge
(cherry picked from commit 67dafa4b56)
2010-09-28 18:34:39 -07:00
Zhenyu Wang
76469d710c i965: fix point size setting in header on sandybridge
(cherry picked from commit fa589770e8)
2010-09-28 18:34:33 -07:00
Zhenyu Wang
6eec2dbfc8 i965: force zero in clipper to ignore RTAIndex on sandybridge
(cherry picked from commit 03218a0093)
2010-09-28 18:34:27 -07:00
Zhenyu Wang
a4f126beaf i965: Fix color interpolation on sandybridge
Don't double store position in vertex attribute. This makes color
interpolation right by using barycentric coordinates.
(cherry picked from commit 41c31c2ebd)
2010-09-28 18:34:22 -07:00
Zhenyu Wang
0d5400519e i965: enable accumulator update in PS kernel too on sandybridge
Accumulator update flag must be set for implicit update on sandybridge.
(cherry picked from commit 8c31a4c3cf)
2010-09-28 18:34:16 -07:00
Zhenyu Wang
36722d7673 i965: new state dump for sandybridge
Dump new state object on sandybridge for cc viewport, clip viewport,
depth stencil, cc and blend state.
(cherry picked from commit b016a78b98)
2010-09-28 18:34:11 -07:00
Zhenyu Wang
3c26f3883a i965: disasm quarter and write enable instruction control on sandybridge
(cherry picked from commit bf60f35934)
2010-09-28 18:34:01 -07:00
Eric Anholt
1975d851e8 i965: Add support for POW in gen6 FS.
Fixes glsl-algebraic-pow-2 in brw_wm_glsl.c mode.
(cherry picked from commit fe2d4a5ea0)
2010-09-28 18:33:49 -07:00
Eric Anholt
f4f1e0d108 i965: Set up inputs to the fragment shader according to FP InputsRead.
Sending down data that doesn't get read doesn't make any sense, and
would make handling things like gl_FrontFacing and gl_PointCoord
harder.
(cherry picked from commit 2f914053bc)
2010-09-28 18:33:34 -07:00
Eric Anholt
cda661fd1c i965: Add support for attribute interpolation on Sandybridge.
Things are simpler these days thanks to barycentric interpolation
parameters being handed in in the payload.
(cherry picked from commit a66e9a4d86)
2010-09-28 18:33:10 -07:00
Eric Anholt
1ab0d45c9f glsl: Fix broadcast_index of lower_variable_index_to_cond_assign.
It's trying to get an int smeared across all channels, not trying to
get a 1:1 mapping of a subset of a vector's channels.  This usually
ended up not mattering with ir_to_mesa, since it just smears floats
into every chan of a vec4.

Fixes:
glsl1-temp array with swizzled variable indexing
(cherry picked from commit 668cdbe129)
2010-09-28 18:31:05 -07:00
Dave Airlie
269dd51162 mesa/mipmap: fix warning since 1acadebd62
1acadebd62 fixed the pointer but not the cast.
(cherry picked from commit 865cf77503)
2010-09-28 18:29:17 -07:00
Eric Anholt
f60c28e083 mesa: Fix type typo in glGenerateMipmap handling of GL_UNSIGNED_INT data.
Fixes ARB_depth_texture/fbo-generatemipmap-formats.
(cherry picked from commit 1acadebd62)
2010-09-28 18:29:10 -07:00
Eric Anholt
aebd78f607 glsl: Fix copy'n'wasted ir_noop_swizzle conditions.
It considered .xyyy a noop for vec4 instead of .xyzw, and similar for vec3.
(cherry picked from commit 3ffab36768)
2010-09-28 18:26:25 -07:00
Eric Anholt
0391eaf088 glsl: Rework assignments with write_masks to have LHS chan count match RHS.
It turns out that most people new to this IR are surprised when an
assignment to (say) 3 components on the LHS takes 4 components on the
RHS.  It also makes for quite strange IR output:

(assign (constant bool (1)) (x) (var_ref color) (swiz x (var_ref v) ))
(assign (constant bool (1)) (y) (var_ref color) (swiz yy (var_ref v) ))
(assign (constant bool (1)) (z) (var_ref color) (swiz zzz (var_ref v) ))

But even worse, even we get it wrong, as shown by this line of our
current step(float, vec4):

(assign (constant bool (1)) (w)
	(var_ref t)
	(expression float b2f (expression bool >=
		    (swiz w (var_ref x))(var_ref edge))))

where we try to assign a float to the writemasked-out x channel and
don't supply anything for the actual w channel we're writing.  Drivers
right now just get lucky since ir_to_mesa spams the float value across
all the source channels of a vec4.

Instead, the RHS will now have a number of components equal to the
number of components actually being written.  Hopefully this confuses
everyone less, and it also makes codegen for a scalar target simpler.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit b39e6f33b6)
2010-09-28 18:25:57 -07:00
Marek Olšák
7436872dce configure.ac: do not build xorg-r300g by default
NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 11eb422a16)
2010-09-28 19:52:22 +02:00
Marek Olšák
7a297bd926 configure.ac: look for libdrm_radeon before building gallium/r300,r600
NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit a1aec2e2be)
2010-09-28 19:52:16 +02:00
Marek Olšák
6c3a8aeb27 r300g: fix swizzling of texture border color
NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit e4fd65e9d7)
2010-09-28 19:51:38 +02:00
Marek Olšák
5c3be43cd2 r300/compiler: fix shadow sampling with swizzled coords
Taking the W component from coords directly ignores swizzling. Instead,
take the component which is mapped to W in the TEX instruction parameter.
The same for Z.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 7128e1625b)

Conflicts:

	src/mesa/drivers/dri/r300/compiler/radeon_program_tex.c
2010-09-28 19:51:19 +02:00
Marek Olšák
5906b9ac51 r300/compiler: do not use copy propagation if SaturateMode is used
NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit c2ea7ffb0a)
2010-09-28 19:48:31 +02:00
Marek Olšák
89ed679a9e r300/compiler: fix projective mapping of 2D NPOT textures
NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 6f747567ec)

Conflicts:

	src/mesa/drivers/dri/r300/compiler/radeon_program_tex.c
2010-09-28 19:48:00 +02:00
Kristian Høgsberg
7d154de048 glx: Invalidate buffers after binding a drawable
If the server doesn't send invalidate events, we may miss a
resize before the rendering starts.  Invalidate the buffers now
so the driver will recheck before rendering starts.

https://bugs.freedesktop.org/show_bug.cgi?id=29984
https://bugs.freedesktop.org/show_bug.cgi?id=30155
2010-09-28 08:32:45 -04:00
Ian Romanick
2a7ed4ba96 dri: Ensure that DRI driver cpp files are in tarballs
(cherry picked from commit 5717093f15fd7259441b2ee5583432fb9833c9a1)
2010-09-27 14:13:46 -07:00
Ian Romanick
8ad0396108 Add 7.9-rc1 md5sums 2010-09-27 10:20:25 -07:00
Ian Romanick
61653b488d mesa: set version string to 7.9-rc1 2010-09-27 10:17:11 -07:00
Marek Olšák
7e62335f35 r300g: fix macrotiling on R350
MACRO_SWITCH on R350 appears to use the RV350 mode by default. Who knew?

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 311ab3d468)

https://bugs.freedesktop.org/show_bug.cgi?id=30313
2010-09-26 23:00:16 +02:00
Joakim Sindholt
b8a60c84aa radeong: fix leaks
(cherry picked from commit 16baa465a2)
2010-09-26 19:59:11 +02:00
Joakim Sindholt
9f221e738d util/u_blitter: fix leak
(cherry picked from commit b51f6e7c23)
2010-09-26 19:53:47 +02:00
Tom Stellard
5c78e931c2 r300/compiler: Fix two mistakes in the presubtract optimization pass.
1. We can't turn an instruction into a presubtract operation if it
writes to one of the registers it reads from.
2. If we turn an instruction into a presubtract operation, we can't
remove that intruction unless all readers can use the presubtract
operation.

This fixes fdo bug 30337.
This is a candidate for the 7.9 branch.
(cherry picked from commit 522e994a22)
2010-09-25 15:53:20 -07:00
Marek Olšák
361084ac4b util: fix util_pack_color for B4G4R4A4
NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 8619495790)
2010-09-25 01:54:12 +02:00
Marek Olšák
9409b544cc Build r300g by default
NOTE: This will go to 7.9 as well.
(cherry picked from commit 85a45dcd5d)
2010-09-25 01:54:12 +02:00
Marek Olšák
7261c3973d r300g: fix the border color for every format other than PIPE_FORMAT_B8G8R8A8
TX_BORDER_COLOR should be formatted according to the texture format.
Also the interaction with ARB_texture_swizzle should be fixed too.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 9f35dcd24c)
2010-09-25 01:54:12 +02:00
Marek Olšák
bb4c9e0969 r300g: make accessing map_list and buffer_handles thread-safe
NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit a333485386)
2010-09-25 01:54:12 +02:00
Marek Olšák
aa05753059 r300g: fixup long-lived BO maps being incorrectly unmapped when flushing
Based on commit 3ddc714b20 by Dave Airlie.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 206d92912c)
2010-09-25 01:54:12 +02:00
Marek Olšák
09085ada5d util: make calling remove_from_list multiple times in a row safe
This commit fixes an infinite loop in foreach_s if remove_from_list is used
more than once on the same item with other list operations in between.

NOTE: This is a candidate for the 7.9 branch because the commit
"r300g: fixup long-lived BO maps being incorrectly unmapped when flushing"
depends on it.
(cherry picked from commit 68afbe89c7)
2010-09-25 01:54:11 +02:00
Alex Deucher
77cc2a4e17 r600c: fix mipmap stride on evergreen
taken from Dave's r600g fix
2010-09-24 13:27:21 -04:00
Brian Paul
96085dc41a st/mesa: use the wrapped renderbuffer in CopyPixels()
Fixes assertion failures when copying stencil pixels.

NOTE: this is a candidate for the 7.9 branch.
(cherry picked from commit f5c810c42f)
2010-09-24 10:18:12 -06:00
Brian Paul
a128b62a35 st/mesa: add missing MESA_FORMAT_S8 case in st_mesa_format_to_pipe_format()
NOTE: this is a candidate for the 7.9 branch.
(cherry picked from commit 10dcc989ab)
2010-09-24 10:17:40 -06:00
Brian Paul
b756100892 mesa: fix assertions to handle srgb formats
http://bugs.freedesktop.org/show_bug.cgi?id=30333

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 9f7c8053e0)
2010-09-24 10:16:52 -06:00
Tom Stellard
3ad0279344 r300g: Always try to build libr300compiler.a
Make libr300compiler.a a PHONY target so that this library will always be
built.  This fixes the problem of libr300compiler.a not being updated
when r300g is being built and r300c is not.

(cherry picked from commit 92762842a0)
2010-09-23 15:42:40 -07:00
Kristian Høgsberg
97cfa1240c intel: Fix GL_ARB_shading_language_120 commit
Fix commit e7087175f8.  Move the reference to
GL_VERSION_2_1_functions to intel_extensions.c where it's available,
don't try to enable a non-existing extension and advertise 1.20 for all
intel chipsets, not just GEN4 and up.
(cherry picked from commit b91dba49e0)
2010-09-23 08:47:01 -07:00
Brian Paul
6b0c280660 mesa: don't advertise bogus GL_ARB_shading_language_120 extension
Instead of using the invalid GL_ARB_shading_language_120 extension to
determine the GLSL version, use a new ctx->Const.GLSLVersion field.
Updated the intel and r600 drivers, but untested.

See fd.o bug 29910

NOTE: This is a candidate for the 7.9 branch (but let's wait and see if
there's any regressions).
(cherry picked from commit e7087175f8)
2010-09-23 08:47:01 -07:00
Eric Anholt
b437c9102f i965: Share the KIL_NV implementation between glsl and non-glsl.
(cherry picked from commit 23c507f135)
2010-09-23 08:47:01 -07:00
Kenneth Graunke
4ec818ea4a glsl: Fix broken handling of ir_binop_equal and ir_binop_nequal.
When ir_binop_all_equal and ir_binop_any_nequal were introduced, the
meaning of these two opcodes changed to return vectors rather than a
single scalar, but the constant expression handling code was incorrectly
written and only worked for scalars.  As a result, only the first
component of the returned vector would be properly initialized.
(cherry picked from commit 6ea16b6c51)
2010-09-23 08:47:01 -07:00
Kenneth Graunke
80a6cbfea3 glsl: Add comments to clarify the types of comparison binops.
(cherry picked from commit 14eea26828)
2010-09-23 08:47:01 -07:00
Brian Paul
ec2031563a glsl2: silence compiler warnings in printf() calls
Such as: "ir_validate.cpp:143: warning: format ‘%p’ expects type ‘void*’,
but argument 2 has type ‘ir_variable*’"
(cherry picked from commit 1739124159)
2010-09-23 08:47:01 -07:00
Ian Romanick
b7d4f08f4e glsl: Add doxygen comments
(cherry picked from commit e053d62aa5)
2010-09-23 08:47:01 -07:00
Kenneth Graunke
2fc2b89680 glsl/builtins: Switch comparison functions to just return an expression.
(cherry picked from commit dbd2480507)
2010-09-23 08:47:01 -07:00
Kenneth Graunke
b29289a61c glsl/builtins: Fix equal and notEqual builtins.
Commit 309cd4115b incorrectly converted
these to all_equal and any_nequal, which is the wrong operation.
(cherry picked from commit 52f9156e88)
2010-09-23 08:47:01 -07:00
Kenneth Graunke
728829ab63 glsl: Properly handle nested structure types.
Fixes piglit test CorrectFull.frag.
(cherry picked from commit ca92ae2699)
2010-09-23 08:47:01 -07:00
Tilman Sauerbeck
9753ee3f3e glsl2: Fixed cloning of ir_call error instructions.
Those have the callee field set to the null pointer, so
calling the public constructor will segfault.

Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
(cherry picked from commit 3894fddccc)
2010-09-23 08:47:01 -07:00
Vinson Lee
2db5398449 glsl: Fix 'control reaches end of non-void function' warning.
Fixes this GCC warning.

lower_variable_index_to_cond_assign.cpp:
In member function
'bool variable_index_to_cond_assign_visitor::needs_lowering(ir_dereference_array*) const':

lower_variable_index_to_cond_assign.cpp:261:
warning: control reaches end of non-void function
(cherry picked from commit a822ae3f1a)
2010-09-23 08:47:01 -07:00
Vinson Lee
ee8a96707a ir_to_mesa: Remove unused member array_indexed from struct statevar_element.
Fixes this GCC warning.
warning: missing initializer for member 'statevar_element::array_indexed'
(cherry picked from commit ff78d6dcc0)
2010-09-23 08:47:01 -07:00
Tilman Sauerbeck
22dcee8da2 glsl2: Empty functions can be inlined.
Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
(cherry picked from commit 19f8f32a96)
2010-09-23 08:47:00 -07:00
Ian Romanick
1b622bfbee glsl2: Add flags to enable variable index lowering
(cherry picked from commit a6ecd1c372)
2010-09-23 08:47:00 -07:00
Ian Romanick
2bb2efd7b2 glsl2: Refactor testing for whether a deref is of a matrix or array
(cherry picked from commit 6e4fe39da2)
2010-09-23 08:47:00 -07:00
Luca Barbieri
a19f0fb50b glsl: add pass to lower variable array indexing to conditional assignments
Currenly GLSL happily generates indirect addressing of any kind of
arrays.

Unfortunately DirectX 9 GPUs are not guaranteed to support any of them in
general.

This pass fixes that by lowering such constructs to a binary search on the
values, followed at the end by vectorized generation of equality masks, and
4 conditional assignments for each mask generation.

Note that this requires the ir_binop_equal change so that we can emit SEQ
to generate the boolean masks.

Unfortunately, ir_structure_splitting is too dumb to turn the resulting
constant array references to individual variables, so this will need to
be added too before this pass can actually be effective for temps.

Several patches in the glsl2-lower-variable-indexing were squashed
into this commit.  These patches fix bugs in Luca's original
implementation, and the individual patches can be seen in that branch.
This was done to aid bisecting in the future.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
(cherry picked from commit a47539c7a1)
2010-09-23 08:47:00 -07:00
Kenneth Graunke
8df8bb43c7 glsl: Don't print blank (function ...) headers for built-ins.
Fixes a regression caused when I added my GLSL ES support.
(cherry picked from commit 8fbe968a62)
2010-09-23 08:47:00 -07:00
Kenneth Graunke
d2b6502c93 glsl: Change from has_builtin_signature to has_user_signature.
The print visitor needs this, and the only existing user can work with
has_user_signature just as well.
(cherry picked from commit 81f0339398)
2010-09-23 08:47:00 -07:00
Brian Paul
e68ec2212f glsl2: fix signed/unsigned comparison warning
(cherry picked from commit 7545514fb6)
2010-09-23 08:47:00 -07:00
Vinson Lee
c0f82e619c glsl: Fix 'format not a string literal and no format arguments' warning.
Fix the following GCC warning.
loop_controls.cpp: In function 'int calculate_iterations(ir_rvalue*, ir_rvalue*, ir_rvalue*, ir_expression_operation)':
loop_controls.cpp:88: warning: format not a string literal and no format arguments
(cherry picked from commit f20f2cc330)
2010-09-23 08:47:00 -07:00
José Fonseca
7f95c59509 draw: Prevent clipped vertices overflow.
Some pathological triangles cause a theoritically impossible number of
clipped vertices.

The clipper will still assert, but at least release builds will not
crash, while this problem is further investigated.
2010-09-23 16:48:59 +01:00
Keith Whitwell
6f310f379b draw: don't apply flatshading to clipped tris with <3 verts
If a triangle was completely culled by clipping, we would still try to
fix up its provoking vertex.
2010-09-23 16:28:27 +01:00
Dave Airlie
f279480df2 egl: fix build since 17eace581d
looks like mesa st didn't get updated.
2010-09-23 07:41:05 -04:00
Kristian Høgsberg
d382c51a23 dri: Pass the __DRIscreen and the __DRIscreen private back to image lookup
We will typically have a current context when we need to lookup the image,
but the lookup implementation don't need it so drop it.
2010-09-23 07:40:59 -04:00
Kristian Høgsberg
cdb3f23d8a glx: Hold on to drawables if we're just switching to another context
https://bugs.freedesktop.org/show_bug.cgi?id=30234
2010-09-23 07:40:27 -04:00
Eric Anholt
8f4c99a9af i965: Also enable CC statistics when doing OQs.
This is required by the spec, so respect that.

(oh, and it was required to actually fix occlusion-query-discard,
though I didn't know at the time of the commit since o-q-discard was
hanging the GPU)

(cherry picked from commit 4a0bc4716d)
2010-09-22 13:45:50 -07:00
Eric Anholt
5285100502 ir_to_mesa: Only compare vector_elements present for any_nequal/all_equal
Fixes: glsl-mat-from-int-ctor-03
(cherry picked from commit eaa6bf59db)
2010-09-22 13:43:10 -07:00
Eric Anholt
56dc7ca916 i965: Track the windowizer's dispatch for kill pixel, promoted, and OQ
Looks like the problem was we weren't passing the depth to the render
target as expected, so the chip would wedge.  Fixes GPU hang in
occlusion-query-discard.

Bug #30097
(cherry picked from commit dd9a88f4dd)
2010-09-22 13:15:59 -07:00
Eric Anholt
902dfdaa86 glsl: Add definition of gl_TextureMatrix inverse/transpose builtins.
Fixes glsl2/builtin-texturematrix.
Bug #30196.
(cherry picked from commit b5bb215629)
2010-09-22 13:15:59 -07:00
Kristian Høgsberg
c98ce41de3 dri2: Make createImageFromName() take a __DRIscreen instead of __DRIcontext
We can't expect to have a context when this is called, and we don't need one
so just require a __DRIscreen instead.

Reported by Yu Dai <yu.dai@intel.com>
2010-09-22 15:27:04 -04:00
Brian Paul
3cdfc62523 llvmpipe: use draw module to draw sprites
For the 7.9 branch, this is simpler than merging the sprite-coord branch
and less likely to break other drivers.

Fixes fd.o bug 29849.
2010-09-22 08:22:36 -06:00
Brian Paul
2606105cc5 glsl2: fix typo in error msg
(cherry picked from commit 2b95525429)
2010-09-21 14:58:24 -06:00
Keith Whitwell
141dbbfc78 llvmpipe: fix flatshading in new line code
Calculate interpolants before rearranging the vertices.
2010-09-21 14:40:22 +01:00
Brian Paul
e678c3f4e1 docs: mark as obsolete, remove dead links
(cherry picked from commit 0bc3e1f4f4)
2010-09-20 10:56:53 -06:00
Brian Paul
1b90889eb4 docs: remove old broken link
(cherry picked from commit a8fde1ba4d)
2010-09-20 10:56:41 -06:00
Brian Paul
a9fceb6979 llvmpipe: fix query bug when no there's no scene
(cherry picked from commit 7888a2f822)
2010-09-20 10:56:10 -06:00
Brian Paul
febd353bcd gallivm: fix incorrect vector shuffle datatype
The permutation vector must always be a vector of int32 values.

(cherry picked from commit 3c9f4c7b75)
2010-09-20 10:55:35 -06:00
Marek Olšák
0f1faa9c7b st/mesa: fix assertion failure in GetTexImage for cubemaps
Can be reproduced with mesa/demos/src/tests/blitfb.

NOTE: This is a candidate for the 7.9 branch.
(cherry picked from commit 168554904b)
2010-09-20 18:21:24 +02:00
Jakob Bornecrantz
b247620854 i915g: Link with wrapper sw winsys with scons
cherry picked from commit 208f1f3810
2010-09-20 15:45:35 +02:00
Michal Krol
d36344f900 svga: Integer constant register file has a separate namespace.
Count int and float constants independently. Since there are only
few i# constants available and hundreds of c# constants, it would
be too easy to end up with an i# declaration out of its range.
2010-09-20 09:29:26 +01:00
Michal Krol
202c1279c6 svga: Fix relative addressing translation for pixel shaders.
Pixel shaders do not have address registers a#, only one
loop register aL. Our only hope is to assume the address
register is in fact a loop counter and replace it with aL.

Do not translate ARL instruction for pixel shaders -- MOVA
instruction is only valid for vertex saders.

Make it more explicit relative addressing of inputs is only valid
for pixel shaders and constants for vertex shaders.
2010-09-20 09:29:24 +01:00
Jakob Bornecrantz
f6e811a7d0 scons: Link against talloc in the Gallium DRI drivers
cherry picked from commit b83156d42f
2010-09-20 00:08:06 +02:00
Luca Barbieri
bd009df1cc auxiliary: fix depth-only and stencil-only clears
Depth-only and stencil-only clears should mask out depth/stencil from the
output, mask out stencil/input from input, and OR or ADD them together.

However, due to a typo they were being ANDed, resulting in zeroing the buffer.
2010-09-19 21:51:25 +02:00
Chia-I Wu
d6df0cdcca egl: Enable drm platform by default.
This enables EGL_MESA_drm_display for st/egl in the default setup.
(cherry picked from commit 2a910b3396)
2010-09-19 17:37:00 +08:00
Chia-I Wu
52f6be0688 st/egl: s/kms/drm/ on the drm backend.
s/kms/drm/, s/kdpy/drmdpy/, and so forth.
(cherry picked from commit e4513e7fb9)
2010-09-19 17:30:00 +08:00
Chia-I Wu
b73cfde5ce st/egl: Rename kms backend to drm.
The main use of the backend is to support EGL_MESA_drm_display.  drm
should be a better name.
(cherry picked from commit e7424d7240)
2010-09-19 17:29:55 +08:00
Chia-I Wu
d7500ac4df st/egl: Split modeset code support to modeset.c.
The modeset code supports now obsolete EGL_MESA_screen_surface.  Move it
to a file of its own.
(cherry picked from commit c7c2e7d0ce)
2010-09-19 17:29:50 +08:00
Brian Paul
83917660f1 tgsi/sse: fix aos_to_soa() loop to handle num_inputs==0
Basically, change the loop from:
  do {...} while (--num_inputs != 0)
into:
  while (num_inputs != 0) { ... --num_inputs; }

Fixes fd.o bug 29987.
(cherry picked from commit 4b27c614cf)
2010-09-17 10:51:50 -07:00
Chia-I Wu
69f67f5ded libgl-xlib: Remove unused st_api_create_OpenGL.
st/egl no longer relies on libGL for OpenGL support.
2010-09-17 12:57:15 +08:00
Chia-I Wu
a42527be33 targets/egl: Use C++ compiler to link GL/ES state trackers.
Otherwise, applications compiled with C compiler might have trouble
using them.
2010-09-17 12:57:06 +08:00
Brian Paul
d169a67ad1 mesa: include mfeatures.h in formats.c
Otherwise, FEATURE_EXT_texture_sRGB was undefined.
This is (part of?) the fix for fd.o bug 30177.

(cherry picked from commit 3a6f9d0f47)
2010-09-16 12:42:53 -06:00
Marek Olšák
be16a35438 r300g/swtcl: fix CS overrun
https://bugs.freedesktop.org/show_bug.cgi?id=29901
(cherry picked from commit d4b2de13bc)
2010-09-16 20:35:39 +02:00
Brian Paul
d46b7c9f6f glx: add const qualifiers to __indirect_glMultiDrawArraysEXT()
(cherry picked from commit 4cd751bcc4)
2010-09-16 14:34:06 +08:00
Chia-I Wu
5d02503467 st/xlib: Notify the context when the front/back buffers are swapped.
The current context should be notified when the the front/back buffers
of the current drawable are swapped.  The notification was skipped when
xmesa_strict_invalidate is false (the default).

This fixes fdo bug #29774.
(cherry picked from commit 03224f492d)
2010-09-16 14:24:46 +08:00
Chia-I Wu
e3b89388ff mesa: Update ES APIspec.xml.
Enable some extensions now that the needed tokens are defined in
GLES/glext.h and GLES2/glext.h.  Update the prototype of MultiDrawArrays
now that the prototype of _mesa_MultiDrawArraysEXT has been updated.
(cherry picked from commit 9ca59b2427)
2010-09-16 14:24:45 +08:00
Chia-I Wu
d6538baa3b glapi: Fix ES build errors again.
This fixes an error in GLAPI ES.  My build is ok with or without this
patch, and the error affects others' setups.

[Patch from Francesco Marella]
(cherry picked from commit e3c46cf586)
2010-09-16 14:24:45 +08:00
Chia-I Wu
16a023ad7d glapi: Fix build errors for ES.
The latest glext.h defines GL_FIXED.  Test GL_OES_fixed_point instead to
decide whether to define GLfixed and GLclampx.

This fixes fdo bug #30205.
(cherry picked from commit cad87ebc3a)
2010-09-16 14:24:45 +08:00
Andre Maasikas
029eb8e764 r600c: fix buffer height setting in dri2 case
fbHeight is 0 in this case

uncovered by changes in b0bc026c and should fix kernel rejecting command
streams after that commit
2010-09-16 09:09:43 +03:00
Brian Paul
71b4054a6c tgsi: fix incorrect usage_mask for shadow tex instructions
The shadow versions of the texture targets use an extra component
(Z) to express distance from light source to the fragment.
Fixes the shadowtex demo with llvmpipe.

(cherry picked from commit 95254bbd2d)
2010-09-15 13:57:50 -06:00
Keith Whitwell
8e8a42ffc5 llvmpipe: return zero from floor_pot(zero) 2010-09-15 16:41:59 +01:00
Dave Airlie
fd4c6bd135 r300g: fix buffer reuse issue caused by previous commit
caused by 0b9eb5c9bb

test run glxgears, resize.
(cherry picked from commit 09ef8e9283)
2010-09-15 13:32:29 +02:00
Marek Olšák
5d4409e0e5 r300g: prevent creating multiple winsys BOs for the same handle
This fixes a DRM deadlock in the cubestorm xscreensaver, because somehow
there must not be 2 different BOs relocated in one CS if both BOs back
the same handle. I was told it is impossible to happen, but apparently
it is not, or there is something else wrong.
(cherry picked from commit 0b9eb5c9bb)
2010-09-15 04:34:36 +02:00
Luca Barbieri
26e63a431d auxiliary: fix unintended fallthrough 2010-09-14 23:22:16 +02:00
1962 changed files with 92658 additions and 147959 deletions

View File

@@ -180,7 +180,7 @@ ultrix-gcc:
# Rules for making release tarballs
VERSION=7.10.2
VERSION=7.9.2
DIRECTORY = Mesa-$(VERSION)
LIB_NAME = MesaLib-$(VERSION)
GLUT_NAME = MesaGLUT-$(VERSION)
@@ -209,6 +209,7 @@ MAIN_FILES = \
$(DIRECTORY)/docs/README.* \
$(DIRECTORY)/docs/RELNOTES* \
$(DIRECTORY)/docs/*.spec \
$(DIRECTORY)/include/GL/internal/glcore.h \
$(DIRECTORY)/include/GL/gl.h \
$(DIRECTORY)/include/GL/glext.h \
$(DIRECTORY)/include/GL/gl_mangle.h \
@@ -307,7 +308,8 @@ MAPI_FILES = \
$(DIRECTORY)/src/mapi/mapi/*.[ch] \
$(DIRECTORY)/src/mapi/vgapi/Makefile \
$(DIRECTORY)/src/mapi/vgapi/vgapi.csv \
$(DIRECTORY)/src/mapi/vgapi/vg.pc.in
$(DIRECTORY)/src/mapi/vgapi/vg.pc.in \
$(DIRECTORY)/src/mapi/vgapi/*.h
EGL_FILES = \
$(DIRECTORY)/include/KHR/*.h \

View File

@@ -3,14 +3,14 @@
#
# For example, invoke scons as
#
# scons build=debug llvm=yes machine=x86
# scons debug=1 dri=0 machine=x86
#
# to set configuration variables. Or you can write those options to a file
# named config.py:
#
# # config.py
# build='debug'
# llvm=True
# debug=1
# dri=0
# machine='x86'
#
# Invoke
@@ -30,8 +30,55 @@ import common
#######################################################################
# Configuration options
default_statetrackers = 'mesa'
default_targets = 'graw-null'
if common.default_platform in ('linux', 'freebsd', 'darwin'):
default_drivers = 'softpipe,galahad,failover,svga,i915,i965,trace,identity,llvmpipe'
default_winsys = 'xlib'
elif common.default_platform in ('winddk',):
default_drivers = 'softpipe,svga,i915,i965,trace,identity'
default_winsys = 'all'
elif common.default_platform in ('embedded',):
default_drivers = 'softpipe,llvmpipe'
default_winsys = 'xlib'
else:
default_drivers = 'all'
default_winsys = 'all'
opts = Variables('config.py')
common.AddOptions(opts)
opts.Add(ListVariable('statetrackers', 'state trackers to build', default_statetrackers,
['mesa', 'python', 'xorg', 'egl']))
opts.Add(ListVariable('drivers', 'pipe drivers to build', default_drivers,
['softpipe', 'galahad', 'failover', 'svga', 'i915', 'i965', 'trace', 'r300', 'r600', 'identity', 'llvmpipe', 'nouveau', 'nv50', 'nvfx']))
opts.Add(ListVariable('winsys', 'winsys drivers to build', default_winsys,
['xlib', 'vmware', 'i915', 'i965', 'gdi', 'radeon', 'r600', 'graw-xlib']))
opts.Add(ListVariable('targets', 'driver targets to build', default_targets,
['dri-i915',
'dri-i965',
'dri-nouveau',
'dri-radeong',
'dri-swrast',
'dri-vmwgfx',
'egl-i915',
'egl-i965',
'egl-nouveau',
'egl-radeon',
'egl-swrast',
'egl-vmwgfx',
'graw-xlib',
'graw-null',
'libgl-gdi',
'libgl-xlib',
'xorg-i915',
'xorg-i965',
'xorg-nouveau',
'xorg-radeon',
'xorg-vmwgfx']))
opts.Add(EnumVariable('MSVS_VERSION', 'MS Visual C++ version', None, allowed_values=('7.1', '8.0', '9.0')))
env = Environment(
options = opts,
@@ -40,26 +87,61 @@ env = Environment(
ENV = os.environ,
)
# Backwards compatability with old target configuration variable
try:
targets = ARGUMENTS['targets']
except KeyError:
pass
else:
targets = targets.split(',')
print 'scons: warning: targets option is deprecated; pass the targets on their own such as'
print
print ' scons %s' % ' '.join(targets)
print
COMMAND_LINE_TARGETS.append(targets)
if os.environ.has_key('CC'):
env['CC'] = os.environ['CC']
if os.environ.has_key('CFLAGS'):
env['CCFLAGS'] += SCons.Util.CLVar(os.environ['CFLAGS'])
if os.environ.has_key('CXX'):
env['CXX'] = os.environ['CXX']
if os.environ.has_key('CXXFLAGS'):
env['CXXFLAGS'] += SCons.Util.CLVar(os.environ['CXXFLAGS'])
if os.environ.has_key('LDFLAGS'):
env['LINKFLAGS'] += SCons.Util.CLVar(os.environ['LDFLAGS'])
Help(opts.GenerateHelpText(env))
# replicate options values in local variables
debug = env['debug']
dri = env['dri']
machine = env['machine']
platform = env['platform']
# derived options
x86 = machine == 'x86'
ppc = machine == 'ppc'
gcc = platform in ('linux', 'freebsd', 'darwin', 'embedded')
msvc = platform in ('windows', 'winddk')
Export([
'debug',
'x86',
'ppc',
'dri',
'platform',
'gcc',
'msvc',
])
#######################################################################
# Environment setup
# Always build trace, rbug, identity, softpipe, and llvmpipe (where possible)
if 'trace' not in env['drivers']:
env['drivers'].append('trace')
if 'rbug' not in env['drivers']:
env['drivers'].append('rbug')
if 'galahad' not in env['drivers']:
env['drivers'].append('galahad')
if 'identity' not in env['drivers']:
env['drivers'].append('identity')
if 'softpipe' not in env['drivers']:
env['drivers'].append('softpipe')
if env['llvm'] and 'llvmpipe' not in env['drivers']:
env['drivers'].append('llvmpipe')
if 'sw' not in env['drivers']:
env['drivers'].append('sw')
# Includes
env.Prepend(CPPPATH = [
'#/include',
@@ -75,7 +157,7 @@ if env['msvc']:
env.Append(CPPPATH = ['#include/c99'])
# Embedded
if env['platform'] == 'embedded':
if platform == 'embedded':
env.Append(CPPDEFINES = [
'_POSIX_SOURCE',
('_POSIX_C_SOURCE', '199309L'),
@@ -92,7 +174,7 @@ if env['platform'] == 'embedded':
])
# Posix
if env['platform'] in ('posix', 'linux', 'freebsd', 'darwin'):
if platform in ('posix', 'linux', 'freebsd', 'darwin'):
env.Append(CPPDEFINES = [
'_POSIX_SOURCE',
('_POSIX_C_SOURCE', '199309L'),
@@ -102,9 +184,9 @@ if env['platform'] in ('posix', 'linux', 'freebsd', 'darwin'):
'PTHREADS',
'HAVE_POSIX_MEMALIGN',
])
if env['gcc']:
if gcc:
env.Append(CFLAGS = ['-fvisibility=hidden'])
if env['platform'] == 'darwin':
if platform == 'darwin':
env.Append(CPPDEFINES = ['_DARWIN_C_SOURCE'])
env.Append(LIBS = [
'm',
@@ -126,7 +208,9 @@ Export('env')
SConscript(
'src/SConscript',
variant_dir = env['build_dir'],
variant_dir = env['build'],
duplicate = 0 # http://www.scons.org/doc/0.97/HTML/scons-user/x2261.html
)
env.Default('src')

View File

@@ -8,8 +8,6 @@ import subprocess
import sys
import platform as _platform
import SCons.Script.SConscript
#######################################################################
# Defaults
@@ -22,15 +20,6 @@ _platform_map = {
default_platform = sys.platform
default_platform = _platform_map.get(default_platform, default_platform)
# Search sys.argv[] for a "platform=foo" argument since we don't have
# an 'env' variable at this point.
if 'platform' in SCons.Script.ARGUMENTS:
selected_platform = SCons.Script.ARGUMENTS['platform']
else:
selected_platform = default_platform
cross_compiling = selected_platform != default_platform
_machine_map = {
'x86': 'x86',
'i386': 'x86',
@@ -48,26 +37,38 @@ if 'PROCESSOR_ARCHITECTURE' in os.environ:
else:
default_machine = _platform.machine()
default_machine = _machine_map.get(default_machine, 'generic')
default_toolchain = 'default'
if selected_platform == 'windows' and cross_compiling:
default_machine = 'x86'
default_toolchain = 'crossmingw'
# find default_llvm value
if 'LLVM' in os.environ:
default_llvm = 'yes'
else:
# Search sys.argv[] for a "platform=foo" argument since we don't have
# an 'env' variable at this point.
platform = default_platform
pattern = re.compile("(platform=)(.*)")
for arg in sys.argv:
m = pattern.match(arg)
if m:
platform = m.group(2)
default_llvm = 'no'
try:
if selected_platform != 'windows' and \
subprocess.call(['llvm-config', '--version'], stdout=subprocess.PIPE) == 0:
if platform != 'windows' and subprocess.call(['llvm-config', '--version'], stdout=subprocess.PIPE) == 0:
default_llvm = 'yes'
except:
pass
# find default_dri value
if default_platform in ('linux', 'freebsd'):
default_dri = 'yes'
elif default_platform in ('winddk', 'windows', 'wince', 'darwin'):
default_dri = 'no'
else:
default_dri = 'no'
#######################################################################
# Common options
@@ -80,15 +81,13 @@ def AddOptions(opts):
from SCons.Variables.EnumVariable import EnumVariable as EnumOption
except ImportError:
from SCons.Options.EnumOption import EnumOption
opts.Add(EnumOption('build', 'build type', 'debug',
allowed_values=('debug', 'checked', 'profile', 'release')))
opts.Add(BoolOption('debug', 'debug build', 'yes'))
opts.Add(BoolOption('profile', 'profile build', 'no'))
opts.Add(BoolOption('quiet', 'quiet command lines', 'yes'))
opts.Add(EnumOption('machine', 'use machine-specific assembly code', default_machine,
allowed_values=('generic', 'ppc', 'x86', 'x86_64')))
opts.Add(EnumOption('platform', 'target platform', default_platform,
allowed_values=('linux', 'cell', 'windows', 'winddk', 'wince', 'darwin', 'embedded', 'cygwin', 'sunos5', 'freebsd8')))
opts.Add('toolchain', 'compiler toolchain', default_toolchain)
opts.Add('toolchain', 'compiler toolchain', 'default')
opts.Add(BoolOption('llvm', 'use LLVM', default_llvm))
opts.Add(BoolOption('debug', 'DEPRECATED: debug build', 'yes'))
opts.Add(BoolOption('profile', 'DEPRECATED: profile build', 'no'))
opts.Add(EnumOption('MSVS_VERSION', 'MS Visual C++ version', None, allowed_values=('7.1', '8.0', '9.0')))
opts.Add(BoolOption('dri', 'build DRI drivers', default_dri))

View File

@@ -9,7 +9,7 @@ CONFIG_NAME = default
# Version info
MESA_MAJOR=7
MESA_MINOR=10
MESA_MINOR=9
MESA_TINY=0
MESA_VERSION = $(MESA_MAJOR).$(MESA_MINOR).$(MESA_TINY)

View File

@@ -58,7 +58,7 @@ EGL_DRIVERS_DIRS = glx
DRIVER_DIRS = dri
GALLIUM_WINSYS_DIRS = sw sw/xlib drm/vmware drm/intel drm/i965
GALLIUM_TARGET_DIRS =
GALLIUM_TARGET_DIRS = egl-swrast
GALLIUM_STATE_TRACKERS_DIRS = egl
DRI_DIRS = i810 i915 i965 mach64 mga r128 r200 r300 radeon \

View File

@@ -15,7 +15,7 @@ ARCH_FLAGS = -mmmx -msse -msse2 -mstackrealign
DEFINES += -DNDEBUG -DGALLIUM_LLVMPIPE -DHAVE_UDIS86
# override -std=c99
CFLAGS += -std=gnu99
CFLAGS += -std=gnu99 -D__STDC_CONSTANT_MACROS
LLVM_VERSION := $(shell llvm-config --version)
@@ -30,7 +30,7 @@ else
endif
ifeq ($(MESA_LLVM),1)
LLVM_CFLAGS=`llvm-config --cppflags`
# LLVM_CFLAGS=`llvm-config --cflags`
LLVM_CXXFLAGS=`llvm-config --cxxflags backend bitreader engine ipo interpreter instrumentation` -Wno-long-long
LLVM_LDFLAGS = $(shell llvm-config --ldflags backend bitreader engine ipo interpreter instrumentation)
LLVM_LIBS = $(shell llvm-config --libs backend bitwriter bitreader engine ipo interpreter instrumentation)

View File

@@ -34,10 +34,6 @@ AC_CHECK_PROGS([PYTHON2], [python2 python])
AC_PATH_PROG([MKDEP], [makedepend])
AC_PATH_PROG([SED], [sed])
if test "x$MKDEP" = "x"; then
AC_MSG_ERROR([makedepend is required to build Mesa])
fi
dnl Our fallback install-sh is a symlink to minstall. Use the existing
dnl configuration in that case.
AC_PROG_INSTALL
@@ -465,71 +461,6 @@ if test "x$enable_selinux" = "xyes"; then
DEFINES="$DEFINES -DMESA_SELINUX"
fi
dnl Determine which APIs to support
AC_ARG_ENABLE([opengl],
[AS_HELP_STRING([--disable-opengl],
[disable support for standard OpenGL API @<:@default=no@:>@])],
[enable_opengl="$enableval"],
[enable_opengl=yes])
AC_ARG_ENABLE([gles1],
[AS_HELP_STRING([--enable-gles1],
[enable support for OpenGL ES 1.x API @<:@default=no@:>@])],
[enable_gles1="$enableval"],
[enable_gles1=no])
AC_ARG_ENABLE([gles2],
[AS_HELP_STRING([--enable-gles2],
[enable support for OpenGL ES 2.x API @<:@default=no@:>@])],
[enable_gles2="$enableval"],
[enable_gles2=no])
AC_ARG_ENABLE([gles-overlay],
[AS_HELP_STRING([--enable-gles-overlay],
[build separate OpenGL ES only libraries @<:@default=no@:>@])],
[enable_gles_overlay="$enableval"],
[enable_gles_overlay=no])
AC_ARG_ENABLE([openvg],
[AS_HELP_STRING([--enable-openvg],
[enable support for OpenVG API @<:@default=no@:>@])],
[enable_openvg="$enableval"],
[enable_openvg=no])
dnl smooth the transition; should be removed eventually
if test "x$enable_openvg" = xno; then
case "x$with_state_trackers" in
x*vega*)
AC_MSG_WARN([vega state tracker is enabled without --enable-openvg])
enable_openvg=yes
;;
esac
fi
if test "x$enable_opengl" = xno -a \
"x$enable_gles1" = xno -a \
"x$enable_gles2" = xno -a \
"x$enable_gles_overlay" = xno -a \
"x$enable_openvg" = xno; then
AC_MSG_ERROR([at least one API should be enabled])
fi
API_DEFINES=""
GLES_OVERLAY=0
if test "x$enable_opengl" = xno; then
API_DEFINES="$API_DEFINES -DFEATURE_GL=0"
else
API_DEFINES="$API_DEFINES -DFEATURE_GL=1"
fi
if test "x$enable_gles1" = xyes; then
API_DEFINES="$API_DEFINES -DFEATURE_ES1=1"
fi
if test "x$enable_gles2" = xyes; then
API_DEFINES="$API_DEFINES -DFEATURE_ES2=1"
fi
if test "x$enable_gles_overlay" = xyes; then
GLES_OVERLAY=1
fi
AC_SUBST([API_DEFINES])
AC_SUBST([GLES_OVERLAY])
dnl
dnl Driver configuration. Options are xlib, dri and osmesa right now.
dnl More later: fbdev, ...
@@ -549,10 +480,6 @@ linux*)
;;
esac
if test "x$enable_opengl" = xno; then
default_driver="no"
fi
AC_ARG_WITH([driver],
[AS_HELP_STRING([--with-driver=DRIVER],
[driver for Mesa: xlib,dri,osmesa @<:@default=dri when available, or xlib@:>@])],
@@ -561,11 +488,6 @@ AC_ARG_WITH([driver],
dnl Check for valid option
case "x$mesa_driver" in
xxlib|xdri|xosmesa)
if test "x$enable_opengl" = xno; then
AC_MSG_ERROR([Driver '$mesa_driver' requires OpenGL enabled])
fi
;;
xno)
;;
*)
AC_MSG_ERROR([Driver '$mesa_driver' is not a valid option])
@@ -577,7 +499,7 @@ dnl Driver specific build directories
dnl
dnl this variable will be prepended to SRC_DIRS and is not exported
CORE_DIRS=""
CORE_DIRS="mapi/glapi glsl mesa"
SRC_DIRS=""
GLU_DIRS="sgi"
@@ -587,30 +509,6 @@ GALLIUM_WINSYS_DIRS="sw"
GALLIUM_DRIVERS_DIRS="softpipe failover galahad trace rbug identity"
GALLIUM_STATE_TRACKERS_DIRS=""
# build glapi if OpenGL is enabled
if test "x$enable_opengl" = xyes; then
CORE_DIRS="$CORE_DIRS mapi/glapi"
fi
# build es1api and es2api if OpenGL ES is enabled
case "x$enable_gles1$enable_gles2$enable_gles_overlay" in
x*yes*)
CORE_DIRS="$CORE_DIRS mapi/es1api mapi/es2api"
;;
esac
# build vgapi if OpenVG is enabled
if test "x$enable_openvg" = xyes; then
CORE_DIRS="$CORE_DIRS mapi/vgapi"
fi
# build glsl and mesa if OpenGL or OpenGL ES is enabled
case "x$enable_opengl$enable_gles1$enable_gles2$enable_gles_overlay" in
x*yes*)
CORE_DIRS="$CORE_DIRS glsl mesa"
;;
esac
case "$mesa_driver" in
xlib)
DRIVER_DIRS="x11"
@@ -625,9 +523,6 @@ dri)
osmesa)
DRIVER_DIRS="osmesa"
;;
no)
DRIVER_DRIS=""
;;
esac
AC_SUBST([SRC_DIRS])
AC_SUBST([GLU_DIRS])
@@ -720,7 +615,7 @@ xlib)
GL_LIB_DEPS=""
fi
;;
dri|no) # these checks are still desired when there is no mesa_driver
dri)
# DRI must be shared, I think
if test "$enable_static" = yes; then
AC_MSG_ERROR([Can't use static libraries for DRI drivers])
@@ -845,6 +740,51 @@ if test "x$with_dri_drivers" = x; then
with_dri_drivers=no
fi
dnl Determine which APIs to support
AC_ARG_ENABLE([opengl],
[AS_HELP_STRING([--disable-opengl],
[disable support for standard OpenGL API @<:@default=no@:>@])],
[enable_opengl="$enableval"],
[enable_opengl=yes])
AC_ARG_ENABLE([gles1],
[AS_HELP_STRING([--enable-gles1],
[enable support for OpenGL ES 1.x API @<:@default=no@:>@])],
[enable_gles1="$enableval"],
[enable_gles1=no])
AC_ARG_ENABLE([gles2],
[AS_HELP_STRING([--enable-gles2],
[enable support for OpenGL ES 2.x API @<:@default=no@:>@])],
[enable_gles2="$enableval"],
[enable_gles2=no])
AC_ARG_ENABLE([gles-overlay],
[AS_HELP_STRING([--enable-gles-overlay],
[build separate OpenGL ES only libraries @<:@default=no@:>@])],
[enable_gles_overlay="$enableval"],
[enable_gles_overlay=no])
API_DEFINES=""
GLES_OVERLAY=0
if test "x$enable_opengl" = xno; then
API_DEFINES="$API_DEFINES -DFEATURE_GL=0"
else
API_DEFINES="$API_DEFINES -DFEATURE_GL=1"
fi
if test "x$enable_gles1" = xyes; then
API_DEFINES="$API_DEFINES -DFEATURE_ES1=1"
fi
if test "x$enable_gles2" = xyes; then
API_DEFINES="$API_DEFINES -DFEATURE_ES2=1"
fi
if test "x$enable_gles_overlay" = xyes -o \
"x$enable_gles1" = xyes -o "x$enable_gles2" = xyes; then
CORE_DIRS="mapi/es1api mapi/es2api $CORE_DIRS"
if test "x$enable_gles_overlay" = xyes; then
GLES_OVERLAY=1
fi
fi
AC_SUBST([API_DEFINES])
AC_SUBST([GLES_OVERLAY])
dnl If $with_dri_drivers is yes, directories will be added through
dnl platform checks
DRI_DIRS=""
@@ -865,7 +805,7 @@ yes)
esac
dnl Set DRI_DIRS, DEFINES and LIB_DEPS
if test "$mesa_driver" = dri -o "$mesa_driver" = no; then
if test "$mesa_driver" = dri; then
# Use TLS in GLX?
if test "x$GLX_USE_TLS" = xyes; then
DEFINES="$DEFINES -DGLX_USE_TLS -DPTHREADS"
@@ -943,21 +883,19 @@ if test "$mesa_driver" = dri -o "$mesa_driver" = no; then
DRI_DIRS=`echo "$DRI_DIRS" | $SED 's/ */ /g'`
# Check for expat
if test "$mesa_driver" = dri; then
EXPAT_INCLUDES=""
EXPAT_LIB=-lexpat
AC_ARG_WITH([expat],
[AS_HELP_STRING([--with-expat=DIR],
[expat install directory])],[
EXPAT_INCLUDES="-I$withval/include"
CPPFLAGS="$CPPFLAGS $EXPAT_INCLUDES"
LDFLAGS="$LDFLAGS -L$withval/$LIB_DIR"
EXPAT_LIB="-L$withval/$LIB_DIR -lexpat"
])
AC_CHECK_HEADER([expat.h],[],[AC_MSG_ERROR([Expat required for DRI.])])
AC_CHECK_LIB([expat],[XML_ParserCreate],[],
[AC_MSG_ERROR([Expat required for DRI.])])
fi
EXPAT_INCLUDES=""
EXPAT_LIB=-lexpat
AC_ARG_WITH([expat],
[AS_HELP_STRING([--with-expat=DIR],
[expat install directory])],[
EXPAT_INCLUDES="-I$withval/include"
CPPFLAGS="$CPPFLAGS $EXPAT_INCLUDES"
LDFLAGS="$LDFLAGS -L$withval/$LIB_DIR"
EXPAT_LIB="-L$withval/$LIB_DIR -lexpat"
])
AC_CHECK_HEADER([expat.h],[],[AC_MSG_ERROR([Expat required for DRI.])])
AC_CHECK_LIB([expat],[XML_ParserCreate],[],
[AC_MSG_ERROR([Expat required for DRI.])])
# put all the necessary libs together
DRI_LIB_DEPS="$SELINUX_LIBS $LIBDRM_LIBS $EXPAT_LIB -lm -lpthread $DLOPEN_LIBS"
@@ -998,9 +936,6 @@ AC_ARG_ENABLE([gl-osmesa],
[gl_osmesa="$enableval"],
[gl_osmesa="$default_gl_osmesa"])
if test "x$gl_osmesa" = xyes; then
if test "x$enable_opengl" = xno; then
AC_MSG_ERROR([OpenGL is not available for OSMesa driver])
fi
if test "$mesa_driver" = osmesa; then
AC_MSG_ERROR([libGL is not available for OSMesa driver])
else
@@ -1057,21 +992,13 @@ AC_ARG_ENABLE([egl],
[disable EGL library @<:@default=enabled@:>@])],
[enable_egl="$enableval"],
[enable_egl=yes])
if test "x$enable_egl" = xno; then
if test "x$mesa_driver" = xno; then
AC_MSG_ERROR([cannot disable EGL when there is no mesa driver])
fi
if test "x$enable_openvg" = xyes; then
AC_MSG_ERROR([cannot enable OpenVG without EGL])
fi
fi
if test "x$enable_egl" = xyes; then
SRC_DIRS="$SRC_DIRS egl"
EGL_LIB_DEPS="$DLOPEN_LIBS -lpthread"
EGL_DRIVERS_DIRS=""
if test "$enable_static" != yes; then
# build egl_glx when libGL is built
if test "$mesa_driver" = xlib -o "$mesa_driver" = dri; then
if test "$mesa_driver" != osmesa; then
EGL_DRIVERS_DIRS="glx"
fi
@@ -1105,12 +1032,6 @@ AC_ARG_ENABLE([glu],
[enable OpenGL Utility library @<:@default=enabled@:>@])],
[enable_glu="$enableval"],
[enable_glu=yes])
if test "x$enable_glu" = xyes -a "x$mesa_driver" = xno; then
AC_MSG_NOTICE([Disabling GLU since there is no OpenGL driver])
enable_glu=no
fi
if test "x$enable_glu" = xyes; then
SRC_DIRS="$SRC_DIRS glu"
@@ -1160,13 +1081,9 @@ AC_ARG_ENABLE([glw],
[enable_glw="$enableval"],
[enable_glw=yes])
dnl Don't build GLw on osmesa
if test "x$enable_glw" = xyes; then
case "$mesa_driver" in
osmesa|no)
AC_MSG_NOTICE([Disabling GLw since there is no OpenGL driver])
enable_glw=no
;;
esac
if test "x$enable_glw" = xyes && test "$mesa_driver" = osmesa; then
AC_MSG_WARN([Disabling GLw since the driver is OSMesa])
enable_glw=no
fi
AC_ARG_ENABLE([motif],
[AS_HELP_STRING([--enable-motif],
@@ -1240,20 +1157,16 @@ AC_ARG_ENABLE([glut],
[enable_glut="$enableval"],
[enable_glut="$default_glut"])
dnl Don't build glut on osmesa
if test "x$enable_glut" = xyes; then
case "$mesa_driver" in
osmesa|no)
AC_MSG_NOTICE([Disabling glut since there is no OpenGL driver])
enable_glut=no
;;
esac
fi
dnl Can't build glut if GLU not available
if test "x$enable_glu$enable_glut" = xnoyes; then
AC_MSG_WARN([Disabling glut since GLU is disabled])
enable_glut=no
fi
dnl Don't build glut on osmesa
if test "x$enable_glut" = xyes && test "$mesa_driver" = osmesa; then
AC_MSG_WARN([Disabling glut since the driver is OSMesa])
enable_glut=no
fi
if test "x$enable_glut" = xyes; then
SRC_DIRS="$SRC_DIRS glut/glx"
@@ -1318,9 +1231,6 @@ AC_ARG_ENABLE([gallium],
[build gallium @<:@default=enabled@:>@])],
[enable_gallium="$enableval"],
[enable_gallium=yes])
if test "x$enable_gallium" = xno -a "x$enable_openvg" = xyes; then
AC_MSG_ERROR([cannot enable OpenVG without Gallium])
fi
if test "x$enable_gallium" = xyes; then
SRC_DIRS="$SRC_DIRS gallium gallium/winsys gallium/targets"
AC_CHECK_HEADER([udis86.h], [HAS_UDIS86="yes"],
@@ -1333,30 +1243,15 @@ AC_SUBST([LLVM_LIBS])
AC_SUBST([LLVM_LDFLAGS])
AC_SUBST([LLVM_VERSION])
VG_LIB_DEPS=""
EGL_CLIENT_APIS='$(GL_LIB)'
if test "x$enable_gles_overlay" = xyes; then
EGL_CLIENT_APIS="$EGL_CLIENT_APIS "'$(GLESv1_CM_LIB) $(GLESv2_LIB)'
fi
dnl
dnl Gallium state trackers configuration
dnl
AC_ARG_ENABLE([gallium-egl],
[AS_HELP_STRING([--enable-gallium-egl],
[enable gallium EGL state tracker @<:@default=auto@:>@])],
[enable_gallium_egl="$enableval"],
[enable_gallium_egl=auto])
if test "x$enable_gallium_egl" = xauto; then
case "$mesa_driver" in
dri|no)
enable_gallium_egl=$enable_egl
;;
*)
enable_gallium_egl=$enable_openvg
;;
esac
fi
case "x$enable_egl$enable_gallium_egl" in
xnoyes)
AC_MSG_ERROR([cannot build Gallium EGL state tracker without EGL])
esac
AC_ARG_WITH([state-trackers],
[AS_HELP_STRING([--with-state-trackers@<:@=DIRS...@:>@],
[comma delimited state_trackers list, e.g.
@@ -1377,24 +1272,16 @@ yes)
dri)
GALLIUM_STATE_TRACKERS_DIRS="dri"
HAVE_ST_DRI="yes"
if test "x$enable_egl" = xyes; then
GALLIUM_STATE_TRACKERS_DIRS="$GALLIUM_STATE_TRACKERS_DIRS egl"
HAVE_ST_EGL="yes"
fi
# Have only tested st/xorg on 1.6.0 servers
PKG_CHECK_MODULES(XORG, [xorg-server >= 1.6.0 libdrm >= $LIBDRM_XORG_REQUIRED libkms >= $LIBKMS_XORG_REQUIRED],
HAVE_ST_XORG="yes"; GALLIUM_STATE_TRACKERS_DIRS="$GALLIUM_STATE_TRACKERS_DIRS xorg",
HAVE_ST_XORG="no")
;;
esac
if test "x$enable_egl" = xyes; then
if test "$enable_openvg" = yes; then
GALLIUM_STATE_TRACKERS_DIRS="$GALLIUM_STATE_TRACKERS_DIRS vega"
st_egl="yes"
fi
if test "$enable_gallium_egl" = yes; then
GALLIUM_STATE_TRACKERS_DIRS="$GALLIUM_STATE_TRACKERS_DIRS egl"
HAVE_ST_EGL="yes"
fi
fi
;;
*)
# verify the requested state tracker exist
@@ -1420,10 +1307,22 @@ yes)
PKG_CHECK_MODULES([LIBKMS_XORG], [libkms >= $LIBKMS_XORG_REQUIRED])
HAVE_ST_XORG="yes"
;;
vega)
if test "x$enable_openvg" != xyes; then
AC_MSG_ERROR([cannot build vega state tracker without --enable-openvg])
es)
AC_MSG_WARN([state tracker 'es' has been replaced by --enable-gles-overlay])
if test "x$enable_gles_overlay" != xyes; then
if test "x$enable_gles1" != xyes -a "x$enable_gles2" != xyes; then
CORE_DIRS="mapi/es1api mapi/es2api $CORE_DIRS"
fi
GLES_OVERLAY=1
EGL_CLIENT_APIS="$EGL_CLIENT_APIS "'$(GLESv1_CM_LIB) $(GLESv2_LIB)'
fi
tracker=""
;;
vega)
CORE_DIRS="$CORE_DIRS mapi/vgapi"
VG_LIB_DEPS="$VG_LIB_DEPS -lpthread"
EGL_CLIENT_APIS="$EGL_CLIENT_APIS "'$(VG_LIB)'
;;
esac
@@ -1441,28 +1340,15 @@ yes)
;;
esac
EGL_CLIENT_APIS=""
VG_LIB_DEPS=""
case "x$enable_opengl$enable_gles1$enable_gles2" in
x*yes*)
EGL_CLIENT_APIS="$EGL_CLIENT_APIS "'$(GL_LIB)'
;;
esac
if test "x$enable_gles_overlay" = xyes; then
EGL_CLIENT_APIS="$EGL_CLIENT_APIS "'$(GLESv1_CM_LIB) $(GLESv2_LIB)'
fi
if test "x$enable_openvg" = xyes; then
EGL_CLIENT_APIS="$EGL_CLIENT_APIS "'$(VG_LIB)'
VG_LIB_DEPS="$VG_LIB_DEPS -lpthread"
fi
AC_SUBST([VG_LIB_DEPS])
AC_SUBST([EGL_CLIENT_APIS])
if test "x$HAVE_ST_EGL" = xyes; then
GALLIUM_TARGET_DIRS="$GALLIUM_TARGET_DIRS egl"
# define GLX_DIRECT_RENDERING even when the driver is not dri
if test "x$mesa_driver" != xdri -a "x$driglx_direct" = xyes; then
DEFINES="$DEFINES -DGLX_DIRECT_RENDERING"
fi
fi
if test "x$HAVE_ST_XORG" = xyes; then
@@ -1550,7 +1436,7 @@ AC_ARG_ENABLE([gallium-llvm],
if test "x$enable_gallium_llvm" = xyes; then
if test "x$LLVM_CONFIG" != xno; then
LLVM_VERSION=`$LLVM_CONFIG --version`
LLVM_CFLAGS=`$LLVM_CONFIG --cppflags`
LLVM_CFLAGS=`$LLVM_CONFIG --cflags`
LLVM_LIBS="`$LLVM_CONFIG --libs jit interpreter nativecodegen bitwriter` -lstdc++"
if test "x$HAS_UDIS86" != xno; then
@@ -1700,19 +1586,6 @@ if test "x$enable_gallium_swrast" = xyes || test "x$enable_gallium_swrast" = xau
fi
fi
dnl
dnl Gallium noop configuration
dnl
AC_ARG_ENABLE([gallium-noop],
[AS_HELP_STRING([--enable-gallium-noop],
[build gallium radeon @<:@default=disabled@:>@])],
[enable_gallium_noop="$enableval"],
[enable_gallium_noop=auto])
if test "x$enable_gallium_noop" = xyes; then
GALLIUM_DRIVERS_DIRS="$GALLIUM_DRIVERS_DIRS noop"
GALLIUM_TARGET_DIRS="$GALLIUM_TARGET_DIRS dri-noop"
fi
dnl prepend CORE_DIRS to SRC_DIRS
SRC_DIRS="$CORE_DIRS $SRC_DIRS"
@@ -1742,56 +1615,25 @@ echo " exec_prefix: $exec_prefix"
echo " libdir: $libdir"
echo " includedir: $includedir"
dnl API info
echo ""
echo " OpenGL: $enable_opengl (ES1: $enable_gles1 ES2: $enable_gles2)"
echo " GLES overlay: $enable_gles_overlay"
echo " OpenVG: $enable_openvg"
dnl Driver info
echo ""
echo " Driver: $mesa_driver"
if test "$mesa_driver" != no; then
if echo "$DRIVER_DIRS" | grep 'osmesa' >/dev/null 2>&1; then
echo " OSMesa: lib$OSMESA_LIB"
else
echo " OSMesa: no"
fi
if test "$mesa_driver" = dri; then
# cleanup the drivers var
dri_dirs=`echo $DRI_DIRS | $SED 's/^ *//;s/ */ /;s/ *$//'`
if test "x$DRI_DIRS" = x; then
echo " DRI drivers: no"
else
echo " DRI drivers: $dri_dirs"
fi
echo " DRI driver dir: $DRI_DRIVER_INSTALL_DIR"
echo " Use XCB: $enable_xcb"
fi
if echo "$DRIVER_DIRS" | grep 'osmesa' >/dev/null 2>&1; then
echo " OSMesa: lib$OSMESA_LIB"
else
echo " OSMesa: no"
fi
echo ""
echo " GLU: $enable_glu"
echo " GLw: $enable_glw (Motif: $enable_motif)"
echo " glut: $enable_glut"
dnl EGL
echo ""
echo " EGL: $enable_egl"
if test "$enable_egl" = yes; then
echo " EGL platforms: $EGL_PLATFORMS"
egl_drivers=""
for d in $EGL_DRIVERS_DIRS; do
egl_drivers="$egl_drivers egl_$d"
done
if test "$enable_gallium" = yes -a "$HAVE_ST_EGL" = yes; then
echo " EGL drivers: ${egl_drivers} egl_gallium"
echo " EGL Gallium STs:$EGL_CLIENT_APIS"
else
echo " EGL drivers: $egl_drivers"
fi
if test "$mesa_driver" = dri; then
# cleanup the drivers var
dri_dirs=`echo $DRI_DIRS | $SED 's/^ *//;s/ */ /;s/ *$//'`
if test "x$DRI_DIRS" = x; then
echo " DRI drivers: no"
else
echo " DRI drivers: $dri_dirs"
fi
echo " DRI driver dir: $DRI_DRIVER_INSTALL_DIR"
fi
echo " Use XCB: $enable_xcb"
echo ""
if test "x$MESA_LLVM" = x1; then
@@ -1810,6 +1652,9 @@ if echo "$SRC_DIRS" | grep 'gallium' >/dev/null 2>&1; then
echo " Winsys dirs: $GALLIUM_WINSYS_DIRS"
echo " Driver dirs: $GALLIUM_DRIVERS_DIRS"
echo " Trackers dirs: $GALLIUM_STATE_TRACKERS_DIRS"
if test "x$HAVE_ST_EGL" = xyes; then
echo " EGL client APIs: $EGL_CLIENT_APIS"
fi
else
echo " Gallium: no"
fi
@@ -1818,6 +1663,15 @@ dnl Libraries
echo ""
echo " Shared libs: $enable_shared"
echo " Static libs: $enable_static"
if test "$enable_egl" = yes; then
echo " EGL: $EGL_DRIVERS_DIRS"
echo " EGL platforms: $EGL_PLATFORMS"
else
echo " EGL: no"
fi
echo " GLU: $enable_glu"
echo " GLw: $enable_glw (Motif: $enable_motif)"
echo " glut: $enable_glut"
dnl Compiler options
# cleanup the CFLAGS/CXXFLAGS/DEFINES vars

View File

@@ -20,22 +20,21 @@ Float textures, renderbuffers some infrastructure done
Framebuffer objects (GL_EXT_framebuffer_object) DONE
Half-float some infrastructure done
Multisample blit DONE
Non-normalized Integer texture/framebuffer formats ~50% done
Non-normalized Integer texture/framebuffer formats not started
1D/2D Texture arrays core Mesa, swrast done
Packed depth/stencil formats DONE
Per-buffer blend and masks (GL_EXT_draw_buffers2) DONE
GL_EXT_texture_compression_rgtc not started
Red and red/green texture formats DONE (swrast, i965, gallium)
Red and red/green texture formats Ian?
Transform feedback (GL_EXT_transform_feedback) ~50% done
glBindFragDataLocation, glGetFragDataLocation,
glBindBufferRange, glBindBufferBase commands
Vertex array objects (GL_APPLE_vertex_array_object) DONE
sRGB framebuffer format (GL_EXT_framebuffer_sRGB) not started
glClearBuffer commands DONE
glGetStringi command DONE
glTexParameterI, glGetTexParameterI commands DONE
glVertexAttribI commands DONE (but converts int
values to floats)
glClearBuffer commands DONE, except for dispatch
glGetStringi command DONE, except for dispatch
glTexParameterI, glGetTexParameterI commands DONE, except for dispatch
glVertexAttribI commands not started
GL 3.1:
@@ -43,9 +42,9 @@ GL 3.1:
GLSL 1.30 and 1.40 not started
Instanced drawing (GL_ARB_draw_instanced) ~50% done
Buffer copying (GL_ARB_copy_buffer) DONE
Primitive restart (GL_NV_primitive_restart) DONE (gallium)
Primitive restart (GL_NV_primitive_restart) not started
16 vertex texture image units not started
Texture buffer objs (GL_ARB_texture_buffer_object) not started
Texture buffer objs (GL_ARB_textur_buffer_object) not started
Rectangular textures (GL_ARB_texture_rectangle) DONE
Uniform buffer objs (GL_ARB_uniform_buffer_object) not started
Signed normalized texture formats ~50% done
@@ -70,7 +69,7 @@ GL 3.3:
GLSL 3.30 not started
GL_ARB_blend_func_extended not started
GL_ARB_explicit_attrib_location DONE (swrast, i915, i965)
GL_ARB_explicit_attrib_location not started
GL_ARB_occlusion_query2 not started
GL_ARB_sampler_objects not started
GL_ARB_texture_rgb10_a2ui not started
@@ -94,18 +93,6 @@ GL_ARB_texture_buffer_object_rgb32 not started
GL_ARB_texture_cube_map_array not started
GL_ARB_texture_gather not started
GL_ARB_transform_feedback2 not started
GL_ARB_transform_feedback3 not started
GL 4.1:
GLSL 4.1 not started
GL_ARB_ES2_compatibility not started
GL_ARB_get_program_binary not started
GL_ARB_separate_shader_objects some infrastructure done
GL_ARB_shader_precision not started
GL_ARB_vertex_attrib_64bit not started
GL_ARB_viewport_array not started

View File

@@ -83,7 +83,7 @@ Additions to the EGL 1.4 Specification:
EGLImageKHR eglCreateDRMImageMESA(EGLDisplay dpy,
const EGLint *attrib_list);
In the attribute list, pass EGL_WIDTH, EGL_HEIGHT and format and
In the attribute list, pass EGL_WIDTH, EGL_EIGHT and format and
use in the attrib list using EGL_DRM_BUFFER_FORMAT_MESA and
EGL_DRM_BUFFER_USE_MESA. The only format specified by this
extension is EGL_DRM_BUFFER_FORMAT_ARGB32_MESA, where each pixel

View File

@@ -21,52 +21,76 @@ When a new release is coming, release candidates (betas) may be found
<p>
The Mesa package is named MesaLib-x.y.z.{tar.bz2, tar.gz, zip} where x.y.z
is the version. There are three types of compressed archives.
Mesa is distributed in several parts:
</p>
<ul>
<li><b>MesaLib-x.y.z</b> - the main Mesa library source code, drivers
and documentation.
</li>
<li><b>MesaDemos-x.y.z</b> - OpenGL demonstration and test programs.
Most of the programs require GLUT (either the
<a href="http://www.opengl.org/resources/libraries/glut"
target="_parent">original GLUT by Mark Kilgard</a> or
<a href="http://freeglut.sourceforge.net" target="_parent">freeglut</a> or
<a href="http://openglut.sourceforge.net" target="_parent">OpenGLUT</a>).
</li>
<li><b>MesaGLUT-x.y.z</b> - Mark Kilgard's GLUT, easily compiled and used
with Mesa. Plus, other implementation of GLUT for DOS, OS/2, BeOS, etc.
</li>
</ul>
<p>
There's also the MesaGLUT-x.y.z.{tar.bz2, tar.gz, zip} packages which
contain Mark Kilgard's GLUT library.
This is optional.
Most Linux distributions include an implementation of GLUT (such as freeglut).
</p>
<p>
In the past, the Mesa demos collection was distributed as
MesaDemos-x.y.z.{tar.bz2, tar.gz, zip}.
Now, the
<a href="ftp://ftp.freedesktop.org/pub/mesa/demos/" target="_parent">
Mesa demos</a> are distributed separately.
If you're not interested in running the demos, you'll only need the first
package.
</p>
<p>
If you're new to this and not sure what you're doing, grab all three packages.
</p>
<p>
The packages are available in .tar.gz, .tar.bz2 and .zip formats.
Other sites might offer additional package formats.
</p>
<H1>Unpacking</H1>
<p>
To unpack .tar.gz files:
All the packages should be in the same directory prior to unpacking.
</p>
<ul>
<li>To unpack .tar.gz files:
<pre>
tar zxf MesaLib-x.y.z.tar.gz
tar zxf MesaLib-X.Y.tar.gz
tar zxf MesaDemos-X.Y.tar.gz
tar zxf MesaGLUT-X.Y.tar.gz
</pre>
or
<pre>
gzcat MesaLib-x.y.z.tar.gz | tar xf -
gzcat MesaLib-X.Y.tar.gz | tar xf -
gzcat MesaDemos-X.Y.tar.gz | tar xf -
gzcat MesaGLUT-X.Y.tar.gz | tar xf -
</pre>
or
<pre>
gunzip MesaLib-x.y.z.tar.gz ; tar xf MesaLib-x.y.z.tar
gunzip MesaLib-X.Y.tar.gz ; tar xf MesaLib-X.Y.tar
gunzip MesaDemos-X.Y.tar.gz ; tar xf MesaDemos-X.Y.tar
gunzip MesaGLUT-X.Y.tar.gz ; tar xf MesaGLUT-X.Y.tar
</pre>
<p>
To unpack .tar.bz2 files:
</p>
<li>To unpack .tar.bz2 files:
<pre>
bunzip2 -c MesaLib-x.y.z.tar.gz | tar xf -
bunzip2 -c MesaLib-X.Y.tar.gz | tar xf -
bunzip2 -c MesaDemos-X.Y.tar.gz | tar xf -
bunzip2 -c MesaGLUT-X.Y.tar.gz | tar xf -
</pre>
<p>
To unpack .zip files:
</p>
<li>To unpack .zip files:
<pre>
unzip MesaLib-x.y.z.zip
unzip MesaLib-X.Y.zip
unzip MesaDemos-X.Y.zip
unzip MesaGLUT-X.Y.zip
</pre>
</ul>
<h1>Contents</h1>
@@ -82,13 +106,22 @@ bin/ - shell scripts for making shared libraries, etc
docs/ - documentation
src/ - source code for libraries
src/mesa - sources for the main Mesa library and device drivers
src/gallium - sources for Gallium and Gallium drivers
src/glu - libGLU source code
src/glx - sources for building libGL with full GLX and DRI support
src/glw - Xt/Motif/OpenGL widget code
</pre>
If you downloaded and unpacked the MesaGLUT.x.y.z package:
If you downloaded and unpacked the MesaDemos.X.Y package:
<pre>
progs/demos - original Mesa demos
progs/xdemos - GLX OpenGL/Mesa demos
progs/redbook - examples from the OpenGL Programming Guide
progs/samples - examples from SGI
progs/images/ - image files
</pre>
If you downloaded and unpacked the MesaGLUT.X.Y package:
<pre>
src/glut - GLUT library source code
</pre>

View File

@@ -19,23 +19,27 @@ API entry points and helper functions for use by the drivers. Drivers are
dynamically loaded by the main library and most of the EGL API calls are
directly dispatched to the drivers.</p>
<p>The driver in use decides the window system to support.</p>
<p>The driver in use decides the window system to support. For drivers that
support hardware rendering, there are usually multiple drivers supporting the
same window system. Each one of of them supports a certain range of graphics
cards.</p>
<h2>Build EGL</h2>
<ol>
<li>
<p>Run <code>configure</code> with the desired client APIs and enable
the driver for your hardware. For example</p>
<p>Run <code>configure</code> with the desired state trackers and enable
the Gallium driver for your hardware. For example</p>
<pre>
$ ./configure --enable-gles2 --enable-openvg --enable-gallium-nouveau
$ ./configure --enable-gles-overlay --with-state-trackers=egl,vega --enable-gallium-intel
</pre>
<p>The main library and OpenGL is enabled by default. The first option above
enables <a href="opengles.html">OpenGL ES 2.x</a>. The second option enables
<a href="openvg.html">OpenVG</a>.</p>
<p>The main library and OpenGL is enabled by default. The first option enables
<a href="opengles.html">OpenGL ES 1.x and 2.x</a>. The <code>egl</code> state
tracker is needed by a number of EGL drivers. EGL drivers will be covered
later. The <a href="openvg.html">vega state tracker</a> provides OpenVG
1.x.</p>
</li>
<li>Build and install Mesa as usual.</li>
@@ -76,38 +80,31 @@ types such as <code>EGLNativeDisplayType</code> or
<p>The available platforms are <code>x11</code>, <code>drm</code>,
<code>fbdev</code>, and <code>gdi</code>. The <code>gdi</code> platform can
only be built with SCons. Unless for special needs, the build system should
select the right platforms automatically.</p>
only be built with SCons.</p>
</li>
<li><code>--enable-gles1</code> and <code>--enable-gles2</code>
<li><code>--with-state-trackers</code>
<p>These options enable OpenGL ES support in OpenGL. The result is one big
internal library that supports multiple APIs.</p>
<p>The argument is a comma separated string. It is usually used to specify the
rendering APIs, such as OpenVG, to build. But it is also used to specify
<code>egl</code> state tracker that <code>egl_gallium</code> depends on.</p>
</li>
<li><code>--enable-gles-overlay</code>
<p>This option enables OpenGL ES as separate internal libraries. This is an
alternative approach to enable OpenGL ES.</p>
<p>This is only supported by <code>egl_gallium</code>. For systems using DRI
drivers, <code>--enable-gles1</code> and <code>--enable-gles2</code> are
suggested instead as all drivers will benefit.</p>
<p>OpenGL and OpenGL ES are not controlled by
<code>--with-state-trackers</code>. OpenGL is always built. To build OpenGL
ES, this option must be explicitly given.</p>
</li>
<li><code>--enable-openvg</code>
<li><code>--enable-gles1</code> and <code>--enable-gles2</code>
<p>OpenVG must be explicitly enabled by this option.</p>
</li>
<li><code>--enable-gallium-egl</code>
<p>Explicitly enable or disable <code>egl_gallium</code>.</p>
<p>Unlike <code>--enable-gles-overlay</code>, which builds one library for each
rendering API, these options enable OpenGL ES support in OpenGL. The result is
one big library that supports multiple APIs.</p>
</li>
@@ -134,16 +131,6 @@ colon-separated directories where the main library will look for drivers, in
addition to the default directory. This variable is ignored for setuid/setgid
binaries.</p>
<p>This variable is usually set to test an uninstalled build. For example, one
may set</p>
<pre>
$ export LD_LIBRARY_PATH=$mesa/lib
$ export EGL_DRIVERS_PATH=$mesa/lib/egl
</pre>
<p>to test a build without installation</p>
</li>
<li><code>EGL_DRIVER</code>
@@ -152,6 +139,10 @@ may set</p>
specified EGL driver to be loaded. It comes in handy when one wants to test a
specific driver. This variable is ignored for setuid/setgid binaries.</p>
<p><code>egl_gallium</code> dynamically loads hardware drivers and client API
modules found in <code>EGL_DRIVERS_PATH</code>. Thus, specifying this variable
alone is not sufficient for <code>egl_gallium</code> for uninstalled build.</p>
</li>
<li><code>EGL_PLATFORM</code>
@@ -159,12 +150,7 @@ specific driver. This variable is ignored for setuid/setgid binaries.</p>
<p>This variable specifies the native platform. The valid values are the same
as those for <code>--with-egl-platforms</code>. When the variable is not set,
the main library uses the first platform listed in
<code>--with-egl-platforms</code> as the native platform.</p>
<p>Extensions like <code>EGL_MESA_drm_display</code> define new functions to
create displays for non-native platforms. These extensions are usually used by
applications that support non-native platforms. Setting this variable is
probably required only for some of the demos found in mesa/demo repository.</p>
<code>--with-egl-platforms</code> as the native platform</p>
</li>
@@ -187,25 +173,11 @@ variable to true forces the use of software rendering.</p>
<h2>EGL Drivers</h2>
<ul>
<li><code>egl_dri2</code>
<p>This driver supports both <code>x11</code> and <code>drm</code> platforms.
It functions as a DRI driver loader. For <code>x11</code> support, it talks to
the X server directly using (XCB-)DRI2 protocol.</p>
<p>This driver can share DRI drivers with <code>libGL</code>.</p>
</li>
<li><code>egl_gallium</code>
<p>This driver is based on Gallium3D. It supports all rendering APIs and
hardwares supported by Gallium3D. It is the only driver that supports OpenVG.
The supported platforms are X11, DRM, FBDEV, and GDI.</p>
<p>This driver comes with its own hardware drivers
(<code>pipe_&lt;hw&gt;</code>) and client API modules
(<code>st_&lt;api&gt;</code>).</p>
The supported platforms are X11, KMS, FBDEV, and GDI.</p>
</li>
@@ -215,24 +187,26 @@ The supported platforms are X11, DRM, FBDEV, and GDI.</p>
the EGL API. It supports both direct and indirect rendering when the GLX does.
It is accelerated when the GLX is. As such, it cannot provide functions that
is not available in GLX or GLX extensions.</p>
</li>
<li><code>egl_dri2</code>
<p>This driver supports the X Window System as its window system. It functions
as a DRI2 driver loader. Unlike <code>egl_glx</code>, it has no dependency on
<code>libGL</code>. It talks to the X server directly using DRI2 protocol.</p>
</li>
<li><code>egl_dri</code>
<p>This driver lacks maintenance and does <em>not</em> build. It is similiar
to <code>egl_dri2</code> in that it functions as a DRI(1) driver loader. But
unlike <code>egl_dri2</code>, it supports Linux framebuffer devices as its
window system and supports EGL_MESA_screen_surface extension. As DRI1 drivers
are phasing out, it might eventually be replaced by <code>egl_dri2</code>.</p>
</li>
</ul>
<h2>Packaging</h2>
<p>The ABI between the main library and its drivers are not stable. Nor is
there a plan to stabilize it at the moment. Of the EGL drivers,
<code>egl_gallium</code> has its own hardware drivers and client API modules.
They are considered internal to <code>egl_gallium</code> and there is also no
stable ABI between them. These should be kept in mind when packaging for
distribution.</p>
<p>Generally, <code>egl_dri2</code> is preferred over <code>egl_gallium</code>
when the system already has DRI drivers. As <code>egl_gallium</code> is loaded
before <code>egl_dri2</code> when both are available, <code>egl_gallium</code>
may either be disabled with <code>--disable-gallium-egl</code> or packaged
separately.</p>
<h2>Developers</h2>
<p>The sources of the main library and the classic drivers can be found at
@@ -321,6 +295,7 @@ should as well lock the display before using it.
<ul>
<li>Pass the conformance tests</li>
<li>Reference counting in main library?</li>
<li>Mixed use of OpenGL, OpenGL ES 1.1, and OpenGL ES 2.0 is supported. But
which one of <code>libGL.so</code>, <code>libGLESv1_CM.so</code>, and
<code>libGLESv2.so</code> should an application link to? Bad things may happen

View File

@@ -12,16 +12,16 @@
<ol>
<li><a href="#unix-x11">Unix / X11</a>
<ul>
<li><a href="#prereq-general">General prerequisites for building</a>
<li><a href="#prereq-dri">Prerequisites for DRI and hardware acceleration</a>
<li><a href="#prereq">Prerequisites for DRI and hardware acceleration</a>
<li><a href="#autoconf">Building with autoconf</a>
<li><a href="#traditional">Building with traditional Makefiles</a>
<li><a href="#libs">The Libraries</a>
<li><a href="#demos">Running the demos
<li><a href="#install">Installing the header and library files
<li><a href="#pkg-config">Building OpenGL programs with pkg-config
</ul>
<li><a href="#windows">Windows</a>
<li><a href="#scons">Building with SCons</a>
<li><a href="#scons">SCons</a>
<li><a href="#other">Other</a>
</ol>
<br>
@@ -31,22 +31,8 @@
<H2>1. Unix/X11 Compilation and Installation</H1>
<a name="prereq-general">
<h3>1.1 General prerequisites for building</h3>
<ul>
<li>lex / yacc - for building the GLSL compiler.
On Linux systems, flex and bison are used.
Versions 2.5.35 and 2.4.1, respectively, (or later) should work.
</li>
<li>python - Python is needed for building the Gallium components.
Version 2.6.4 or later should work.
</li>
</ul>
<a name="prereq-dri">
<h3>1.2 Prerequisites for DRI and hardware acceleration</h3>
<a name="prereq">
<h3>1.1 Prerequisites for DRI and hardware acceleration</h3>
<p>
The following are required for DRI-based hardware acceleration with Mesa:
@@ -63,7 +49,7 @@ version 2.4.15 or later
<a name="autoconf">
<h3>1.3 Building with Autoconf</h3>
<h3>1.2 Building with Autoconf</h3>
<p>
Mesa may be <a href="autoconf.html">built using autoconf</a>.
@@ -73,7 +59,7 @@ If that fails the traditional Mesa build system is available.
<a name="traditional">
<h3>1.4 Building with traditional Makefiles</h3>
<h3>1.3 Building with traditional Makefiles</h3>
<p>
The traditional Mesa build system is based on a collection of pre-defined
@@ -140,7 +126,7 @@ Later, if you want to rebuild for a different configuration run
<a name="libs">
<h3>1.5 The libraries</h3>
<h3>1.4 The libraries</h3>
<p>
When compilation has finished, look in the top-level <code>lib/</code>
@@ -199,11 +185,81 @@ If you built the DRI hardware drivers, you'll also see the DRI drivers:
-rwxr-xr-x 1 brian users 10997120 Jul 21 12:13 unichrome_dri.so
</pre>
<a name="demos">
<h3>1.5 Running the demos</h3>
<p>
If you built with Gallium support, look in lib/gallium/ for Gallium-based
versions of libGL and device drivers.
If you downloaded/unpacked the MesaDemos-x.y.z.tar.gz archive or
obtained Mesa from CVS, the <b>progs/</b> directory will contain a
bunch of demonstration programs.
</p>
<p>
Before running a demo, you'll probably have to set two environment variables
to indicate where the libraries are located. For example:
<p>
<blockquote>
<b>cd lib/</b>
<br>
<b>export LD_LIBRARY_PATH=${PWD}</b>
<br>
<b>export LIBGL_DRIVERS_PATH=${PWD}</b> (if using DRI drivers)
</blockquote>
<p>
Next, change to the Mesa/demos/ directory:
</p>
<blockquote>
<b>cd ../progs/demos</b>
</blockquote>
<p>
Run a demo such as gears:
</p>
<blockquote>
<b>./gears</b>
</blockquote>
<p>
If this doesn't work, try the <b>Mesa/progs/xdemos/glxinfo</b> program
and see that it prints the expected Mesa version number.
</p>
<p>
If you're using Linux or a similar OS, verify that the demo program is
being linked with the proper library files:
</p>
<blockquote>
<b>ldd gears</b>
</blockquote>
<p>
You should see something like this:
</p>
<pre>
libglut.so.3 => /home/brian/Mesa/lib/libglut.so.3 (0x40013000)
libGLU.so.1 => /home/brian/Mesa/lib/libGLU.so.1 (0x40051000)
libGL.so.1 => /home/brian/Mesa/lib/libGL.so.1 (0x400e0000)
libc.so.6 => /lib/i686/libc.so.6 (0x42000000)
libm.so.6 => /lib/i686/libm.so.6 (0x403da000)
libX11.so.6 => /usr/X11R6/lib/libX11.so.6 (0x403fc000)
libXmu.so.6 => /usr/X11R6/lib/libXmu.so.6 (0x404da000)
libXt.so.6 => /usr/X11R6/lib/libXt.so.6 (0x404f1000)
libXi.so.6 => /usr/X11R6/lib/libXi.so.6 (0x40543000)
libstdc++.so.5 => /usr/lib/libstdc++.so.5 (0x4054b000)
libgcc_s.so.1 => /lib/libgcc_s.so.1 (0x405fd000)
libXext.so.6 => /usr/X11R6/lib/libXext.so.6 (0x40605000)
libpthread.so.0 => /lib/i686/libpthread.so.0 (0x40613000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
libdl.so.2 => /lib/libdl.so.2 (0x40644000)
libSM.so.6 => /usr/X11R6/lib/libSM.so.6 (0x40647000)
libICE.so.6 => /usr/X11R6/lib/libICE.so.6 (0x40650000)
</pre>
<p>
Retrace your steps if this doesn't look right.
</p>
<a name="install">

View File

@@ -11,39 +11,6 @@
<H1>News</H1>
<h2>April 6, 2011</h2>
<p>
<a href="relnotes-7.10.2.html">Mesa 7.10.2</a> is released. This is a bug
fix release release.
</p>
<h2>March 2, 2011</h2>
<p>
<a href="relnotes-7.10.1.html">Mesa 7.10.1</a> is released. This is a bug
fix release release.
</p>
<p>
Also, <a href="relnotes-7.9.2.html">Mesa 7.9.2</a> is released.
This is a bug fix release.
</p>
<h2>January 7, 2011</h2>
<p>
<a href="relnotes-7.10.html">Mesa 7.10</a> (final) is released. This is a new
development release.
</p>
<p>
Also, <a href="relnotes-7.9.1.html">Mesa 7.9.1</a> (final) is released.
This is a bug fix release.
</p>
<h2>October 4, 2010</h2>
<p>

View File

@@ -26,27 +26,36 @@ Please refer to <a href="egl.html">Mesa EGL</a> for more information about EGL.
<h2>Building the library</h2>
<ol>
<li>Run <code>configure</code> with <code>--enable-openvg</code>. If you do
not need OpenGL, you can add <code>--disable-opengl</code> to save the
compilation time.</li>
<li>Build and install Mesa as usual.</li>
<li>Build Mesa3D with Gallium3D. Any build that builds Gallium3D libraries, EGL, and Gallium EGL drivers will suffice</li>
<li>cd src/gallium/state_trackers/vega; make</li>
<li>The last step will build libOpenVG library. You can add the libdir to LD_LIBRARY_PATH or install libOpenVG</li>
</ol>
<h3>Sample build</h3>
A sample build looks as follows:
<pre>
$ ./configure --disable-opengl --enable-openvg
$ ./configure --with-state-trackers=egl,vega --enable-gallium-intel
$ make
$ make install
</pre>
<p>It will install <code>libOpenVG.so</code>, <code>libEGL.so</code>, and one
or more EGL drivers.</p>
<h2>OpenVG Demos</h2>
<p>OpenVG demos can be found in mesa/demos repository.</p>
<p>
To build the OpenVG demos:
</p>
<pre>
cd progs/openvg
make
</pre>
<p>
To run a demo:
</p>
<pre>
cd openvg/demos
./lion
</pre>
</body>
</html>

View File

@@ -1,380 +0,0 @@
<HTML>
<head>
<TITLE>Mesa Release Notes</TITLE>
<link rel="stylesheet" type="text/css" href="mesa.css">
<meta http-equiv="content-type" content="text/html; charset=utf-8" />
</head>
<BODY>
<body bgcolor="#eeeeee">
<H1>Mesa 7.10.1 Release Notes / March 2, 2011</H1>
<p>
Mesa 7.10.1 is a bug fix release which fixes bugs found since the 7.10 release.
</p>
<p>
Mesa 7.10.1 implements the OpenGL 2.1 API, but the version reported by
glGetString(GL_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 2.1.
</p>
<p>
See the <a href="install.html">Compiling/Installing page</a> for prerequisites
for DRI hardware acceleration.
</p>
<h2>MD5 checksums</h2>
<pre>
4b4cee19f3bf16eb78bd4cc278ccf812 MesaLib-7.10.1.tar.gz
efe8da4d80c2a5d32a800770b8ce5dfa MesaLib-7.10.1.tar.bz2
0fd2b1a025934de3f8cecf9fb9b57f4c MesaLib-7.10.1.zip
42beb0f5188d544476c19496f725fa67 MesaGLUT-7.10.1.tar.gz
637bb8a20fdad89f7382b4ea83f896e3 MesaGLUT-7.10.1.tar.bz2
bdbf3ffb2606d6aa8afabb6c6243b91b MesaGLUT-7.10.1.zip
</pre>
<h2>New features</h2>
<p>None.</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li>Fix an off-by-one bug in a vsplit assertion.</li>
<li>Fix incorrect handling of <tt>layout</tt> qualifier
with <tt>in</tt>, <tt>out</tt>, <tt>attribute</tt>, and <tt>varying</tt>.</li>
<li>Fix an i965 shader bug where the negative absolute value was generated instead of the absolute value of a negation.</li>
<li>Fix numerous issues handling precision qualifiers in GLSL ES.</li>
<li>Fixed a few GLX protocol encoder bugs (Julien Cristau)</li>
<li>Assorted Gallium llvmpipe driver bug fixes</li>
<li>Assorted Mesa/Gallium state tracker bug fixes</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=26795">Bug 26795</a> - gl_FragCoord off by one in Gallium drivers.</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=29164">Bug 29164</a> - [GLSL 1.20] invariant variable shouldn't be used before declaration</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=29823">Bug 29823</a> - GetUniform[if]v busted</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=29927">Bug 29927</a> - [glsl2] fail to compile shader with constructor for array of struct type</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=30156">Bug 30156</a> - [i965] After updating to Mesa 7.9, Civilization IV starts to show garbage</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=31923">Bug 31923</a> - [GLSL 1.20] allowing inconsistent centroid declaration between two vertex shaders</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=31925">Bug 31925</a> - [GLSL 1.20] "#pragma STDGL invariant(all)" fail</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=32214">Bug 32214</a> - [gles2]no link error happens when missing vertex shader or frag shader</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=32375">Bug 32375</a> - [gl gles2] Not able to get the attribute by function glGetVertexAttribfv</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=32541">Bug 32541</a> - Segmentation Fault while running an HDR (high dynamic range) rendering demo</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=32569">Bug 32569</a> - [gles2] glGetShaderPrecisionFormat not implemented yet</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=32695">Bug 32695</a> - [glsl] SIGSEGV glcpp/glcpp-parse.y:833</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=32831">Bug 32831</a> - [glsl] division by zero crashes GLSL compiler</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=32910">Bug 32910</a> - Keywords 'in' and 'out' not handled properly for GLSL 1.20 shaders</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=33219">Bug 33219</a> -[GLSL bisected] implicit sized array triggers segfault in ir_to_mesa_visitor::copy_propagate</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=33306">Bug 33306</a> - GLSL integer division by zero crashes GLSL compiler</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=33308">Bug 33308</a> -[glsl] ast_to_hir.cpp:3016: virtual ir_rvalue* ast_jump_statement::hir(exec_list*, _mesa_glsl_parse_state*): Assertion `ret != __null' failed.</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=33316">Bug 33316</a> - uniform array will be allocate one line more and initialize it when it was freed will abort</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=33386">Bug 33386</a> - Dubious assembler in read_rgba_span_x86.S</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=33388">Bug 33388</a> - Dubious assembler in xform4.S</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=33433">Bug 33433</a> - Error in x86-64 API dispatch code.</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=33507">Bug 33507</a> - [glsl] GLSL preprocessor modulus by zero crash</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=33508">Bug 33508</a> - [glsl] GLSL compiler modulus by zero crash</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=33916">Bug 33916</a> - Compiler accepts reserved operators % and %=</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=34030">Bug 34030</a> - [bisected] Starcraft 2: some effects are corrupted or too big</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=34047">Bug 34047</a> - Assert in _tnl_import_array() when using GLfixed vertex datatypes with GLESv2</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=34114">Bug 34114</a> - Sun Studio build fails due to standard library functions not being in global namespace</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=34179">Bug 34179</a> - Nouveau 3D driver: nv50_pc_emit.c:863 assertion error kills Compiz</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=34198">Bug 34198</a> - [GLSL] implicit sized array with index 0 used gets assertion</li>
<li><a href="https://bugs.launchpad.net/ubuntu/+source/mesa/+bug/691653">Ubuntu bug 691653</a> - compiz crashes when using alt-tab (the radeon driver kills it) </li>
<li><a href="https://bugs.meego.com/show_bug.cgi?id=13005">Meego bug 13005</a> - Graphics GLSL issue lead to camera preview fail on Pinetrail</li>
<!-- <li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=">Bug </a> - </li> -->
</ul>
<h2>Changes</h2>
<p>The full set of changes can be viewed by using the following GIT command:</p>
<pre>
git log mesa-7.10..mesa-7.10.1
</pre>
<p>Alberto Milone (1):
<ul>
<li>r600c: add evergreen ARL support.</li>
</ul></p>
<p>Brian Paul (21):
<ul>
<li>draw: Fix an off-by-one bug in a vsplit assertion.</li>
<li>docs: add links to 7.9.1 and 7.10 release notes</li>
<li>docs: added news item for 7.9.1 and 7.10 release</li>
<li>gallivm: work around LLVM 2.6 bug when calling C functions</li>
<li>gallivm: fix copy&amp;paste error from previous commit</li>
<li>mesa: fix a few format table mistakes, assertions</li>
<li>mesa: fix num_draw_buffers==0 in fixed-function fragment program generation</li>
<li>mesa: don't assert in GetIntegerIndexed, etc</li>
<li>mesa: check for dummy renderbuffer in _mesa_FramebufferRenderbufferEXT()</li>
<li>llvmpipe: make sure binning is active when we begin/end a query</li>
<li>st/mesa: fix incorrect fragcoord.x translation</li>
<li>softpipe: fix off-by-one error in setup_fragcoord_coeff()</li>
<li>cso: fix loop bound in cso_set_vertex_samplers()</li>
<li>st/mesa: fix incorrect glCopyPixels position on fallback path</li>
<li>st/mesa: set renderbuffer _BaseFormat in a few places</li>
<li>st/mesa: fix the default case in st_format_datatype()</li>
<li>st/mesa: need to translate clear color according to surface's base format</li>
<li>docs: update 7.9.2 release notes with Brian's cherry-picks</li>
<li>docs: add link to 7.10.1 release notes</li>
<li>mesa: implement glGetShaderPrecisionFormat()</li>
<li>docs: updated environment variable list</li>
</ul></p>
<p>Bryce Harrington (1):
<ul>
<li>r300g: Null pointer check for buffer deref in gallium winsys</li>
</ul></p>
<p>Chad Versace (20):
<ul>
<li>glsl: At link-time, check that globals have matching centroid qualifiers</li>
<li>glcpp: Fix segfault when validating macro redefinitions</li>
<li>glsl: Fix parser rule for type_specifier</li>
<li>glsl: Change default value of ast_type_specifier::precision</li>
<li>glsl: Add semantic checks for precision qualifiers</li>
<li>glsl: Add support for default precision statements</li>
<li>glsl: Remove redundant semantic check in parser</li>
<li>glsl: Fix semantic checks on precision qualifiers</li>
<li>glsl: Fix segfault due to missing printf argument</li>
<li>glsl: Mark 'in' variables at global scope as read-only</li>
<li>mesa: Refactor handling of extension strings</li>
<li>mesa: Add/remove extensions in extension string</li>
<li>mesa: Change dependencies of some OES extension strings</li>
<li>mesa: Change OES_point_sprite to depend on ARB_point_sprite</li>
<li>mesa: Change OES_standard_derivatives to be stand-alone extension</li>
<li>i915: Disable extension OES_standard_derivatives</li>
<li>glcpp: Raise error when modulus is zero</li>
<li>glsl: Set operators '%' and '%=' to be reserved when GLSL &lt 1.30</li>
<li>glsl: Reinstate constant-folding for division by zero</li>
<li>tnl: Add support for datatype GL_FIXED in vertex arrays</li>
</ul></p>
<p>Chia-I Wu (1):
<ul>
<li>mesa: Add glDepthRangef and glClearDepthf to APIspec.xml.</li>
</ul></p>
<p>Christoph Bumiller (1):
<ul>
<li>nv50,nvc0: do not forget to apply sign mode to saved TGSI inputs</li>
</ul></p>
<p>Cyril Brulebois (1):
<ul>
<li>Point to bugs.freedesktop.org rather than bugzilla.freedesktop.org</li>
</ul></p>
<p>Dave Airlie (3):
<ul>
<li>radeon/r200: fix fbo-clearmipmap + gen-teximage</li>
<li>radeon: calculate complete texture state inside TFP function</li>
<li>radeon: avoid segfault on 3D textures.</li>
</ul></p>
<p>Dimitry Andric (4):
<ul>
<li>mesa: s/movzx/movzbl/</li>
<li>mesa: s/movzxw/movzwl/ in read_rgba_span_x86.S</li>
<li>glapi: adding @ char before type specifier in glapi_x86.S</li>
<li>glapi: add @GOTPCREL relocation type</li>
</ul></p>
<p>Eric Anholt (16):
<ul>
<li>glsl: Fix the lowering of variable array indexing to not lose write_masks.</li>
<li>i965/fs: When producing ir_unop_abs of an operand, strip negate.</li>
<li>i965/vs: When MOVing to produce ABS, strip negate of the operand.</li>
<li>i965/fs: Do flat shading when appropriate.</li>
<li>i965: Avoid double-negation of immediate values in the VS.</li>
<li>intel: Make renderbuffer tiling choice match texture tiling choice.</li>
<li>i965: Fix dead pointers to fp-&gt;Parameters-&gt;ParameterValues[] after realloc.</li>
<li>docs: Add a relnote for the Civ IV on i965.</li>
<li>glapi: Add entrypoints and enums for GL_ARB_ES2_compatibility.</li>
<li>mesa: Add extension enable bit for GL_ARB_ES2_compatibility.</li>
<li>mesa: Add actual support for glReleaseShaderCompiler from ES2.</li>
<li>mesa: Add support for glDepthRangef and glClearDepthf.</li>
<li>mesa: Add getters for ARB_ES2_compatibility MAX_*_VECTORS.</li>
<li>mesa: Add getter for GL_SHADER_COMPILER with ARB_ES2_compatibility.</li>
<li>i965: Fix a bug in i965 compute-to-MRF.</li>
<li>i965/fs: Add a helper function for detecting math opcodes.</li>
</ul></p>
<p>Fredrik Höglund (1):
<ul>
<li>st/mesa: fix a regression from cae2bb76</li>
</ul></p>
<p>Ian Romanick (42):
<ul>
<li>docs: Add 7.10 md5sums</li>
<li>glsl: Support the 'invariant(all)' pragma</li>
<li>glcpp: Generate an error for division by zero</li>
<li>glsl: Add version_string containing properly formatted GLSL version</li>
<li>glsl &amp; glcpp: Refresh autogenerated lexer and parser files.</li>
<li>glsl: Disallow 'in' and 'out' on globals in GLSL 1.20</li>
<li>glsl: Track variable usage, use that to enforce semantics</li>
<li>glsl: Allow 'in' and 'out' when 'layout' is also available</li>
<li>docs: Initial bits of 7.10.1 release notes</li>
<li>mesa: bump version to 7.10.1-devel</li>
<li>doc: Update 7.10.1 release notes</li>
<li>glsl: Emit errors or warnings when 'layout' is used with 'attribute' or 'varying'</li>
<li>docs: Update 7.10.1 release notes</li>
<li>glsl: Refresh autogenerated lexer and parser files.</li>
<li>glsl: Don't assert when the value returned by a function has no rvalue</li>
<li>linker: Set sizes for non-global arrays as well</li>
<li>linker: Propagate max_array_access while linking functions</li>
<li>docs: Update 7.10.1 release notes</li>
<li>mesa: glGetUniform only returns a single element of an array</li>
<li>linker: Generate link errors when ES shaders are missing stages</li>
<li>mesa: Fix error checks in GetVertexAttrib functions</li>
<li>Use C-style system headers in C++ code to avoid issues with std:: namespace</li>
<li>docs: Update 7.10.1 release notes</li>
<li>glapi: Regenerate for GL_ARB_ES2_compatibility.</li>
<li>mesa: Connect glGetShaderPrecisionFormat into the dispatch table</li>
<li>i965: Set correct values for range/precision of fragment shader types</li>
<li>i915: Set correct values for range/precision of fragment shader types</li>
<li>intel: Fix typeos from 3d028024 and 790ff232</li>
<li>glsl: Ensure that all GLSL versions are supported in the stand-alone compiler</li>
<li>glsl: Reject shader versions not supported by the implementation</li>
<li>mesa: Initial size for secondary color array is 3</li>
<li>glsl: Finish out the reduce/reduce error fixes</li>
<li>glsl: Regenerate compiler and glcpp files from cherry picks</li>
<li>linker: Fix off-by-one error implicit array sizing</li>
<li>docs: update 7.10.1 release notes with Ian's recent cherry picks</li>
<li>i915: Only mark a register as available if all components are written</li>
<li>i915: Calculate partial result to temp register first</li>
<li>i915: Force lowering of all types of indirect array accesses in the FS</li>
<li>docs: Update 7.10.1 with (hopefully) the last of the cherry picks</li>
<li>docs: Clean up bug fixes list</li>
<li>intel: Remove driver date and related bits from renderer string</li>
<li>mesa: set version string to 7.10.1 (final)</li>
</ul></p>
<p>Jian Zhao (1):
<ul>
<li>mesa: fix an error in uniform arrays in row calculating.</li>
</ul></p>
<p>Julien Cristau (3):
<ul>
<li>glx: fix request lengths</li>
<li>glx: fix GLXChangeDrawableAttributesSGIX request</li>
<li>glx: fix length of GLXGetFBConfigsSGIX</li>
</ul></p>
<p>Keith Packard (1):
<ul>
<li>glsl: Eliminate reduce/reduce conflicts in glsl grammar</li>
</ul></p>
<p>Kenneth Graunke (20):
<ul>
<li>glsl: Expose a public glsl_type::void_type const pointer.</li>
<li>glsl: Don't bother unsetting a destructor that was never set.</li>
<li>glsl, i965: Remove unnecessary talloc includes.</li>
<li>glcpp: Remove use of talloc reference counting.</li>
<li>ralloc: Add a fake implementation of ralloc based on talloc.</li>
<li>Convert everything from the talloc API to the ralloc API.</li>
<li>ralloc: a new MIT-licensed recursive memory allocator.</li>
<li>Remove talloc from the make and automake build systems.</li>
<li>Remove talloc from the SCons build system.</li>
<li>Remove the talloc sources from the Mesa repository.</li>
<li>glsl: Fix use of uninitialized values in _mesa_glsl_parse_state ctor.</li>
<li>i965/fs: Apply source modifier workarounds to POW as well.</li>
<li>i965: Fix shaders that write to gl_PointSize on Sandybridge.</li>
<li>i965/fs: Avoid register coalescing away gen6 MATH workarounds.</li>
<li>i965/fs: Correctly set up gl_FragCoord.w on Sandybridge.</li>
<li>i965: Increase Sandybridge point size clamp.</li>
<li>i965/fs: Refactor control flow stack handling.</li>
<li>i965: Increase Sandybridge point size clamp in the clip state.</li>
<li>glsl: Use reralloc instead of plain realloc.</li>
<li>Revert "i965/fs: Correctly set up gl_FragCoord.w on Sandybridge."</li>
</ul></p>
<p>Marek Olšák (4):
<ul>
<li>docs: fix messed up names with special characters in relnotes-7.10</li>
<li>docs: fix messed up names with special characters in relnotes-7.9.1</li>
<li>mesa: fix texture3D mipmap generation for UNSIGNED_BYTE_3_3_2</li>
<li>st/dri: Track drawable context bindings</li>
</ul></p>
<p>Paulo Zanoni (1):
<ul>
<li>dri_util: fail driCreateNewScreen if InitScreen is NULL</li>
</ul></p>
<p>Sam Hocevar (2):
<ul>
<li>docs: add glsl info</li>
<li>docs: fix glsl_compiler name</li>
</ul></p>
<p>Tom Fogal (1):
<ul>
<li>Regenerate gl_mangle.h.</li>
</ul></p>
<p>Tom Stellard (2):
<ul>
<li>r300/compiler: Disable register rename pass on r500</li>
<li>r300/compiler: Don't erase sources when converting RGB-&gt;Alpha</li>
</ul></p>
<p>Vinson Lee (3):
<ul>
<li>ralloc: Add missing va_end following va_copy.</li>
<li>mesa: Move declaration before code in extensions.c.</li>
<li>mesa: Move loop variable declarations outside for loop in extensions.c.</li>
</ul></p>
<p>nobled (1):
<ul>
<li>glx: Put null check before use</li>
</ul></p>
</p>
</body>
</html>

View File

@@ -1,201 +0,0 @@
<HTML>
<head>
<TITLE>Mesa Release Notes</TITLE>
<link rel="stylesheet" type="text/css" href="mesa.css">
<meta http-equiv="content-type" content="text/html; charset=utf-8" />
</head>
<BODY>
<body bgcolor="#eeeeee">
<H1>Mesa 7.10.2 Release Notes / TBD</H1>
<p>
Mesa 7.10.2 is a bug fix release which fixes bugs found since the 7.10 release.
</p>
<p>
Mesa 7.10.2 implements the OpenGL 2.1 API, but the version reported by
glGetString(GL_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 2.1.
</p>
<p>
See the <a href="install.html">Compiling/Installing page</a> for prerequisites
for DRI hardware acceleration.
</p>
<h2>MD5 checksums</h2>
<pre>
TBD
</pre>
<h2>New features</h2>
<p>None.</p>
<h2>Bug fixes</h2>
<p>This list is likely incomplete.</p>
<ul>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=29172">Bug 29172</a> - Arrandale - Pill Popper Pops Pills</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=31159">Bug 31159</a> - shadow problem in 0ad game</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=32688">Bug 32688</a> - [RADEON:KMS:R300G] some games have a wireframe or outline visible</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=32949">Bug 32949</a> - [glsl wine] Need for Speed renders incorrectly with GLSL enabled</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=34203">Bug 34203</a> - [GLSL] fail to call long chains across shaders</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=34376">Bug 34376</a> - [GLSL] allowing assignment to unsized array
<ul>
<li>The commit message incorrectly
lists <a href="https://bugs.freedesktop.org/show_bug.cgi?id=34367">bug
34367</a>.</li>
</ul>
</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=34370">Bug 34370</a> - [GLSL] "i&lt;5 &amp;&amp; i&lt;4" in for loop fails</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=34374">Bug 34374</a> - [GLSL] fail to redeclare an array using initializer</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=35073">Bug 35073</a> - [GM45] Alpha test is broken when rendering to FBO with no color attachment</li>
<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=35483">Bug 35483</a> - util_blit_pixels_writemask: crash in line 322 of src/gallium/auxiliary/util/u_blit.c</li>
<!-- <li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=">Bug </a> - </li> -->
</ul>
<h2>Changes</h2>
<p>The full set of changes can be viewed by using the following GIT command:</p>
<pre>
git log mesa-7.10.1..mesa-7.10.2
</pre>
<p><em>Note:</em> Reverted commits and the reverts are not included in this list.</p>
<p>Alex Deucher (2):
<ul>
<li>r600c: add new ontario pci ids</li>
<li>r600g: add some additional ontario pci ids</li>
</ul></p>
<p>Benjamin Franzke (1):
<ul>
<li>st/dri: Fix surfaceless gl using contexts with previous bound surfaces</li>
</ul></p>
<p>Brian Paul (9):
<ul>
<li>docs: pull 7.9.2 release notes into 7.10 branch</li>
<li>docs: update news.html with 7.10.1 and 7.9.2 releases</li>
<li>docs: fill in 7.10.1 release data</li>
<li>docs: add, fix release notes links</li>
<li>docs: update info about Mesa packaging/contents</li>
<li>docs: update prerequisites, remove old demo info</li>
<li>mesa: Guard against null pointer deref in fbo validation</li>
<li>st/mesa: Apply LOD bias from correct texture unit</li>
<li>glsl: silence warning in printf() with a cast</li>
</ul></p>
<p>Chad Versace (1):
<ul>
<li>i965: Fix tex_swizzle when depth mode is GL_RED</li>
</ul></p>
<p>Dave Airlie (1):
<ul>
<li>r600: don't close fd on failed load</li>
</ul></p>
<p>Eric Anholt (2):
<ul>
<li>i965: Apply a workaround for the Ironlake "vertex flashing".</li>
<li>i965: Fix alpha testing when there is no color buffer in the FBO.</li>
</ul></p>
<p>Fabian Bieler (1):
<ul>
<li>st/mesa: Apply LOD from texture object</li>
</ul></p>
<p>Henri Verbeet (1):
<ul>
<li>st/mesa: Validate state before doing blits.</li>
</ul></p>
<p>Ian Romanick (13):
<ul>
<li>docs: Add 7.10.1 md5sums</li>
<li>glsl: Refactor AST-to-HIR code handling variable initializers</li>
<li>glsl: Refactor AST-to-HIR code handling variable redeclarations</li>
<li>glsl: Process redeclarations before initializers</li>
<li>glsl: Function signatures cannot have NULL return type</li>
<li>glsl: Add several function / call related validations</li>
<li>linker: Add imported functions to the linked IR</li>
<li>glsl: Use insert_before for lists instead of open coding it</li>
<li>glsl: Only allow unsized array assignment in an initializer</li>
<li>glcpp: Refresh autogenerated lexer files</li>
<li>docs: Initial bits of 7.10.2 release notes</li>
<li>mesa: set version string to 7.10.2</li>
<li>mesa: Remove nonexistant files from _FILES lists</li>
</ul></p>
<p>Jerome Glisse (1):
<ul>
<li>r600g: move user fence into base radeon structure</li>
</ul></p>
<p>José Fonseca (2):
<ul>
<li>mesa: Fix typo glGet*v(GL_TEXTURE_COORD_ARRAY_*).</li>
<li>mesa: More glGet* fixes.</li>
</ul></p>
<p>Kenneth Graunke (4):
<ul>
<li>glcpp: Rework lexer to use a SKIP state rather than REJECT.</li>
<li>glcpp: Remove trailing contexts from #if rules.</li>
<li>i965/fs: Fix linear gl_Color interpolation on pre-gen6 hardware.</li>
<li>glsl: Accept precision qualifiers on sampler types, but only in ES.</li>
</ul></p>
<p>Marek Olšák (15):
<ul>
<li>st/mesa: fix crash when DrawBuffer-&gt;_ColorDrawBuffers[0] is NULL</li>
<li>st/mesa: fail to alloc a renderbuffer if st_choose_renderbuffer_format fails</li>
<li>r300/compiler: fix the saturate modifier when applied to TEX instructions</li>
<li>r300/compiler: fix translating the src negate bits in pair_translate</li>
<li>r300/compiler: Abs doesn't cancel Negate (in the conversion to native swizzles)</li>
<li>r300/compiler: TEX instructions don't support negation on source arguments</li>
<li>r300/compiler: do not set TEX_IGNORE_UNCOVERED on r500</li>
<li>r300/compiler: saturate Z before the shadow comparison</li>
<li>r300/compiler: fix equal and notequal shadow compare functions</li>
<li>r300/compiler: remove unused variables</li>
<li>st/mesa: fix crash when using both user and vbo buffers with the same stride</li>
<li>r300g: fix alpha-test with no colorbuffer</li>
<li>r300g: tell the GLSL compiler to lower the continue opcode</li>
<li>r300/compiler: propagate SaturateMode down to the result of shadow comparison</li>
<li>r300/compiler: apply the texture swizzle to shadow pass and fail values too</li>
</ul></p>
<p>Michel Dänzer (1):
<ul>
<li>Use proper source row stride when getting depth/stencil texels.</li>
</ul></p>
<p>Tom Stellard (4):
<ul>
<li>r300/compiler: Use a 4-bit writemask in pair instructions</li>
<li>prog_optimize: Fix reallocating registers for shaders with loops</li>
<li>r300/compiler: Fix vertex shader MAD instructions with constant swizzles</li>
<li>r300/compiler: Don't try to convert RGB to Alpha in full instructions</li>
</ul></p>
</body>
</html>

File diff suppressed because it is too large Load Diff

View File

@@ -10,7 +10,7 @@
<body bgcolor="#eeeeee">
<H1>Mesa 7.9.2 Release Notes / March 2, 2011</H1>
<H1>Mesa 7.9.2 Release Notes / TBD</H1>
<p>
Mesa 7.9.2 is a bug fix release which fixes bugs found since the 7.9.1 release.
@@ -28,12 +28,7 @@ for DRI hardware acceleration.
<h2>MD5 checksums</h2>
<pre>
eb4ab8c1a03386def3ea34b1358e9cda MesaLib-7.9.2.tar.gz
8f6d1474912787ce13bd35f3bae9938a MesaLib-7.9.2.tar.bz2
427a81dd43ac97603768dc5c6af3df26 MesaLib-7.9.2.zip
aacb8f4db997e346db40c6066942140a MesaGLUT-7.9.2.tar.gz
18abe6cff4fad8ad4752c7b7ab548e5d MesaGLUT-7.9.2.tar.bz2
3189e5732d636c71baf3d8bc23ce7b11 MesaGLUT-7.9.2.zip
TBD
</pre>

View File

@@ -13,8 +13,6 @@ The release notes summarize what's new or changed in each Mesa release.
</p>
<UL>
<LI><A HREF="relnotes-7.10.1.html">7.10.1 release notes</A>
<LI><A HREF="relnotes-7.10.html">7.10 release notes</A>
<LI><A HREF="relnotes-7.9.2.html">7.9.2 release notes</A>
<LI><A HREF="relnotes-7.9.1.html">7.9.1 release notes</A>
<LI><A HREF="relnotes-7.9.html">7.9 release notes</A>

View File

@@ -31,7 +31,6 @@
#define glAccum MANGLE(Accum)
#define glActiveProgramEXT MANGLE(ActiveProgramEXT)
#define glActiveShaderProgram MANGLE(ActiveShaderProgram)
#define glActiveStencilFaceEXT MANGLE(ActiveStencilFaceEXT)
#define glActiveTextureARB MANGLE(ActiveTextureARB)
#define glActiveTexture MANGLE(ActiveTexture)
@@ -57,7 +56,6 @@
#define glBeginOcclusionQueryNV MANGLE(BeginOcclusionQueryNV)
#define glBeginPerfMonitorAMD MANGLE(BeginPerfMonitorAMD)
#define glBeginQueryARB MANGLE(BeginQueryARB)
#define glBeginQueryIndexed MANGLE(BeginQueryIndexed)
#define glBeginQuery MANGLE(BeginQuery)
#define glBeginTransformFeedbackEXT MANGLE(BeginTransformFeedbackEXT)
#define glBeginTransformFeedback MANGLE(BeginTransformFeedback)
@@ -77,27 +75,22 @@
#define glBindBufferRange MANGLE(BindBufferRange)
#define glBindBufferRangeNV MANGLE(BindBufferRangeNV)
#define glBindFragDataLocationEXT MANGLE(BindFragDataLocationEXT)
#define glBindFragDataLocationIndexed MANGLE(BindFragDataLocationIndexed)
#define glBindFragDataLocation MANGLE(BindFragDataLocation)
#define glBindFragmentShaderATI MANGLE(BindFragmentShaderATI)
#define glBindFramebufferEXT MANGLE(BindFramebufferEXT)
#define glBindFramebuffer MANGLE(BindFramebuffer)
#define glBindImageTextureEXT MANGLE(BindImageTextureEXT)
#define glBindLightParameterEXT MANGLE(BindLightParameterEXT)
#define glBindMaterialParameterEXT MANGLE(BindMaterialParameterEXT)
#define glBindMultiTextureEXT MANGLE(BindMultiTextureEXT)
#define glBindParameterEXT MANGLE(BindParameterEXT)
#define glBindProgramARB MANGLE(BindProgramARB)
#define glBindProgramNV MANGLE(BindProgramNV)
#define glBindProgramPipeline MANGLE(BindProgramPipeline)
#define glBindRenderbufferEXT MANGLE(BindRenderbufferEXT)
#define glBindRenderbuffer MANGLE(BindRenderbuffer)
#define glBindSampler MANGLE(BindSampler)
#define glBindTexGenParameterEXT MANGLE(BindTexGenParameterEXT)
#define glBindTextureEXT MANGLE(BindTextureEXT)
#define glBindTexture MANGLE(BindTexture)
#define glBindTextureUnitParameterEXT MANGLE(BindTextureUnitParameterEXT)
#define glBindTransformFeedback MANGLE(BindTransformFeedback)
#define glBindTransformFeedbackNV MANGLE(BindTransformFeedbackNV)
#define glBindVertexArrayAPPLE MANGLE(BindVertexArrayAPPLE)
#define glBindVertexArray MANGLE(BindVertexArray)
@@ -119,22 +112,18 @@
#define glBlendColorEXT MANGLE(BlendColorEXT)
#define glBlendColor MANGLE(BlendColor)
#define glBlendEquationEXT MANGLE(BlendEquationEXT)
#define glBlendEquationiARB MANGLE(BlendEquationiARB)
#define glBlendEquationi MANGLE(BlendEquationi)
#define glBlendEquationIndexedAMD MANGLE(BlendEquationIndexedAMD)
#define glBlendEquation MANGLE(BlendEquation)
#define glBlendEquationSeparateATI MANGLE(BlendEquationSeparateATI)
#define glBlendEquationSeparateEXT MANGLE(BlendEquationSeparateEXT)
#define glBlendEquationSeparateiARB MANGLE(BlendEquationSeparateiARB)
#define glBlendEquationSeparatei MANGLE(BlendEquationSeparatei)
#define glBlendEquationSeparateIndexedAMD MANGLE(BlendEquationSeparateIndexedAMD)
#define glBlendEquationSeparate MANGLE(BlendEquationSeparate)
#define glBlendFunciARB MANGLE(BlendFunciARB)
#define glBlendFunci MANGLE(BlendFunci)
#define glBlendFuncIndexedAMD MANGLE(BlendFuncIndexedAMD)
#define glBlendFunc MANGLE(BlendFunc)
#define glBlendFuncSeparateEXT MANGLE(BlendFuncSeparateEXT)
#define glBlendFuncSeparateiARB MANGLE(BlendFuncSeparateiARB)
#define glBlendFuncSeparatei MANGLE(BlendFuncSeparatei)
#define glBlendFuncSeparateIndexedAMD MANGLE(BlendFuncSeparateIndexedAMD)
#define glBlendFuncSeparateINGR MANGLE(BlendFuncSeparateINGR)
@@ -164,7 +153,6 @@
#define glClearColor MANGLE(ClearColor)
#define glClearDebugLogMESA MANGLE(ClearDebugLogMESA)
#define glClearDepthdNV MANGLE(ClearDepthdNV)
#define glClearDepthf MANGLE(ClearDepthf)
#define glClearDepth MANGLE(ClearDepth)
#define glClearIndex MANGLE(ClearIndex)
#define glClear MANGLE(Clear)
@@ -227,10 +215,6 @@
#define glColorMaskIndexedEXT MANGLE(ColorMaskIndexedEXT)
#define glColorMask MANGLE(ColorMask)
#define glColorMaterial MANGLE(ColorMaterial)
#define glColorP3ui MANGLE(ColorP3ui)
#define glColorP3uiv MANGLE(ColorP3uiv)
#define glColorP4ui MANGLE(ColorP4ui)
#define glColorP4uiv MANGLE(ColorP4uiv)
#define glColorPointerEXT MANGLE(ColorPointerEXT)
#define glColorPointerListIBM MANGLE(ColorPointerListIBM)
#define glColorPointer MANGLE(ColorPointer)
@@ -252,7 +236,6 @@
#define glCombinerParameterivNV MANGLE(CombinerParameterivNV)
#define glCombinerStageParameterfvNV MANGLE(CombinerStageParameterfvNV)
#define glCompileShaderARB MANGLE(CompileShaderARB)
#define glCompileShaderIncludeARB MANGLE(CompileShaderIncludeARB)
#define glCompileShader MANGLE(CompileShader)
#define glCompressedMultiTexImage1DEXT MANGLE(CompressedMultiTexImage1DEXT)
#define glCompressedMultiTexImage2DEXT MANGLE(CompressedMultiTexImage2DEXT)
@@ -327,18 +310,10 @@
#define glCreateShader MANGLE(CreateShader)
#define glCreateShaderObjectARB MANGLE(CreateShaderObjectARB)
#define glCreateShaderProgramEXT MANGLE(CreateShaderProgramEXT)
#define glCreateShaderProgramv MANGLE(CreateShaderProgramv)
#define glCreateSyncFromCLeventARB MANGLE(CreateSyncFromCLeventARB)
#define glCullFace MANGLE(CullFace)
#define glCullParameterdvEXT MANGLE(CullParameterdvEXT)
#define glCullParameterfvEXT MANGLE(CullParameterfvEXT)
#define glCurrentPaletteMatrixARB MANGLE(CurrentPaletteMatrixARB)
#define glDebugMessageCallbackAMD MANGLE(DebugMessageCallbackAMD)
#define glDebugMessageCallbackARB MANGLE(DebugMessageCallbackARB)
#define glDebugMessageControlARB MANGLE(DebugMessageControlARB)
#define glDebugMessageEnableAMD MANGLE(DebugMessageEnableAMD)
#define glDebugMessageInsertAMD MANGLE(DebugMessageInsertAMD)
#define glDebugMessageInsertARB MANGLE(DebugMessageInsertARB)
#define glDeformationMap3dSGIX MANGLE(DeformationMap3dSGIX)
#define glDeformationMap3fSGIX MANGLE(DeformationMap3fSGIX)
#define glDeformSGIX MANGLE(DeformSGIX)
@@ -351,25 +326,20 @@
#define glDeleteFramebuffersEXT MANGLE(DeleteFramebuffersEXT)
#define glDeleteFramebuffers MANGLE(DeleteFramebuffers)
#define glDeleteLists MANGLE(DeleteLists)
#define glDeleteNamedStringARB MANGLE(DeleteNamedStringARB)
#define glDeleteNamesAMD MANGLE(DeleteNamesAMD)
#define glDeleteObjectARB MANGLE(DeleteObjectARB)
#define glDeleteOcclusionQueriesNV MANGLE(DeleteOcclusionQueriesNV)
#define glDeletePerfMonitorsAMD MANGLE(DeletePerfMonitorsAMD)
#define glDeleteProgram MANGLE(DeleteProgram)
#define glDeleteProgramPipelines MANGLE(DeleteProgramPipelines)
#define glDeleteProgramsARB MANGLE(DeleteProgramsARB)
#define glDeleteProgramsNV MANGLE(DeleteProgramsNV)
#define glDeleteQueriesARB MANGLE(DeleteQueriesARB)
#define glDeleteQueries MANGLE(DeleteQueries)
#define glDeleteRenderbuffersEXT MANGLE(DeleteRenderbuffersEXT)
#define glDeleteRenderbuffers MANGLE(DeleteRenderbuffers)
#define glDeleteSamplers MANGLE(DeleteSamplers)
#define glDeleteShader MANGLE(DeleteShader)
#define glDeleteSync MANGLE(DeleteSync)
#define glDeleteTexturesEXT MANGLE(DeleteTexturesEXT)
#define glDeleteTextures MANGLE(DeleteTextures)
#define glDeleteTransformFeedbacks MANGLE(DeleteTransformFeedbacks)
#define glDeleteTransformFeedbacksNV MANGLE(DeleteTransformFeedbacksNV)
#define glDeleteVertexArraysAPPLE MANGLE(DeleteVertexArraysAPPLE)
#define glDeleteVertexArrays MANGLE(DeleteVertexArrays)
@@ -378,10 +348,7 @@
#define glDepthBoundsEXT MANGLE(DepthBoundsEXT)
#define glDepthFunc MANGLE(DepthFunc)
#define glDepthMask MANGLE(DepthMask)
#define glDepthRangeArrayv MANGLE(DepthRangeArrayv)
#define glDepthRangedNV MANGLE(DepthRangedNV)
#define glDepthRangef MANGLE(DepthRangef)
#define glDepthRangeIndexed MANGLE(DepthRangeIndexed)
#define glDepthRange MANGLE(DepthRange)
#define glDetachObjectARB MANGLE(DetachObjectARB)
#define glDetachShader MANGLE(DetachShader)
@@ -396,7 +363,6 @@
#define glDisableVertexAttribArrayARB MANGLE(DisableVertexAttribArrayARB)
#define glDisableVertexAttribArray MANGLE(DisableVertexAttribArray)
#define glDrawArraysEXT MANGLE(DrawArraysEXT)
#define glDrawArraysIndirect MANGLE(DrawArraysIndirect)
#define glDrawArraysInstancedARB MANGLE(DrawArraysInstancedARB)
#define glDrawArraysInstancedEXT MANGLE(DrawArraysInstancedEXT)
#define glDrawArraysInstanced MANGLE(DrawArraysInstanced)
@@ -408,7 +374,6 @@
#define glDrawElementArrayAPPLE MANGLE(DrawElementArrayAPPLE)
#define glDrawElementArrayATI MANGLE(DrawElementArrayATI)
#define glDrawElementsBaseVertex MANGLE(DrawElementsBaseVertex)
#define glDrawElementsIndirect MANGLE(DrawElementsIndirect)
#define glDrawElementsInstancedARB MANGLE(DrawElementsInstancedARB)
#define glDrawElementsInstancedBaseVertex MANGLE(DrawElementsInstancedBaseVertex)
#define glDrawElementsInstancedEXT MANGLE(DrawElementsInstancedEXT)
@@ -421,9 +386,7 @@
#define glDrawRangeElementsBaseVertex MANGLE(DrawRangeElementsBaseVertex)
#define glDrawRangeElementsEXT MANGLE(DrawRangeElementsEXT)
#define glDrawRangeElements MANGLE(DrawRangeElements)
#define glDrawTransformFeedback MANGLE(DrawTransformFeedback)
#define glDrawTransformFeedbackNV MANGLE(DrawTransformFeedbackNV)
#define glDrawTransformFeedbackStream MANGLE(DrawTransformFeedbackStream)
#define glEdgeFlagFormatNV MANGLE(EdgeFlagFormatNV)
#define glEdgeFlag MANGLE(EdgeFlag)
#define glEdgeFlagPointerEXT MANGLE(EdgeFlagPointerEXT)
@@ -451,7 +414,6 @@
#define glEndOcclusionQueryNV MANGLE(EndOcclusionQueryNV)
#define glEndPerfMonitorAMD MANGLE(EndPerfMonitorAMD)
#define glEndQueryARB MANGLE(EndQueryARB)
#define glEndQueryIndexed MANGLE(EndQueryIndexed)
#define glEndQuery MANGLE(EndQuery)
#define glEndTransformFeedbackEXT MANGLE(EndTransformFeedbackEXT)
#define glEndTransformFeedback MANGLE(EndTransformFeedback)
@@ -485,7 +447,6 @@
#define glFlush MANGLE(Flush)
#define glFlushMappedBufferRangeAPPLE MANGLE(FlushMappedBufferRangeAPPLE)
#define glFlushMappedBufferRange MANGLE(FlushMappedBufferRange)
#define glFlushMappedNamedBufferRangeEXT MANGLE(FlushMappedNamedBufferRangeEXT)
#define glFlushPixelDataRangeNV MANGLE(FlushPixelDataRangeNV)
#define glFlushRasterSGIX MANGLE(FlushRasterSGIX)
#define glFlushVertexArrayRangeAPPLE MANGLE(FlushVertexArrayRangeAPPLE)
@@ -537,6 +498,7 @@
#define glFramebufferTextureEXT MANGLE(FramebufferTextureEXT)
#define glFramebufferTextureFaceARB MANGLE(FramebufferTextureFaceARB)
#define glFramebufferTextureFaceEXT MANGLE(FramebufferTextureFaceEXT)
#define glFramebufferTextureFace MANGLE(FramebufferTextureFace)
#define glFramebufferTextureLayerARB MANGLE(FramebufferTextureLayerARB)
#define glFramebufferTextureLayerEXT MANGLE(FramebufferTextureLayerEXT)
#define glFramebufferTextureLayer MANGLE(FramebufferTextureLayer)
@@ -559,30 +521,23 @@
#define glGenFramebuffersEXT MANGLE(GenFramebuffersEXT)
#define glGenFramebuffers MANGLE(GenFramebuffers)
#define glGenLists MANGLE(GenLists)
#define glGenNamesAMD MANGLE(GenNamesAMD)
#define glGenOcclusionQueriesNV MANGLE(GenOcclusionQueriesNV)
#define glGenPerfMonitorsAMD MANGLE(GenPerfMonitorsAMD)
#define glGenProgramPipelines MANGLE(GenProgramPipelines)
#define glGenProgramsARB MANGLE(GenProgramsARB)
#define glGenProgramsNV MANGLE(GenProgramsNV)
#define glGenQueriesARB MANGLE(GenQueriesARB)
#define glGenQueries MANGLE(GenQueries)
#define glGenRenderbuffersEXT MANGLE(GenRenderbuffersEXT)
#define glGenRenderbuffers MANGLE(GenRenderbuffers)
#define glGenSamplers MANGLE(GenSamplers)
#define glGenSymbolsEXT MANGLE(GenSymbolsEXT)
#define glGenTexturesEXT MANGLE(GenTexturesEXT)
#define glGenTextures MANGLE(GenTextures)
#define glGenTransformFeedbacks MANGLE(GenTransformFeedbacks)
#define glGenTransformFeedbacksNV MANGLE(GenTransformFeedbacksNV)
#define glGenVertexArraysAPPLE MANGLE(GenVertexArraysAPPLE)
#define glGenVertexArrays MANGLE(GenVertexArrays)
#define glGenVertexShadersEXT MANGLE(GenVertexShadersEXT)
#define glGetActiveAttribARB MANGLE(GetActiveAttribARB)
#define glGetActiveAttrib MANGLE(GetActiveAttrib)
#define glGetActiveSubroutineName MANGLE(GetActiveSubroutineName)
#define glGetActiveSubroutineUniformiv MANGLE(GetActiveSubroutineUniformiv)
#define glGetActiveSubroutineUniformName MANGLE(GetActiveSubroutineUniformName)
#define glGetActiveUniformARB MANGLE(GetActiveUniformARB)
#define glGetActiveUniformBlockiv MANGLE(GetActiveUniformBlockiv)
#define glGetActiveUniformBlockName MANGLE(GetActiveUniformBlockName)
@@ -634,21 +589,16 @@
#define glGetConvolutionParameteriv MANGLE(GetConvolutionParameteriv)
#define glGetDebugLogLengthMESA MANGLE(GetDebugLogLengthMESA)
#define glGetDebugLogMESA MANGLE(GetDebugLogMESA)
#define glGetDebugMessageLogAMD MANGLE(GetDebugMessageLogAMD)
#define glGetDebugMessageLogARB MANGLE(GetDebugMessageLogARB)
#define glGetDetailTexFuncSGIS MANGLE(GetDetailTexFuncSGIS)
#define glGetDoubleIndexedvEXT MANGLE(GetDoubleIndexedvEXT)
#define glGetDoublei_v MANGLE(GetDoublei_v)
#define glGetDoublev MANGLE(GetDoublev)
#define glGetError MANGLE(GetError)
#define glGetFenceivNV MANGLE(GetFenceivNV)
#define glGetFinalCombinerInputParameterfvNV MANGLE(GetFinalCombinerInputParameterfvNV)
#define glGetFinalCombinerInputParameterivNV MANGLE(GetFinalCombinerInputParameterivNV)
#define glGetFloatIndexedvEXT MANGLE(GetFloatIndexedvEXT)
#define glGetFloati_v MANGLE(GetFloati_v)
#define glGetFloatv MANGLE(GetFloatv)
#define glGetFogFuncSGIS MANGLE(GetFogFuncSGIS)
#define glGetFragDataIndex MANGLE(GetFragDataIndex)
#define glGetFragDataLocationEXT MANGLE(GetFragDataLocationEXT)
#define glGetFragDataLocation MANGLE(GetFragDataLocation)
#define glGetFragmentLightfvSGIX MANGLE(GetFragmentLightfvSGIX)
@@ -658,7 +608,6 @@
#define glGetFramebufferAttachmentParameterivEXT MANGLE(GetFramebufferAttachmentParameterivEXT)
#define glGetFramebufferAttachmentParameteriv MANGLE(GetFramebufferAttachmentParameteriv)
#define glGetFramebufferParameterivEXT MANGLE(GetFramebufferParameterivEXT)
#define glGetGraphicsResetStatusARB MANGLE(GetGraphicsResetStatusARB)
#define glGetHandleARB MANGLE(GetHandleARB)
#define glGetHistogramEXT MANGLE(GetHistogramEXT)
#define glGetHistogram MANGLE(GetHistogram)
@@ -729,26 +678,6 @@
#define glGetNamedProgramLocalParameterIuivEXT MANGLE(GetNamedProgramLocalParameterIuivEXT)
#define glGetNamedProgramStringEXT MANGLE(GetNamedProgramStringEXT)
#define glGetNamedRenderbufferParameterivEXT MANGLE(GetNamedRenderbufferParameterivEXT)
#define glGetNamedStringARB MANGLE(GetNamedStringARB)
#define glGetNamedStringivARB MANGLE(GetNamedStringivARB)
#define glGetnColorTableARB MANGLE(GetnColorTableARB)
#define glGetnCompressedTexImageARB MANGLE(GetnCompressedTexImageARB)
#define glGetnConvolutionFilterARB MANGLE(GetnConvolutionFilterARB)
#define glGetnHistogramARB MANGLE(GetnHistogramARB)
#define glGetnMapdvARB MANGLE(GetnMapdvARB)
#define glGetnMapfvARB MANGLE(GetnMapfvARB)
#define glGetnMapivARB MANGLE(GetnMapivARB)
#define glGetnMinmaxARB MANGLE(GetnMinmaxARB)
#define glGetnPixelMapfvARB MANGLE(GetnPixelMapfvARB)
#define glGetnPixelMapuivARB MANGLE(GetnPixelMapuivARB)
#define glGetnPixelMapusvARB MANGLE(GetnPixelMapusvARB)
#define glGetnPolygonStippleARB MANGLE(GetnPolygonStippleARB)
#define glGetnSeparableFilterARB MANGLE(GetnSeparableFilterARB)
#define glGetnTexImageARB MANGLE(GetnTexImageARB)
#define glGetnUniformdvARB MANGLE(GetnUniformdvARB)
#define glGetnUniformfvARB MANGLE(GetnUniformfvARB)
#define glGetnUniformivARB MANGLE(GetnUniformivARB)
#define glGetnUniformuivARB MANGLE(GetnUniformuivARB)
#define glGetObjectBufferfvATI MANGLE(GetObjectBufferfvATI)
#define glGetObjectBufferivATI MANGLE(GetObjectBufferivATI)
#define glGetObjectParameterfvARB MANGLE(GetObjectParameterfvARB)
@@ -771,7 +700,6 @@
#define glGetPointervEXT MANGLE(GetPointervEXT)
#define glGetPointerv MANGLE(GetPointerv)
#define glGetPolygonStipple MANGLE(GetPolygonStipple)
#define glGetProgramBinary MANGLE(GetProgramBinary)
#define glGetProgramEnvParameterdvARB MANGLE(GetProgramEnvParameterdvARB)
#define glGetProgramEnvParameterfvARB MANGLE(GetProgramEnvParameterfvARB)
#define glGetProgramEnvParameterIivNV MANGLE(GetProgramEnvParameterIivNV)
@@ -788,42 +716,28 @@
#define glGetProgramNamedParameterfvNV MANGLE(GetProgramNamedParameterfvNV)
#define glGetProgramParameterdvNV MANGLE(GetProgramParameterdvNV)
#define glGetProgramParameterfvNV MANGLE(GetProgramParameterfvNV)
#define glGetProgramPipelineInfoLog MANGLE(GetProgramPipelineInfoLog)
#define glGetProgramPipelineiv MANGLE(GetProgramPipelineiv)
#define glGetProgramRegisterfvMESA MANGLE(GetProgramRegisterfvMESA)
#define glGetProgramStageiv MANGLE(GetProgramStageiv)
#define glGetProgramStringARB MANGLE(GetProgramStringARB)
#define glGetProgramStringNV MANGLE(GetProgramStringNV)
#define glGetProgramSubroutineParameteruivNV MANGLE(GetProgramSubroutineParameteruivNV)
#define glGetQueryIndexediv MANGLE(GetQueryIndexediv)
#define glGetQueryivARB MANGLE(GetQueryivARB)
#define glGetQueryiv MANGLE(GetQueryiv)
#define glGetQueryObjecti64vEXT MANGLE(GetQueryObjecti64vEXT)
#define glGetQueryObjecti64v MANGLE(GetQueryObjecti64v)
#define glGetQueryObjectivARB MANGLE(GetQueryObjectivARB)
#define glGetQueryObjectiv MANGLE(GetQueryObjectiv)
#define glGetQueryObjectui64vEXT MANGLE(GetQueryObjectui64vEXT)
#define glGetQueryObjectui64v MANGLE(GetQueryObjectui64v)
#define glGetQueryObjectuivARB MANGLE(GetQueryObjectuivARB)
#define glGetQueryObjectuiv MANGLE(GetQueryObjectuiv)
#define glGetRenderbufferParameterivEXT MANGLE(GetRenderbufferParameterivEXT)
#define glGetRenderbufferParameteriv MANGLE(GetRenderbufferParameteriv)
#define glGetSamplerParameterfv MANGLE(GetSamplerParameterfv)
#define glGetSamplerParameterIiv MANGLE(GetSamplerParameterIiv)
#define glGetSamplerParameterIuiv MANGLE(GetSamplerParameterIuiv)
#define glGetSamplerParameteriv MANGLE(GetSamplerParameteriv)
#define glGetSeparableFilterEXT MANGLE(GetSeparableFilterEXT)
#define glGetSeparableFilter MANGLE(GetSeparableFilter)
#define glGetShaderInfoLog MANGLE(GetShaderInfoLog)
#define glGetShaderiv MANGLE(GetShaderiv)
#define glGetShaderPrecisionFormat MANGLE(GetShaderPrecisionFormat)
#define glGetShaderSourceARB MANGLE(GetShaderSourceARB)
#define glGetShaderSource MANGLE(GetShaderSource)
#define glGetSharpenTexFuncSGIS MANGLE(GetSharpenTexFuncSGIS)
#define glGetStringi MANGLE(GetStringi)
#define glGetString MANGLE(GetString)
#define glGetSubroutineIndex MANGLE(GetSubroutineIndex)
#define glGetSubroutineUniformLocation MANGLE(GetSubroutineUniformLocation)
#define glGetSynciv MANGLE(GetSynciv)
#define glGetTexBumpParameterfvATI MANGLE(GetTexBumpParameterfvATI)
#define glGetTexBumpParameterivATI MANGLE(GetTexBumpParameterivATI)
@@ -856,17 +770,14 @@
#define glGetTransformFeedbackVaryingNV MANGLE(GetTransformFeedbackVaryingNV)
#define glGetUniformBlockIndex MANGLE(GetUniformBlockIndex)
#define glGetUniformBufferSizeEXT MANGLE(GetUniformBufferSizeEXT)
#define glGetUniformdv MANGLE(GetUniformdv)
#define glGetUniformfvARB MANGLE(GetUniformfvARB)
#define glGetUniformfv MANGLE(GetUniformfv)
#define glGetUniformi64vNV MANGLE(GetUniformi64vNV)
#define glGetUniformIndices MANGLE(GetUniformIndices)
#define glGetUniformivARB MANGLE(GetUniformivARB)
#define glGetUniformiv MANGLE(GetUniformiv)
#define glGetUniformLocationARB MANGLE(GetUniformLocationARB)
#define glGetUniformLocation MANGLE(GetUniformLocation)
#define glGetUniformOffsetEXT MANGLE(GetUniformOffsetEXT)
#define glGetUniformSubroutineuiv MANGLE(GetUniformSubroutineuiv)
#define glGetUniformui64vNV MANGLE(GetUniformui64vNV)
#define glGetUniformuivEXT MANGLE(GetUniformuivEXT)
#define glGetUniformuiv MANGLE(GetUniformuiv)
@@ -892,10 +803,6 @@
#define glGetVertexAttribivARB MANGLE(GetVertexAttribivARB)
#define glGetVertexAttribiv MANGLE(GetVertexAttribiv)
#define glGetVertexAttribivNV MANGLE(GetVertexAttribivNV)
#define glGetVertexAttribLdvEXT MANGLE(GetVertexAttribLdvEXT)
#define glGetVertexAttribLdv MANGLE(GetVertexAttribLdv)
#define glGetVertexAttribLi64vNV MANGLE(GetVertexAttribLi64vNV)
#define glGetVertexAttribLui64vNV MANGLE(GetVertexAttribLui64vNV)
#define glGetVertexAttribPointervARB MANGLE(GetVertexAttribPointervARB)
#define glGetVertexAttribPointerv MANGLE(GetVertexAttribPointerv)
#define glGetVertexAttribPointervNV MANGLE(GetVertexAttribPointervNV)
@@ -957,25 +864,20 @@
#define glIsFramebufferEXT MANGLE(IsFramebufferEXT)
#define glIsFramebuffer MANGLE(IsFramebuffer)
#define glIsList MANGLE(IsList)
#define glIsNameAMD MANGLE(IsNameAMD)
#define glIsNamedBufferResidentNV MANGLE(IsNamedBufferResidentNV)
#define glIsNamedStringARB MANGLE(IsNamedStringARB)
#define glIsObjectBufferATI MANGLE(IsObjectBufferATI)
#define glIsOcclusionQueryNV MANGLE(IsOcclusionQueryNV)
#define glIsProgramARB MANGLE(IsProgramARB)
#define glIsProgram MANGLE(IsProgram)
#define glIsProgramNV MANGLE(IsProgramNV)
#define glIsProgramPipeline MANGLE(IsProgramPipeline)
#define glIsQueryARB MANGLE(IsQueryARB)
#define glIsQuery MANGLE(IsQuery)
#define glIsRenderbufferEXT MANGLE(IsRenderbufferEXT)
#define glIsRenderbuffer MANGLE(IsRenderbuffer)
#define glIsSampler MANGLE(IsSampler)
#define glIsShader MANGLE(IsShader)
#define glIsSync MANGLE(IsSync)
#define glIsTextureEXT MANGLE(IsTextureEXT)
#define glIsTexture MANGLE(IsTexture)
#define glIsTransformFeedback MANGLE(IsTransformFeedback)
#define glIsTransformFeedbackNV MANGLE(IsTransformFeedbackNV)
#define glIsVariantEnabledEXT MANGLE(IsVariantEnabledEXT)
#define glIsVertexArrayAPPLE MANGLE(IsVertexArrayAPPLE)
@@ -1013,8 +915,6 @@
#define glLogicOp MANGLE(LogicOp)
#define glMakeBufferNonResidentNV MANGLE(MakeBufferNonResidentNV)
#define glMakeBufferResidentNV MANGLE(MakeBufferResidentNV)
#define glMakeNamedBufferNonResidentNV MANGLE(MakeNamedBufferNonResidentNV)
#define glMakeNamedBufferResidentNV MANGLE(MakeNamedBufferResidentNV)
#define glMap1d MANGLE(Map1d)
#define glMap1f MANGLE(Map1f)
#define glMap2d MANGLE(Map2d)
@@ -1028,7 +928,6 @@
#define glMapGrid2d MANGLE(MapGrid2d)
#define glMapGrid2f MANGLE(MapGrid2f)
#define glMapNamedBufferEXT MANGLE(MapNamedBufferEXT)
#define glMapNamedBufferRangeEXT MANGLE(MapNamedBufferRangeEXT)
#define glMapObjectBufferATI MANGLE(MapObjectBufferATI)
#define glMapParameterfvNV MANGLE(MapParameterfvNV)
#define glMapParameterivNV MANGLE(MapParameterivNV)
@@ -1064,10 +963,8 @@
#define glMatrixScalefEXT MANGLE(MatrixScalefEXT)
#define glMatrixTranslatedEXT MANGLE(MatrixTranslatedEXT)
#define glMatrixTranslatefEXT MANGLE(MatrixTranslatefEXT)
#define glMemoryBarrierEXT MANGLE(MemoryBarrierEXT)
#define glMinmaxEXT MANGLE(MinmaxEXT)
#define glMinmax MANGLE(Minmax)
#define glMinSampleShadingARB MANGLE(MinSampleShadingARB)
#define glMinSampleShading MANGLE(MinSampleShading)
#define glMultiDrawArraysEXT MANGLE(MultiDrawArraysEXT)
#define glMultiDrawArrays MANGLE(MultiDrawArrays)
@@ -1151,14 +1048,6 @@
#define glMultiTexCoord4s MANGLE(MultiTexCoord4s)
#define glMultiTexCoord4svARB MANGLE(MultiTexCoord4svARB)
#define glMultiTexCoord4sv MANGLE(MultiTexCoord4sv)
#define glMultiTexCoordP1ui MANGLE(MultiTexCoordP1ui)
#define glMultiTexCoordP1uiv MANGLE(MultiTexCoordP1uiv)
#define glMultiTexCoordP2ui MANGLE(MultiTexCoordP2ui)
#define glMultiTexCoordP2uiv MANGLE(MultiTexCoordP2uiv)
#define glMultiTexCoordP3ui MANGLE(MultiTexCoordP3ui)
#define glMultiTexCoordP3uiv MANGLE(MultiTexCoordP3uiv)
#define glMultiTexCoordP4ui MANGLE(MultiTexCoordP4ui)
#define glMultiTexCoordP4uiv MANGLE(MultiTexCoordP4uiv)
#define glMultiTexCoordPointerEXT MANGLE(MultiTexCoordPointerEXT)
#define glMultiTexEnvfEXT MANGLE(MultiTexEnvfEXT)
#define glMultiTexEnvfvEXT MANGLE(MultiTexEnvfvEXT)
@@ -1191,7 +1080,6 @@
#define glMultTransposeMatrixf MANGLE(MultTransposeMatrixf)
#define glNamedBufferDataEXT MANGLE(NamedBufferDataEXT)
#define glNamedBufferSubDataEXT MANGLE(NamedBufferSubDataEXT)
#define glNamedCopyBufferSubDataEXT MANGLE(NamedCopyBufferSubDataEXT)
#define glNamedFramebufferRenderbufferEXT MANGLE(NamedFramebufferRenderbufferEXT)
#define glNamedFramebufferTexture1DEXT MANGLE(NamedFramebufferTexture1DEXT)
#define glNamedFramebufferTexture2DEXT MANGLE(NamedFramebufferTexture2DEXT)
@@ -1199,6 +1087,8 @@
#define glNamedFramebufferTextureEXT MANGLE(NamedFramebufferTextureEXT)
#define glNamedFramebufferTextureFaceEXT MANGLE(NamedFramebufferTextureFaceEXT)
#define glNamedFramebufferTextureLayerEXT MANGLE(NamedFramebufferTextureLayerEXT)
#define glNamedMakeBufferNonResidentNV MANGLE(NamedMakeBufferNonResidentNV)
#define glNamedMakeBufferResidentNV MANGLE(NamedMakeBufferResidentNV)
#define glNamedProgramLocalParameter4dEXT MANGLE(NamedProgramLocalParameter4dEXT)
#define glNamedProgramLocalParameter4dvEXT MANGLE(NamedProgramLocalParameter4dvEXT)
#define glNamedProgramLocalParameter4fEXT MANGLE(NamedProgramLocalParameter4fEXT)
@@ -1214,7 +1104,6 @@
#define glNamedRenderbufferStorageEXT MANGLE(NamedRenderbufferStorageEXT)
#define glNamedRenderbufferStorageMultisampleCoverageEXT MANGLE(NamedRenderbufferStorageMultisampleCoverageEXT)
#define glNamedRenderbufferStorageMultisampleEXT MANGLE(NamedRenderbufferStorageMultisampleEXT)
#define glNamedStringARB MANGLE(NamedStringARB)
#define glNewList MANGLE(NewList)
#define glNewObjectBufferATI MANGLE(NewObjectBufferATI)
#define glNormal3b MANGLE(Normal3b)
@@ -1232,8 +1121,6 @@
#define glNormal3s MANGLE(Normal3s)
#define glNormal3sv MANGLE(Normal3sv)
#define glNormalFormatNV MANGLE(NormalFormatNV)
#define glNormalP3ui MANGLE(NormalP3ui)
#define glNormalP3uiv MANGLE(NormalP3uiv)
#define glNormalPointerEXT MANGLE(NormalPointerEXT)
#define glNormalPointerListIBM MANGLE(NormalPointerListIBM)
#define glNormalPointer MANGLE(NormalPointer)
@@ -1253,9 +1140,6 @@
#define glOrtho MANGLE(Ortho)
#define glPassTexCoordATI MANGLE(PassTexCoordATI)
#define glPassThrough MANGLE(PassThrough)
#define glPatchParameterfv MANGLE(PatchParameterfv)
#define glPatchParameteri MANGLE(PatchParameteri)
#define glPauseTransformFeedback MANGLE(PauseTransformFeedback)
#define glPauseTransformFeedbackNV MANGLE(PauseTransformFeedbackNV)
#define glPixelDataRangeNV MANGLE(PixelDataRangeNV)
#define glPixelMapfv MANGLE(PixelMapfv)
@@ -1307,7 +1191,6 @@
#define glPrimitiveRestartNV MANGLE(PrimitiveRestartNV)
#define glPrioritizeTexturesEXT MANGLE(PrioritizeTexturesEXT)
#define glPrioritizeTextures MANGLE(PrioritizeTextures)
#define glProgramBinary MANGLE(ProgramBinary)
#define glProgramBufferParametersfvNV MANGLE(ProgramBufferParametersfvNV)
#define glProgramBufferParametersIivNV MANGLE(ProgramBufferParametersIivNV)
#define glProgramBufferParametersIuivNV MANGLE(ProgramBufferParametersIuivNV)
@@ -1348,123 +1231,39 @@
#define glProgramParameters4dvNV MANGLE(ProgramParameters4dvNV)
#define glProgramParameters4fvNV MANGLE(ProgramParameters4fvNV)
#define glProgramStringARB MANGLE(ProgramStringARB)
#define glProgramSubroutineParametersuivNV MANGLE(ProgramSubroutineParametersuivNV)
#define glProgramUniform1dEXT MANGLE(ProgramUniform1dEXT)
#define glProgramUniform1d MANGLE(ProgramUniform1d)
#define glProgramUniform1dvEXT MANGLE(ProgramUniform1dvEXT)
#define glProgramUniform1dv MANGLE(ProgramUniform1dv)
#define glProgramUniform1fEXT MANGLE(ProgramUniform1fEXT)
#define glProgramUniform1f MANGLE(ProgramUniform1f)
#define glProgramUniform1fvEXT MANGLE(ProgramUniform1fvEXT)
#define glProgramUniform1fv MANGLE(ProgramUniform1fv)
#define glProgramUniform1i64NV MANGLE(ProgramUniform1i64NV)
#define glProgramUniform1i64vNV MANGLE(ProgramUniform1i64vNV)
#define glProgramUniform1iEXT MANGLE(ProgramUniform1iEXT)
#define glProgramUniform1i MANGLE(ProgramUniform1i)
#define glProgramUniform1ivEXT MANGLE(ProgramUniform1ivEXT)
#define glProgramUniform1iv MANGLE(ProgramUniform1iv)
#define glProgramUniform1ui64NV MANGLE(ProgramUniform1ui64NV)
#define glProgramUniform1ui64vNV MANGLE(ProgramUniform1ui64vNV)
#define glProgramUniform1uiEXT MANGLE(ProgramUniform1uiEXT)
#define glProgramUniform1ui MANGLE(ProgramUniform1ui)
#define glProgramUniform1uivEXT MANGLE(ProgramUniform1uivEXT)
#define glProgramUniform1uiv MANGLE(ProgramUniform1uiv)
#define glProgramUniform2dEXT MANGLE(ProgramUniform2dEXT)
#define glProgramUniform2d MANGLE(ProgramUniform2d)
#define glProgramUniform2dvEXT MANGLE(ProgramUniform2dvEXT)
#define glProgramUniform2dv MANGLE(ProgramUniform2dv)
#define glProgramUniform2fEXT MANGLE(ProgramUniform2fEXT)
#define glProgramUniform2f MANGLE(ProgramUniform2f)
#define glProgramUniform2fvEXT MANGLE(ProgramUniform2fvEXT)
#define glProgramUniform2fv MANGLE(ProgramUniform2fv)
#define glProgramUniform2i64NV MANGLE(ProgramUniform2i64NV)
#define glProgramUniform2i64vNV MANGLE(ProgramUniform2i64vNV)
#define glProgramUniform2iEXT MANGLE(ProgramUniform2iEXT)
#define glProgramUniform2i MANGLE(ProgramUniform2i)
#define glProgramUniform2ivEXT MANGLE(ProgramUniform2ivEXT)
#define glProgramUniform2iv MANGLE(ProgramUniform2iv)
#define glProgramUniform2ui64NV MANGLE(ProgramUniform2ui64NV)
#define glProgramUniform2ui64vNV MANGLE(ProgramUniform2ui64vNV)
#define glProgramUniform2uiEXT MANGLE(ProgramUniform2uiEXT)
#define glProgramUniform2ui MANGLE(ProgramUniform2ui)
#define glProgramUniform2uivEXT MANGLE(ProgramUniform2uivEXT)
#define glProgramUniform2uiv MANGLE(ProgramUniform2uiv)
#define glProgramUniform3dEXT MANGLE(ProgramUniform3dEXT)
#define glProgramUniform3d MANGLE(ProgramUniform3d)
#define glProgramUniform3dvEXT MANGLE(ProgramUniform3dvEXT)
#define glProgramUniform3dv MANGLE(ProgramUniform3dv)
#define glProgramUniform3fEXT MANGLE(ProgramUniform3fEXT)
#define glProgramUniform3f MANGLE(ProgramUniform3f)
#define glProgramUniform3fvEXT MANGLE(ProgramUniform3fvEXT)
#define glProgramUniform3fv MANGLE(ProgramUniform3fv)
#define glProgramUniform3i64NV MANGLE(ProgramUniform3i64NV)
#define glProgramUniform3i64vNV MANGLE(ProgramUniform3i64vNV)
#define glProgramUniform3iEXT MANGLE(ProgramUniform3iEXT)
#define glProgramUniform3i MANGLE(ProgramUniform3i)
#define glProgramUniform3ivEXT MANGLE(ProgramUniform3ivEXT)
#define glProgramUniform3iv MANGLE(ProgramUniform3iv)
#define glProgramUniform3ui64NV MANGLE(ProgramUniform3ui64NV)
#define glProgramUniform3ui64vNV MANGLE(ProgramUniform3ui64vNV)
#define glProgramUniform3uiEXT MANGLE(ProgramUniform3uiEXT)
#define glProgramUniform3ui MANGLE(ProgramUniform3ui)
#define glProgramUniform3uivEXT MANGLE(ProgramUniform3uivEXT)
#define glProgramUniform3uiv MANGLE(ProgramUniform3uiv)
#define glProgramUniform4dEXT MANGLE(ProgramUniform4dEXT)
#define glProgramUniform4d MANGLE(ProgramUniform4d)
#define glProgramUniform4dvEXT MANGLE(ProgramUniform4dvEXT)
#define glProgramUniform4dv MANGLE(ProgramUniform4dv)
#define glProgramUniform4fEXT MANGLE(ProgramUniform4fEXT)
#define glProgramUniform4f MANGLE(ProgramUniform4f)
#define glProgramUniform4fvEXT MANGLE(ProgramUniform4fvEXT)
#define glProgramUniform4fv MANGLE(ProgramUniform4fv)
#define glProgramUniform4i64NV MANGLE(ProgramUniform4i64NV)
#define glProgramUniform4i64vNV MANGLE(ProgramUniform4i64vNV)
#define glProgramUniform4iEXT MANGLE(ProgramUniform4iEXT)
#define glProgramUniform4i MANGLE(ProgramUniform4i)
#define glProgramUniform4ivEXT MANGLE(ProgramUniform4ivEXT)
#define glProgramUniform4iv MANGLE(ProgramUniform4iv)
#define glProgramUniform4ui64NV MANGLE(ProgramUniform4ui64NV)
#define glProgramUniform4ui64vNV MANGLE(ProgramUniform4ui64vNV)
#define glProgramUniform4uiEXT MANGLE(ProgramUniform4uiEXT)
#define glProgramUniform4ui MANGLE(ProgramUniform4ui)
#define glProgramUniform4uivEXT MANGLE(ProgramUniform4uivEXT)
#define glProgramUniform4uiv MANGLE(ProgramUniform4uiv)
#define glProgramUniformMatrix2dvEXT MANGLE(ProgramUniformMatrix2dvEXT)
#define glProgramUniformMatrix2dv MANGLE(ProgramUniformMatrix2dv)
#define glProgramUniformMatrix2fvEXT MANGLE(ProgramUniformMatrix2fvEXT)
#define glProgramUniformMatrix2fv MANGLE(ProgramUniformMatrix2fv)
#define glProgramUniformMatrix2x3dvEXT MANGLE(ProgramUniformMatrix2x3dvEXT)
#define glProgramUniformMatrix2x3dv MANGLE(ProgramUniformMatrix2x3dv)
#define glProgramUniformMatrix2x3fvEXT MANGLE(ProgramUniformMatrix2x3fvEXT)
#define glProgramUniformMatrix2x3fv MANGLE(ProgramUniformMatrix2x3fv)
#define glProgramUniformMatrix2x4dvEXT MANGLE(ProgramUniformMatrix2x4dvEXT)
#define glProgramUniformMatrix2x4dv MANGLE(ProgramUniformMatrix2x4dv)
#define glProgramUniformMatrix2x4fvEXT MANGLE(ProgramUniformMatrix2x4fvEXT)
#define glProgramUniformMatrix2x4fv MANGLE(ProgramUniformMatrix2x4fv)
#define glProgramUniformMatrix3dvEXT MANGLE(ProgramUniformMatrix3dvEXT)
#define glProgramUniformMatrix3dv MANGLE(ProgramUniformMatrix3dv)
#define glProgramUniformMatrix3fvEXT MANGLE(ProgramUniformMatrix3fvEXT)
#define glProgramUniformMatrix3fv MANGLE(ProgramUniformMatrix3fv)
#define glProgramUniformMatrix3x2dvEXT MANGLE(ProgramUniformMatrix3x2dvEXT)
#define glProgramUniformMatrix3x2dv MANGLE(ProgramUniformMatrix3x2dv)
#define glProgramUniformMatrix3x2fvEXT MANGLE(ProgramUniformMatrix3x2fvEXT)
#define glProgramUniformMatrix3x2fv MANGLE(ProgramUniformMatrix3x2fv)
#define glProgramUniformMatrix3x4dvEXT MANGLE(ProgramUniformMatrix3x4dvEXT)
#define glProgramUniformMatrix3x4dv MANGLE(ProgramUniformMatrix3x4dv)
#define glProgramUniformMatrix3x4fvEXT MANGLE(ProgramUniformMatrix3x4fvEXT)
#define glProgramUniformMatrix3x4fv MANGLE(ProgramUniformMatrix3x4fv)
#define glProgramUniformMatrix4dvEXT MANGLE(ProgramUniformMatrix4dvEXT)
#define glProgramUniformMatrix4dv MANGLE(ProgramUniformMatrix4dv)
#define glProgramUniformMatrix4fvEXT MANGLE(ProgramUniformMatrix4fvEXT)
#define glProgramUniformMatrix4fv MANGLE(ProgramUniformMatrix4fv)
#define glProgramUniformMatrix4x2dvEXT MANGLE(ProgramUniformMatrix4x2dvEXT)
#define glProgramUniformMatrix4x2dv MANGLE(ProgramUniformMatrix4x2dv)
#define glProgramUniformMatrix4x2fvEXT MANGLE(ProgramUniformMatrix4x2fvEXT)
#define glProgramUniformMatrix4x2fv MANGLE(ProgramUniformMatrix4x2fv)
#define glProgramUniformMatrix4x3dvEXT MANGLE(ProgramUniformMatrix4x3dvEXT)
#define glProgramUniformMatrix4x3dv MANGLE(ProgramUniformMatrix4x3dv)
#define glProgramUniformMatrix4x3fvEXT MANGLE(ProgramUniformMatrix4x3fvEXT)
#define glProgramUniformMatrix4x3fv MANGLE(ProgramUniformMatrix4x3fv)
#define glProgramUniformui64NV MANGLE(ProgramUniformui64NV)
#define glProgramUniformui64vNV MANGLE(ProgramUniformui64vNV)
#define glProgramVertexLimitNV MANGLE(ProgramVertexLimitNV)
@@ -1475,7 +1274,6 @@
#define glPushClientAttrib MANGLE(PushClientAttrib)
#define glPushMatrix MANGLE(PushMatrix)
#define glPushName MANGLE(PushName)
#define glQueryCounter MANGLE(QueryCounter)
#define glRasterPos2d MANGLE(RasterPos2d)
#define glRasterPos2dv MANGLE(RasterPos2dv)
#define glRasterPos2f MANGLE(RasterPos2f)
@@ -1502,7 +1300,6 @@
#define glRasterPos4sv MANGLE(RasterPos4sv)
#define glReadBuffer MANGLE(ReadBuffer)
#define glReadInstrumentsSGIX MANGLE(ReadInstrumentsSGIX)
#define glReadnPixelsARB MANGLE(ReadnPixelsARB)
#define glReadPixels MANGLE(ReadPixels)
#define glRectd MANGLE(Rectd)
#define glRectdv MANGLE(Rectdv)
@@ -1513,7 +1310,6 @@
#define glRects MANGLE(Rects)
#define glRectsv MANGLE(Rectsv)
#define glReferencePlaneSGIX MANGLE(ReferencePlaneSGIX)
#define glReleaseShaderCompiler MANGLE(ReleaseShaderCompiler)
#define glRenderbufferStorageEXT MANGLE(RenderbufferStorageEXT)
#define glRenderbufferStorage MANGLE(RenderbufferStorage)
#define glRenderbufferStorageMultisampleCoverageNV MANGLE(RenderbufferStorageMultisampleCoverageNV)
@@ -1549,7 +1345,6 @@
#define glResetMinmaxEXT MANGLE(ResetMinmaxEXT)
#define glResetMinmax MANGLE(ResetMinmax)
#define glResizeBuffersMESA MANGLE(ResizeBuffersMESA)
#define glResumeTransformFeedback MANGLE(ResumeTransformFeedback)
#define glResumeTransformFeedbackNV MANGLE(ResumeTransformFeedbackNV)
#define glRotated MANGLE(Rotated)
#define glRotatef MANGLE(Rotatef)
@@ -1562,17 +1357,8 @@
#define glSampleMaskSGIS MANGLE(SampleMaskSGIS)
#define glSamplePatternEXT MANGLE(SamplePatternEXT)
#define glSamplePatternSGIS MANGLE(SamplePatternSGIS)
#define glSamplerParameterf MANGLE(SamplerParameterf)
#define glSamplerParameterfv MANGLE(SamplerParameterfv)
#define glSamplerParameterIiv MANGLE(SamplerParameterIiv)
#define glSamplerParameteri MANGLE(SamplerParameteri)
#define glSamplerParameterIuiv MANGLE(SamplerParameterIuiv)
#define glSamplerParameteriv MANGLE(SamplerParameteriv)
#define glScaled MANGLE(Scaled)
#define glScalef MANGLE(Scalef)
#define glScissorArrayv MANGLE(ScissorArrayv)
#define glScissorIndexed MANGLE(ScissorIndexed)
#define glScissorIndexedv MANGLE(ScissorIndexedv)
#define glScissor MANGLE(Scissor)
#define glSecondaryColor3bEXT MANGLE(SecondaryColor3bEXT)
#define glSecondaryColor3b MANGLE(SecondaryColor3b)
@@ -1609,8 +1395,6 @@
#define glSecondaryColor3usvEXT MANGLE(SecondaryColor3usvEXT)
#define glSecondaryColor3usv MANGLE(SecondaryColor3usv)
#define glSecondaryColorFormatNV MANGLE(SecondaryColorFormatNV)
#define glSecondaryColorP3ui MANGLE(SecondaryColorP3ui)
#define glSecondaryColorP3uiv MANGLE(SecondaryColorP3uiv)
#define glSecondaryColorPointerEXT MANGLE(SecondaryColorPointerEXT)
#define glSecondaryColorPointerListIBM MANGLE(SecondaryColorPointerListIBM)
#define glSecondaryColorPointer MANGLE(SecondaryColorPointer)
@@ -1624,7 +1408,6 @@
#define glSetInvariantEXT MANGLE(SetInvariantEXT)
#define glSetLocalConstantEXT MANGLE(SetLocalConstantEXT)
#define glShadeModel MANGLE(ShadeModel)
#define glShaderBinary MANGLE(ShaderBinary)
#define glShaderOp1EXT MANGLE(ShaderOp1EXT)
#define glShaderOp2EXT MANGLE(ShaderOp2EXT)
#define glShaderOp3EXT MANGLE(ShaderOp3EXT)
@@ -1726,14 +1509,6 @@
#define glTexCoord4s MANGLE(TexCoord4s)
#define glTexCoord4sv MANGLE(TexCoord4sv)
#define glTexCoordFormatNV MANGLE(TexCoordFormatNV)
#define glTexCoordP1ui MANGLE(TexCoordP1ui)
#define glTexCoordP1uiv MANGLE(TexCoordP1uiv)
#define glTexCoordP2ui MANGLE(TexCoordP2ui)
#define glTexCoordP2uiv MANGLE(TexCoordP2uiv)
#define glTexCoordP3ui MANGLE(TexCoordP3ui)
#define glTexCoordP3uiv MANGLE(TexCoordP3uiv)
#define glTexCoordP4ui MANGLE(TexCoordP4ui)
#define glTexCoordP4uiv MANGLE(TexCoordP4uiv)
#define glTexCoordPointerEXT MANGLE(TexCoordPointerEXT)
#define glTexCoordPointerListIBM MANGLE(TexCoordPointerListIBM)
#define glTexCoordPointer MANGLE(TexCoordPointer)
@@ -1794,108 +1569,73 @@
#define glTextureSubImage3DEXT MANGLE(TextureSubImage3DEXT)
#define glTrackMatrixNV MANGLE(TrackMatrixNV)
#define glTransformFeedbackAttribsNV MANGLE(TransformFeedbackAttribsNV)
#define glTransformFeedbackStreamAttribsNV MANGLE(TransformFeedbackStreamAttribsNV)
#define glTransformFeedbackVaryingsEXT MANGLE(TransformFeedbackVaryingsEXT)
#define glTransformFeedbackVaryings MANGLE(TransformFeedbackVaryings)
#define glTransformFeedbackVaryingsNV MANGLE(TransformFeedbackVaryingsNV)
#define glTranslated MANGLE(Translated)
#define glTranslatef MANGLE(Translatef)
#define glUniform1d MANGLE(Uniform1d)
#define glUniform1dv MANGLE(Uniform1dv)
#define glUniform1fARB MANGLE(Uniform1fARB)
#define glUniform1f MANGLE(Uniform1f)
#define glUniform1fvARB MANGLE(Uniform1fvARB)
#define glUniform1fv MANGLE(Uniform1fv)
#define glUniform1i64NV MANGLE(Uniform1i64NV)
#define glUniform1i64vNV MANGLE(Uniform1i64vNV)
#define glUniform1iARB MANGLE(Uniform1iARB)
#define glUniform1i MANGLE(Uniform1i)
#define glUniform1ivARB MANGLE(Uniform1ivARB)
#define glUniform1iv MANGLE(Uniform1iv)
#define glUniform1ui64NV MANGLE(Uniform1ui64NV)
#define glUniform1ui64vNV MANGLE(Uniform1ui64vNV)
#define glUniform1uiEXT MANGLE(Uniform1uiEXT)
#define glUniform1ui MANGLE(Uniform1ui)
#define glUniform1uivEXT MANGLE(Uniform1uivEXT)
#define glUniform1uiv MANGLE(Uniform1uiv)
#define glUniform2d MANGLE(Uniform2d)
#define glUniform2dv MANGLE(Uniform2dv)
#define glUniform2fARB MANGLE(Uniform2fARB)
#define glUniform2f MANGLE(Uniform2f)
#define glUniform2fvARB MANGLE(Uniform2fvARB)
#define glUniform2fv MANGLE(Uniform2fv)
#define glUniform2i64NV MANGLE(Uniform2i64NV)
#define glUniform2i64vNV MANGLE(Uniform2i64vNV)
#define glUniform2iARB MANGLE(Uniform2iARB)
#define glUniform2i MANGLE(Uniform2i)
#define glUniform2ivARB MANGLE(Uniform2ivARB)
#define glUniform2iv MANGLE(Uniform2iv)
#define glUniform2ui64NV MANGLE(Uniform2ui64NV)
#define glUniform2ui64vNV MANGLE(Uniform2ui64vNV)
#define glUniform2uiEXT MANGLE(Uniform2uiEXT)
#define glUniform2ui MANGLE(Uniform2ui)
#define glUniform2uivEXT MANGLE(Uniform2uivEXT)
#define glUniform2uiv MANGLE(Uniform2uiv)
#define glUniform3d MANGLE(Uniform3d)
#define glUniform3dv MANGLE(Uniform3dv)
#define glUniform3fARB MANGLE(Uniform3fARB)
#define glUniform3f MANGLE(Uniform3f)
#define glUniform3fvARB MANGLE(Uniform3fvARB)
#define glUniform3fv MANGLE(Uniform3fv)
#define glUniform3i64NV MANGLE(Uniform3i64NV)
#define glUniform3i64vNV MANGLE(Uniform3i64vNV)
#define glUniform3iARB MANGLE(Uniform3iARB)
#define glUniform3i MANGLE(Uniform3i)
#define glUniform3ivARB MANGLE(Uniform3ivARB)
#define glUniform3iv MANGLE(Uniform3iv)
#define glUniform3ui64NV MANGLE(Uniform3ui64NV)
#define glUniform3ui64vNV MANGLE(Uniform3ui64vNV)
#define glUniform3uiEXT MANGLE(Uniform3uiEXT)
#define glUniform3ui MANGLE(Uniform3ui)
#define glUniform3uivEXT MANGLE(Uniform3uivEXT)
#define glUniform3uiv MANGLE(Uniform3uiv)
#define glUniform4d MANGLE(Uniform4d)
#define glUniform4dv MANGLE(Uniform4dv)
#define glUniform4fARB MANGLE(Uniform4fARB)
#define glUniform4f MANGLE(Uniform4f)
#define glUniform4fvARB MANGLE(Uniform4fvARB)
#define glUniform4fv MANGLE(Uniform4fv)
#define glUniform4i64NV MANGLE(Uniform4i64NV)
#define glUniform4i64vNV MANGLE(Uniform4i64vNV)
#define glUniform4iARB MANGLE(Uniform4iARB)
#define glUniform4i MANGLE(Uniform4i)
#define glUniform4ivARB MANGLE(Uniform4ivARB)
#define glUniform4iv MANGLE(Uniform4iv)
#define glUniform4ui64NV MANGLE(Uniform4ui64NV)
#define glUniform4ui64vNV MANGLE(Uniform4ui64vNV)
#define glUniform4uiEXT MANGLE(Uniform4uiEXT)
#define glUniform4ui MANGLE(Uniform4ui)
#define glUniform4uivEXT MANGLE(Uniform4uivEXT)
#define glUniform4uiv MANGLE(Uniform4uiv)
#define glUniformBlockBinding MANGLE(UniformBlockBinding)
#define glUniformBufferEXT MANGLE(UniformBufferEXT)
#define glUniformMatrix2dv MANGLE(UniformMatrix2dv)
#define glUniformMatrix2fvARB MANGLE(UniformMatrix2fvARB)
#define glUniformMatrix2fv MANGLE(UniformMatrix2fv)
#define glUniformMatrix2x3dv MANGLE(UniformMatrix2x3dv)
#define glUniformMatrix2x3fv MANGLE(UniformMatrix2x3fv)
#define glUniformMatrix2x4dv MANGLE(UniformMatrix2x4dv)
#define glUniformMatrix2x4fv MANGLE(UniformMatrix2x4fv)
#define glUniformMatrix3dv MANGLE(UniformMatrix3dv)
#define glUniformMatrix3fvARB MANGLE(UniformMatrix3fvARB)
#define glUniformMatrix3fv MANGLE(UniformMatrix3fv)
#define glUniformMatrix3x2dv MANGLE(UniformMatrix3x2dv)
#define glUniformMatrix3x2fv MANGLE(UniformMatrix3x2fv)
#define glUniformMatrix3x4dv MANGLE(UniformMatrix3x4dv)
#define glUniformMatrix3x4fv MANGLE(UniformMatrix3x4fv)
#define glUniformMatrix4dv MANGLE(UniformMatrix4dv)
#define glUniformMatrix4fvARB MANGLE(UniformMatrix4fvARB)
#define glUniformMatrix4fv MANGLE(UniformMatrix4fv)
#define glUniformMatrix4x2dv MANGLE(UniformMatrix4x2dv)
#define glUniformMatrix4x2fv MANGLE(UniformMatrix4x2fv)
#define glUniformMatrix4x3dv MANGLE(UniformMatrix4x3dv)
#define glUniformMatrix4x3fv MANGLE(UniformMatrix4x3fv)
#define glUniformSubroutinesuiv MANGLE(UniformSubroutinesuiv)
#define glUniformui64NV MANGLE(Uniformui64NV)
#define glUniformui64vNV MANGLE(Uniformui64vNV)
#define glUnlockArraysEXT MANGLE(UnlockArraysEXT)
@@ -1906,11 +1646,9 @@
#define glUpdateObjectBufferATI MANGLE(UpdateObjectBufferATI)
#define glUseProgram MANGLE(UseProgram)
#define glUseProgramObjectARB MANGLE(UseProgramObjectARB)
#define glUseProgramStages MANGLE(UseProgramStages)
#define glUseShaderProgramEXT MANGLE(UseShaderProgramEXT)
#define glValidateProgramARB MANGLE(ValidateProgramARB)
#define glValidateProgram MANGLE(ValidateProgram)
#define glValidateProgramPipeline MANGLE(ValidateProgramPipeline)
#define glVariantArrayObjectATI MANGLE(VariantArrayObjectATI)
#define glVariantbvEXT MANGLE(VariantbvEXT)
#define glVariantdvEXT MANGLE(VariantdvEXT)
@@ -1921,16 +1659,6 @@
#define glVariantubvEXT MANGLE(VariantubvEXT)
#define glVariantuivEXT MANGLE(VariantuivEXT)
#define glVariantusvEXT MANGLE(VariantusvEXT)
#define glVDPAUFiniNV MANGLE(VDPAUFiniNV)
#define glVDPAUGetSurfaceivNV MANGLE(VDPAUGetSurfaceivNV)
#define glVDPAUInitNV MANGLE(VDPAUInitNV)
#define glVDPAUIsSurfaceNV MANGLE(VDPAUIsSurfaceNV)
#define glVDPAUMapSurfacesNV MANGLE(VDPAUMapSurfacesNV)
#define glVDPAURegisterOutputSurfaceNV MANGLE(VDPAURegisterOutputSurfaceNV)
#define glVDPAURegisterVideoSurfaceNV MANGLE(VDPAURegisterVideoSurfaceNV)
#define glVDPAUSurfaceAccessNV MANGLE(VDPAUSurfaceAccessNV)
#define glVDPAUUnmapSurfacesNV MANGLE(VDPAUUnmapSurfacesNV)
#define glVDPAUUnregisterSurfaceNV MANGLE(VDPAUUnregisterSurfaceNV)
#define glVertex2d MANGLE(Vertex2d)
#define glVertex2dv MANGLE(Vertex2dv)
#define glVertex2f MANGLE(Vertex2f)
@@ -1964,7 +1692,6 @@
#define glVertexArrayParameteriAPPLE MANGLE(VertexArrayParameteriAPPLE)
#define glVertexArrayRangeAPPLE MANGLE(VertexArrayRangeAPPLE)
#define glVertexArrayRangeNV MANGLE(VertexArrayRangeNV)
#define glVertexArrayVertexAttribLOffsetEXT MANGLE(VertexArrayVertexAttribLOffsetEXT)
#define glVertexAttrib1dARB MANGLE(VertexAttrib1dARB)
#define glVertexAttrib1d MANGLE(VertexAttrib1d)
#define glVertexAttrib1dNV MANGLE(VertexAttrib1dNV)
@@ -2073,7 +1800,6 @@
#define glVertexAttrib4usv MANGLE(VertexAttrib4usv)
#define glVertexAttribArrayObjectATI MANGLE(VertexAttribArrayObjectATI)
#define glVertexAttribDivisorARB MANGLE(VertexAttribDivisorARB)
#define glVertexAttribDivisor MANGLE(VertexAttribDivisor)
#define glVertexAttribFormatNV MANGLE(VertexAttribFormatNV)
#define glVertexAttribI1iEXT MANGLE(VertexAttribI1iEXT)
#define glVertexAttribI1i MANGLE(VertexAttribI1i)
@@ -2118,49 +1844,6 @@
#define glVertexAttribIFormatNV MANGLE(VertexAttribIFormatNV)
#define glVertexAttribIPointerEXT MANGLE(VertexAttribIPointerEXT)
#define glVertexAttribIPointer MANGLE(VertexAttribIPointer)
#define glVertexAttribL1dEXT MANGLE(VertexAttribL1dEXT)
#define glVertexAttribL1d MANGLE(VertexAttribL1d)
#define glVertexAttribL1dvEXT MANGLE(VertexAttribL1dvEXT)
#define glVertexAttribL1dv MANGLE(VertexAttribL1dv)
#define glVertexAttribL1i64NV MANGLE(VertexAttribL1i64NV)
#define glVertexAttribL1i64vNV MANGLE(VertexAttribL1i64vNV)
#define glVertexAttribL1ui64NV MANGLE(VertexAttribL1ui64NV)
#define glVertexAttribL1ui64vNV MANGLE(VertexAttribL1ui64vNV)
#define glVertexAttribL2dEXT MANGLE(VertexAttribL2dEXT)
#define glVertexAttribL2d MANGLE(VertexAttribL2d)
#define glVertexAttribL2dvEXT MANGLE(VertexAttribL2dvEXT)
#define glVertexAttribL2dv MANGLE(VertexAttribL2dv)
#define glVertexAttribL2i64NV MANGLE(VertexAttribL2i64NV)
#define glVertexAttribL2i64vNV MANGLE(VertexAttribL2i64vNV)
#define glVertexAttribL2ui64NV MANGLE(VertexAttribL2ui64NV)
#define glVertexAttribL2ui64vNV MANGLE(VertexAttribL2ui64vNV)
#define glVertexAttribL3dEXT MANGLE(VertexAttribL3dEXT)
#define glVertexAttribL3d MANGLE(VertexAttribL3d)
#define glVertexAttribL3dvEXT MANGLE(VertexAttribL3dvEXT)
#define glVertexAttribL3dv MANGLE(VertexAttribL3dv)
#define glVertexAttribL3i64NV MANGLE(VertexAttribL3i64NV)
#define glVertexAttribL3i64vNV MANGLE(VertexAttribL3i64vNV)
#define glVertexAttribL3ui64NV MANGLE(VertexAttribL3ui64NV)
#define glVertexAttribL3ui64vNV MANGLE(VertexAttribL3ui64vNV)
#define glVertexAttribL4dEXT MANGLE(VertexAttribL4dEXT)
#define glVertexAttribL4d MANGLE(VertexAttribL4d)
#define glVertexAttribL4dvEXT MANGLE(VertexAttribL4dvEXT)
#define glVertexAttribL4dv MANGLE(VertexAttribL4dv)
#define glVertexAttribL4i64NV MANGLE(VertexAttribL4i64NV)
#define glVertexAttribL4i64vNV MANGLE(VertexAttribL4i64vNV)
#define glVertexAttribL4ui64NV MANGLE(VertexAttribL4ui64NV)
#define glVertexAttribL4ui64vNV MANGLE(VertexAttribL4ui64vNV)
#define glVertexAttribLFormatNV MANGLE(VertexAttribLFormatNV)
#define glVertexAttribLPointerEXT MANGLE(VertexAttribLPointerEXT)
#define glVertexAttribLPointer MANGLE(VertexAttribLPointer)
#define glVertexAttribP1ui MANGLE(VertexAttribP1ui)
#define glVertexAttribP1uiv MANGLE(VertexAttribP1uiv)
#define glVertexAttribP2ui MANGLE(VertexAttribP2ui)
#define glVertexAttribP2uiv MANGLE(VertexAttribP2uiv)
#define glVertexAttribP3ui MANGLE(VertexAttribP3ui)
#define glVertexAttribP3uiv MANGLE(VertexAttribP3uiv)
#define glVertexAttribP4ui MANGLE(VertexAttribP4ui)
#define glVertexAttribP4uiv MANGLE(VertexAttribP4uiv)
#define glVertexAttribPointerARB MANGLE(VertexAttribPointerARB)
#define glVertexAttribPointer MANGLE(VertexAttribPointer)
#define glVertexAttribPointerNV MANGLE(VertexAttribPointerNV)
@@ -2185,12 +1868,6 @@
#define glVertexBlendEnvfATI MANGLE(VertexBlendEnvfATI)
#define glVertexBlendEnviATI MANGLE(VertexBlendEnviATI)
#define glVertexFormatNV MANGLE(VertexFormatNV)
#define glVertexP2ui MANGLE(VertexP2ui)
#define glVertexP2uiv MANGLE(VertexP2uiv)
#define glVertexP3ui MANGLE(VertexP3ui)
#define glVertexP3uiv MANGLE(VertexP3uiv)
#define glVertexP4ui MANGLE(VertexP4ui)
#define glVertexP4uiv MANGLE(VertexP4uiv)
#define glVertexPointerEXT MANGLE(VertexPointerEXT)
#define glVertexPointerListIBM MANGLE(VertexPointerListIBM)
#define glVertexPointer MANGLE(VertexPointer)
@@ -2236,9 +1913,6 @@
#define glVideoCaptureStreamParameterdvNV MANGLE(VideoCaptureStreamParameterdvNV)
#define glVideoCaptureStreamParameterfvNV MANGLE(VideoCaptureStreamParameterfvNV)
#define glVideoCaptureStreamParameterivNV MANGLE(VideoCaptureStreamParameterivNV)
#define glViewportArrayv MANGLE(ViewportArrayv)
#define glViewportIndexedf MANGLE(ViewportIndexedf)
#define glViewportIndexedfv MANGLE(ViewportIndexedfv)
#define glViewport MANGLE(Viewport)
#define glWaitSync MANGLE(WaitSync)
#define glWeightbvARB MANGLE(WeightbvARB)

View File

@@ -29,9 +29,9 @@ extern "C" {
*/
/* Header file version number, required by OpenGL ABI for Linux */
/* glext.h last updated $Date: 2010-11-03 18:59:30 -0700 (Wed, 03 Nov 2010) $ */
/* glext.h last updated $Date: 2010-08-03 01:30:25 -0700 (Tue, 03 Aug 2010) $ */
/* Current version at http://www.opengl.org/registry/ */
#define GL_GLEXT_VERSION 66
#define GL_GLEXT_VERSION 64
/* Function declaration macros - to move into glplatform.h */
#if defined(_WIN32) && !defined(APIENTRY) && !defined(__CYGWIN__) && !defined(__SCITECH_SNAP__)
@@ -4840,7 +4840,7 @@ extern "C" {
#endif
#ifndef GL_AMD_seamless_cubemap_per_texture
/* reuse GL_TEXTURE_CUBE_MAP_SEAMLESS */
/* reuse GL_TEXTURE_CUBE_MAP_SEAMLESS_ARB */
#endif
#ifndef GL_AMD_conservative_depth
@@ -4925,8 +4925,6 @@ extern "C" {
#define GL_MIN_FRAGMENT_INTERPOLATION_OFFSET_NV 0x8E5B
#define GL_MAX_FRAGMENT_INTERPOLATION_OFFSET_NV 0x8E5C
#define GL_FRAGMENT_PROGRAM_INTERPOLATION_OFFSET_BITS_NV 0x8E5D
#define GL_MIN_PROGRAM_TEXTURE_GATHER_OFFSET_NV 0x8E5E
#define GL_MAX_PROGRAM_TEXTURE_GATHER_OFFSET_NV 0x8E5F
#define GL_MAX_PROGRAM_SUBROUTINE_PARAMETERS_NV 0x8F44
#define GL_MAX_PROGRAM_SUBROUTINE_NUM_NV 0x8F45
#endif
@@ -5021,11 +5019,6 @@ extern "C" {
#ifndef GL_AMD_transform_feedback3_lines_triangles
#endif
#ifndef GL_AMD_depth_clamp_separate
#define GL_DEPTH_CLAMP_NEAR_AMD 0x901E
#define GL_DEPTH_CLAMP_FAR_AMD 0x901F
#endif
/*************************************************************/
@@ -8772,8 +8765,8 @@ GLAPI void APIENTRY glProgramParameter4dNV (GLenum target, GLuint index, GLdoubl
GLAPI void APIENTRY glProgramParameter4dvNV (GLenum target, GLuint index, const GLdouble *v);
GLAPI void APIENTRY glProgramParameter4fNV (GLenum target, GLuint index, GLfloat x, GLfloat y, GLfloat z, GLfloat w);
GLAPI void APIENTRY glProgramParameter4fvNV (GLenum target, GLuint index, const GLfloat *v);
GLAPI void APIENTRY glProgramParameters4dvNV (GLenum target, GLuint index, GLsizei count, const GLdouble *v);
GLAPI void APIENTRY glProgramParameters4fvNV (GLenum target, GLuint index, GLsizei count, const GLfloat *v);
GLAPI void APIENTRY glProgramParameters4dvNV (GLenum target, GLuint index, GLuint count, const GLdouble *v);
GLAPI void APIENTRY glProgramParameters4fvNV (GLenum target, GLuint index, GLuint count, const GLfloat *v);
GLAPI void APIENTRY glRequestResidentProgramsNV (GLsizei n, const GLuint *programs);
GLAPI void APIENTRY glTrackMatrixNV (GLenum target, GLuint address, GLenum matrix, GLenum transform);
GLAPI void APIENTRY glVertexAttribPointerNV (GLuint index, GLint fsize, GLenum type, GLsizei stride, const GLvoid *pointer);
@@ -8837,8 +8830,8 @@ typedef void (APIENTRYP PFNGLPROGRAMPARAMETER4DNVPROC) (GLenum target, GLuint in
typedef void (APIENTRYP PFNGLPROGRAMPARAMETER4DVNVPROC) (GLenum target, GLuint index, const GLdouble *v);
typedef void (APIENTRYP PFNGLPROGRAMPARAMETER4FNVPROC) (GLenum target, GLuint index, GLfloat x, GLfloat y, GLfloat z, GLfloat w);
typedef void (APIENTRYP PFNGLPROGRAMPARAMETER4FVNVPROC) (GLenum target, GLuint index, const GLfloat *v);
typedef void (APIENTRYP PFNGLPROGRAMPARAMETERS4DVNVPROC) (GLenum target, GLuint index, GLsizei count, const GLdouble *v);
typedef void (APIENTRYP PFNGLPROGRAMPARAMETERS4FVNVPROC) (GLenum target, GLuint index, GLsizei count, const GLfloat *v);
typedef void (APIENTRYP PFNGLPROGRAMPARAMETERS4DVNVPROC) (GLenum target, GLuint index, GLuint count, const GLdouble *v);
typedef void (APIENTRYP PFNGLPROGRAMPARAMETERS4FVNVPROC) (GLenum target, GLuint index, GLuint count, const GLfloat *v);
typedef void (APIENTRYP PFNGLREQUESTRESIDENTPROGRAMSNVPROC) (GLsizei n, const GLuint *programs);
typedef void (APIENTRYP PFNGLTRACKMATRIXNVPROC) (GLenum target, GLuint address, GLenum matrix, GLenum transform);
typedef void (APIENTRYP PFNGLVERTEXATTRIBPOINTERNVPROC) (GLuint index, GLint fsize, GLenum type, GLsizei stride, const GLvoid *pointer);
@@ -11027,10 +11020,6 @@ typedef void (APIENTRYP PFNGLVDPAUUNMAPSURFACESNVPROC) (GLsizei numSurface, cons
#define GL_AMD_transform_feedback3_lines_triangles 1
#endif
#ifndef GL_AMD_depth_clamp_separate
#define GL_AMD_depth_clamp_separate 1
#endif
#ifdef __cplusplus
}

View File

@@ -1,8 +1,8 @@
/* $Revision: 9203 $ on $Date:: 2009-10-07 02:21:52 -0700 #$ */
/* $Revision: 6822 $ on $Date:: 2008-10-30 05:14:19 -0400 #$ */
/*------------------------------------------------------------------------
*
* OpenVG 1.1 Reference Implementation
* OpenVG 1.0.1 Reference Implementation
* -------------------------------------
*
* Copyright (c) 2008 The Khronos Group Inc.
@@ -28,7 +28,7 @@
*
*//**
* \file
* \brief OpenVG 1.1 API.
* \brief OpenVG 1.0.1 API.
*//*-------------------------------------------------------------------*/
#ifndef _OPENVG_H
@@ -42,7 +42,6 @@ extern "C" {
#define OPENVG_VERSION_1_0 1
#define OPENVG_VERSION_1_0_1 1
#define OPENVG_VERSION_1_1 2
#ifndef VG_MAXSHORT
#define VG_MAXSHORT 0x7FFF
@@ -56,12 +55,10 @@ extern "C" {
#define VG_MAX_ENUM 0x7FFFFFFF
#endif
typedef VGuint VGHandle;
typedef long VGHandle;
typedef VGHandle VGPath;
typedef VGHandle VGImage;
typedef VGHandle VGMaskLayer;
typedef VGHandle VGFont;
typedef VGHandle VGPaint;
#define VG_INVALID_HANDLE ((VGHandle)0)
@@ -99,10 +96,6 @@ typedef enum {
/* Scissoring rectangles */
VG_SCISSOR_RECTS = 0x1106,
/* Color Transformation */
VG_COLOR_TRANSFORM = 0x1170,
VG_COLOR_TRANSFORM_VALUES = 0x1171,
/* Stroke parameters */
VG_STROKE_LINE_WIDTH = 0x1110,
VG_STROKE_CAP_STYLE = 0x1111,
@@ -118,9 +111,6 @@ typedef enum {
/* Color for vgClear */
VG_CLEAR_COLOR = 0x1121,
/* Glyph origin */
VG_GLYPH_ORIGIN = 0x1122,
/* Enable/disable alpha masking and scissoring */
VG_MASKING = 0x1130,
VG_SCISSORING = 0x1131,
@@ -175,7 +165,6 @@ typedef enum {
VG_MATRIX_IMAGE_USER_TO_SURFACE = 0x1401,
VG_MATRIX_FILL_PAINT_TO_USER = 0x1402,
VG_MATRIX_STROKE_PAINT_TO_USER = 0x1403,
VG_MATRIX_GLYPH_USER_TO_SURFACE = 0x1404,
VG_MATRIX_MODE_FORCE_SIZE = VG_MAX_ENUM
} VGMatrixMode;
@@ -376,8 +365,6 @@ typedef enum {
VG_lL_8 = 10,
VG_A_8 = 11,
VG_BW_1 = 12,
VG_A_1 = 13,
VG_A_4 = 14,
/* {A,X}RGB channel ordering */
VG_sXRGB_8888 = 0 | (1 << 6),
@@ -461,12 +448,6 @@ typedef enum {
VG_BLEND_MODE_FORCE_SIZE = VG_MAX_ENUM
} VGBlendMode;
typedef enum {
VG_FONT_NUM_GLYPHS = 0x2F00,
VG_FONT_PARAM_TYPE_FORCE_SIZE = VG_MAX_ENUM
} VGFontParamType;
typedef enum {
VG_IMAGE_FORMAT_QUERY = 0x2100,
VG_PATH_DATATYPE_QUERY = 0x2101,
@@ -560,22 +541,8 @@ VG_API_CALL void VG_API_ENTRY vgShear(VGfloat shx, VGfloat shy) VG_API_EXIT;
VG_API_CALL void VG_API_ENTRY vgRotate(VGfloat angle) VG_API_EXIT;
/* Masking and Clearing */
VG_API_CALL void VG_API_ENTRY vgMask(VGHandle mask, VGMaskOperation operation,
VGint x, VGint y,
VGint width, VGint height) VG_API_EXIT;
VG_API_CALL void VG_API_ENTRY vgRenderToMask(VGPath path,
VGbitfield paintModes,
VGMaskOperation operation) VG_API_EXIT;
VG_API_CALL VGMaskLayer VG_API_ENTRY vgCreateMaskLayer(VGint width, VGint height) VG_API_EXIT;
VG_API_CALL void VG_API_ENTRY vgDestroyMaskLayer(VGMaskLayer maskLayer) VG_API_EXIT;
VG_API_CALL void VG_API_ENTRY vgFillMaskLayer(VGMaskLayer maskLayer,
VGint x, VGint y,
VGint width, VGint height,
VGfloat value) VG_API_EXIT;
VG_API_CALL void VG_API_ENTRY vgCopyMask(VGMaskLayer maskLayer,
VGint dx, VGint dy,
VGint sx, VGint sy,
VGint width, VGint height) VG_API_EXIT;
VG_API_CALL void VG_API_ENTRY vgMask(VGImage mask, VGMaskOperation operation,
VGint x, VGint y, VGint width, VGint height) VG_API_EXIT;
VG_API_CALL void VG_API_ENTRY vgClear(VGint x, VGint y, VGint width, VGint height) VG_API_EXIT;
/* Paths */
@@ -669,33 +636,6 @@ VG_API_CALL void VG_API_ENTRY vgCopyPixels(VGint dx, VGint dy,
VGint sx, VGint sy,
VGint width, VGint height) VG_API_EXIT;
/* Text */
VG_API_CALL VGFont VG_API_ENTRY vgCreateFont(VGint glyphCapacityHint) VG_API_EXIT;
VG_API_CALL void VG_API_ENTRY vgDestroyFont(VGFont font) VG_API_EXIT;
VG_API_CALL void VG_API_ENTRY vgSetGlyphToPath(VGFont font,
VGuint glyphIndex,
VGPath path,
VGboolean isHinted,
const VGfloat glyphOrigin [2],
const VGfloat escapement[2]) VG_API_EXIT;
VG_API_CALL void VG_API_ENTRY vgSetGlyphToImage(VGFont font,
VGuint glyphIndex,
VGImage image,
const VGfloat glyphOrigin [2],
const VGfloat escapement[2]) VG_API_EXIT;
VG_API_CALL void VG_API_ENTRY vgClearGlyph(VGFont font,VGuint glyphIndex) VG_API_EXIT;
VG_API_CALL void VG_API_ENTRY vgDrawGlyph(VGFont font,
VGuint glyphIndex,
VGbitfield paintModes,
VGboolean allowAutoHinting) VG_API_EXIT;
VG_API_CALL void VG_API_ENTRY vgDrawGlyphs(VGFont font,
VGint glyphCount,
const VGuint *glyphIndices,
const VGfloat *adjustments_x,
const VGfloat *adjustments_y,
VGbitfield paintModes,
VGboolean allowAutoHinting) VG_API_EXIT;
/* Image Filters */
VG_API_CALL void VG_API_ENTRY vgColorMatrix(VGImage dst, VGImage src,
const VGfloat * matrix) VG_API_EXIT;

View File

@@ -1,233 +1,233 @@
/* $Revision: 6810 $ on $Date:: 2008-10-29 07:31:37 -0700 #$ */
/*------------------------------------------------------------------------
*
* VG extensions Reference Implementation
* -------------------------------------
*
* Copyright (c) 2008 The Khronos Group Inc.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and /or associated documentation files
* (the "Materials "), to deal in the Materials without restriction,
* including without limitation the rights to use, copy, modify, merge,
* publish, distribute, sublicense, and/or sell copies of the Materials,
* and to permit persons to whom the Materials are furnished to do so,
* subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included
* in all copies or substantial portions of the Materials.
*
* THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
* DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
* OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE MATERIALS OR
* THE USE OR OTHER DEALINGS IN THE MATERIALS.
*
*//**
* \file
* \brief VG extensions
*//*-------------------------------------------------------------------*/
#ifndef _VGEXT_H
#define _VGEXT_H
#ifdef __cplusplus
extern "C" {
#endif
#include <VG/openvg.h>
#include <VG/vgu.h>
#ifndef VG_API_ENTRYP
# define VG_API_ENTRYP VG_API_ENTRY*
#endif
#ifndef VGU_API_ENTRYP
# define VGU_API_ENTRYP VGU_API_ENTRY*
#endif
/*-------------------------------------------------------------------------------
* KHR extensions
*------------------------------------------------------------------------------*/
typedef enum {
#ifndef VG_KHR_iterative_average_blur
VG_MAX_AVERAGE_BLUR_DIMENSION_KHR = 0x116B,
VG_AVERAGE_BLUR_DIMENSION_RESOLUTION_KHR = 0x116C,
VG_MAX_AVERAGE_BLUR_ITERATIONS_KHR = 0x116D,
#endif
VG_PARAM_TYPE_KHR_FORCE_SIZE = VG_MAX_ENUM
} VGParamTypeKHR;
#ifndef VG_KHR_EGL_image
#define VG_KHR_EGL_image 1
/* VGEGLImageKHR is an opaque handle to an EGLImage */
typedef void* VGeglImageKHR;
#ifdef VG_VGEXT_PROTOTYPES
VG_API_CALL VGImage VG_API_ENTRY vgCreateEGLImageTargetKHR(VGeglImageKHR image);
#endif
typedef VGImage (VG_API_ENTRYP PFNVGCREATEEGLIMAGETARGETKHRPROC) (VGeglImageKHR image);
#endif
#ifndef VG_KHR_iterative_average_blur
#define VG_KHR_iterative_average_blur 1
#ifdef VG_VGEXT_PROTOTYPES
VG_API_CALL void vgIterativeAverageBlurKHR(VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGTilingMode tilingMode);
#endif
typedef void (VG_API_ENTRYP PFNVGITERATIVEAVERAGEBLURKHRPROC) (VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGTilingMode tilingMode);
#endif
#ifndef VG_KHR_advanced_blending
#define VG_KHR_advanced_blending 1
typedef enum {
VG_BLEND_OVERLAY_KHR = 0x2010,
VG_BLEND_HARDLIGHT_KHR = 0x2011,
VG_BLEND_SOFTLIGHT_SVG_KHR = 0x2012,
VG_BLEND_SOFTLIGHT_KHR = 0x2013,
VG_BLEND_COLORDODGE_KHR = 0x2014,
VG_BLEND_COLORBURN_KHR = 0x2015,
VG_BLEND_DIFFERENCE_KHR = 0x2016,
VG_BLEND_SUBTRACT_KHR = 0x2017,
VG_BLEND_INVERT_KHR = 0x2018,
VG_BLEND_EXCLUSION_KHR = 0x2019,
VG_BLEND_LINEARDODGE_KHR = 0x201a,
VG_BLEND_LINEARBURN_KHR = 0x201b,
VG_BLEND_VIVIDLIGHT_KHR = 0x201c,
VG_BLEND_LINEARLIGHT_KHR = 0x201d,
VG_BLEND_PINLIGHT_KHR = 0x201e,
VG_BLEND_HARDMIX_KHR = 0x201f,
VG_BLEND_CLEAR_KHR = 0x2020,
VG_BLEND_DST_KHR = 0x2021,
VG_BLEND_SRC_OUT_KHR = 0x2022,
VG_BLEND_DST_OUT_KHR = 0x2023,
VG_BLEND_SRC_ATOP_KHR = 0x2024,
VG_BLEND_DST_ATOP_KHR = 0x2025,
VG_BLEND_XOR_KHR = 0x2026,
VG_BLEND_MODE_KHR_FORCE_SIZE= VG_MAX_ENUM
} VGBlendModeKHR;
#endif
#ifndef VG_KHR_parametric_filter
#define VG_KHR_parametric_filter 1
typedef enum {
VG_PF_OBJECT_VISIBLE_FLAG_KHR = (1 << 0),
VG_PF_KNOCKOUT_FLAG_KHR = (1 << 1),
VG_PF_OUTER_FLAG_KHR = (1 << 2),
VG_PF_INNER_FLAG_KHR = (1 << 3),
VG_PF_TYPE_KHR_FORCE_SIZE = VG_MAX_ENUM
} VGPfTypeKHR;
typedef enum {
VGU_IMAGE_IN_USE_ERROR = 0xF010,
VGU_ERROR_CODE_KHR_FORCE_SIZE = VG_MAX_ENUM
} VGUErrorCodeKHR;
#ifdef VG_VGEXT_PROTOTYPES
VG_API_CALL void VG_API_ENTRY vgParametricFilterKHR(VGImage dst,VGImage src,VGImage blur,VGfloat strength,VGfloat offsetX,VGfloat offsetY,VGbitfield filterFlags,VGPaint highlightPaint,VGPaint shadowPaint);
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguDropShadowKHR(VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint shadowColorRGBA);
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguGlowKHR(VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint glowColorRGBA) ;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguBevelKHR(VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint highlightColorRGBA,VGuint shadowColorRGBA);
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguGradientGlowKHR(VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint stopsCount,const VGfloat* glowColorRampStops);
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguGradientBevelKHR(VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint stopsCount,const VGfloat* bevelColorRampStops);
#endif
typedef void (VG_API_ENTRYP PFNVGPARAMETRICFILTERKHRPROC) (VGImage dst,VGImage src,VGImage blur,VGfloat strength,VGfloat offsetX,VGfloat offsetY,VGbitfield filterFlags,VGPaint highlightPaint,VGPaint shadowPaint);
typedef VGUErrorCode (VGU_API_ENTRYP PFNVGUDROPSHADOWKHRPROC) (VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint shadowColorRGBA);
typedef VGUErrorCode (VGU_API_ENTRYP PFNVGUGLOWKHRPROC) (VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint glowColorRGBA);
typedef VGUErrorCode (VGU_API_ENTRYP PFNVGUBEVELKHRPROC) (VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint highlightColorRGBA,VGuint shadowColorRGBA);
typedef VGUErrorCode (VGU_API_ENTRYP PFNVGUGRADIENTGLOWKHRPROC) (VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint stopsCount,const VGfloat* glowColorRampStops);
typedef VGUErrorCode (VGU_API_ENTRYP PFNVGUGRADIENTBEVELKHRPROC) (VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint stopsCount,const VGfloat* bevelColorRampStops);
#endif
/*-------------------------------------------------------------------------------
* NDS extensions
*------------------------------------------------------------------------------*/
#ifndef VG_NDS_paint_generation
#define VG_NDS_paint_generation 1
typedef enum {
VG_PAINT_COLOR_RAMP_LINEAR_NDS = 0x1A10,
VG_COLOR_MATRIX_NDS = 0x1A11,
VG_PAINT_COLOR_TRANSFORM_LINEAR_NDS = 0x1A12,
VG_PAINT_PARAM_TYPE_NDS_FORCE_SIZE = VG_MAX_ENUM
} VGPaintParamTypeNds;
typedef enum {
VG_DRAW_IMAGE_COLOR_MATRIX_NDS = 0x1F10,
VG_IMAGE_MODE_NDS_FORCE_SIZE = VG_MAX_ENUM
} VGImageModeNds;
#endif
#ifndef VG_NDS_projective_geometry
#define VG_NDS_projective_geometry 1
typedef enum {
VG_CLIP_MODE_NDS = 0x1180,
VG_CLIP_LINES_NDS = 0x1181,
VG_MAX_CLIP_LINES_NDS = 0x1182,
VG_PARAM_TYPE_NDS_FORCE_SIZE = VG_MAX_ENUM
} VGParamTypeNds;
typedef enum {
VG_CLIPMODE_NONE_NDS = 0x3000,
VG_CLIPMODE_CLIP_CLOSED_NDS = 0x3001,
VG_CLIPMODE_CLIP_OPEN_NDS = 0x3002,
VG_CLIPMODE_CULL_NDS = 0x3003,
VG_CLIPMODE_NDS_FORCE_SIZE = VG_MAX_ENUM
} VGClipModeNds;
typedef enum {
VG_RQUAD_TO_NDS = ( 13 << 1 ),
VG_RCUBIC_TO_NDS = ( 14 << 1 ),
VG_PATH_SEGMENT_NDS_FORCE_SIZE = VG_MAX_ENUM
} VGPathSegmentNds;
typedef enum {
VG_RQUAD_TO_ABS_NDS = (VG_RQUAD_TO_NDS | VG_ABSOLUTE),
VG_RQUAD_TO_REL_NDS = (VG_RQUAD_TO_NDS | VG_RELATIVE),
VG_RCUBIC_TO_ABS_NDS = (VG_RCUBIC_TO_NDS | VG_ABSOLUTE),
VG_RCUBIC_TO_REL_NDS = (VG_RCUBIC_TO_NDS | VG_RELATIVE),
VG_PATH_COMMAND_NDS_FORCE_SIZE = VG_MAX_ENUM
} VGPathCommandNds;
#ifdef VG_VGEXT_PROTOTYPES
VG_API_CALL void VG_API_ENTRY vgProjectiveMatrixNDS(VGboolean enable) ;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguTransformClipLineNDS(const VGfloat Ain,const VGfloat Bin,const VGfloat Cin,const VGfloat* matrix,const VGboolean inverse,VGfloat* Aout,VGfloat* Bout,VGfloat* Cout);
#endif
typedef void (VG_API_ENTRYP PFNVGPROJECTIVEMATRIXNDSPROC) (VGboolean enable) ;
typedef VGUErrorCode (VGU_API_ENTRYP PFNVGUTRANSFORMCLIPLINENDSPROC) (const VGfloat Ain,const VGfloat Bin,const VGfloat Cin,const VGfloat* matrix,const VGboolean inverse,VGfloat* Aout,VGfloat* Bout,VGfloat* Cout);
#endif
#ifdef __cplusplus
} /* extern "C" */
#endif
#endif /* _VGEXT_H */
/* $Revision: 6810 $ on $Date:: 2008-10-29 10:31:37 -0400 #$ */
/*------------------------------------------------------------------------
*
* VG extensions Reference Implementation
* -------------------------------------
*
* Copyright (c) 2008 The Khronos Group Inc.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and /or associated documentation files
* (the "Materials "), to deal in the Materials without restriction,
* including without limitation the rights to use, copy, modify, merge,
* publish, distribute, sublicense, and/or sell copies of the Materials,
* and to permit persons to whom the Materials are furnished to do so,
* subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included
* in all copies or substantial portions of the Materials.
*
* THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
* DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
* OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE MATERIALS OR
* THE USE OR OTHER DEALINGS IN THE MATERIALS.
*
*//**
* \file
* \brief VG extensions
*//*-------------------------------------------------------------------*/
#ifndef _VGEXT_H
#define _VGEXT_H
#ifdef __cplusplus
extern "C" {
#endif
#include <VG/openvg.h>
#include <VG/vgu.h>
#ifndef VG_API_ENTRYP
# define VG_API_ENTRYP VG_API_ENTRY*
#endif
#ifndef VGU_API_ENTRYP
# define VGU_API_ENTRYP VGU_API_ENTRY*
#endif
/*-------------------------------------------------------------------------------
* KHR extensions
*------------------------------------------------------------------------------*/
typedef enum {
#ifndef VG_KHR_iterative_average_blur
VG_MAX_AVERAGE_BLUR_DIMENSION_KHR = 0x116B,
VG_AVERAGE_BLUR_DIMENSION_RESOLUTION_KHR = 0x116C,
VG_MAX_AVERAGE_BLUR_ITERATIONS_KHR = 0x116D,
#endif
VG_PARAM_TYPE_KHR_FORCE_SIZE = VG_MAX_ENUM
} VGParamTypeKHR;
#ifndef VG_KHR_EGL_image
#define VG_KHR_EGL_image 1
/* VGEGLImageKHR is an opaque handle to an EGLImage */
typedef void* VGeglImageKHR;
#ifdef VG_VGEXT_PROTOTYPES
VG_API_CALL VGImage VG_API_ENTRY vgCreateEGLImageTargetKHR(VGeglImageKHR image);
#endif
typedef VGImage (VG_API_ENTRYP PFNVGCREATEEGLIMAGETARGETKHRPROC) (VGeglImageKHR image);
#endif
#ifndef VG_KHR_iterative_average_blur
#define VG_KHR_iterative_average_blur 1
#ifdef VG_VGEXT_PROTOTYPES
VG_API_CALL void vgIterativeAverageBlurKHR(VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGTilingMode tilingMode);
#endif
typedef void (VG_API_ENTRYP PFNVGITERATIVEAVERAGEBLURKHRPROC) (VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGTilingMode tilingMode);
#endif
#ifndef VG_KHR_advanced_blending
#define VG_KHR_advanced_blending 1
typedef enum {
VG_BLEND_OVERLAY_KHR = 0x2010,
VG_BLEND_HARDLIGHT_KHR = 0x2011,
VG_BLEND_SOFTLIGHT_SVG_KHR = 0x2012,
VG_BLEND_SOFTLIGHT_KHR = 0x2013,
VG_BLEND_COLORDODGE_KHR = 0x2014,
VG_BLEND_COLORBURN_KHR = 0x2015,
VG_BLEND_DIFFERENCE_KHR = 0x2016,
VG_BLEND_SUBTRACT_KHR = 0x2017,
VG_BLEND_INVERT_KHR = 0x2018,
VG_BLEND_EXCLUSION_KHR = 0x2019,
VG_BLEND_LINEARDODGE_KHR = 0x201a,
VG_BLEND_LINEARBURN_KHR = 0x201b,
VG_BLEND_VIVIDLIGHT_KHR = 0x201c,
VG_BLEND_LINEARLIGHT_KHR = 0x201d,
VG_BLEND_PINLIGHT_KHR = 0x201e,
VG_BLEND_HARDMIX_KHR = 0x201f,
VG_BLEND_CLEAR_KHR = 0x2020,
VG_BLEND_DST_KHR = 0x2021,
VG_BLEND_SRC_OUT_KHR = 0x2022,
VG_BLEND_DST_OUT_KHR = 0x2023,
VG_BLEND_SRC_ATOP_KHR = 0x2024,
VG_BLEND_DST_ATOP_KHR = 0x2025,
VG_BLEND_XOR_KHR = 0x2026,
VG_BLEND_MODE_KHR_FORCE_SIZE= VG_MAX_ENUM
} VGBlendModeKHR;
#endif
#ifndef VG_KHR_parametric_filter
#define VG_KHR_parametric_filter 1
typedef enum {
VG_PF_OBJECT_VISIBLE_FLAG_KHR = (1 << 0),
VG_PF_KNOCKOUT_FLAG_KHR = (1 << 1),
VG_PF_OUTER_FLAG_KHR = (1 << 2),
VG_PF_INNER_FLAG_KHR = (1 << 3),
VG_PF_TYPE_KHR_FORCE_SIZE = VG_MAX_ENUM
} VGPfTypeKHR;
typedef enum {
VGU_IMAGE_IN_USE_ERROR = 0xF010,
VGU_ERROR_CODE_KHR_FORCE_SIZE = VG_MAX_ENUM
} VGUErrorCodeKHR;
#ifdef VG_VGEXT_PROTOTYPES
VG_API_CALL void VG_API_ENTRY vgParametricFilterKHR(VGImage dst,VGImage src,VGImage blur,VGfloat strength,VGfloat offsetX,VGfloat offsetY,VGbitfield filterFlags,VGPaint highlightPaint,VGPaint shadowPaint);
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguDropShadowKHR(VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint shadowColorRGBA);
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguGlowKHR(VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint glowColorRGBA) ;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguBevelKHR(VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint highlightColorRGBA,VGuint shadowColorRGBA);
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguGradientGlowKHR(VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint stopsCount,const VGfloat* glowColorRampStops);
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguGradientBevelKHR(VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint stopsCount,const VGfloat* bevelColorRampStops);
#endif
typedef void (VG_API_ENTRYP PFNVGPARAMETRICFILTERKHRPROC) (VGImage dst,VGImage src,VGImage blur,VGfloat strength,VGfloat offsetX,VGfloat offsetY,VGbitfield filterFlags,VGPaint highlightPaint,VGPaint shadowPaint);
typedef VGUErrorCode (VGU_API_ENTRYP PFNVGUDROPSHADOWKHRPROC) (VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint shadowColorRGBA);
typedef VGUErrorCode (VGU_API_ENTRYP PFNVGUGLOWKHRPROC) (VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint glowColorRGBA);
typedef VGUErrorCode (VGU_API_ENTRYP PFNVGUBEVELKHRPROC) (VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint highlightColorRGBA,VGuint shadowColorRGBA);
typedef VGUErrorCode (VGU_API_ENTRYP PFNVGUGRADIENTGLOWKHRPROC) (VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint stopsCount,const VGfloat* glowColorRampStops);
typedef VGUErrorCode (VGU_API_ENTRYP PFNVGUGRADIENTBEVELKHRPROC) (VGImage dst,VGImage src,VGfloat dimX,VGfloat dimY,VGuint iterative,VGfloat strength,VGfloat distance,VGfloat angle,VGbitfield filterFlags,VGbitfield allowedQuality,VGuint stopsCount,const VGfloat* bevelColorRampStops);
#endif
/*-------------------------------------------------------------------------------
* NDS extensions
*------------------------------------------------------------------------------*/
#ifndef VG_NDS_paint_generation
#define VG_NDS_paint_generation 1
typedef enum {
VG_PAINT_COLOR_RAMP_LINEAR_NDS = 0x1A10,
VG_COLOR_MATRIX_NDS = 0x1A11,
VG_PAINT_COLOR_TRANSFORM_LINEAR_NDS = 0x1A12,
VG_PAINT_PARAM_TYPE_NDS_FORCE_SIZE = VG_MAX_ENUM
} VGPaintParamTypeNds;
typedef enum {
VG_DRAW_IMAGE_COLOR_MATRIX_NDS = 0x1F10,
VG_IMAGE_MODE_NDS_FORCE_SIZE = VG_MAX_ENUM
} VGImageModeNds;
#endif
#ifndef VG_NDS_projective_geometry
#define VG_NDS_projective_geometry 1
typedef enum {
VG_CLIP_MODE_NDS = 0x1180,
VG_CLIP_LINES_NDS = 0x1181,
VG_MAX_CLIP_LINES_NDS = 0x1182,
VG_PARAM_TYPE_NDS_FORCE_SIZE = VG_MAX_ENUM
} VGParamTypeNds;
typedef enum {
VG_CLIPMODE_NONE_NDS = 0x3000,
VG_CLIPMODE_CLIP_CLOSED_NDS = 0x3001,
VG_CLIPMODE_CLIP_OPEN_NDS = 0x3002,
VG_CLIPMODE_CULL_NDS = 0x3003,
VG_CLIPMODE_NDS_FORCE_SIZE = VG_MAX_ENUM
} VGClipModeNds;
typedef enum {
VG_RQUAD_TO_NDS = ( 13 << 1 ),
VG_RCUBIC_TO_NDS = ( 14 << 1 ),
VG_PATH_SEGMENT_NDS_FORCE_SIZE = VG_MAX_ENUM
} VGPathSegmentNds;
typedef enum {
VG_RQUAD_TO_ABS_NDS = (VG_RQUAD_TO_NDS | VG_ABSOLUTE),
VG_RQUAD_TO_REL_NDS = (VG_RQUAD_TO_NDS | VG_RELATIVE),
VG_RCUBIC_TO_ABS_NDS = (VG_RCUBIC_TO_NDS | VG_ABSOLUTE),
VG_RCUBIC_TO_REL_NDS = (VG_RCUBIC_TO_NDS | VG_RELATIVE),
VG_PATH_COMMAND_NDS_FORCE_SIZE = VG_MAX_ENUM
} VGPathCommandNds;
#ifdef VG_VGEXT_PROTOTYPES
VG_API_CALL void VG_API_ENTRY vgProjectiveMatrixNDS(VGboolean enable) ;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguTransformClipLineNDS(const VGfloat Ain,const VGfloat Bin,const VGfloat Cin,const VGfloat* matrix,const VGboolean inverse,VGfloat* Aout,VGfloat* Bout,VGfloat* Cout);
#endif
typedef void (VG_API_ENTRYP PFNVGPROJECTIVEMATRIXNDSPROC) (VGboolean enable) ;
typedef VGUErrorCode (VGU_API_ENTRYP PFNVGUTRANSFORMCLIPLINENDSPROC) (const VGfloat Ain,const VGfloat Bin,const VGfloat Cin,const VGfloat* matrix,const VGboolean inverse,VGfloat* Aout,VGfloat* Bout,VGfloat* Cout);
#endif
#ifdef __cplusplus
} /* extern "C" */
#endif
#endif /* _VGEXT_H */

View File

@@ -1,92 +1,92 @@
/* $Revision: 6810 $ on $Date:: 2008-10-29 07:31:37 -0700 #$ */
/*------------------------------------------------------------------------
*
* VG platform specific header Reference Implementation
* ----------------------------------------------------
*
* Copyright (c) 2008 The Khronos Group Inc.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and /or associated documentation files
* (the "Materials "), to deal in the Materials without restriction,
* including without limitation the rights to use, copy, modify, merge,
* publish, distribute, sublicense, and/or sell copies of the Materials,
* and to permit persons to whom the Materials are furnished to do so,
* subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included
* in all copies or substantial portions of the Materials.
*
* THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
* DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
* OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE MATERIALS OR
* THE USE OR OTHER DEALINGS IN THE MATERIALS.
*
*//**
* \file
* \brief VG platform specific header
*//*-------------------------------------------------------------------*/
#ifndef _VGPLATFORM_H
#define _VGPLATFORM_H
#include <KHR/khrplatform.h>
#ifdef __cplusplus
extern "C" {
#endif
#ifndef VG_API_CALL
#if defined(OPENVG_STATIC_LIBRARY)
# define VG_API_CALL
#else
# define VG_API_CALL KHRONOS_APICALL
#endif /* defined OPENVG_STATIC_LIBRARY */
#endif /* ifndef VG_API_CALL */
#ifndef VGU_API_CALL
#if defined(OPENVG_STATIC_LIBRARY)
# define VGU_API_CALL
#else
# define VGU_API_CALL KHRONOS_APICALL
#endif /* defined OPENVG_STATIC_LIBRARY */
#endif /* ifndef VGU_API_CALL */
#ifndef VG_API_ENTRY
#define VG_API_ENTRY
#endif
#ifndef VG_API_EXIT
#define VG_API_EXIT
#endif
#ifndef VGU_API_ENTRY
#define VGU_API_ENTRY
#endif
#ifndef VGU_API_EXIT
#define VGU_API_EXIT
#endif
typedef float VGfloat;
typedef signed char VGbyte;
typedef unsigned char VGubyte;
typedef signed short VGshort;
typedef signed int VGint;
typedef unsigned int VGuint;
typedef unsigned int VGbitfield;
#ifndef VG_VGEXT_PROTOTYPES
#define VG_VGEXT_PROTOTYPES
#endif
#ifdef __cplusplus
} /* extern "C" */
#endif
#endif /* _VGPLATFORM_H */
/* $Revision: 6810 $ on $Date:: 2008-10-29 10:31:37 -0400 #$ */
/*------------------------------------------------------------------------
*
* VG platform specific header Reference Implementation
* ----------------------------------------------------
*
* Copyright (c) 2008 The Khronos Group Inc.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and /or associated documentation files
* (the "Materials "), to deal in the Materials without restriction,
* including without limitation the rights to use, copy, modify, merge,
* publish, distribute, sublicense, and/or sell copies of the Materials,
* and to permit persons to whom the Materials are furnished to do so,
* subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included
* in all copies or substantial portions of the Materials.
*
* THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
* DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
* OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE MATERIALS OR
* THE USE OR OTHER DEALINGS IN THE MATERIALS.
*
*//**
* \file
* \brief VG platform specific header
*//*-------------------------------------------------------------------*/
#ifndef _VGPLATFORM_H
#define _VGPLATFORM_H
#include <KHR/khrplatform.h>
#ifdef __cplusplus
extern "C" {
#endif
#ifndef VG_API_CALL
#if defined(OPENVG_STATIC_LIBRARY)
# define VG_API_CALL
#else
# define VG_API_CALL KHRONOS_APICALL
#endif /* defined OPENVG_STATIC_LIBRARY */
#endif /* ifndef VG_API_CALL */
#ifndef VGU_API_CALL
#if defined(OPENVG_STATIC_LIBRARY)
# define VGU_API_CALL
#else
# define VGU_API_CALL KHRONOS_APICALL
#endif /* defined OPENVG_STATIC_LIBRARY */
#endif /* ifndef VGU_API_CALL */
#ifndef VG_API_ENTRY
#define VG_API_ENTRY
#endif
#ifndef VG_API_EXIT
#define VG_API_EXIT
#endif
#ifndef VGU_API_ENTRY
#define VGU_API_ENTRY
#endif
#ifndef VGU_API_EXIT
#define VGU_API_EXIT
#endif
typedef float VGfloat;
typedef signed char VGbyte;
typedef unsigned char VGubyte;
typedef signed short VGshort;
typedef signed int VGint;
typedef unsigned int VGuint;
typedef unsigned int VGbitfield;
#ifndef VG_VGEXT_PROTOTYPES
#define VG_VGEXT_PROTOTYPES
#endif
#ifdef __cplusplus
} /* extern "C" */
#endif
#endif /* _VGPLATFORM_H */

View File

@@ -1,131 +1,130 @@
/* $Revision: 6810 $ on $Date:: 2008-10-29 07:31:37 -0700 #$ */
/*------------------------------------------------------------------------
*
* VGU 1.1 Reference Implementation
* -------------------------------------
*
* Copyright (c) 2008 The Khronos Group Inc.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and /or associated documentation files
* (the "Materials "), to deal in the Materials without restriction,
* including without limitation the rights to use, copy, modify, merge,
* publish, distribute, sublicense, and/or sell copies of the Materials,
* and to permit persons to whom the Materials are furnished to do so,
* subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included
* in all copies or substantial portions of the Materials.
*
* THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
* DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
* OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE MATERIALS OR
* THE USE OR OTHER DEALINGS IN THE MATERIALS.
*
*//**
* \file
* \brief VGU 1.1 API.
*//*-------------------------------------------------------------------*/
#ifndef _VGU_H
#define _VGU_H
#ifdef __cplusplus
extern "C" {
#endif
#include <VG/openvg.h>
#define VGU_VERSION_1_0 1
#define VGU_VERSION_1_1 2
#ifndef VGU_API_CALL
# error VGU_API_CALL must be defined
#endif
#ifndef VGU_API_ENTRY
# error VGU_API_ENTRY must be defined
#endif
#ifndef VGU_API_EXIT
# error VGU_API_EXIT must be defined
#endif
typedef enum {
VGU_NO_ERROR = 0,
VGU_BAD_HANDLE_ERROR = 0xF000,
VGU_ILLEGAL_ARGUMENT_ERROR = 0xF001,
VGU_OUT_OF_MEMORY_ERROR = 0xF002,
VGU_PATH_CAPABILITY_ERROR = 0xF003,
VGU_BAD_WARP_ERROR = 0xF004,
VGU_ERROR_CODE_FORCE_SIZE = VG_MAX_ENUM
} VGUErrorCode;
typedef enum {
VGU_ARC_OPEN = 0xF100,
VGU_ARC_CHORD = 0xF101,
VGU_ARC_PIE = 0xF102,
VGU_ARC_TYPE_FORCE_SIZE = VG_MAX_ENUM
} VGUArcType;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguLine(VGPath path,
VGfloat x0, VGfloat y0,
VGfloat x1, VGfloat y1) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguPolygon(VGPath path,
const VGfloat * points, VGint count,
VGboolean closed) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguRect(VGPath path,
VGfloat x, VGfloat y,
VGfloat width, VGfloat height) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguRoundRect(VGPath path,
VGfloat x, VGfloat y,
VGfloat width, VGfloat height,
VGfloat arcWidth, VGfloat arcHeight) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguEllipse(VGPath path,
VGfloat cx, VGfloat cy,
VGfloat width, VGfloat height) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguArc(VGPath path,
VGfloat x, VGfloat y,
VGfloat width, VGfloat height,
VGfloat startAngle, VGfloat angleExtent,
VGUArcType arcType) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguComputeWarpQuadToSquare(VGfloat sx0, VGfloat sy0,
VGfloat sx1, VGfloat sy1,
VGfloat sx2, VGfloat sy2,
VGfloat sx3, VGfloat sy3,
VGfloat * matrix) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguComputeWarpSquareToQuad(VGfloat dx0, VGfloat dy0,
VGfloat dx1, VGfloat dy1,
VGfloat dx2, VGfloat dy2,
VGfloat dx3, VGfloat dy3,
VGfloat * matrix) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguComputeWarpQuadToQuad(VGfloat dx0, VGfloat dy0,
VGfloat dx1, VGfloat dy1,
VGfloat dx2, VGfloat dy2,
VGfloat dx3, VGfloat dy3,
VGfloat sx0, VGfloat sy0,
VGfloat sx1, VGfloat sy1,
VGfloat sx2, VGfloat sy2,
VGfloat sx3, VGfloat sy3,
VGfloat * matrix) VGU_API_EXIT;
#ifdef __cplusplus
} /* extern "C" */
#endif
#endif /* #ifndef _VGU_H */
/* $Revision: 6810 $ on $Date:: 2008-10-29 10:31:37 -0400 #$ */
/*------------------------------------------------------------------------
*
* VGU 1.0.1 Reference Implementation
* -------------------------------------
*
* Copyright (c) 2008 The Khronos Group Inc.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and /or associated documentation files
* (the "Materials "), to deal in the Materials without restriction,
* including without limitation the rights to use, copy, modify, merge,
* publish, distribute, sublicense, and/or sell copies of the Materials,
* and to permit persons to whom the Materials are furnished to do so,
* subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included
* in all copies or substantial portions of the Materials.
*
* THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
* DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
* OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE MATERIALS OR
* THE USE OR OTHER DEALINGS IN THE MATERIALS.
*
*//**
* \file
* \brief VGU 1.0.1 API.
*//*-------------------------------------------------------------------*/
#ifndef _VGU_H
#define _VGU_H
#ifdef __cplusplus
extern "C" {
#endif
#include <VG/openvg.h>
#define VGU_VERSION_1_0 1
#ifndef VGU_API_CALL
# error VGU_API_CALL must be defined
#endif
#ifndef VGU_API_ENTRY
# error VGU_API_ENTRY must be defined
#endif
#ifndef VGU_API_EXIT
# error VGU_API_EXIT must be defined
#endif
typedef enum {
VGU_NO_ERROR = 0,
VGU_BAD_HANDLE_ERROR = 0xF000,
VGU_ILLEGAL_ARGUMENT_ERROR = 0xF001,
VGU_OUT_OF_MEMORY_ERROR = 0xF002,
VGU_PATH_CAPABILITY_ERROR = 0xF003,
VGU_BAD_WARP_ERROR = 0xF004,
VGU_ERROR_CODE_FORCE_SIZE = VG_MAX_ENUM
} VGUErrorCode;
typedef enum {
VGU_ARC_OPEN = 0xF100,
VGU_ARC_CHORD = 0xF101,
VGU_ARC_PIE = 0xF102,
VGU_ARC_TYPE_FORCE_SIZE = VG_MAX_ENUM
} VGUArcType;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguLine(VGPath path,
VGfloat x0, VGfloat y0,
VGfloat x1, VGfloat y1) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguPolygon(VGPath path,
const VGfloat * points, VGint count,
VGboolean closed) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguRect(VGPath path,
VGfloat x, VGfloat y,
VGfloat width, VGfloat height) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguRoundRect(VGPath path,
VGfloat x, VGfloat y,
VGfloat width, VGfloat height,
VGfloat arcWidth, VGfloat arcHeight) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguEllipse(VGPath path,
VGfloat cx, VGfloat cy,
VGfloat width, VGfloat height) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguArc(VGPath path,
VGfloat x, VGfloat y,
VGfloat width, VGfloat height,
VGfloat startAngle, VGfloat angleExtent,
VGUArcType arcType) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguComputeWarpQuadToSquare(VGfloat sx0, VGfloat sy0,
VGfloat sx1, VGfloat sy1,
VGfloat sx2, VGfloat sy2,
VGfloat sx3, VGfloat sy3,
VGfloat * matrix) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguComputeWarpSquareToQuad(VGfloat dx0, VGfloat dy0,
VGfloat dx1, VGfloat dy1,
VGfloat dx2, VGfloat dy2,
VGfloat dx3, VGfloat dy3,
VGfloat * matrix) VGU_API_EXIT;
VGU_API_CALL VGUErrorCode VGU_API_ENTRY vguComputeWarpQuadToQuad(VGfloat dx0, VGfloat dy0,
VGfloat dx1, VGfloat dy1,
VGfloat dx2, VGfloat dy2,
VGfloat dx3, VGfloat dy3,
VGfloat sx0, VGfloat sy0,
VGfloat sx1, VGfloat sy1,
VGfloat sx2, VGfloat sy2,
VGfloat sx3, VGfloat sy3,
VGfloat * matrix) VGU_API_EXIT;
#ifdef __cplusplus
} /* extern "C" */
#endif
#endif /* #ifndef _VGU_H */

View File

@@ -54,13 +54,11 @@ prefixes32 = SCons.Util.Split("""
i586-mingw32msvc-
i686-mingw32msvc-
i686-pc-mingw32-
i686-w64-mingw32-
""")
prefixes64 = SCons.Util.Split("""
amd64-mingw32-
amd64-mingw32msvc-
amd64-pc-mingw32-
x86_64-w64-mingw32-
""")
def find(env):

View File

@@ -49,35 +49,30 @@ def symlink(target, source, env):
os.symlink(os.path.basename(source), target)
def install(env, source, subdir):
target_dir = os.path.join(env.Dir('#.').srcnode().abspath, env['build_dir'], subdir)
return env.Install(target_dir, source)
target_dir = os.path.join(env.Dir('#.').srcnode().abspath, env['build'], subdir)
env.Install(target_dir, source)
def install_program(env, source):
return install(env, source, 'bin')
install(env, source, 'bin')
def install_shared_library(env, sources, version = ()):
targets = []
install_dir = os.path.join(env.Dir('#.').srcnode().abspath, env['build_dir'])
install_dir = os.path.join(env.Dir('#.').srcnode().abspath, env['build'])
version = tuple(map(str, version))
if env['SHLIBSUFFIX'] == '.dll':
dlls = env.FindIxes(sources, 'SHLIBPREFIX', 'SHLIBSUFFIX')
targets += install(env, dlls, 'bin')
install(env, dlls, 'bin')
libs = env.FindIxes(sources, 'LIBPREFIX', 'LIBSUFFIX')
targets += install(env, libs, 'lib')
install(env, libs, 'lib')
else:
for source in sources:
target_dir = os.path.join(install_dir, 'lib')
target_name = '.'.join((str(source),) + version)
last = env.InstallAs(os.path.join(target_dir, target_name), source)
targets += last
while len(version):
version = version[:-1]
target_name = '.'.join((str(source),) + version)
action = SCons.Action.Action(symlink, "$TARGET -> $SOURCE")
last = env.Command(os.path.join(target_dir, target_name), last, action)
targets += last
return targets
def createInstallMethods(env):
env.AddMethod(install_program, 'InstallProgram')
@@ -103,41 +98,6 @@ def num_jobs():
return 1
def pkg_config_modules(env, name, modules):
'''Simple wrapper for pkg-config.'''
env[name] = False
if env['platform'] == 'windows':
return
if not env.Detect('pkg-config'):
return
if subprocess.call(["pkg-config", "--exists", ' '.join(modules)]) != 0:
return
# Put -I and -L flags directly into the environment, as these don't affect
# the compilation of targets that do not use them
try:
env.ParseConfig('pkg-config --cflags-only-I --libs-only-L ' + ' '.join(modules))
except OSError:
return
# Other flags may affect the compilation of unrelated targets, so store
# them with a prefix, (e.g., XXX_CFLAGS, XXX_LIBS, etc)
try:
flags = env.ParseFlags('!pkg-config --cflags-only-other --libs-only-l --libs-only-other ' + ' '.join(modules))
except OSError:
return
prefix = name.upper() + '_'
for flag_name, flag_value in flags.iteritems():
env[prefix + flag_name] = flag_value
env[name] = True
def generate(env):
"""Common environment generation code"""
@@ -150,32 +110,27 @@ def generate(env):
env['toolchain'] = 'wcesdk'
env.Tool(env['toolchain'])
# Allow override compiler and specify additional flags from environment
if os.environ.has_key('CC'):
env['CC'] = os.environ['CC']
# Update CCVERSION to match
pipe = SCons.Action._subproc(env, [env['CC'], '--version'],
stdin = 'devnull',
stderr = 'devnull',
stdout = subprocess.PIPE)
if pipe.wait() == 0:
line = pipe.stdout.readline()
match = re.search(r'[0-9]+(\.[0-9]+)+', line)
if match:
env['CCVERSION'] = match.group(0)
if os.environ.has_key('CFLAGS'):
env['CCFLAGS'] += SCons.Util.CLVar(os.environ['CFLAGS'])
if os.environ.has_key('CXX'):
env['CXX'] = os.environ['CXX']
if os.environ.has_key('CXXFLAGS'):
env['CXXFLAGS'] += SCons.Util.CLVar(os.environ['CXXFLAGS'])
if os.environ.has_key('LDFLAGS'):
env['LINKFLAGS'] += SCons.Util.CLVar(os.environ['LDFLAGS'])
if env['platform'] == 'embedded':
# Allow overriding compiler from environment
if os.environ.has_key('CC'):
env['CC'] = os.environ['CC']
# Update CCVERSION to match
pipe = SCons.Action._subproc(env, [env['CC'], '--version'],
stdin = 'devnull',
stderr = 'devnull',
stdout = subprocess.PIPE)
if pipe.wait() == 0:
line = pipe.stdout.readline()
match = re.search(r'[0-9]+(\.[0-9]+)+', line)
if match:
env['CCVERSION'] = match.group(0)
env['gcc'] = 'gcc' in os.path.basename(env['CC']).split('-')
env['msvc'] = env['CC'] == 'cl'
# shortcuts
debug = env['debug']
machine = env['machine']
platform = env['platform']
x86 = env['machine'] == 'x86'
@@ -183,48 +138,20 @@ def generate(env):
gcc = env['gcc']
msvc = env['msvc']
# Backwards compatability with the debug= profile= options
if env['build'] == 'debug':
if not env['debug']:
print 'scons: warning: debug option is deprecated and will be removed eventually; use instead'
print
print ' scons build=release'
print
env['build'] = 'release'
if env['profile']:
print 'scons: warning: profile option is deprecated and will be removed eventually; use instead'
print
print ' scons build=profile'
print
env['build'] = 'profile'
if False:
# Enforce SConscripts to use the new build variable
env.popitem('debug')
env.popitem('profile')
else:
# Backwards portability with older sconscripts
if env['build'] in ('debug', 'checked'):
env['debug'] = True
env['profile'] = False
if env['build'] == 'profile':
env['debug'] = False
env['profile'] = True
if env['build'] == 'release':
env['debug'] = False
env['profile'] = False
# Put build output in a separate dir, which depends on the current
# configuration. See also http://www.scons.org/wiki/AdvancedBuildExample
build_topdir = 'build'
build_subdir = env['platform']
if env['machine'] != 'generic':
build_subdir += '-' + env['machine']
if env['build'] != 'release':
build_subdir += '-' + env['build']
if env['debug']:
build_subdir += "-debug"
if env['profile']:
build_subdir += "-profile"
build_dir = os.path.join(build_topdir, build_subdir)
# Place the .sconsign file in the build dir too, to avoid issues with
# different scons versions building the same source file
env['build_dir'] = build_dir
env['build'] = build_dir
env.SConsignFile(os.path.join(build_dir, '.sconsign'))
if 'SCONS_CACHE_DIR' in os.environ:
print 'scons: Using build cache in %s.' % (os.environ['SCONS_CACHE_DIR'],)
@@ -236,16 +163,13 @@ def generate(env):
if env.GetOption('num_jobs') <= 1:
env.SetOption('num_jobs', num_jobs())
env.Decider('MD5-timestamp')
env.SetOption('max_drift', 60)
# C preprocessor options
cppdefines = []
if env['build'] in ('debug', 'checked'):
if debug:
cppdefines += ['DEBUG']
else:
cppdefines += ['NDEBUG']
if env['build'] == 'profile':
if env['profile']:
cppdefines += ['PROFILE']
if platform == 'windows':
cppdefines += [
@@ -266,7 +190,7 @@ def generate(env):
'_SCL_SECURE_NO_WARNINGS',
'_SCL_SECURE_NO_DEPRECATE',
]
if env['build'] in ('debug', 'checked'):
if debug:
cppdefines += ['_DEBUG']
if env['toolchain'] == 'winddk':
# Mimic WINDDK's builtin flags. See also:
@@ -293,7 +217,7 @@ def generate(env):
('__BUILDMACHINE__', 'WinDDK'),
('FPO', '0'),
]
if env['build'] in ('debug', 'checked'):
if debug:
cppdefines += [('DBG', 1)]
if platform == 'wince':
cppdefines += [
@@ -329,16 +253,15 @@ def generate(env):
ccflags = [] # C & C++
if gcc:
ccversion = env['CCVERSION']
if env['build'] == 'debug':
ccflags += ['-O0']
if debug:
ccflags += ['-O0', '-g3']
elif ccversion.startswith('4.2.'):
# gcc 4.2.x optimizer is broken
print "warning: gcc 4.2.x optimizer is broken -- disabling optimizations"
ccflags += ['-O0']
ccflags += ['-O0', '-g3']
else:
ccflags += ['-O3']
ccflags += ['-g3']
if env['build'] in ('checked', 'profile'):
ccflags += ['-O3', '-g3']
if env['profile']:
# See http://code.google.com/p/jrfonseca/wiki/Gprof2Dot#Which_options_should_I_pass_to_gcc_when_compiling_for_profiling?
ccflags += [
'-fno-omit-frame-pointer',
@@ -397,7 +320,7 @@ def generate(env):
# See also:
# - http://msdn.microsoft.com/en-us/library/19z1t1wy.aspx
# - cl /?
if env['build'] == 'debug':
if debug:
ccflags += [
'/Od', # disable optimizations
'/Oi', # enable intrinsic functions
@@ -466,7 +389,7 @@ def generate(env):
if env['platform'] == 'windows' and msvc:
# Choose the appropriate MSVC CRT
# http://msdn.microsoft.com/en-us/library/2kzt1wy3.aspx
if env['build'] in ('debug', 'checked'):
if env['debug']:
env.Append(CCFLAGS = ['/MTd'])
env.Append(SHCCFLAGS = ['/LDd'])
else:
@@ -498,7 +421,7 @@ def generate(env):
else:
env['_LIBFLAGS'] = '-Wl,--start-group ' + env['_LIBFLAGS'] + ' -Wl,--end-group'
if msvc:
if env['build'] != 'debug':
if not env['debug']:
# enable Link-time Code Generation
linkflags += ['/LTCG']
env.Append(ARFLAGS = ['/LTCG'])
@@ -537,7 +460,7 @@ def generate(env):
'/entry:DrvEnableDriver',
]
if env['build'] != 'release':
if env['debug'] or env['profile']:
linkflags += [
'/MAP', # http://msdn.microsoft.com/en-us/library/k7xkk3e2.aspx
]
@@ -554,19 +477,9 @@ def generate(env):
# Default libs
env.Append(LIBS = [])
# Load tools
# Load LLVM
if env['llvm']:
env.Tool('llvm')
env.Tool('udis86')
pkg_config_modules(env, 'x11', ['x11', 'xext'])
pkg_config_modules(env, 'drm', ['libdrm'])
pkg_config_modules(env, 'drm_intel', ['libdrm_intel'])
pkg_config_modules(env, 'drm_radeon', ['libdrm_radeon'])
pkg_config_modules(env, 'xorg', ['xorg-server'])
pkg_config_modules(env, 'kms', ['libkms'])
env['dri'] = env['x11'] and env['drm']
# Custom builders and methods
env.Tool('custom')

View File

@@ -38,8 +38,6 @@ import SCons.Util
def generate(env):
env['llvm'] = False
try:
llvm_dir = os.environ['LLVM']
except KeyError:
@@ -66,13 +64,13 @@ def generate(env):
# XXX: There is no llvm-config on Windows, so assume a standard layout
if llvm_dir is None:
print 'scons: LLVM environment variable must be specified when building for windows'
return
env.Exit(1)
# Try to determine the LLVM version from llvm/Config/config.h
llvm_config = os.path.join(llvm_dir, 'include/llvm/Config/config.h')
if not os.path.exists(llvm_config):
print 'scons: could not find %s' % llvm_config
return
env.Exit(1)
llvm_version_re = re.compile(r'^#define PACKAGE_VERSION "([^"]*)"')
llvm_version = None
for line in open(llvm_config, 'rt'):
@@ -83,7 +81,7 @@ def generate(env):
break
if llvm_version is None:
print 'scons: could not determine the LLVM version from %s' % llvm_config
return
env.Exit(1)
env.Prepend(CPPPATH = [os.path.join(llvm_dir, 'include')])
env.AppendUnique(CPPDEFINES = [
@@ -126,7 +124,7 @@ def generate(env):
# Some of the LLVM C headers use the inline keyword without
# defining it.
env.Append(CPPDEFINES = [('inline', '__inline')])
if env['build'] in ('debug', 'checked'):
if env['debug']:
# LLVM libraries are static, build with /MT, and they
# automatically link agains LIBCMT. When we're doing a
# debug build we'll be linking against LIBCMTD, so disable
@@ -135,7 +133,7 @@ def generate(env):
else:
if not env.Detect('llvm-config'):
print 'scons: llvm-config script not found' % llvm_version
return
env.Exit(1)
llvm_version = env.backtick('llvm-config --version').rstrip()
llvm_version = distutils.version.LooseVersion(llvm_version)
@@ -146,12 +144,11 @@ def generate(env):
env.ParseConfig('llvm-config --ldflags')
except OSError:
print 'scons: llvm-config version %s failed' % llvm_version
return
env.Exit(1)
else:
env['LINK'] = env['CXX']
assert llvm_version is not None
env['llvm'] = True
print 'scons: Found LLVM version %s' % llvm_version
env['LLVM_VERSION'] = llvm_version

View File

@@ -31,10 +31,8 @@ def generate(env):
conf = env.Configure()
if conf.CheckHeader('udis86.h'): # and conf.CheckLib('udis86'):
env['UDIS86'] = True
env.Append(CPPDEFINES = [('HAVE_UDIS86', '1')])
env.Prepend(LIBS = ['udis86'])
else:
env['UDIS86'] = False
conf.Finish()

View File

@@ -122,7 +122,7 @@ def get_wce600_paths(env):
host_cpu = os.environ.get('_HOSTCPUTYPE', 'i386')
target_cpu = os.environ.get('_TGTCPU', 'x86')
if env['build'] == 'debug':
if env['debug']:
build = 'debug'
else:
build = 'retail'

View File

@@ -1,16 +1,19 @@
Import('*')
SConscript('mapi/vgapi/SConscript')
if env['platform'] == 'windows':
if 'egl' in env['statetrackers']:
SConscript('mapi/vgapi/SConscript')
SConscript('egl/main/SConscript')
SConscript('glsl/SConscript')
SConscript('mapi/glapi/SConscript')
SConscript('mesa/SConscript')
if 'mesa' in env['statetrackers']:
if platform == 'windows':
SConscript('talloc/SConscript')
if env['platform'] != 'embedded':
SConscript('glut/glx/SConscript')
SConscript('glsl/SConscript')
SConscript('mapi/glapi/SConscript')
SConscript('mesa/SConscript')
if platform != 'embedded':
SConscript('glut/glx/SConscript')
SConscript('gallium/SConscript')

View File

@@ -24,8 +24,8 @@ $(EGL_DRIVER_PATH): $(EGL_DRIVER)
$(EGL_DRIVER): $(EGL_OBJECTS) Makefile $(TOP)/src/egl/drivers/Makefile.template
@$(MKLIB) -o $(EGL_DRIVER) -noprefix \
-linker '$(CC)' -ldflags '-L$(TOP)/$(LIB_DIR) $(LDFLAGS)' \
$(MKLIB_OPTIONS) \
-linker '$(CC)' -ldflags '$(LDFLAGS)' \
-L$(TOP)/$(LIB_DIR) $(MKLIB_OPTIONS) \
$(EGL_OBJECTS) $(EGL_LIBS) -l$(EGL_LIB)
.c.o:

View File

@@ -248,20 +248,21 @@ dri2_add_config(_EGLDisplay *disp, const __DRIconfig *dri_config, int id,
if (double_buffer)
return NULL;
if (depth > 0 && depth != base.BufferSize)
if (depth > 0 && depth != _eglGetConfigKey(&base, EGL_BUFFER_SIZE))
return NULL;
base.NativeRenderable = EGL_TRUE;
_eglSetConfigKey(&base, EGL_NATIVE_RENDERABLE, EGL_TRUE);
base.SurfaceType = surface_type;
_eglSetConfigKey(&base, EGL_SURFACE_TYPE, surface_type);
if (surface_type & (EGL_PIXMAP_BIT | EGL_PBUFFER_BIT)) {
base.BindToTextureRGB = bind_to_texture_rgb;
if (base.AlphaSize > 0)
base.BindToTextureRGBA = bind_to_texture_rgba;
_eglSetConfigKey(&base, EGL_BIND_TO_TEXTURE_RGB, bind_to_texture_rgb);
if (_eglGetConfigKey(&base, EGL_ALPHA_SIZE) > 0)
_eglSetConfigKey(&base,
EGL_BIND_TO_TEXTURE_RGBA, bind_to_texture_rgba);
}
base.RenderableType = disp->ClientAPIsMask;
base.Conformant = disp->ClientAPIsMask;
_eglSetConfigKey(&base, EGL_RENDERABLE_TYPE, disp->ClientAPIsMask);
_eglSetConfigKey(&base, EGL_CONFORMANT, disp->ClientAPIsMask);
if (!_eglValidateConfig(&base, EGL_FALSE)) {
_eglLog(_EGL_DEBUG, "DRI2: failed to validate config %d", id);
@@ -272,7 +273,7 @@ dri2_add_config(_EGLDisplay *disp, const __DRIconfig *dri_config, int id,
if (conf != NULL) {
memcpy(&conf->base, &base, sizeof base);
conf->dri_config = dri_config;
_eglLinkConfig(&conf->base);
_eglAddConfig(disp, &conf->base);
}
return conf;
@@ -291,7 +292,7 @@ dri2_process_buffers(struct dri2_egl_surface *dri2_surf,
struct dri2_egl_display *dri2_dpy =
dri2_egl_display(dri2_surf->base.Resource.Display);
xcb_rectangle_t rectangle;
unsigned i;
int i;
dri2_surf->buffer_count = count;
dri2_surf->have_fake_front = 0;
@@ -338,8 +339,6 @@ dri2_get_buffers(__DRIdrawable * driDrawable,
xcb_dri2_get_buffers_reply_t *reply;
xcb_dri2_get_buffers_cookie_t cookie;
(void) driDrawable;
cookie = xcb_dri2_get_buffers_unchecked (dri2_dpy->conn,
dri2_surf->drawable,
count, count, attachments);
@@ -361,16 +360,12 @@ dri2_get_buffers(__DRIdrawable * driDrawable,
static void
dri2_flush_front_buffer(__DRIdrawable * driDrawable, void *loaderPrivate)
{
(void) driDrawable;
/* FIXME: Does EGL support front buffer rendering at all? */
#if 0
struct dri2_egl_surface *dri2_surf = loaderPrivate;
dri2WaitGL(dri2_surf);
#else
(void) loaderPrivate;
#endif
}
@@ -381,8 +376,6 @@ dri2_lookup_egl_image(__DRIscreen *screen, void *image, void *data)
struct dri2_egl_image *dri2_img;
_EGLImage *img;
(void) screen;
img = _eglLookupImage(image, disp);
if (img == NULL) {
_eglError(EGL_BAD_PARAMETER, "dri2_lookup_egl_image");
@@ -413,8 +406,6 @@ dri2_get_buffers_with_format(__DRIdrawable * driDrawable,
xcb_dri2_get_buffers_with_format_cookie_t cookie;
xcb_dri2_attach_format_t *format_attachments;
(void) driDrawable;
format_attachments = (xcb_dri2_attach_format_t *) attachments;
cookie = xcb_dri2_get_buffers_with_format_unchecked (dri2_dpy->conn,
dri2_surf->drawable,
@@ -448,14 +439,14 @@ struct dri2_extension_match {
static struct dri2_extension_match dri2_driver_extensions[] = {
{ __DRI_CORE, 1, offsetof(struct dri2_egl_display, core) },
{ __DRI_DRI2, 1, offsetof(struct dri2_egl_display, dri2) },
{ NULL, 0, 0 }
{ NULL }
};
static struct dri2_extension_match dri2_core_extensions[] = {
{ __DRI2_FLUSH, 1, offsetof(struct dri2_egl_display, flush) },
{ __DRI_TEX_BUFFER, 2, offsetof(struct dri2_egl_display, tex_buffer) },
{ __DRI_IMAGE, 1, offsetof(struct dri2_egl_display, image) },
{ NULL, 0, 0 }
{ NULL }
};
static EGLBoolean
@@ -780,8 +771,6 @@ dri2_initialize_x11(_EGLDriver *drv, _EGLDisplay *disp,
{
struct dri2_egl_display *dri2_dpy;
(void) drv;
dri2_dpy = malloc(sizeof *dri2_dpy);
if (!dri2_dpy)
return _eglError(EGL_BAD_ALLOC, "eglInitialize");
@@ -928,436 +917,9 @@ const int i965_chip_ids[] = {
0x2e92, /* PCI_CHIP_B43_G1 */
};
const int r100_chip_ids[] = {
0x4C57, /* PCI_CHIP_RADEON_LW */
0x4C58, /* PCI_CHIP_RADEON_LX */
0x4C59, /* PCI_CHIP_RADEON_LY */
0x4C5A, /* PCI_CHIP_RADEON_LZ */
0x5144, /* PCI_CHIP_RADEON_QD */
0x5145, /* PCI_CHIP_RADEON_QE */
0x5146, /* PCI_CHIP_RADEON_QF */
0x5147, /* PCI_CHIP_RADEON_QG */
0x5159, /* PCI_CHIP_RADEON_QY */
0x515A, /* PCI_CHIP_RADEON_QZ */
0x5157, /* PCI_CHIP_RV200_QW */
0x5158, /* PCI_CHIP_RV200_QX */
0x515E, /* PCI_CHIP_RN50_515E */
0x5969, /* PCI_CHIP_RN50_5969 */
0x4136, /* PCI_CHIP_RS100_4136 */
0x4336, /* PCI_CHIP_RS100_4336 */
0x4137, /* PCI_CHIP_RS200_4137 */
0x4337, /* PCI_CHIP_RS200_4337 */
0x4237, /* PCI_CHIP_RS250_4237 */
0x4437, /* PCI_CHIP_RS250_4437 */
};
const int r200_chip_ids[] = {
0x5148, /* PCI_CHIP_R200_QH */
0x514C, /* PCI_CHIP_R200_QL */
0x514D, /* PCI_CHIP_R200_QM */
0x4242, /* PCI_CHIP_R200_BB */
0x4243, /* PCI_CHIP_R200_BC */
0x4966, /* PCI_CHIP_RV250_If */
0x4967, /* PCI_CHIP_RV250_Ig */
0x4C64, /* PCI_CHIP_RV250_Ld */
0x4C66, /* PCI_CHIP_RV250_Lf */
0x4C67, /* PCI_CHIP_RV250_Lg */
0x5960, /* PCI_CHIP_RV280_5960 */
0x5961, /* PCI_CHIP_RV280_5961 */
0x5962, /* PCI_CHIP_RV280_5962 */
0x5964, /* PCI_CHIP_RV280_5964 */
0x5965, /* PCI_CHIP_RV280_5965 */
0x5C61, /* PCI_CHIP_RV280_5C61 */
0x5C63, /* PCI_CHIP_RV280_5C63 */
0x5834, /* PCI_CHIP_RS300_5834 */
0x5835, /* PCI_CHIP_RS300_5835 */
0x7834, /* PCI_CHIP_RS350_7834 */
0x7835, /* PCI_CHIP_RS350_7835 */
};
const int r300_chip_ids[] = {
0x4144, /* PCI_CHIP_R300_AD */
0x4145, /* PCI_CHIP_R300_AE */
0x4146, /* PCI_CHIP_R300_AF */
0x4147, /* PCI_CHIP_R300_AG */
0x4E44, /* PCI_CHIP_R300_ND */
0x4E45, /* PCI_CHIP_R300_NE */
0x4E46, /* PCI_CHIP_R300_NF */
0x4E47, /* PCI_CHIP_R300_NG */
0x4E48, /* PCI_CHIP_R350_NH */
0x4E49, /* PCI_CHIP_R350_NI */
0x4E4B, /* PCI_CHIP_R350_NK */
0x4148, /* PCI_CHIP_R350_AH */
0x4149, /* PCI_CHIP_R350_AI */
0x414A, /* PCI_CHIP_R350_AJ */
0x414B, /* PCI_CHIP_R350_AK */
0x4E4A, /* PCI_CHIP_R360_NJ */
0x4150, /* PCI_CHIP_RV350_AP */
0x4151, /* PCI_CHIP_RV350_AQ */
0x4152, /* PCI_CHIP_RV350_AR */
0x4153, /* PCI_CHIP_RV350_AS */
0x4154, /* PCI_CHIP_RV350_AT */
0x4155, /* PCI_CHIP_RV350_AU */
0x4156, /* PCI_CHIP_RV350_AV */
0x4E50, /* PCI_CHIP_RV350_NP */
0x4E51, /* PCI_CHIP_RV350_NQ */
0x4E52, /* PCI_CHIP_RV350_NR */
0x4E53, /* PCI_CHIP_RV350_NS */
0x4E54, /* PCI_CHIP_RV350_NT */
0x4E56, /* PCI_CHIP_RV350_NV */
0x5460, /* PCI_CHIP_RV370_5460 */
0x5462, /* PCI_CHIP_RV370_5462 */
0x5464, /* PCI_CHIP_RV370_5464 */
0x5B60, /* PCI_CHIP_RV370_5B60 */
0x5B62, /* PCI_CHIP_RV370_5B62 */
0x5B63, /* PCI_CHIP_RV370_5B63 */
0x5B64, /* PCI_CHIP_RV370_5B64 */
0x5B65, /* PCI_CHIP_RV370_5B65 */
0x3150, /* PCI_CHIP_RV380_3150 */
0x3152, /* PCI_CHIP_RV380_3152 */
0x3154, /* PCI_CHIP_RV380_3154 */
0x3155, /* PCI_CHIP_RV380_3155 */
0x3E50, /* PCI_CHIP_RV380_3E50 */
0x3E54, /* PCI_CHIP_RV380_3E54 */
0x4A48, /* PCI_CHIP_R420_JH */
0x4A49, /* PCI_CHIP_R420_JI */
0x4A4A, /* PCI_CHIP_R420_JJ */
0x4A4B, /* PCI_CHIP_R420_JK */
0x4A4C, /* PCI_CHIP_R420_JL */
0x4A4D, /* PCI_CHIP_R420_JM */
0x4A4E, /* PCI_CHIP_R420_JN */
0x4A4F, /* PCI_CHIP_R420_JO */
0x4A50, /* PCI_CHIP_R420_JP */
0x4A54, /* PCI_CHIP_R420_JT */
0x5548, /* PCI_CHIP_R423_UH */
0x5549, /* PCI_CHIP_R423_UI */
0x554A, /* PCI_CHIP_R423_UJ */
0x554B, /* PCI_CHIP_R423_UK */
0x5550, /* PCI_CHIP_R423_5550 */
0x5551, /* PCI_CHIP_R423_UQ */
0x5552, /* PCI_CHIP_R423_UR */
0x5554, /* PCI_CHIP_R423_UT */
0x5D57, /* PCI_CHIP_R423_5D57 */
0x554C, /* PCI_CHIP_R430_554C */
0x554D, /* PCI_CHIP_R430_554D */
0x554E, /* PCI_CHIP_R430_554E */
0x554F, /* PCI_CHIP_R430_554F */
0x5D48, /* PCI_CHIP_R430_5D48 */
0x5D49, /* PCI_CHIP_R430_5D49 */
0x5D4A, /* PCI_CHIP_R430_5D4A */
0x5D4C, /* PCI_CHIP_R480_5D4C */
0x5D4D, /* PCI_CHIP_R480_5D4D */
0x5D4E, /* PCI_CHIP_R480_5D4E */
0x5D4F, /* PCI_CHIP_R480_5D4F */
0x5D50, /* PCI_CHIP_R480_5D50 */
0x5D52, /* PCI_CHIP_R480_5D52 */
0x4B49, /* PCI_CHIP_R481_4B49 */
0x4B4A, /* PCI_CHIP_R481_4B4A */
0x4B4B, /* PCI_CHIP_R481_4B4B */
0x4B4C, /* PCI_CHIP_R481_4B4C */
0x564A, /* PCI_CHIP_RV410_564A */
0x564B, /* PCI_CHIP_RV410_564B */
0x564F, /* PCI_CHIP_RV410_564F */
0x5652, /* PCI_CHIP_RV410_5652 */
0x5653, /* PCI_CHIP_RV410_5653 */
0x5657, /* PCI_CHIP_RV410_5657 */
0x5E48, /* PCI_CHIP_RV410_5E48 */
0x5E4A, /* PCI_CHIP_RV410_5E4A */
0x5E4B, /* PCI_CHIP_RV410_5E4B */
0x5E4C, /* PCI_CHIP_RV410_5E4C */
0x5E4D, /* PCI_CHIP_RV410_5E4D */
0x5E4F, /* PCI_CHIP_RV410_5E4F */
0x5A41, /* PCI_CHIP_RS400_5A41 */
0x5A42, /* PCI_CHIP_RS400_5A42 */
0x5A61, /* PCI_CHIP_RC410_5A61 */
0x5A62, /* PCI_CHIP_RC410_5A62 */
0x5954, /* PCI_CHIP_RS480_5954 */
0x5955, /* PCI_CHIP_RS480_5955 */
0x5974, /* PCI_CHIP_RS482_5974 */
0x5975, /* PCI_CHIP_RS482_5975 */
0x7100, /* PCI_CHIP_R520_7100 */
0x7101, /* PCI_CHIP_R520_7101 */
0x7102, /* PCI_CHIP_R520_7102 */
0x7103, /* PCI_CHIP_R520_7103 */
0x7104, /* PCI_CHIP_R520_7104 */
0x7105, /* PCI_CHIP_R520_7105 */
0x7106, /* PCI_CHIP_R520_7106 */
0x7108, /* PCI_CHIP_R520_7108 */
0x7109, /* PCI_CHIP_R520_7109 */
0x710A, /* PCI_CHIP_R520_710A */
0x710B, /* PCI_CHIP_R520_710B */
0x710C, /* PCI_CHIP_R520_710C */
0x710E, /* PCI_CHIP_R520_710E */
0x710F, /* PCI_CHIP_R520_710F */
0x7140, /* PCI_CHIP_RV515_7140 */
0x7141, /* PCI_CHIP_RV515_7141 */
0x7142, /* PCI_CHIP_RV515_7142 */
0x7143, /* PCI_CHIP_RV515_7143 */
0x7144, /* PCI_CHIP_RV515_7144 */
0x7145, /* PCI_CHIP_RV515_7145 */
0x7146, /* PCI_CHIP_RV515_7146 */
0x7147, /* PCI_CHIP_RV515_7147 */
0x7149, /* PCI_CHIP_RV515_7149 */
0x714A, /* PCI_CHIP_RV515_714A */
0x714B, /* PCI_CHIP_RV515_714B */
0x714C, /* PCI_CHIP_RV515_714C */
0x714D, /* PCI_CHIP_RV515_714D */
0x714E, /* PCI_CHIP_RV515_714E */
0x714F, /* PCI_CHIP_RV515_714F */
0x7151, /* PCI_CHIP_RV515_7151 */
0x7152, /* PCI_CHIP_RV515_7152 */
0x7153, /* PCI_CHIP_RV515_7153 */
0x715E, /* PCI_CHIP_RV515_715E */
0x715F, /* PCI_CHIP_RV515_715F */
0x7180, /* PCI_CHIP_RV515_7180 */
0x7181, /* PCI_CHIP_RV515_7181 */
0x7183, /* PCI_CHIP_RV515_7183 */
0x7186, /* PCI_CHIP_RV515_7186 */
0x7187, /* PCI_CHIP_RV515_7187 */
0x7188, /* PCI_CHIP_RV515_7188 */
0x718A, /* PCI_CHIP_RV515_718A */
0x718B, /* PCI_CHIP_RV515_718B */
0x718C, /* PCI_CHIP_RV515_718C */
0x718D, /* PCI_CHIP_RV515_718D */
0x718F, /* PCI_CHIP_RV515_718F */
0x7193, /* PCI_CHIP_RV515_7193 */
0x7196, /* PCI_CHIP_RV515_7196 */
0x719B, /* PCI_CHIP_RV515_719B */
0x719F, /* PCI_CHIP_RV515_719F */
0x7200, /* PCI_CHIP_RV515_7200 */
0x7210, /* PCI_CHIP_RV515_7210 */
0x7211, /* PCI_CHIP_RV515_7211 */
0x71C0, /* PCI_CHIP_RV530_71C0 */
0x71C1, /* PCI_CHIP_RV530_71C1 */
0x71C2, /* PCI_CHIP_RV530_71C2 */
0x71C3, /* PCI_CHIP_RV530_71C3 */
0x71C4, /* PCI_CHIP_RV530_71C4 */
0x71C5, /* PCI_CHIP_RV530_71C5 */
0x71C6, /* PCI_CHIP_RV530_71C6 */
0x71C7, /* PCI_CHIP_RV530_71C7 */
0x71CD, /* PCI_CHIP_RV530_71CD */
0x71CE, /* PCI_CHIP_RV530_71CE */
0x71D2, /* PCI_CHIP_RV530_71D2 */
0x71D4, /* PCI_CHIP_RV530_71D4 */
0x71D5, /* PCI_CHIP_RV530_71D5 */
0x71D6, /* PCI_CHIP_RV530_71D6 */
0x71DA, /* PCI_CHIP_RV530_71DA */
0x71DE, /* PCI_CHIP_RV530_71DE */
0x7281, /* PCI_CHIP_RV560_7281 */
0x7283, /* PCI_CHIP_RV560_7283 */
0x7287, /* PCI_CHIP_RV560_7287 */
0x7290, /* PCI_CHIP_RV560_7290 */
0x7291, /* PCI_CHIP_RV560_7291 */
0x7293, /* PCI_CHIP_RV560_7293 */
0x7297, /* PCI_CHIP_RV560_7297 */
0x7280, /* PCI_CHIP_RV570_7280 */
0x7288, /* PCI_CHIP_RV570_7288 */
0x7289, /* PCI_CHIP_RV570_7289 */
0x728B, /* PCI_CHIP_RV570_728B */
0x728C, /* PCI_CHIP_RV570_728C */
0x7240, /* PCI_CHIP_R580_7240 */
0x7243, /* PCI_CHIP_R580_7243 */
0x7244, /* PCI_CHIP_R580_7244 */
0x7245, /* PCI_CHIP_R580_7245 */
0x7246, /* PCI_CHIP_R580_7246 */
0x7247, /* PCI_CHIP_R580_7247 */
0x7248, /* PCI_CHIP_R580_7248 */
0x7249, /* PCI_CHIP_R580_7249 */
0x724A, /* PCI_CHIP_R580_724A */
0x724B, /* PCI_CHIP_R580_724B */
0x724C, /* PCI_CHIP_R580_724C */
0x724D, /* PCI_CHIP_R580_724D */
0x724E, /* PCI_CHIP_R580_724E */
0x724F, /* PCI_CHIP_R580_724F */
0x7284, /* PCI_CHIP_R580_7284 */
0x793F, /* PCI_CHIP_RS600_793F */
0x7941, /* PCI_CHIP_RS600_7941 */
0x7942, /* PCI_CHIP_RS600_7942 */
0x791E, /* PCI_CHIP_RS690_791E */
0x791F, /* PCI_CHIP_RS690_791F */
0x796C, /* PCI_CHIP_RS740_796C */
0x796D, /* PCI_CHIP_RS740_796D */
0x796E, /* PCI_CHIP_RS740_796E */
0x796F, /* PCI_CHIP_RS740_796F */
};
const int r600_chip_ids[] = {
0x9400, /* PCI_CHIP_R600_9400 */
0x9401, /* PCI_CHIP_R600_9401 */
0x9402, /* PCI_CHIP_R600_9402 */
0x9403, /* PCI_CHIP_R600_9403 */
0x9405, /* PCI_CHIP_R600_9405 */
0x940A, /* PCI_CHIP_R600_940A */
0x940B, /* PCI_CHIP_R600_940B */
0x940F, /* PCI_CHIP_R600_940F */
0x94C0, /* PCI_CHIP_RV610_94C0 */
0x94C1, /* PCI_CHIP_RV610_94C1 */
0x94C3, /* PCI_CHIP_RV610_94C3 */
0x94C4, /* PCI_CHIP_RV610_94C4 */
0x94C5, /* PCI_CHIP_RV610_94C5 */
0x94C6, /* PCI_CHIP_RV610_94C6 */
0x94C7, /* PCI_CHIP_RV610_94C7 */
0x94C8, /* PCI_CHIP_RV610_94C8 */
0x94C9, /* PCI_CHIP_RV610_94C9 */
0x94CB, /* PCI_CHIP_RV610_94CB */
0x94CC, /* PCI_CHIP_RV610_94CC */
0x94CD, /* PCI_CHIP_RV610_94CD */
0x9580, /* PCI_CHIP_RV630_9580 */
0x9581, /* PCI_CHIP_RV630_9581 */
0x9583, /* PCI_CHIP_RV630_9583 */
0x9586, /* PCI_CHIP_RV630_9586 */
0x9587, /* PCI_CHIP_RV630_9587 */
0x9588, /* PCI_CHIP_RV630_9588 */
0x9589, /* PCI_CHIP_RV630_9589 */
0x958A, /* PCI_CHIP_RV630_958A */
0x958B, /* PCI_CHIP_RV630_958B */
0x958C, /* PCI_CHIP_RV630_958C */
0x958D, /* PCI_CHIP_RV630_958D */
0x958E, /* PCI_CHIP_RV630_958E */
0x958F, /* PCI_CHIP_RV630_958F */
0x9500, /* PCI_CHIP_RV670_9500 */
0x9501, /* PCI_CHIP_RV670_9501 */
0x9504, /* PCI_CHIP_RV670_9504 */
0x9505, /* PCI_CHIP_RV670_9505 */
0x9506, /* PCI_CHIP_RV670_9506 */
0x9507, /* PCI_CHIP_RV670_9507 */
0x9508, /* PCI_CHIP_RV670_9508 */
0x9509, /* PCI_CHIP_RV670_9509 */
0x950F, /* PCI_CHIP_RV670_950F */
0x9511, /* PCI_CHIP_RV670_9511 */
0x9515, /* PCI_CHIP_RV670_9515 */
0x9517, /* PCI_CHIP_RV670_9517 */
0x9519, /* PCI_CHIP_RV670_9519 */
0x95C0, /* PCI_CHIP_RV620_95C0 */
0x95C2, /* PCI_CHIP_RV620_95C2 */
0x95C4, /* PCI_CHIP_RV620_95C4 */
0x95C5, /* PCI_CHIP_RV620_95C5 */
0x95C6, /* PCI_CHIP_RV620_95C6 */
0x95C7, /* PCI_CHIP_RV620_95C7 */
0x95C9, /* PCI_CHIP_RV620_95C9 */
0x95CC, /* PCI_CHIP_RV620_95CC */
0x95CD, /* PCI_CHIP_RV620_95CD */
0x95CE, /* PCI_CHIP_RV620_95CE */
0x95CF, /* PCI_CHIP_RV620_95CF */
0x9590, /* PCI_CHIP_RV635_9590 */
0x9591, /* PCI_CHIP_RV635_9591 */
0x9593, /* PCI_CHIP_RV635_9593 */
0x9595, /* PCI_CHIP_RV635_9595 */
0x9596, /* PCI_CHIP_RV635_9596 */
0x9597, /* PCI_CHIP_RV635_9597 */
0x9598, /* PCI_CHIP_RV635_9598 */
0x9599, /* PCI_CHIP_RV635_9599 */
0x959B, /* PCI_CHIP_RV635_959B */
0x9610, /* PCI_CHIP_RS780_9610 */
0x9611, /* PCI_CHIP_RS780_9611 */
0x9612, /* PCI_CHIP_RS780_9612 */
0x9613, /* PCI_CHIP_RS780_9613 */
0x9614, /* PCI_CHIP_RS780_9614 */
0x9615, /* PCI_CHIP_RS780_9615 */
0x9616, /* PCI_CHIP_RS780_9616 */
0x9710, /* PCI_CHIP_RS880_9710 */
0x9711, /* PCI_CHIP_RS880_9711 */
0x9712, /* PCI_CHIP_RS880_9712 */
0x9713, /* PCI_CHIP_RS880_9713 */
0x9714, /* PCI_CHIP_RS880_9714 */
0x9715, /* PCI_CHIP_RS880_9715 */
0x9440, /* PCI_CHIP_RV770_9440 */
0x9441, /* PCI_CHIP_RV770_9441 */
0x9442, /* PCI_CHIP_RV770_9442 */
0x9443, /* PCI_CHIP_RV770_9443 */
0x9444, /* PCI_CHIP_RV770_9444 */
0x9446, /* PCI_CHIP_RV770_9446 */
0x944A, /* PCI_CHIP_RV770_944A */
0x944B, /* PCI_CHIP_RV770_944B */
0x944C, /* PCI_CHIP_RV770_944C */
0x944E, /* PCI_CHIP_RV770_944E */
0x9450, /* PCI_CHIP_RV770_9450 */
0x9452, /* PCI_CHIP_RV770_9452 */
0x9456, /* PCI_CHIP_RV770_9456 */
0x945A, /* PCI_CHIP_RV770_945A */
0x945B, /* PCI_CHIP_RV770_945B */
0x945E, /* PCI_CHIP_RV770_945E */
0x9460, /* PCI_CHIP_RV790_9460 */
0x9462, /* PCI_CHIP_RV790_9462 */
0x946A, /* PCI_CHIP_RV770_946A */
0x946B, /* PCI_CHIP_RV770_946B */
0x947A, /* PCI_CHIP_RV770_947A */
0x947B, /* PCI_CHIP_RV770_947B */
0x9480, /* PCI_CHIP_RV730_9480 */
0x9487, /* PCI_CHIP_RV730_9487 */
0x9488, /* PCI_CHIP_RV730_9488 */
0x9489, /* PCI_CHIP_RV730_9489 */
0x948A, /* PCI_CHIP_RV730_948A */
0x948F, /* PCI_CHIP_RV730_948F */
0x9490, /* PCI_CHIP_RV730_9490 */
0x9491, /* PCI_CHIP_RV730_9491 */
0x9495, /* PCI_CHIP_RV730_9495 */
0x9498, /* PCI_CHIP_RV730_9498 */
0x949C, /* PCI_CHIP_RV730_949C */
0x949E, /* PCI_CHIP_RV730_949E */
0x949F, /* PCI_CHIP_RV730_949F */
0x9540, /* PCI_CHIP_RV710_9540 */
0x9541, /* PCI_CHIP_RV710_9541 */
0x9542, /* PCI_CHIP_RV710_9542 */
0x954E, /* PCI_CHIP_RV710_954E */
0x954F, /* PCI_CHIP_RV710_954F */
0x9552, /* PCI_CHIP_RV710_9552 */
0x9553, /* PCI_CHIP_RV710_9553 */
0x9555, /* PCI_CHIP_RV710_9555 */
0x9557, /* PCI_CHIP_RV710_9557 */
0x955F, /* PCI_CHIP_RV710_955F */
0x94A0, /* PCI_CHIP_RV740_94A0 */
0x94A1, /* PCI_CHIP_RV740_94A1 */
0x94A3, /* PCI_CHIP_RV740_94A3 */
0x94B1, /* PCI_CHIP_RV740_94B1 */
0x94B3, /* PCI_CHIP_RV740_94B3 */
0x94B4, /* PCI_CHIP_RV740_94B4 */
0x94B5, /* PCI_CHIP_RV740_94B5 */
0x94B9, /* PCI_CHIP_RV740_94B9 */
0x68E0, /* PCI_CHIP_CEDAR_68E0 */
0x68E1, /* PCI_CHIP_CEDAR_68E1 */
0x68E4, /* PCI_CHIP_CEDAR_68E4 */
0x68E5, /* PCI_CHIP_CEDAR_68E5 */
0x68E8, /* PCI_CHIP_CEDAR_68E8 */
0x68E9, /* PCI_CHIP_CEDAR_68E9 */
0x68F1, /* PCI_CHIP_CEDAR_68F1 */
0x68F8, /* PCI_CHIP_CEDAR_68F8 */
0x68F9, /* PCI_CHIP_CEDAR_68F9 */
0x68FE, /* PCI_CHIP_CEDAR_68FE */
0x68C0, /* PCI_CHIP_REDWOOD_68C0 */
0x68C1, /* PCI_CHIP_REDWOOD_68C1 */
0x68C8, /* PCI_CHIP_REDWOOD_68C8 */
0x68C9, /* PCI_CHIP_REDWOOD_68C9 */
0x68D8, /* PCI_CHIP_REDWOOD_68D8 */
0x68D9, /* PCI_CHIP_REDWOOD_68D9 */
0x68DA, /* PCI_CHIP_REDWOOD_68DA */
0x68DE, /* PCI_CHIP_REDWOOD_68DE */
0x68A0, /* PCI_CHIP_JUNIPER_68A0 */
0x68A1, /* PCI_CHIP_JUNIPER_68A1 */
0x68A8, /* PCI_CHIP_JUNIPER_68A8 */
0x68A9, /* PCI_CHIP_JUNIPER_68A9 */
0x68B0, /* PCI_CHIP_JUNIPER_68B0 */
0x68B8, /* PCI_CHIP_JUNIPER_68B8 */
0x68B9, /* PCI_CHIP_JUNIPER_68B9 */
0x68BE, /* PCI_CHIP_JUNIPER_68BE */
0x6880, /* PCI_CHIP_CYPRESS_6880 */
0x6888, /* PCI_CHIP_CYPRESS_6888 */
0x6889, /* PCI_CHIP_CYPRESS_6889 */
0x688A, /* PCI_CHIP_CYPRESS_688A */
0x6898, /* PCI_CHIP_CYPRESS_6898 */
0x6899, /* PCI_CHIP_CYPRESS_6899 */
0x689E, /* PCI_CHIP_CYPRESS_689E */
0x689C, /* PCI_CHIP_HEMLOCK_689C */
0x689D, /* PCI_CHIP_HEMLOCK_689D */
};
const struct dri2_driver_map driver_map[] = {
{ 0x8086, "i915", i915_chip_ids, ARRAY_SIZE(i915_chip_ids) },
{ 0x8086, "i965", i965_chip_ids, ARRAY_SIZE(i965_chip_ids) },
{ 0x1002, "radeon", r100_chip_ids, ARRAY_SIZE(r100_chip_ids) },
{ 0x1002, "r200", r200_chip_ids, ARRAY_SIZE(r200_chip_ids) },
{ 0x1002, "r300", r300_chip_ids, ARRAY_SIZE(r300_chip_ids) },
{ 0x1002, "r600", r600_chip_ids, ARRAY_SIZE(r600_chip_ids) },
};
static char *
@@ -1524,8 +1086,6 @@ dri2_create_context(_EGLDriver *drv, _EGLDisplay *disp, _EGLConfig *conf,
const __DRIconfig *dri_config;
int api;
(void) drv;
dri2_ctx = malloc(sizeof *dri2_ctx);
if (!dri2_ctx) {
_eglError(EGL_BAD_ALLOC, "eglCreateContext");
@@ -1597,9 +1157,7 @@ dri2_destroy_surface(_EGLDriver *drv, _EGLDisplay *disp, _EGLSurface *surf)
struct dri2_egl_display *dri2_dpy = dri2_egl_display(disp);
struct dri2_egl_surface *dri2_surf = dri2_egl_surface(surf);
(void) drv;
if (!_eglPutSurface(surf))
if (_eglIsSurfaceBound(surf))
return EGL_TRUE;
(*dri2_dpy->core->destroyDrawable)(dri2_surf->dri_drawable);
@@ -1626,17 +1184,15 @@ dri2_make_current(_EGLDriver *drv, _EGLDisplay *disp, _EGLSurface *dsurf,
struct dri2_egl_surface *dri2_dsurf = dri2_egl_surface(dsurf);
struct dri2_egl_surface *dri2_rsurf = dri2_egl_surface(rsurf);
struct dri2_egl_context *dri2_ctx = dri2_egl_context(ctx);
_EGLContext *old_ctx;
_EGLSurface *old_dsurf, *old_rsurf;
__DRIdrawable *ddraw, *rdraw;
__DRIcontext *cctx;
/* make new bindings */
if (!_eglBindContext(ctx, dsurf, rsurf, &old_ctx, &old_dsurf, &old_rsurf))
/* bind the new context and return the "orphaned" one */
if (!_eglBindContext(&ctx, &dsurf, &rsurf))
return EGL_FALSE;
/* flush before context switch */
if (old_ctx && dri2_drv->glFlush)
if (ctx && dri2_drv->glFlush)
dri2_drv->glFlush();
ddraw = (dri2_dsurf) ? dri2_dsurf->dri_drawable : NULL;
@@ -1645,33 +1201,16 @@ dri2_make_current(_EGLDriver *drv, _EGLDisplay *disp, _EGLSurface *dsurf,
if ((cctx == NULL && ddraw == NULL && rdraw == NULL) ||
dri2_dpy->core->bindContext(cctx, ddraw, rdraw)) {
dri2_destroy_surface(drv, disp, old_dsurf);
dri2_destroy_surface(drv, disp, old_rsurf);
if (old_ctx) {
/* unbind the old context only when there is no new context bound */
if (!ctx) {
__DRIcontext *old_cctx = dri2_egl_context(old_ctx)->dri_context;
dri2_dpy->core->unbindContext(old_cctx);
}
/* no destroy? */
_eglPutContext(old_ctx);
}
if (dsurf && !_eglIsSurfaceLinked(dsurf))
dri2_destroy_surface(drv, disp, dsurf);
if (rsurf && rsurf != dsurf && !_eglIsSurfaceLinked(dsurf))
dri2_destroy_surface(drv, disp, rsurf);
if (ctx != NULL && !_eglIsContextLinked(ctx))
dri2_dpy->core->unbindContext(dri2_egl_context(ctx)->dri_context);
return EGL_TRUE;
} else {
/* undo the previous _eglBindContext */
_eglBindContext(old_ctx, old_dsurf, old_rsurf, &ctx, &dsurf, &rsurf);
assert(&dri2_ctx->base == ctx &&
&dri2_dsurf->base == dsurf &&
&dri2_rsurf->base == rsurf);
_eglPutSurface(dsurf);
_eglPutSurface(rsurf);
_eglPutContext(ctx);
_eglPutSurface(old_dsurf);
_eglPutSurface(old_rsurf);
_eglPutContext(old_ctx);
_eglBindContext(&ctx, &dsurf, &rsurf);
return EGL_FALSE;
}
@@ -1693,8 +1232,6 @@ dri2_create_surface(_EGLDriver *drv, _EGLDisplay *disp, EGLint type,
xcb_screen_iterator_t s;
xcb_generic_error_t *error;
(void) drv;
dri2_surf = malloc(sizeof *dri2_surf);
if (!dri2_surf) {
_eglError(EGL_BAD_ALLOC, "dri2_create_surface");
@@ -1708,7 +1245,8 @@ dri2_create_surface(_EGLDriver *drv, _EGLDisplay *disp, EGLint type,
if (type == EGL_PBUFFER_BIT) {
dri2_surf->drawable = xcb_generate_id(dri2_dpy->conn);
s = xcb_setup_roots_iterator(xcb_get_setup(dri2_dpy->conn));
xcb_create_pixmap(dri2_dpy->conn, conf->BufferSize,
xcb_create_pixmap(dri2_dpy->conn,
_eglGetConfigKey(conf, EGL_BUFFER_SIZE),
dri2_surf->drawable, s.data->root,
dri2_surf->base.Width, dri2_surf->base.Height);
} else {
@@ -1842,7 +1380,7 @@ dri2_swap_buffers_region(_EGLDriver *drv, _EGLDisplay *disp, _EGLSurface *draw,
xcb_rectangle_t rectangles[16];
int i;
if (numRects > (int)ARRAY_SIZE(rectangles))
if (numRects > ARRAY_SIZE(rectangles))
return dri2_copy_region(drv, disp, draw, dri2_surf->region);
/* FIXME: Invert y here? */
@@ -1867,8 +1405,6 @@ dri2_swap_buffers_region(_EGLDriver *drv, _EGLDisplay *disp, _EGLSurface *draw,
static _EGLProc
dri2_get_proc_address(_EGLDriver *drv, const char *procname)
{
(void) drv;
/* FIXME: Do we need to support lookup of EGL symbols too? */
return (_EGLProc) _glapi_get_proc_address(procname);
@@ -1880,8 +1416,6 @@ dri2_wait_client(_EGLDriver *drv, _EGLDisplay *disp, _EGLContext *ctx)
struct dri2_egl_display *dri2_dpy = dri2_egl_display(disp);
struct dri2_egl_surface *dri2_surf = dri2_egl_surface(ctx->DrawSurface);
(void) drv;
/* FIXME: If EGL allows frontbuffer rendering for window surfaces,
* we need to copy fake to real here.*/
@@ -1893,9 +1427,6 @@ dri2_wait_client(_EGLDriver *drv, _EGLDisplay *disp, _EGLContext *ctx)
static EGLBoolean
dri2_wait_native(_EGLDriver *drv, _EGLDisplay *disp, EGLint engine)
{
(void) drv;
(void) disp;
if (engine != EGL_CORE_NATIVE_ENGINE)
return _eglError(EGL_BAD_PARAMETER, "eglWaitNative");
/* glXWaitX(); */
@@ -1918,8 +1449,6 @@ dri2_copy_buffers(_EGLDriver *drv, _EGLDisplay *disp, _EGLSurface *surf,
struct dri2_egl_surface *dri2_surf = dri2_egl_surface(surf);
xcb_gcontext_t gc;
(void) drv;
(*dri2_dpy->flush->flush)(dri2_surf->dri_drawable);
gc = xcb_generate_id(dri2_dpy->conn);
@@ -1983,11 +1512,6 @@ static EGLBoolean
dri2_release_tex_image(_EGLDriver *drv,
_EGLDisplay *disp, _EGLSurface *surf, EGLint buffer)
{
(void) drv;
(void) disp;
(void) surf;
(void) buffer;
return EGL_TRUE;
}
@@ -2007,8 +1531,6 @@ dri2_create_image_khr_pixmap(_EGLDisplay *disp, _EGLContext *ctx,
xcb_generic_error_t *error;
int stride, format;
(void) ctx;
drawable = (xcb_drawable_t) buffer;
xcb_dri2_create_drawable (dri2_dpy->conn, drawable);
attachments[0] = XCB_DRI2_ATTACHMENT_BUFFER_FRONT_LEFT;
@@ -2057,7 +1579,7 @@ dri2_create_image_khr_pixmap(_EGLDisplay *disp, _EGLContext *ctx,
return EGL_NO_IMAGE_KHR;
}
if (!_eglInitImage(&dri2_img->base, disp)) {
if (!_eglInitImage(&dri2_img->base, disp, attr_list)) {
free(buffers_reply);
free(geometry_reply);
return EGL_NO_IMAGE_KHR;
@@ -2100,7 +1622,7 @@ dri2_create_image_khr_renderbuffer(_EGLDisplay *disp, _EGLContext *ctx,
return EGL_NO_IMAGE_KHR;
}
if (!_eglInitImage(&dri2_img->base, disp))
if (!_eglInitImage(&dri2_img->base, disp, attr_list))
return EGL_NO_IMAGE_KHR;
dri2_img->dri_image =
@@ -2117,28 +1639,54 @@ dri2_create_image_mesa_drm_buffer(_EGLDisplay *disp, _EGLContext *ctx,
{
struct dri2_egl_display *dri2_dpy = dri2_egl_display(disp);
struct dri2_egl_image *dri2_img;
EGLint format, name, pitch, err;
_EGLImageAttribs attrs;
(void) ctx;
EGLint width, height, format, name, stride, pitch, i, err;
name = (EGLint) buffer;
err = _eglParseImageAttribList(&attrs, disp, attr_list);
if (err != EGL_SUCCESS)
return NULL;
err = EGL_SUCCESS;
width = 0;
height = 0;
format = 0;
stride = 0;
if (attrs.Width <= 0 || attrs.Height <= 0 ||
attrs.DRMBufferStrideMESA <= 0) {
for (i = 0; attr_list[i] != EGL_NONE; i++) {
EGLint attr = attr_list[i++];
EGLint val = attr_list[i];
switch (attr) {
case EGL_WIDTH:
width = val;
break;
case EGL_HEIGHT:
height = val;
break;
case EGL_DRM_BUFFER_FORMAT_MESA:
format = val;
break;
case EGL_DRM_BUFFER_STRIDE_MESA:
stride = val;
break;
default:
err = EGL_BAD_ATTRIBUTE;
break;
}
if (err != EGL_SUCCESS) {
_eglLog(_EGL_WARNING, "bad image attribute 0x%04x", attr);
return NULL;
}
}
if (width <= 0 || height <= 0 || stride <= 0) {
_eglError(EGL_BAD_PARAMETER,
"bad width, height or stride");
return NULL;
}
switch (attrs.DRMBufferFormatMESA) {
switch (format) {
case EGL_DRM_BUFFER_FORMAT_ARGB32_MESA:
format = __DRI_IMAGE_FORMAT_ARGB8888;
pitch = attrs.DRMBufferStrideMESA;
pitch = stride;
break;
default:
_eglError(EGL_BAD_PARAMETER,
@@ -2152,15 +1700,15 @@ dri2_create_image_mesa_drm_buffer(_EGLDisplay *disp, _EGLContext *ctx,
return NULL;
}
if (!_eglInitImage(&dri2_img->base, disp)) {
if (!_eglInitImage(&dri2_img->base, disp, attr_list)) {
free(dri2_img);
return NULL;
}
dri2_img->dri_image =
dri2_dpy->image->createImageFromName(dri2_dpy->dri_screen,
attrs.Width,
attrs.Height,
width,
height,
format,
name,
pitch,
@@ -2179,8 +1727,6 @@ dri2_create_image_khr(_EGLDriver *drv, _EGLDisplay *disp,
_EGLContext *ctx, EGLenum target,
EGLClientBuffer buffer, const EGLint *attr_list)
{
(void) drv;
switch (target) {
case EGL_NATIVE_PIXMAP_KHR:
return dri2_create_image_khr_pixmap(disp, ctx, buffer, attr_list);
@@ -2200,8 +1746,6 @@ dri2_destroy_image_khr(_EGLDriver *drv, _EGLDisplay *disp, _EGLImage *image)
struct dri2_egl_display *dri2_dpy = dri2_egl_display(disp);
struct dri2_egl_image *dri2_img = dri2_egl_image(image);
(void) drv;
dri2_dpy->image->destroyImage(dri2_img->dri_image);
free(dri2_img);
@@ -2214,13 +1758,10 @@ dri2_create_drm_image_mesa(_EGLDriver *drv, _EGLDisplay *disp,
{
struct dri2_egl_display *dri2_dpy = dri2_egl_display(disp);
struct dri2_egl_image *dri2_img;
_EGLImageAttribs attrs;
unsigned int dri_use, valid_mask;
int format;
int width, height, format, i;
unsigned int use, dri_use, valid_mask;
EGLint err = EGL_SUCCESS;
(void) drv;
dri2_img = malloc(sizeof *dri2_img);
if (!dri2_img) {
_eglError(EGL_BAD_ALLOC, "dri2_create_image_khr");
@@ -2232,50 +1773,74 @@ dri2_create_drm_image_mesa(_EGLDriver *drv, _EGLDisplay *disp,
goto cleanup_img;
}
if (!_eglInitImage(&dri2_img->base, disp)) {
if (!_eglInitImage(&dri2_img->base, disp, attr_list)) {
err = EGL_BAD_PARAMETER;
goto cleanup_img;
}
err = _eglParseImageAttribList(&attrs, disp, attr_list);
if (err != EGL_SUCCESS)
goto cleanup_img;
width = 0;
height = 0;
format = 0;
use = 0;
for (i = 0; attr_list[i] != EGL_NONE; i++) {
EGLint attr = attr_list[i++];
EGLint val = attr_list[i];
if (attrs.Width <= 0 || attrs.Height <= 0) {
_eglLog(_EGL_WARNING, "bad width or height (%dx%d)",
attrs.Width, attrs.Height);
switch (attr) {
case EGL_WIDTH:
width = val;
break;
case EGL_HEIGHT:
height = val;
break;
case EGL_DRM_BUFFER_FORMAT_MESA:
format = val;
break;
case EGL_DRM_BUFFER_USE_MESA:
use = val;
break;
default:
err = EGL_BAD_ATTRIBUTE;
break;
}
if (err != EGL_SUCCESS) {
_eglLog(_EGL_WARNING, "bad image attribute 0x%04x", attr);
goto cleanup_img;
}
}
if (width <= 0 || height <= 0) {
_eglLog(_EGL_WARNING, "bad width or height (%dx%d)", width, height);
goto cleanup_img;
}
switch (attrs.DRMBufferFormatMESA) {
switch (format) {
case EGL_DRM_BUFFER_FORMAT_ARGB32_MESA:
format = __DRI_IMAGE_FORMAT_ARGB8888;
break;
default:
_eglLog(_EGL_WARNING, "bad image format value 0x%04x",
attrs.DRMBufferFormatMESA);
_eglLog(_EGL_WARNING, "bad image format value 0x%04x", format);
goto cleanup_img;
}
valid_mask =
EGL_DRM_BUFFER_USE_SCANOUT_MESA |
EGL_DRM_BUFFER_USE_SHARE_MESA;
if (attrs.DRMBufferUseMESA & ~valid_mask) {
_eglLog(_EGL_WARNING, "bad image use bit 0x%04x",
attrs.DRMBufferUseMESA & ~valid_mask);
if (use & ~valid_mask) {
_eglLog(_EGL_WARNING, "bad image use bit 0x%04x", use & ~valid_mask);
goto cleanup_img;
}
dri_use = 0;
if (attrs.DRMBufferUseMESA & EGL_DRM_BUFFER_USE_SHARE_MESA)
if (use & EGL_DRM_BUFFER_USE_SHARE_MESA)
dri_use |= __DRI_IMAGE_USE_SHARE;
if (attrs.DRMBufferUseMESA & EGL_DRM_BUFFER_USE_SCANOUT_MESA)
if (use & EGL_DRM_BUFFER_USE_SCANOUT_MESA)
dri_use |= __DRI_IMAGE_USE_SCANOUT;
dri2_img->dri_image =
dri2_dpy->image->createImage(dri2_dpy->dri_screen,
attrs.Width, attrs.Height,
format, dri_use, dri2_img);
width, height, format, dri_use, dri2_img);
if (dri2_img->dri_image == NULL) {
err = EGL_BAD_ALLOC;
goto cleanup_img;
@@ -2297,8 +1862,6 @@ dri2_export_drm_image_mesa(_EGLDriver *drv, _EGLDisplay *disp, _EGLImage *img,
struct dri2_egl_display *dri2_dpy = dri2_egl_display(disp);
struct dri2_egl_image *dri2_img = dri2_egl_image(img);
(void) drv;
if (name && !dri2_dpy->image->queryImage(dri2_img->dri_image,
__DRI_IMAGE_ATTRIB_NAME, name)) {
_eglError(EGL_BAD_ALLOC, "dri2_export_drm_image_mesa");
@@ -2325,8 +1888,6 @@ _eglMain(const char *args)
{
struct dri2_egl_driver *dri2_drv;
(void) args;
dri2_drv = malloc(sizeof *dri2_drv);
if (!dri2_drv)
return NULL;

View File

@@ -132,38 +132,28 @@ static const struct {
int egl_attr;
} fbconfig_attributes[] = {
/* table 3.1 of GLX 1.4 */
{ GLX_FBCONFIG_ID, 0 },
{ GLX_BUFFER_SIZE, EGL_BUFFER_SIZE },
{ GLX_LEVEL, EGL_LEVEL },
{ GLX_DOUBLEBUFFER, 0 },
{ GLX_STEREO, 0 },
{ GLX_AUX_BUFFERS, 0 },
{ GLX_RED_SIZE, EGL_RED_SIZE },
{ GLX_GREEN_SIZE, EGL_GREEN_SIZE },
{ GLX_BLUE_SIZE, EGL_BLUE_SIZE },
{ GLX_ALPHA_SIZE, EGL_ALPHA_SIZE },
{ GLX_DEPTH_SIZE, EGL_DEPTH_SIZE },
{ GLX_STENCIL_SIZE, EGL_STENCIL_SIZE },
{ GLX_ACCUM_RED_SIZE, 0 },
{ GLX_ACCUM_GREEN_SIZE, 0 },
{ GLX_ACCUM_BLUE_SIZE, 0 },
{ GLX_ACCUM_ALPHA_SIZE, 0 },
{ GLX_SAMPLE_BUFFERS, EGL_SAMPLE_BUFFERS },
{ GLX_SAMPLES, EGL_SAMPLES },
{ GLX_RENDER_TYPE, 0 },
{ GLX_DRAWABLE_TYPE, EGL_SURFACE_TYPE },
{ GLX_X_RENDERABLE, EGL_NATIVE_RENDERABLE },
{ GLX_X_VISUAL_TYPE, EGL_NATIVE_VISUAL_TYPE },
{ GLX_CONFIG_CAVEAT, EGL_CONFIG_CAVEAT },
{ GLX_TRANSPARENT_TYPE, EGL_TRANSPARENT_TYPE },
{ GLX_TRANSPARENT_INDEX_VALUE, 0 },
{ GLX_TRANSPARENT_RED_VALUE, EGL_TRANSPARENT_RED_VALUE },
{ GLX_TRANSPARENT_GREEN_VALUE, EGL_TRANSPARENT_GREEN_VALUE },
{ GLX_TRANSPARENT_BLUE_VALUE, EGL_TRANSPARENT_BLUE_VALUE },
{ GLX_MAX_PBUFFER_WIDTH, EGL_MAX_PBUFFER_WIDTH },
{ GLX_MAX_PBUFFER_HEIGHT, EGL_MAX_PBUFFER_HEIGHT },
{ GLX_MAX_PBUFFER_PIXELS, EGL_MAX_PBUFFER_PIXELS },
{ GLX_VISUAL_ID, EGL_NATIVE_VISUAL_ID }
{ GLX_BUFFER_SIZE, EGL_BUFFER_SIZE },
{ GLX_LEVEL, EGL_LEVEL },
{ GLX_RED_SIZE, EGL_RED_SIZE },
{ GLX_GREEN_SIZE, EGL_GREEN_SIZE },
{ GLX_BLUE_SIZE, EGL_BLUE_SIZE },
{ GLX_ALPHA_SIZE, EGL_ALPHA_SIZE },
{ GLX_DEPTH_SIZE, EGL_DEPTH_SIZE },
{ GLX_STENCIL_SIZE, EGL_STENCIL_SIZE },
{ GLX_SAMPLE_BUFFERS, EGL_SAMPLE_BUFFERS },
{ GLX_SAMPLES, EGL_SAMPLES },
{ GLX_X_RENDERABLE, EGL_NATIVE_RENDERABLE },
{ GLX_X_VISUAL_TYPE, EGL_NATIVE_VISUAL_TYPE },
{ GLX_CONFIG_CAVEAT, EGL_CONFIG_CAVEAT },
{ GLX_TRANSPARENT_TYPE, EGL_TRANSPARENT_TYPE },
{ GLX_TRANSPARENT_RED_VALUE, EGL_TRANSPARENT_RED_VALUE },
{ GLX_TRANSPARENT_GREEN_VALUE, EGL_TRANSPARENT_GREEN_VALUE },
{ GLX_TRANSPARENT_BLUE_VALUE, EGL_TRANSPARENT_BLUE_VALUE },
{ GLX_MAX_PBUFFER_WIDTH, EGL_MAX_PBUFFER_WIDTH },
{ GLX_MAX_PBUFFER_HEIGHT, EGL_MAX_PBUFFER_HEIGHT },
{ GLX_MAX_PBUFFER_PIXELS, EGL_MAX_PBUFFER_PIXELS },
{ GLX_VISUAL_ID, EGL_NATIVE_VISUAL_ID },
{ GLX_X_VISUAL_TYPE, EGL_NATIVE_VISUAL_TYPE },
};
@@ -171,31 +161,12 @@ static EGLBoolean
convert_fbconfig(Display *dpy, GLXFBConfig fbconfig,
struct GLX_egl_config *GLX_conf)
{
int err, attr, val;
unsigned i;
/* must have rgba bit */
err = glXGetFBConfigAttrib(dpy, fbconfig, GLX_RENDER_TYPE, &val);
if (err || !(val & GLX_RGBA_BIT))
return EGL_FALSE;
/* must know whether it is double-buffered */
err = glXGetFBConfigAttrib(dpy, fbconfig, GLX_DOUBLEBUFFER, &val);
if (err)
return EGL_FALSE;
GLX_conf->double_buffered = val;
GLX_conf->Base.RenderableType = EGL_OPENGL_BIT;
GLX_conf->Base.Conformant = EGL_OPENGL_BIT;
int err = 0, attr, egl_attr, val, i;
EGLint conformant, config_caveat, surface_type;
for (i = 0; i < ARRAY_SIZE(fbconfig_attributes); i++) {
EGLint egl_attr, egl_val;
attr = fbconfig_attributes[i].attr;
egl_attr = fbconfig_attributes[i].egl_attr;
if (!egl_attr)
continue;
err = glXGetFBConfigAttrib(dpy, fbconfig, attr, &val);
if (err) {
if (err == GLX_BAD_ATTRIBUTE) {
@@ -205,71 +176,57 @@ convert_fbconfig(Display *dpy, GLXFBConfig fbconfig,
break;
}
switch (egl_attr) {
case EGL_SURFACE_TYPE:
egl_val = 0;
if (val & GLX_WINDOW_BIT)
egl_val |= EGL_WINDOW_BIT;
/* pixmap and pbuffer surfaces must be single-buffered in EGL */
if (!GLX_conf->double_buffered) {
if (val & GLX_PIXMAP_BIT)
egl_val |= EGL_PIXMAP_BIT;
if (val & GLX_PBUFFER_BIT)
egl_val |= EGL_PBUFFER_BIT;
}
break;
case EGL_NATIVE_VISUAL_TYPE:
switch (val) {
case GLX_TRUE_COLOR:
egl_val = TrueColor;
break;
case GLX_DIRECT_COLOR:
egl_val = DirectColor;
break;
case GLX_PSEUDO_COLOR:
egl_val = PseudoColor;
break;
case GLX_STATIC_COLOR:
egl_val = StaticColor;
break;
case GLX_GRAY_SCALE:
egl_val = GrayScale;
break;
case GLX_STATIC_GRAY:
egl_val = StaticGray;
break;
default:
egl_val = EGL_NONE;
break;
}
break;
case EGL_CONFIG_CAVEAT:
egl_val = EGL_NONE;
if (val == GLX_SLOW_CONFIG) {
egl_val = EGL_SLOW_CONFIG;
}
else if (val == GLX_NON_CONFORMANT_CONFIG) {
GLX_conf->Base.Conformant &= ~EGL_OPENGL_BIT;
egl_val = EGL_NONE;
}
break;
case EGL_TRANSPARENT_TYPE:
egl_val = (val == GLX_TRANSPARENT_RGB) ?
EGL_TRANSPARENT_RGB : EGL_NONE;
break;
default:
egl_val = val;
break;
}
_eglSetConfigKey(&GLX_conf->Base, egl_attr, egl_val);
_eglSetConfigKey(&GLX_conf->Base, egl_attr, val);
}
if (err)
return EGL_FALSE;
if (!GLX_conf->Base.SurfaceType)
/* must have rgba bit */
glXGetFBConfigAttrib(dpy, fbconfig, GLX_RENDER_TYPE, &val);
if (!(val & GLX_RGBA_BIT))
return EGL_FALSE;
conformant = EGL_OPENGL_BIT;
glXGetFBConfigAttrib(dpy, fbconfig, GLX_CONFIG_CAVEAT, &val);
if (val == GLX_SLOW_CONFIG)
config_caveat = EGL_SLOW_CONFIG;
if (val == GLX_NON_CONFORMANT_CONFIG)
conformant &= ~EGL_OPENGL_BIT;
if (!(conformant & EGL_OPENGL_ES_BIT))
config_caveat = EGL_NON_CONFORMANT_CONFIG;
_eglSetConfigKey(&GLX_conf->Base, EGL_CONFIG_CAVEAT, config_caveat);
surface_type = 0;
glXGetFBConfigAttrib(dpy, fbconfig, GLX_DRAWABLE_TYPE, &val);
if (val & GLX_WINDOW_BIT)
surface_type |= EGL_WINDOW_BIT;
if (val & GLX_PIXMAP_BIT)
surface_type |= EGL_PIXMAP_BIT;
if (val & GLX_PBUFFER_BIT)
surface_type |= EGL_PBUFFER_BIT;
/* pixmap and pbuffer surfaces must be single-buffered in EGL */
glXGetFBConfigAttrib(dpy, fbconfig, GLX_DOUBLEBUFFER, &val);
GLX_conf->double_buffered = val;
if (GLX_conf->double_buffered) {
surface_type &= ~(EGL_PIXMAP_BIT | EGL_PBUFFER_BIT);
if (!surface_type)
return EGL_FALSE;
}
_eglSetConfigKey(&GLX_conf->Base, EGL_SURFACE_TYPE, surface_type);
_eglSetConfigKey(&GLX_conf->Base, EGL_RENDERABLE_TYPE, EGL_OPENGL_BIT);
glXGetFBConfigAttrib(dpy, fbconfig, GLX_TRANSPARENT_TYPE, &val);
if (val == GLX_NONE) {
_eglSetConfigKey(&GLX_conf->Base, EGL_TRANSPARENT_TYPE, EGL_NONE);
_eglSetConfigKey(&GLX_conf->Base, EGL_TRANSPARENT_RED_VALUE, 0);
_eglSetConfigKey(&GLX_conf->Base, EGL_TRANSPARENT_GREEN_VALUE, 0);
_eglSetConfigKey(&GLX_conf->Base, EGL_TRANSPARENT_BLUE_VALUE, 0);
}
return EGL_TRUE;
}
@@ -278,69 +235,34 @@ static const struct {
int egl_attr;
} visual_attributes[] = {
/* table 3.7 of GLX 1.4 */
{ GLX_USE_GL, 0 },
{ GLX_BUFFER_SIZE, EGL_BUFFER_SIZE },
{ GLX_LEVEL, EGL_LEVEL },
{ GLX_RGBA, 0 },
{ GLX_DOUBLEBUFFER, 0 },
{ GLX_STEREO, 0 },
{ GLX_AUX_BUFFERS, 0 },
{ GLX_RED_SIZE, EGL_RED_SIZE },
{ GLX_GREEN_SIZE, EGL_GREEN_SIZE },
{ GLX_BLUE_SIZE, EGL_BLUE_SIZE },
{ GLX_ALPHA_SIZE, EGL_ALPHA_SIZE },
{ GLX_DEPTH_SIZE, EGL_DEPTH_SIZE },
{ GLX_STENCIL_SIZE, EGL_STENCIL_SIZE },
{ GLX_ACCUM_RED_SIZE, 0 },
{ GLX_ACCUM_GREEN_SIZE, 0 },
{ GLX_ACCUM_BLUE_SIZE, 0 },
{ GLX_ACCUM_ALPHA_SIZE, 0 },
{ GLX_SAMPLE_BUFFERS, EGL_SAMPLE_BUFFERS },
{ GLX_SAMPLES, EGL_SAMPLES },
{ GLX_FBCONFIG_ID, 0 },
/* GLX_EXT_visual_rating */
{ GLX_VISUAL_CAVEAT_EXT, EGL_CONFIG_CAVEAT }
/* no GLX_USE_GL */
{ GLX_BUFFER_SIZE, EGL_BUFFER_SIZE },
{ GLX_LEVEL, EGL_LEVEL },
{ GLX_RED_SIZE, EGL_RED_SIZE },
{ GLX_GREEN_SIZE, EGL_GREEN_SIZE },
{ GLX_BLUE_SIZE, EGL_BLUE_SIZE },
{ GLX_ALPHA_SIZE, EGL_ALPHA_SIZE },
{ GLX_DEPTH_SIZE, EGL_DEPTH_SIZE },
{ GLX_STENCIL_SIZE, EGL_STENCIL_SIZE },
{ GLX_SAMPLE_BUFFERS, EGL_SAMPLE_BUFFERS },
{ GLX_SAMPLES, EGL_SAMPLES },
};
static EGLBoolean
convert_visual(Display *dpy, XVisualInfo *vinfo,
struct GLX_egl_config *GLX_conf)
{
int err, attr, val;
unsigned i;
int err, attr, egl_attr, val, i;
EGLint conformant, config_caveat, surface_type;
/* the visual must support OpenGL and RGBA buffer */
/* the visual must support OpenGL */
err = glXGetConfig(dpy, vinfo, GLX_USE_GL, &val);
if (!err && val)
err = glXGetConfig(dpy, vinfo, GLX_RGBA, &val);
if (err || !val)
return EGL_FALSE;
/* must know whether it is double-buffered */
err = glXGetConfig(dpy, vinfo, GLX_DOUBLEBUFFER, &val);
if (err)
return EGL_FALSE;
GLX_conf->double_buffered = val;
GLX_conf->Base.RenderableType = EGL_OPENGL_BIT;
GLX_conf->Base.Conformant = EGL_OPENGL_BIT;
GLX_conf->Base.SurfaceType = EGL_WINDOW_BIT;
/* pixmap surfaces must be single-buffered in EGL */
if (!GLX_conf->double_buffered)
GLX_conf->Base.SurfaceType |= EGL_PIXMAP_BIT;
GLX_conf->Base.NativeVisualID = vinfo->visualid;
GLX_conf->Base.NativeVisualType = vinfo->class;
GLX_conf->Base.NativeRenderable = EGL_TRUE;
for (i = 0; i < ARRAY_SIZE(visual_attributes); i++) {
EGLint egl_attr, egl_val;
attr = visual_attributes[i].attr;
egl_attr = visual_attributes[i].egl_attr;
if (!egl_attr)
continue;
egl_attr = fbconfig_attributes[i].egl_attr;
err = glXGetConfig(dpy, vinfo, attr, &val);
if (err) {
if (err == GLX_BAD_ATTRIBUTE) {
@@ -350,26 +272,42 @@ convert_visual(Display *dpy, XVisualInfo *vinfo,
break;
}
switch (egl_attr) {
case EGL_CONFIG_CAVEAT:
egl_val = EGL_NONE;
if (val == GLX_SLOW_VISUAL_EXT) {
egl_val = EGL_SLOW_CONFIG;
}
else if (val == GLX_NON_CONFORMANT_VISUAL_EXT) {
GLX_conf->Base.Conformant &= ~EGL_OPENGL_BIT;
egl_val = EGL_NONE;
}
break;
break;
default:
egl_val = val;
break;
}
_eglSetConfigKey(&GLX_conf->Base, egl_attr, egl_val);
_eglSetConfigKey(&GLX_conf->Base, egl_attr, val);
}
if (err)
return EGL_FALSE;
return (err) ? EGL_FALSE : EGL_TRUE;
glXGetConfig(dpy, vinfo, GLX_RGBA, &val);
if (!val)
return EGL_FALSE;
conformant = EGL_OPENGL_BIT;
glXGetConfig(dpy, vinfo, GLX_VISUAL_CAVEAT_EXT, &val);
if (val == GLX_SLOW_CONFIG)
config_caveat = EGL_SLOW_CONFIG;
if (val == GLX_NON_CONFORMANT_CONFIG)
conformant &= ~EGL_OPENGL_BIT;
if (!(conformant & EGL_OPENGL_ES_BIT))
config_caveat = EGL_NON_CONFORMANT_CONFIG;
_eglSetConfigKey(&GLX_conf->Base, EGL_CONFIG_CAVEAT, config_caveat);
_eglSetConfigKey(&GLX_conf->Base, EGL_NATIVE_VISUAL_ID, vinfo->visualid);
_eglSetConfigKey(&GLX_conf->Base, EGL_NATIVE_VISUAL_TYPE, vinfo->class);
/* pixmap and pbuffer surfaces must be single-buffered in EGL */
glXGetConfig(dpy, vinfo, GLX_DOUBLEBUFFER, &val);
GLX_conf->double_buffered = val;
surface_type = EGL_WINDOW_BIT;
/* pixmap surfaces must be single-buffered in EGL */
if (!GLX_conf->double_buffered)
surface_type |= EGL_PIXMAP_BIT;
_eglSetConfigKey(&GLX_conf->Base, EGL_SURFACE_TYPE, surface_type);
_eglSetConfigKey(&GLX_conf->Base, EGL_RENDERABLE_TYPE, EGL_OPENGL_BIT);
_eglSetConfigKey(&GLX_conf->Base, EGL_NATIVE_RENDERABLE, EGL_TRUE);
return EGL_TRUE;
}
@@ -377,31 +315,30 @@ static void
fix_config(struct GLX_egl_display *GLX_dpy, struct GLX_egl_config *GLX_conf)
{
_EGLConfig *conf = &GLX_conf->Base;
EGLint surface_type, r, g, b, a;
surface_type = GET_CONFIG_ATTRIB(conf, EGL_SURFACE_TYPE);
if (!GLX_conf->double_buffered && GLX_dpy->single_buffered_quirk) {
/* some GLX impls do not like single-buffered window surface */
conf->SurfaceType &= ~EGL_WINDOW_BIT;
surface_type &= ~EGL_WINDOW_BIT;
/* pbuffer bit is usually not set */
if (GLX_dpy->have_pbuffer)
conf->SurfaceType |= EGL_PBUFFER_BIT;
surface_type |= EGL_PBUFFER_BIT;
SET_CONFIG_ATTRIB(conf, EGL_SURFACE_TYPE, surface_type);
}
/* no visual attribs unless window bit is set */
if (!(conf->SurfaceType & EGL_WINDOW_BIT)) {
conf->NativeVisualID = 0;
conf->NativeVisualType = EGL_NONE;
}
if (conf->TransparentType != EGL_TRANSPARENT_RGB) {
/* some impls set them to -1 (GLX_DONT_CARE) */
conf->TransparentRedValue = 0;
conf->TransparentGreenValue = 0;
conf->TransparentBlueValue = 0;
if (!(surface_type & EGL_WINDOW_BIT)) {
SET_CONFIG_ATTRIB(conf, EGL_NATIVE_VISUAL_ID, 0);
SET_CONFIG_ATTRIB(conf, EGL_NATIVE_VISUAL_TYPE, EGL_NONE);
}
/* make sure buffer size is set correctly */
conf->BufferSize =
conf->RedSize + conf->GreenSize + conf->BlueSize + conf->AlphaSize;
r = GET_CONFIG_ATTRIB(conf, EGL_RED_SIZE);
g = GET_CONFIG_ATTRIB(conf, EGL_GREEN_SIZE);
b = GET_CONFIG_ATTRIB(conf, EGL_BLUE_SIZE);
a = GET_CONFIG_ATTRIB(conf, EGL_ALPHA_SIZE);
SET_CONFIG_ATTRIB(conf, EGL_BUFFER_SIZE, r + g + b + a);
}
@@ -452,7 +389,7 @@ create_configs(_EGLDisplay *dpy, struct GLX_egl_display *GLX_dpy,
memcpy(GLX_conf, &template, sizeof(template));
GLX_conf->index = i;
_eglLinkConfig(&GLX_conf->Base);
_eglAddConfig(dpy, &GLX_conf->Base);
id++;
}
}
@@ -530,8 +467,6 @@ GLX_eglInitialize(_EGLDriver *drv, _EGLDisplay *disp,
{
struct GLX_egl_display *GLX_dpy;
(void) drv;
if (disp->Platform != _EGL_PLATFORM_X11)
return EGL_FALSE;
@@ -616,8 +551,6 @@ GLX_eglCreateContext(_EGLDriver *drv, _EGLDisplay *disp, _EGLConfig *conf,
struct GLX_egl_display *GLX_dpy = GLX_egl_display(disp);
struct GLX_egl_context *GLX_ctx_shared = GLX_egl_context(share_list);
(void) drv;
if (!GLX_ctx) {
_eglError(EGL_BAD_ALLOC, "eglCreateContext");
return NULL;
@@ -677,16 +610,12 @@ GLX_eglMakeCurrent(_EGLDriver *drv, _EGLDisplay *disp, _EGLSurface *dsurf,
struct GLX_egl_surface *GLX_dsurf = GLX_egl_surface(dsurf);
struct GLX_egl_surface *GLX_rsurf = GLX_egl_surface(rsurf);
struct GLX_egl_context *GLX_ctx = GLX_egl_context(ctx);
_EGLContext *old_ctx;
_EGLSurface *old_dsurf, *old_rsurf;
GLXDrawable ddraw, rdraw;
GLXContext cctx;
EGLBoolean ret = EGL_FALSE;
(void) drv;
/* make new bindings */
if (!_eglBindContext(ctx, dsurf, rsurf, &old_ctx, &old_dsurf, &old_rsurf))
/* bind the new context and return the "orphaned" one */
if (!_eglBindContext(&ctx, &dsurf, &rsurf))
return EGL_FALSE;
ddraw = (GLX_dsurf) ? GLX_dsurf->glx_drawable : None;
@@ -699,27 +628,13 @@ GLX_eglMakeCurrent(_EGLDriver *drv, _EGLDisplay *disp, _EGLSurface *dsurf,
ret = glXMakeCurrent(GLX_dpy->dpy, ddraw, cctx);
if (ret) {
if (_eglPutSurface(old_dsurf))
destroy_surface(disp, old_dsurf);
if (_eglPutSurface(old_rsurf))
destroy_surface(disp, old_rsurf);
/* no destroy? */
_eglPutContext(old_ctx);
if (dsurf && !_eglIsSurfaceLinked(dsurf))
destroy_surface(disp, dsurf);
if (rsurf && rsurf != dsurf && !_eglIsSurfaceLinked(rsurf))
destroy_surface(disp, rsurf);
}
else {
/* undo the previous _eglBindContext */
_eglBindContext(old_ctx, old_dsurf, old_rsurf, &ctx, &dsurf, &rsurf);
assert(&GLX_ctx->Base == ctx &&
&GLX_dsurf->Base == dsurf &&
&GLX_rsurf->Base == rsurf);
_eglPutSurface(dsurf);
_eglPutSurface(rsurf);
_eglPutContext(ctx);
_eglPutSurface(old_dsurf);
_eglPutSurface(old_rsurf);
_eglPutContext(old_ctx);
_eglBindContext(&ctx, &dsurf, &rsurf);
}
return ret;
@@ -751,8 +666,6 @@ GLX_eglCreateWindowSurface(_EGLDriver *drv, _EGLDisplay *disp,
struct GLX_egl_surface *GLX_surf;
uint width, height;
(void) drv;
GLX_surf = CALLOC_STRUCT(GLX_egl_surface);
if (!GLX_surf) {
_eglError(EGL_BAD_ALLOC, "eglCreateWindowSurface");
@@ -799,8 +712,6 @@ GLX_eglCreatePixmapSurface(_EGLDriver *drv, _EGLDisplay *disp,
struct GLX_egl_surface *GLX_surf;
uint width, height;
(void) drv;
GLX_surf = CALLOC_STRUCT(GLX_egl_surface);
if (!GLX_surf) {
_eglError(EGL_BAD_ALLOC, "eglCreatePixmapSurface");
@@ -861,8 +772,6 @@ GLX_eglCreatePbufferSurface(_EGLDriver *drv, _EGLDisplay *disp,
int attribs[5];
int i;
(void) drv;
GLX_surf = CALLOC_STRUCT(GLX_egl_surface);
if (!GLX_surf) {
_eglError(EGL_BAD_ALLOC, "eglCreatePbufferSurface");
@@ -921,9 +830,7 @@ GLX_eglCreatePbufferSurface(_EGLDriver *drv, _EGLDisplay *disp,
static EGLBoolean
GLX_eglDestroySurface(_EGLDriver *drv, _EGLDisplay *disp, _EGLSurface *surf)
{
(void) drv;
if (_eglPutSurface(surf))
if (!_eglIsSurfaceBound(surf))
destroy_surface(disp, surf);
return EGL_TRUE;
@@ -936,8 +843,6 @@ GLX_eglSwapBuffers(_EGLDriver *drv, _EGLDisplay *disp, _EGLSurface *draw)
struct GLX_egl_display *GLX_dpy = GLX_egl_display(disp);
struct GLX_egl_surface *GLX_surf = GLX_egl_surface(draw);
(void) drv;
glXSwapBuffers(GLX_dpy->dpy, GLX_surf->glx_drawable);
return EGL_TRUE;
@@ -949,18 +854,12 @@ GLX_eglSwapBuffers(_EGLDriver *drv, _EGLDisplay *disp, _EGLSurface *draw)
static _EGLProc
GLX_eglGetProcAddress(_EGLDriver *drv, const char *procname)
{
(void) drv;
return (_EGLProc) glXGetProcAddress((const GLubyte *) procname);
}
static EGLBoolean
GLX_eglWaitClient(_EGLDriver *drv, _EGLDisplay *dpy, _EGLContext *ctx)
{
(void) drv;
(void) dpy;
(void) ctx;
glXWaitGL();
return EGL_TRUE;
}
@@ -968,9 +867,6 @@ GLX_eglWaitClient(_EGLDriver *drv, _EGLDisplay *dpy, _EGLContext *ctx)
static EGLBoolean
GLX_eglWaitNative(_EGLDriver *drv, _EGLDisplay *dpy, EGLint engine)
{
(void) drv;
(void) dpy;
if (engine != EGL_CORE_NATIVE_ENGINE)
return _eglError(EGL_BAD_PARAMETER, "eglWaitNative");
glXWaitX();
@@ -994,8 +890,6 @@ _eglMain(const char *args)
{
struct GLX_egl_driver *GLX_drv = CALLOC_STRUCT(GLX_egl_driver);
(void) args;
if (!GLX_drv)
return NULL;

View File

@@ -36,7 +36,6 @@ SOURCES = \
eglcurrent.c \
egldisplay.c \
egldriver.c \
eglfallbacks.c \
eglglobals.c \
eglimage.c \
egllog.c \

View File

@@ -4,49 +4,48 @@
Import('*')
env = env.Clone()
if env['platform'] != 'winddk':
env.Append(CPPDEFINES = [
'_EGL_NATIVE_PLATFORM=_EGL_PLATFORM_WINDOWS',
'_EGL_DRIVER_SEARCH_DIR=\\"\\"',
'_EGL_OS_WINDOWS',
'_EGL_GET_CORE_ADDRESSES',
'KHRONOS_DLL_EXPORTS',
])
env = env.Clone()
env.Append(CPPPATH = [
'#/include',
])
env.Append(CPPDEFINES = [
'_EGL_NATIVE_PLATFORM=_EGL_PLATFORM_WINDOWS',
'_EGL_DRIVER_SEARCH_DIR=\\"\\"',
'_EGL_OS_WINDOWS',
'_EGL_GET_CORE_ADDRESSES',
'KHRONOS_DLL_EXPORTS',
])
egl_sources = [
'eglapi.c',
'eglarray.c',
'eglconfig.c',
'eglcontext.c',
'eglcurrent.c',
'egldisplay.c',
'egldriver.c',
'eglfallbacks.c',
'eglglobals.c',
'eglimage.c',
'egllog.c',
'eglmisc.c',
'eglmode.c',
'eglscreen.c',
'eglstring.c',
'eglsurface.c',
'eglsync.c',
]
env.Append(CPPPATH = [
'#/include',
])
egl = env.SharedLibrary(
target = 'libEGL',
source = egl_sources + ['egl.def'],
)
egl_sources = [
'eglapi.c',
'eglarray.c',
'eglconfig.c',
'eglcontext.c',
'eglcurrent.c',
'egldisplay.c',
'egldriver.c',
'eglglobals.c',
'eglimage.c',
'egllog.c',
'eglmisc.c',
'eglmode.c',
'eglscreen.c',
'eglstring.c',
'eglsurface.c',
'eglsync.c',
]
installed_egl = env.InstallSharedLibrary(egl, version=(1, 4, 0))
egl = env.SharedLibrary(
target = 'libEGL',
source = egl_sources + ['egl.def'],
)
env.Alias('egl', installed_egl)
env.InstallSharedLibrary(egl, version=(1, 4, 0))
egl = [env.FindIxes(egl, 'LIBPREFIX', 'LIBSUFFIX')]
egl = [env.FindIxes(egl, 'LIBPREFIX', 'LIBSUFFIX')]
Export('egl')
Export('egl')

View File

@@ -416,7 +416,7 @@ eglCreateContext(EGLDisplay dpy, EGLConfig config, EGLContext share_list,
RETURN_EGL_ERROR(disp, EGL_BAD_CONTEXT, EGL_NO_CONTEXT);
context = drv->API.CreateContext(drv, disp, conf, share, attrib_list);
ret = (context) ? _eglLinkContext(context) : EGL_NO_CONTEXT;
ret = (context) ? _eglLinkContext(context, disp) : EGL_NO_CONTEXT;
RETURN_EGL_EVAL(disp, ret);
}
@@ -515,7 +515,7 @@ eglCreateWindowSurface(EGLDisplay dpy, EGLConfig config,
RETURN_EGL_ERROR(disp, EGL_BAD_NATIVE_WINDOW, EGL_NO_SURFACE);
surf = drv->API.CreateWindowSurface(drv, disp, conf, window, attrib_list);
ret = (surf) ? _eglLinkSurface(surf) : EGL_NO_SURFACE;
ret = (surf) ? _eglLinkSurface(surf, disp) : EGL_NO_SURFACE;
RETURN_EGL_EVAL(disp, ret);
}
@@ -536,7 +536,7 @@ eglCreatePixmapSurface(EGLDisplay dpy, EGLConfig config,
RETURN_EGL_ERROR(disp, EGL_BAD_NATIVE_PIXMAP, EGL_NO_SURFACE);
surf = drv->API.CreatePixmapSurface(drv, disp, conf, pixmap, attrib_list);
ret = (surf) ? _eglLinkSurface(surf) : EGL_NO_SURFACE;
ret = (surf) ? _eglLinkSurface(surf, disp) : EGL_NO_SURFACE;
RETURN_EGL_EVAL(disp, ret);
}
@@ -555,7 +555,7 @@ eglCreatePbufferSurface(EGLDisplay dpy, EGLConfig config,
_EGL_CHECK_CONFIG(disp, conf, EGL_NO_SURFACE, drv);
surf = drv->API.CreatePbufferSurface(drv, disp, conf, attrib_list);
ret = (surf) ? _eglLinkSurface(surf) : EGL_NO_SURFACE;
ret = (surf) ? _eglLinkSurface(surf, disp) : EGL_NO_SURFACE;
RETURN_EGL_EVAL(disp, ret);
}
@@ -648,12 +648,11 @@ eglSwapInterval(EGLDisplay dpy, EGLint interval)
_EGL_CHECK_DISPLAY(disp, EGL_FALSE, drv);
if (_eglGetContextHandle(ctx) == EGL_NO_CONTEXT ||
ctx->Resource.Display != disp)
if (!ctx || !_eglIsContextLinked(ctx) || ctx->Resource.Display != disp)
RETURN_EGL_ERROR(disp, EGL_BAD_CONTEXT, EGL_FALSE);
surf = ctx->DrawSurface;
if (_eglGetSurfaceHandle(surf) == EGL_NO_SURFACE)
if (!_eglIsSurfaceLinked(surf))
RETURN_EGL_ERROR(disp, EGL_BAD_SURFACE, EGL_FALSE);
ret = drv->API.SwapInterval(drv, disp, surf, interval);
@@ -674,8 +673,7 @@ eglSwapBuffers(EGLDisplay dpy, EGLSurface surface)
_EGL_CHECK_SURFACE(disp, surf, EGL_FALSE, drv);
/* surface must be bound to current context in EGL 1.4 */
if (_eglGetContextHandle(ctx) == EGL_NO_CONTEXT ||
surf != ctx->DrawSurface)
if (!ctx || !_eglIsContextLinked(ctx) || surf != ctx->DrawSurface)
RETURN_EGL_ERROR(disp, EGL_BAD_SURFACE, EGL_FALSE);
ret = drv->API.SwapBuffers(drv, disp, surf);
@@ -716,8 +714,7 @@ eglWaitClient(void)
_eglLockMutex(&disp->Mutex);
/* let bad current context imply bad current surface */
if (_eglGetContextHandle(ctx) == EGL_NO_CONTEXT ||
_eglGetSurfaceHandle(ctx->DrawSurface) == EGL_NO_SURFACE)
if (!_eglIsContextLinked(ctx) || !_eglIsSurfaceLinked(ctx->DrawSurface))
RETURN_EGL_ERROR(disp, EGL_BAD_CURRENT_SURFACE, EGL_FALSE);
/* a valid current context implies an initialized current display */
@@ -766,8 +763,7 @@ eglWaitNative(EGLint engine)
_eglLockMutex(&disp->Mutex);
/* let bad current context imply bad current surface */
if (_eglGetContextHandle(ctx) == EGL_NO_CONTEXT ||
_eglGetSurfaceHandle(ctx->DrawSurface) == EGL_NO_SURFACE)
if (!_eglIsContextLinked(ctx) || !_eglIsSurfaceLinked(ctx->DrawSurface))
RETURN_EGL_ERROR(disp, EGL_BAD_CURRENT_SURFACE, EGL_FALSE);
/* a valid current context implies an initialized current display */
@@ -1047,7 +1043,7 @@ eglCreateScreenSurfaceMESA(EGLDisplay dpy, EGLConfig config,
_EGL_CHECK_CONFIG(disp, conf, EGL_NO_SURFACE, drv);
surf = drv->API.CreateScreenSurfaceMESA(drv, disp, conf, attrib_list);
ret = (surf) ? _eglLinkSurface(surf) : EGL_NO_SURFACE;
ret = (surf) ? _eglLinkSurface(surf, disp) : EGL_NO_SURFACE;
RETURN_EGL_EVAL(disp, ret);
}
@@ -1239,7 +1235,7 @@ eglCreatePbufferFromClientBuffer(EGLDisplay dpy, EGLenum buftype,
surf = drv->API.CreatePbufferFromClientBuffer(drv, disp, buftype, buffer,
conf, attrib_list);
ret = (surf) ? _eglLinkSurface(surf) : EGL_NO_SURFACE;
ret = (surf) ? _eglLinkSurface(surf, disp) : EGL_NO_SURFACE;
RETURN_EGL_EVAL(disp, ret);
}
@@ -1302,7 +1298,7 @@ eglCreateImageKHR(EGLDisplay dpy, EGLContext ctx, EGLenum target,
img = drv->API.CreateImageKHR(drv,
disp, context, target, buffer, attr_list);
ret = (img) ? _eglLinkImage(img) : EGL_NO_IMAGE_KHR;
ret = (img) ? _eglLinkImage(img, disp) : EGL_NO_IMAGE_KHR;
RETURN_EGL_EVAL(disp, ret);
}
@@ -1348,7 +1344,7 @@ eglCreateSyncKHR(EGLDisplay dpy, EGLenum type, const EGLint *attrib_list)
RETURN_EGL_EVAL(disp, EGL_NO_SYNC_KHR);
sync = drv->API.CreateSyncKHR(drv, disp, type, attrib_list);
ret = (sync) ? _eglLinkSync(sync) : EGL_NO_SYNC_KHR;
ret = (sync) ? _eglLinkSync(sync, disp) : EGL_NO_SYNC_KHR;
RETURN_EGL_EVAL(disp, ret);
}
@@ -1441,8 +1437,7 @@ eglSwapBuffersRegionNOK(EGLDisplay dpy, EGLSurface surface,
RETURN_EGL_EVAL(disp, EGL_FALSE);
/* surface must be bound to current context in EGL 1.4 */
if (_eglGetContextHandle(ctx) == EGL_NO_CONTEXT ||
surf != ctx->DrawSurface)
if (!ctx || !_eglIsContextLinked(ctx) || surf != ctx->DrawSurface)
RETURN_EGL_ERROR(disp, EGL_BAD_SURFACE, EGL_FALSE);
ret = drv->API.SwapBuffersRegionNOK(drv, disp, surf, numRects, rects);
@@ -1468,7 +1463,7 @@ eglCreateDRMImageMESA(EGLDisplay dpy, const EGLint *attr_list)
RETURN_EGL_EVAL(disp, EGL_NO_IMAGE_KHR);
img = drv->API.CreateDRMImageMESA(drv, disp, attr_list);
ret = (img) ? _eglLinkImage(img) : EGL_NO_IMAGE_KHR;
ret = (img) ? _eglLinkImage(img, disp) : EGL_NO_IMAGE_KHR;
RETURN_EGL_EVAL(disp, ret);
}

View File

@@ -118,39 +118,38 @@ _eglFindArray(_EGLArray *array, void *elem)
/**
* Filter an array and return the number of filtered elements.
* Filter an array and return the filtered data. The returned data pointer
* should be freed.
*/
EGLint
_eglFilterArray(_EGLArray *array, void **data, EGLint size,
void **
_eglFilterArray(_EGLArray *array, EGLint *size,
_EGLArrayForEach filter, void *filter_data)
{
void **data;
EGLint count = 0, i;
if (!array)
return 0;
if (!array) {
*size = 0;
return malloc(0);
}
data = malloc(array->Size * sizeof(array->Elements[0]));
if (!data)
return NULL;
if (filter) {
for (i = 0; i < array->Size; i++) {
if (filter(array->Elements[i], filter_data)) {
if (data && count < size)
data[count] = array->Elements[i];
count++;
}
if (data && count >= size)
break;
if (filter(array->Elements[i], filter_data))
data[count++] = array->Elements[i];
}
}
else {
if (data) {
count = (size < array->Size) ? size : array->Size;
memcpy(data, array->Elements, count * sizeof(array->Elements[0]));
}
else {
count = array->Size;
}
memcpy(data, array->Elements, array->Size * sizeof(array->Elements[0]));
}
return count;
*size = count;
return data;
}

View File

@@ -37,8 +37,8 @@ void *
_eglFindArray(_EGLArray *array, void *elem);
PUBLIC EGLint
_eglFilterArray(_EGLArray *array, void **data, EGLint size,
void **
_eglFilterArray(_EGLArray *array, EGLint *size,
_EGLArrayForEach filter, void *filter_data);

View File

@@ -24,34 +24,34 @@
* IDs are from 1 to N respectively.
*/
void
_eglInitConfig(_EGLConfig *conf, _EGLDisplay *dpy, EGLint id)
_eglInitConfig(_EGLConfig *config, _EGLDisplay *dpy, EGLint id)
{
memset(conf, 0, sizeof(*conf));
memset(config, 0, sizeof(*config));
conf->Display = dpy;
config->Display = dpy;
/* some attributes take non-zero default values */
conf->ConfigID = id;
conf->ConfigCaveat = EGL_NONE;
conf->TransparentType = EGL_NONE;
conf->NativeVisualType = EGL_NONE;
conf->ColorBufferType = EGL_RGB_BUFFER;
SET_CONFIG_ATTRIB(config, EGL_CONFIG_ID, id);
SET_CONFIG_ATTRIB(config, EGL_CONFIG_CAVEAT, EGL_NONE);
SET_CONFIG_ATTRIB(config, EGL_TRANSPARENT_TYPE, EGL_NONE);
SET_CONFIG_ATTRIB(config, EGL_NATIVE_VISUAL_TYPE, EGL_NONE);
#ifdef EGL_VERSION_1_2
SET_CONFIG_ATTRIB(config, EGL_COLOR_BUFFER_TYPE, EGL_RGB_BUFFER);
#endif /* EGL_VERSION_1_2 */
}
/**
* Link a config to its display and return the handle of the link.
* Link a config to a display and return the handle of the link.
* The handle can be passed to client directly.
*
* Note that we just save the ptr to the config (we don't copy the config).
*/
PUBLIC EGLConfig
_eglLinkConfig(_EGLConfig *conf)
EGLConfig
_eglAddConfig(_EGLDisplay *dpy, _EGLConfig *conf)
{
_EGLDisplay *dpy = conf->Display;
/* sanity check */
assert(dpy && conf->ConfigID > 0);
assert(GET_CONFIG_ATTRIB(conf, EGL_CONFIG_ID) > 0);
if (!dpy->Configs) {
dpy->Configs = _eglCreateArray("Config", 16);
@@ -59,29 +59,23 @@ _eglLinkConfig(_EGLConfig *conf)
return (EGLConfig) NULL;
}
conf->Display = dpy;
_eglAppendArray(dpy->Configs, (void *) conf);
return (EGLConfig) conf;
}
/**
* Lookup a handle to find the linked config.
* Return NULL if the handle has no corresponding linked config.
*/
_EGLConfig *
_eglLookupConfig(EGLConfig config, _EGLDisplay *dpy)
EGLBoolean
_eglCheckConfigHandle(EGLConfig config, _EGLDisplay *dpy)
{
_EGLConfig *conf;
if (!dpy)
return NULL;
conf = (_EGLConfig *) _eglFindArray(dpy->Configs, (void *) config);
if (conf)
assert(conf->Display == dpy);
return conf;
return (conf != NULL);
}
@@ -110,7 +104,6 @@ static const struct {
EGLint default_value;
} _eglValidationTable[] =
{
/* core */
{ EGL_BUFFER_SIZE, ATTRIB_TYPE_INTEGER,
ATTRIB_CRITERION_ATLEAST,
0 },
@@ -207,13 +200,22 @@ static const struct {
{ EGL_TRANSPARENT_BLUE_VALUE, ATTRIB_TYPE_INTEGER,
ATTRIB_CRITERION_EXACT,
EGL_DONT_CARE },
/* these are not real attributes */
{ EGL_MATCH_NATIVE_PIXMAP, ATTRIB_TYPE_PSEUDO,
ATTRIB_CRITERION_SPECIAL,
EGL_NONE },
/* extensions */
/* there is a gap before EGL_SAMPLES */
{ 0x3030, ATTRIB_TYPE_PSEUDO,
ATTRIB_CRITERION_IGNORE,
0 },
{ EGL_NONE, ATTRIB_TYPE_PSEUDO,
ATTRIB_CRITERION_IGNORE,
0 },
{ EGL_Y_INVERTED_NOK, ATTRIB_TYPE_BOOLEAN,
ATTRIB_CRITERION_EXACT,
EGL_DONT_CARE }
EGL_DONT_CARE },
};
@@ -230,13 +232,18 @@ _eglValidateConfig(const _EGLConfig *conf, EGLBoolean for_matching)
{
EGLint i, attr, val;
EGLBoolean valid = EGL_TRUE;
EGLint red_size = 0, green_size = 0, blue_size = 0, luminance_size = 0;
EGLint alpha_size = 0, buffer_size = 0;
/* all attributes should have been listed */
assert(ARRAY_SIZE(_eglValidationTable) == _EGL_CONFIG_NUM_ATTRIBS);
/* check attributes by their types */
for (i = 0; i < ARRAY_SIZE(_eglValidationTable); i++) {
EGLint mask;
attr = _eglValidationTable[i].attr;
val = _eglGetConfigKey(conf, attr);
val = GET_CONFIG_ATTRIB(conf, attr);
switch (_eglValidationTable[i].type) {
case ATTRIB_TYPE_INTEGER:
@@ -248,14 +255,30 @@ _eglValidateConfig(const _EGLConfig *conf, EGLBoolean for_matching)
break;
case EGL_SAMPLE_BUFFERS:
/* there can be at most 1 sample buffer */
if (val > 1 || val < 0)
if (val > 1)
valid = EGL_FALSE;
break;
default:
if (val < 0)
valid = EGL_FALSE;
case EGL_RED_SIZE:
red_size = val;
break;
case EGL_GREEN_SIZE:
green_size = val;
break;
case EGL_BLUE_SIZE:
blue_size = val;
break;
case EGL_LUMINANCE_SIZE:
luminance_size = val;
break;
case EGL_ALPHA_SIZE:
alpha_size = val;
break;
case EGL_BUFFER_SIZE:
buffer_size = val;
break;
}
if (val < 0)
valid = EGL_FALSE;
break;
case ATTRIB_TYPE_BOOLEAN:
if (val != EGL_TRUE && val != EGL_FALSE)
@@ -343,18 +366,17 @@ _eglValidateConfig(const _EGLConfig *conf, EGLBoolean for_matching)
/* now check for conflicting attribute values */
switch (conf->ColorBufferType) {
switch (GET_CONFIG_ATTRIB(conf, EGL_COLOR_BUFFER_TYPE)) {
case EGL_RGB_BUFFER:
if (conf->LuminanceSize)
if (luminance_size)
valid = EGL_FALSE;
if (conf->RedSize + conf->GreenSize +
conf->BlueSize + conf->AlphaSize != conf->BufferSize)
if (red_size + green_size + blue_size + alpha_size != buffer_size)
valid = EGL_FALSE;
break;
case EGL_LUMINANCE_BUFFER:
if (conf->RedSize || conf->GreenSize || conf->BlueSize)
if (red_size || green_size || blue_size)
valid = EGL_FALSE;
if (conf->LuminanceSize + conf->AlphaSize != conf->BufferSize)
if (luminance_size + alpha_size != buffer_size)
valid = EGL_FALSE;
break;
}
@@ -363,19 +385,23 @@ _eglValidateConfig(const _EGLConfig *conf, EGLBoolean for_matching)
return EGL_FALSE;
}
if (!conf->SampleBuffers && conf->Samples)
val = GET_CONFIG_ATTRIB(conf, EGL_SAMPLE_BUFFERS);
if (!val && GET_CONFIG_ATTRIB(conf, EGL_SAMPLES))
valid = EGL_FALSE;
if (!valid) {
_eglLog(_EGL_DEBUG, "conflicting samples and sample buffers");
return EGL_FALSE;
}
if (!(conf->SurfaceType & EGL_WINDOW_BIT)) {
if (conf->NativeVisualID != 0 || conf->NativeVisualType != EGL_NONE)
val = GET_CONFIG_ATTRIB(conf, EGL_SURFACE_TYPE);
if (!(val & EGL_WINDOW_BIT)) {
if (GET_CONFIG_ATTRIB(conf, EGL_NATIVE_VISUAL_ID) != 0 ||
GET_CONFIG_ATTRIB(conf, EGL_NATIVE_VISUAL_TYPE) != EGL_NONE)
valid = EGL_FALSE;
}
if (!(conf->SurfaceType & EGL_PBUFFER_BIT)) {
if (conf->BindToTextureRGB || conf->BindToTextureRGBA)
if (!(val & EGL_PBUFFER_BIT)) {
if (GET_CONFIG_ATTRIB(conf, EGL_BIND_TO_TEXTURE_RGB) ||
GET_CONFIG_ATTRIB(conf, EGL_BIND_TO_TEXTURE_RGBA))
valid = EGL_FALSE;
}
if (!valid) {
@@ -407,11 +433,11 @@ _eglMatchConfig(const _EGLConfig *conf, const _EGLConfig *criteria)
continue;
attr = _eglValidationTable[i].attr;
cmp = _eglGetConfigKey(criteria, attr);
cmp = GET_CONFIG_ATTRIB(criteria, attr);
if (cmp == EGL_DONT_CARE)
continue;
val = _eglGetConfigKey(conf, attr);
val = GET_CONFIG_ATTRIB(conf, attr);
switch (_eglValidationTable[i].criterion) {
case ATTRIB_CRITERION_EXACT:
if (val != cmp)
@@ -452,11 +478,16 @@ _eglMatchConfig(const _EGLConfig *conf, const _EGLConfig *criteria)
static INLINE EGLBoolean
_eglIsConfigAttribValid(_EGLConfig *conf, EGLint attr)
{
if (_eglOffsetOfConfig(attr) < 0)
if (_eglIndexConfig(conf, attr) < 0)
return EGL_FALSE;
/* there are some holes in the range */
switch (attr) {
case 0x3030 /* a gap before EGL_SAMPLES */:
case EGL_NONE:
#ifdef EGL_VERSION_1_4
case EGL_MATCH_NATIVE_PIXMAP:
#endif
return EGL_FALSE;
case EGL_Y_INVERTED_NOK:
return conf->Display->Extensions.NOK_texture_from_pixmap;
@@ -472,18 +503,18 @@ _eglIsConfigAttribValid(_EGLConfig *conf, EGLint attr)
* Return EGL_FALSE if any of the attribute is invalid.
*/
EGLBoolean
_eglParseConfigAttribList(_EGLConfig *conf, _EGLDisplay *dpy,
const EGLint *attrib_list)
_eglParseConfigAttribList(_EGLConfig *conf, const EGLint *attrib_list)
{
EGLint attr, val, i;
_eglInitConfig(conf, dpy, EGL_DONT_CARE);
EGLint config_id = 0, level = 0;
EGLBoolean has_native_visual_type = EGL_FALSE;
EGLBoolean has_transparent_color = EGL_FALSE;
/* reset to default values */
for (i = 0; i < ARRAY_SIZE(_eglValidationTable); i++) {
attr = _eglValidationTable[i].attr;
val = _eglValidationTable[i].default_value;
_eglSetConfigKey(conf, attr, val);
SET_CONFIG_ATTRIB(conf, attr, val);
}
/* parse the list */
@@ -493,33 +524,59 @@ _eglParseConfigAttribList(_EGLConfig *conf, _EGLDisplay *dpy,
if (!_eglIsConfigAttribValid(conf, attr))
return EGL_FALSE;
SET_CONFIG_ATTRIB(conf, attr, val);
_eglSetConfigKey(conf, attr, val);
/* rememeber some attributes for post-processing */
switch (attr) {
case EGL_CONFIG_ID:
config_id = val;
break;
case EGL_LEVEL:
level = val;
break;
case EGL_NATIVE_VISUAL_TYPE:
has_native_visual_type = EGL_TRUE;
break;
case EGL_TRANSPARENT_RED_VALUE:
case EGL_TRANSPARENT_GREEN_VALUE:
case EGL_TRANSPARENT_BLUE_VALUE:
has_transparent_color = EGL_TRUE;
break;
default:
break;
}
}
if (!_eglValidateConfig(conf, EGL_TRUE))
return EGL_FALSE;
/* the spec says that EGL_LEVEL cannot be EGL_DONT_CARE */
if (conf->Level == EGL_DONT_CARE)
if (level == EGL_DONT_CARE)
return EGL_FALSE;
/* ignore other attributes when EGL_CONFIG_ID is given */
if (conf->ConfigID != EGL_DONT_CARE) {
for (i = 0; i < ARRAY_SIZE(_eglValidationTable); i++) {
attr = _eglValidationTable[i].attr;
if (attr != EGL_CONFIG_ID)
_eglSetConfigKey(conf, attr, EGL_DONT_CARE);
}
if (config_id > 0) {
_eglResetConfigKeys(conf, EGL_DONT_CARE);
SET_CONFIG_ATTRIB(conf, EGL_CONFIG_ID, config_id);
}
else {
if (!(conf->SurfaceType & EGL_WINDOW_BIT))
conf->NativeVisualType = EGL_DONT_CARE;
if (has_native_visual_type) {
val = GET_CONFIG_ATTRIB(conf, EGL_SURFACE_TYPE);
if (!(val & EGL_WINDOW_BIT))
SET_CONFIG_ATTRIB(conf, EGL_NATIVE_VISUAL_TYPE, EGL_DONT_CARE);
}
if (conf->TransparentType == EGL_NONE) {
conf->TransparentRedValue = EGL_DONT_CARE;
conf->TransparentGreenValue = EGL_DONT_CARE;
conf->TransparentBlueValue = EGL_DONT_CARE;
if (has_transparent_color) {
val = GET_CONFIG_ATTRIB(conf, EGL_TRANSPARENT_TYPE);
if (val == EGL_NONE) {
SET_CONFIG_ATTRIB(conf, EGL_TRANSPARENT_RED_VALUE,
EGL_DONT_CARE);
SET_CONFIG_ATTRIB(conf, EGL_TRANSPARENT_GREEN_VALUE,
EGL_DONT_CARE);
SET_CONFIG_ATTRIB(conf, EGL_TRANSPARENT_BLUE_VALUE,
EGL_DONT_CARE);
}
}
}
@@ -553,6 +610,7 @@ _eglCompareConfigs(const _EGLConfig *conf1, const _EGLConfig *conf2,
EGL_ALPHA_MASK_SIZE,
};
EGLint val1, val2;
EGLBoolean rgb_buffer;
EGLint i;
if (conf1 == conf2)
@@ -561,41 +619,44 @@ _eglCompareConfigs(const _EGLConfig *conf1, const _EGLConfig *conf2,
/* the enum values have the desired ordering */
assert(EGL_NONE < EGL_SLOW_CONFIG);
assert(EGL_SLOW_CONFIG < EGL_NON_CONFORMANT_CONFIG);
val1 = conf1->ConfigCaveat - conf2->ConfigCaveat;
if (val1)
return val1;
val1 = GET_CONFIG_ATTRIB(conf1, EGL_CONFIG_CAVEAT);
val2 = GET_CONFIG_ATTRIB(conf2, EGL_CONFIG_CAVEAT);
if (val1 != val2)
return (val1 - val2);
/* the enum values have the desired ordering */
assert(EGL_RGB_BUFFER < EGL_LUMINANCE_BUFFER);
val1 = conf1->ColorBufferType - conf2->ColorBufferType;
if (val1)
return val1;
val1 = GET_CONFIG_ATTRIB(conf1, EGL_COLOR_BUFFER_TYPE);
val2 = GET_CONFIG_ATTRIB(conf2, EGL_COLOR_BUFFER_TYPE);
if (val1 != val2)
return (val1 - val2);
rgb_buffer = (val1 == EGL_RGB_BUFFER);
if (criteria) {
val1 = val2 = 0;
if (conf1->ColorBufferType == EGL_RGB_BUFFER) {
if (criteria->RedSize > 0) {
val1 += conf1->RedSize;
val2 += conf2->RedSize;
if (rgb_buffer) {
if (GET_CONFIG_ATTRIB(criteria, EGL_RED_SIZE) > 0) {
val1 += GET_CONFIG_ATTRIB(conf1, EGL_RED_SIZE);
val2 += GET_CONFIG_ATTRIB(conf2, EGL_RED_SIZE);
}
if (criteria->GreenSize > 0) {
val1 += conf1->GreenSize;
val2 += conf2->GreenSize;
if (GET_CONFIG_ATTRIB(criteria, EGL_GREEN_SIZE) > 0) {
val1 += GET_CONFIG_ATTRIB(conf1, EGL_GREEN_SIZE);
val2 += GET_CONFIG_ATTRIB(conf2, EGL_GREEN_SIZE);
}
if (criteria->BlueSize > 0) {
val1 += conf1->BlueSize;
val2 += conf2->BlueSize;
if (GET_CONFIG_ATTRIB(criteria, EGL_BLUE_SIZE) > 0) {
val1 += GET_CONFIG_ATTRIB(conf1, EGL_BLUE_SIZE);
val2 += GET_CONFIG_ATTRIB(conf2, EGL_BLUE_SIZE);
}
}
else {
if (criteria->LuminanceSize > 0) {
val1 += conf1->LuminanceSize;
val2 += conf2->LuminanceSize;
if (GET_CONFIG_ATTRIB(criteria, EGL_LUMINANCE_SIZE) > 0) {
val1 += GET_CONFIG_ATTRIB(conf1, EGL_LUMINANCE_SIZE);
val2 += GET_CONFIG_ATTRIB(conf2, EGL_LUMINANCE_SIZE);
}
}
if (criteria->AlphaSize > 0) {
val1 += conf1->AlphaSize;
val2 += conf2->AlphaSize;
if (GET_CONFIG_ATTRIB(criteria, EGL_ALPHA_SIZE) > 0) {
val1 += GET_CONFIG_ATTRIB(conf1, EGL_ALPHA_SIZE);
val2 += GET_CONFIG_ATTRIB(conf2, EGL_ALPHA_SIZE);
}
}
else {
@@ -608,15 +669,24 @@ _eglCompareConfigs(const _EGLConfig *conf1, const _EGLConfig *conf2,
return (val2 - val1);
for (i = 0; i < ARRAY_SIZE(compare_attribs); i++) {
val1 = _eglGetConfigKey(conf1, compare_attribs[i]);
val2 = _eglGetConfigKey(conf2, compare_attribs[i]);
val1 = GET_CONFIG_ATTRIB(conf1, compare_attribs[i]);
val2 = GET_CONFIG_ATTRIB(conf2, compare_attribs[i]);
if (val1 != val2)
return (val1 - val2);
}
/* EGL_NATIVE_VISUAL_TYPE cannot be compared here */
return (compare_id) ? (conf1->ConfigID - conf2->ConfigID) : 0;
if (compare_id) {
val1 = GET_CONFIG_ATTRIB(conf1, EGL_CONFIG_ID);
val2 = GET_CONFIG_ATTRIB(conf2, EGL_CONFIG_ID);
assert(val1 != val2);
}
else {
val1 = val2 = 0;
}
return (val1 - val2);
}
@@ -694,25 +764,15 @@ _eglChooseConfig(_EGLDriver *drv, _EGLDisplay *disp, const EGLint *attrib_list,
if (!num_configs)
return _eglError(EGL_BAD_PARAMETER, "eglChooseConfigs");
if (!_eglParseConfigAttribList(&criteria, disp, attrib_list))
_eglInitConfig(&criteria, disp, 0);
if (!_eglParseConfigAttribList(&criteria, attrib_list))
return _eglError(EGL_BAD_ATTRIBUTE, "eglChooseConfig");
/* get the number of matched configs */
count = _eglFilterArray(disp->Configs, NULL, 0,
configList = (_EGLConfig **) _eglFilterArray(disp->Configs, &count,
(_EGLArrayForEach) _eglMatchConfig, (void *) &criteria);
if (!count) {
*num_configs = count;
return EGL_TRUE;
}
configList = malloc(sizeof(*configList) * count);
if (!configList)
return _eglError(EGL_BAD_ALLOC, "eglChooseConfig(out of memory)");
/* get the matched configs */
_eglFilterArray(disp->Configs, (void **) configList, count,
(_EGLArrayForEach) _eglMatchConfig, (void *) &criteria);
/* perform sorting of configs */
if (configs && count) {
_eglSortConfigs((const _EGLConfig **) configList, count,
@@ -742,7 +802,7 @@ _eglGetConfigAttrib(_EGLDriver *drv, _EGLDisplay *dpy, _EGLConfig *conf,
if (!value)
return _eglError(EGL_BAD_PARAMETER, "eglGetConfigAttrib");
*value = _eglGetConfigKey(conf, attribute);
*value = GET_CONFIG_ATTRIB(conf, attribute);
return EGL_TRUE;
}

View File

@@ -3,107 +3,72 @@
#include <assert.h>
#include <stddef.h>
#include "egltypedefs.h"
/* update _eglValidationTable and _eglOffsetOfConfig before updating this
* struct */
#define _EGL_CONFIG_FIRST_ATTRIB EGL_BUFFER_SIZE
#define _EGL_CONFIG_LAST_ATTRIB EGL_CONFORMANT
#define _EGL_CONFIG_NUM_CONTIGUOUS_ATTRIBS \
(_EGL_CONFIG_LAST_ATTRIB - _EGL_CONFIG_FIRST_ATTRIB + 1)
/* Attributes outside the contiguous block:
*
* EGL_Y_INVERTED_NOK
*/
#define _EGL_CONFIG_FIRST_EXTRA_ATTRIB _EGL_CONFIG_NUM_CONTIGUOUS_ATTRIBS
#define _EGL_CONFIG_NUM_EXTRA_ATTRIBS 1
#define _EGL_CONFIG_NUM_ATTRIBS \
_EGL_CONFIG_NUM_CONTIGUOUS_ATTRIBS + _EGL_CONFIG_NUM_EXTRA_ATTRIBS
struct _egl_config
{
_EGLDisplay *Display;
/* core */
EGLint BufferSize;
EGLint AlphaSize;
EGLint BlueSize;
EGLint GreenSize;
EGLint RedSize;
EGLint DepthSize;
EGLint StencilSize;
EGLint ConfigCaveat;
EGLint ConfigID;
EGLint Level;
EGLint MaxPbufferHeight;
EGLint MaxPbufferPixels;
EGLint MaxPbufferWidth;
EGLint NativeRenderable;
EGLint NativeVisualID;
EGLint NativeVisualType;
EGLint Samples;
EGLint SampleBuffers;
EGLint SurfaceType;
EGLint TransparentType;
EGLint TransparentBlueValue;
EGLint TransparentGreenValue;
EGLint TransparentRedValue;
EGLint BindToTextureRGB;
EGLint BindToTextureRGBA;
EGLint MinSwapInterval;
EGLint MaxSwapInterval;
EGLint LuminanceSize;
EGLint AlphaMaskSize;
EGLint ColorBufferType;
EGLint RenderableType;
EGLint MatchNativePixmap;
EGLint Conformant;
/* extensions */
EGLint YInvertedNOK;
EGLint Storage[_EGL_CONFIG_NUM_ATTRIBS];
};
/**
* Map an EGL attribute enum to the offset of the member in _EGLConfig.
* Macros for source level compatibility.
*/
#define SET_CONFIG_ATTRIB(CONF, ATTR, VAL) _eglSetConfigKey(CONF, ATTR, VAL)
#define GET_CONFIG_ATTRIB(CONF, ATTR) _eglGetConfigKey(CONF, ATTR)
/**
* Given a key, return an index into the storage of the config.
* Return -1 if the key is invalid.
*/
static INLINE EGLint
_eglOffsetOfConfig(EGLint attr)
_eglIndexConfig(const _EGLConfig *conf, EGLint key)
{
switch (attr) {
#define ATTRIB_MAP(attr, memb) case attr: return offsetof(_EGLConfig, memb)
/* core */
ATTRIB_MAP(EGL_BUFFER_SIZE, BufferSize);
ATTRIB_MAP(EGL_ALPHA_SIZE, AlphaSize);
ATTRIB_MAP(EGL_BLUE_SIZE, BlueSize);
ATTRIB_MAP(EGL_GREEN_SIZE, GreenSize);
ATTRIB_MAP(EGL_RED_SIZE, RedSize);
ATTRIB_MAP(EGL_DEPTH_SIZE, DepthSize);
ATTRIB_MAP(EGL_STENCIL_SIZE, StencilSize);
ATTRIB_MAP(EGL_CONFIG_CAVEAT, ConfigCaveat);
ATTRIB_MAP(EGL_CONFIG_ID, ConfigID);
ATTRIB_MAP(EGL_LEVEL, Level);
ATTRIB_MAP(EGL_MAX_PBUFFER_HEIGHT, MaxPbufferHeight);
ATTRIB_MAP(EGL_MAX_PBUFFER_PIXELS, MaxPbufferPixels);
ATTRIB_MAP(EGL_MAX_PBUFFER_WIDTH, MaxPbufferWidth);
ATTRIB_MAP(EGL_NATIVE_RENDERABLE, NativeRenderable);
ATTRIB_MAP(EGL_NATIVE_VISUAL_ID, NativeVisualID);
ATTRIB_MAP(EGL_NATIVE_VISUAL_TYPE, NativeVisualType);
ATTRIB_MAP(EGL_SAMPLES, Samples);
ATTRIB_MAP(EGL_SAMPLE_BUFFERS, SampleBuffers);
ATTRIB_MAP(EGL_SURFACE_TYPE, SurfaceType);
ATTRIB_MAP(EGL_TRANSPARENT_TYPE, TransparentType);
ATTRIB_MAP(EGL_TRANSPARENT_BLUE_VALUE, TransparentBlueValue);
ATTRIB_MAP(EGL_TRANSPARENT_GREEN_VALUE, TransparentGreenValue);
ATTRIB_MAP(EGL_TRANSPARENT_RED_VALUE, TransparentRedValue);
ATTRIB_MAP(EGL_BIND_TO_TEXTURE_RGB, BindToTextureRGB);
ATTRIB_MAP(EGL_BIND_TO_TEXTURE_RGBA, BindToTextureRGBA);
ATTRIB_MAP(EGL_MIN_SWAP_INTERVAL, MinSwapInterval);
ATTRIB_MAP(EGL_MAX_SWAP_INTERVAL, MaxSwapInterval);
ATTRIB_MAP(EGL_LUMINANCE_SIZE, LuminanceSize);
ATTRIB_MAP(EGL_ALPHA_MASK_SIZE, AlphaMaskSize);
ATTRIB_MAP(EGL_COLOR_BUFFER_TYPE, ColorBufferType);
ATTRIB_MAP(EGL_RENDERABLE_TYPE, RenderableType);
ATTRIB_MAP(EGL_MATCH_NATIVE_PIXMAP, MatchNativePixmap);
ATTRIB_MAP(EGL_CONFORMANT, Conformant);
/* extensions */
ATTRIB_MAP(EGL_Y_INVERTED_NOK, YInvertedNOK);
#undef ATTRIB_MAP
(void) conf;
if (key >= _EGL_CONFIG_FIRST_ATTRIB &&
key < _EGL_CONFIG_FIRST_ATTRIB + _EGL_CONFIG_NUM_CONTIGUOUS_ATTRIBS)
return key - _EGL_CONFIG_FIRST_ATTRIB;
switch (key) {
case EGL_Y_INVERTED_NOK:
return _EGL_CONFIG_FIRST_EXTRA_ATTRIB;
default:
return -1;
}
}
/**
* Reset all keys in the config to a given value.
*/
static INLINE void
_eglResetConfigKeys(_EGLConfig *conf, EGLint val)
{
EGLint i;
for (i = 0; i < _EGL_CONFIG_NUM_ATTRIBS; i++)
conf->Storage[i] = val;
}
/**
* Update a config for a given key.
*
@@ -114,9 +79,9 @@ _eglOffsetOfConfig(EGLint attr)
static INLINE void
_eglSetConfigKey(_EGLConfig *conf, EGLint key, EGLint val)
{
EGLint offset = _eglOffsetOfConfig(key);
assert(offset >= 0);
*((EGLint *) ((char *) conf + offset)) = val;
EGLint idx = _eglIndexConfig(conf, key);
assert(idx >= 0);
conf->Storage[idx] = val;
}
@@ -126,9 +91,9 @@ _eglSetConfigKey(_EGLConfig *conf, EGLint key, EGLint val)
static INLINE EGLint
_eglGetConfigKey(const _EGLConfig *conf, EGLint key)
{
EGLint offset = _eglOffsetOfConfig(key);
assert(offset >= 0);
return *((EGLint *) ((char *) conf + offset));
EGLint idx = _eglIndexConfig(conf, key);
assert(idx >= 0);
return conf->Storage[idx];
}
@@ -137,20 +102,34 @@ _eglInitConfig(_EGLConfig *config, _EGLDisplay *dpy, EGLint id);
PUBLIC EGLConfig
_eglLinkConfig(_EGLConfig *conf);
_eglAddConfig(_EGLDisplay *dpy, _EGLConfig *conf);
extern _EGLConfig *
_eglLookupConfig(EGLConfig config, _EGLDisplay *dpy);
extern EGLBoolean
_eglCheckConfigHandle(EGLConfig config, _EGLDisplay *dpy);
/**
* Return the handle of a linked config.
* Lookup a handle to find the linked config.
* Return NULL if the handle has no corresponding linked config.
*/
static INLINE _EGLConfig *
_eglLookupConfig(EGLConfig config, _EGLDisplay *dpy)
{
_EGLConfig *conf = (_EGLConfig *) config;
if (!dpy || !_eglCheckConfigHandle(config, dpy))
conf = NULL;
return conf;
}
/**
* Return the handle of a linked config, or NULL.
*/
static INLINE EGLConfig
_eglGetConfigHandle(_EGLConfig *conf)
{
return (EGLConfig) conf;
return (EGLConfig) ((conf && conf->Display) ? conf : NULL);
}
@@ -163,8 +142,7 @@ _eglMatchConfig(const _EGLConfig *conf, const _EGLConfig *criteria);
PUBLIC EGLBoolean
_eglParseConfigAttribList(_EGLConfig *conf, _EGLDisplay *dpy,
const EGLint *attrib_list);
_eglParseConfigAttribList(_EGLConfig *conf, const EGLint *attrib_list);
PUBLIC EGLint

View File

@@ -103,7 +103,8 @@ _eglInitContext(_EGLContext *ctx, _EGLDisplay *dpy, _EGLConfig *conf,
return EGL_FALSE;
}
_eglInitResource(&ctx->Resource, sizeof(*ctx), dpy);
memset(ctx, 0, sizeof(_EGLContext));
ctx->Resource.Display = dpy;
ctx->ClientAPI = api;
ctx->Config = conf;
ctx->WindowRenderBuffer = EGL_NONE;
@@ -112,12 +113,13 @@ _eglInitContext(_EGLContext *ctx, _EGLDisplay *dpy, _EGLConfig *conf,
err = _eglParseContextAttribList(ctx, attrib_list);
if (err == EGL_SUCCESS && ctx->Config) {
EGLint api_bit;
EGLint renderable_type, api_bit;
renderable_type = GET_CONFIG_ATTRIB(ctx->Config, EGL_RENDERABLE_TYPE);
api_bit = _eglGetContextAPIBit(ctx);
if (!(ctx->Config->RenderableType & api_bit)) {
if (!(renderable_type & api_bit)) {
_eglLog(_EGL_DEBUG, "context api is 0x%x while config supports 0x%x",
api_bit, ctx->Config->RenderableType);
api_bit, renderable_type);
err = EGL_BAD_CONFIG;
}
}
@@ -128,6 +130,29 @@ _eglInitContext(_EGLContext *ctx, _EGLDisplay *dpy, _EGLConfig *conf,
}
/**
* Just a placeholder/demo function. Real driver will never use this!
*/
_EGLContext *
_eglCreateContext(_EGLDriver *drv, _EGLDisplay *dpy, _EGLConfig *conf,
_EGLContext *share_list, const EGLint *attrib_list)
{
return NULL;
}
/**
* Default fallback routine - drivers should usually override this.
*/
EGLBoolean
_eglDestroyContext(_EGLDriver *drv, _EGLDisplay *dpy, _EGLContext *ctx)
{
if (!_eglIsContextBound(ctx))
free(ctx);
return EGL_TRUE;
}
#ifdef EGL_VERSION_1_2
static EGLint
_eglQueryContextRenderBuffer(_EGLContext *ctx)
@@ -158,9 +183,7 @@ _eglQueryContext(_EGLDriver *drv, _EGLDisplay *dpy, _EGLContext *c,
switch (attribute) {
case EGL_CONFIG_ID:
if (!c->Config)
return _eglError(EGL_BAD_ATTRIBUTE, "eglQueryContext");
*value = c->Config->ConfigID;
*value = GET_CONFIG_ATTRIB(c->Config, EGL_CONFIG_ID);
break;
case EGL_CONTEXT_CLIENT_VERSION:
*value = c->ClientVersion;
@@ -303,65 +326,79 @@ _eglCheckMakeCurrent(_EGLContext *ctx, _EGLSurface *draw, _EGLSurface *read)
/**
* Bind the context to the current thread and given surfaces. Return the
* previous bound context and surfaces. The caller should unreference the
* returned context and surfaces.
*
* Making a second call with the resources returned by the first call
* unsurprisingly undoes the first call, except for the resouce reference
* counts.
* "orphaned" context and surfaces. Each argument is both input and output.
*/
EGLBoolean
_eglBindContext(_EGLContext *ctx, _EGLSurface *draw, _EGLSurface *read,
_EGLContext **old_ctx,
_EGLSurface **old_draw, _EGLSurface **old_read)
_eglBindContext(_EGLContext **ctx, _EGLSurface **draw, _EGLSurface **read)
{
_EGLThreadInfo *t = _eglGetCurrentThread();
_EGLContext *prev_ctx;
_EGLSurface *prev_draw, *prev_read;
_EGLContext *newCtx = *ctx, *oldCtx;
_EGLSurface *newDraw = *draw, *newRead = *read;
if (!_eglCheckMakeCurrent(ctx, draw, read))
if (!_eglCheckMakeCurrent(newCtx, newDraw, newRead))
return EGL_FALSE;
/* increment refcounts before binding */
_eglGetContext(ctx);
_eglGetSurface(draw);
_eglGetSurface(read);
/* bind the new context */
prev_ctx = _eglBindContextToThread(ctx, t);
oldCtx = _eglBindContextToThread(newCtx, t);
/* break previous bindings */
if (prev_ctx) {
prev_draw = prev_ctx->DrawSurface;
prev_read = prev_ctx->ReadSurface;
/* break old bindings */
if (oldCtx) {
*ctx = oldCtx;
*draw = oldCtx->DrawSurface;
*read = oldCtx->ReadSurface;
if (prev_draw)
prev_draw->CurrentContext = NULL;
if (prev_read)
prev_read->CurrentContext = NULL;
if (*draw)
(*draw)->CurrentContext = NULL;
if (*read)
(*read)->CurrentContext = NULL;
prev_ctx->DrawSurface = NULL;
prev_ctx->ReadSurface = NULL;
}
else {
prev_draw = prev_read = NULL;
oldCtx->DrawSurface = NULL;
oldCtx->ReadSurface = NULL;
}
/* establish new bindings */
if (ctx) {
if (draw)
draw->CurrentContext = ctx;
if (read)
read->CurrentContext = ctx;
if (newCtx) {
if (newDraw)
newDraw->CurrentContext = newCtx;
if (newRead)
newRead->CurrentContext = newCtx;
ctx->DrawSurface = draw;
ctx->ReadSurface = read;
newCtx->DrawSurface = newDraw;
newCtx->ReadSurface = newRead;
}
assert(old_ctx && old_draw && old_read);
*old_ctx = prev_ctx;
*old_draw = prev_draw;
*old_read = prev_read;
/* an old context or surface is not orphaned if it is still bound */
if (*ctx == newCtx)
*ctx = NULL;
if (*draw == newDraw || *draw == newRead)
*draw = NULL;
if (*read == newDraw || *read == newRead)
*read = NULL;
return EGL_TRUE;
}
/**
* Just a placeholder/demo function. Drivers should override this.
*/
EGLBoolean
_eglMakeCurrent(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *draw,
_EGLSurface *read, _EGLContext *ctx)
{
return EGL_FALSE;
}
/**
* This is defined by the EGL_MESA_copy_context extension.
*/
EGLBoolean
_eglCopyContextMESA(_EGLDriver *drv, EGLDisplay dpy, EGLContext source,
EGLContext dest, EGLint mask)
{
/* This function will always have to be overridden/implemented in the
* device driver. If the driver is based on Mesa, use _mesa_copy_context().
*/
return EGL_FALSE;
}

View File

@@ -34,46 +34,51 @@ _eglInitContext(_EGLContext *ctx, _EGLDisplay *dpy,
_EGLConfig *config, const EGLint *attrib_list);
extern _EGLContext *
_eglCreateContext(_EGLDriver *drv, _EGLDisplay *dpy, _EGLConfig *conf, _EGLContext *share_list, const EGLint *attrib_list);
extern EGLBoolean
_eglDestroyContext(_EGLDriver *drv, _EGLDisplay *dpy, _EGLContext *ctx);
extern EGLBoolean
_eglQueryContext(_EGLDriver *drv, _EGLDisplay *dpy, _EGLContext *ctx, EGLint attribute, EGLint *value);
PUBLIC EGLBoolean
_eglBindContext(_EGLContext *ctx, _EGLSurface *draw, _EGLSurface *read,
_EGLContext **old_ctx,
_EGLSurface **old_draw, _EGLSurface **old_read);
_eglBindContext(_EGLContext **ctx, _EGLSurface **draw, _EGLSurface **read);
extern EGLBoolean
_eglMakeCurrent(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *draw, _EGLSurface *read, _EGLContext *ctx);
extern EGLBoolean
_eglCopyContextMESA(_EGLDriver *drv, EGLDisplay dpy, EGLContext source, EGLContext dest, EGLint mask);
/**
* Increment reference count for the context.
*/
static INLINE _EGLContext *
_eglGetContext(_EGLContext *ctx)
{
if (ctx)
_eglGetResource(&ctx->Resource);
return ctx;
}
/**
* Decrement reference count for the context.
* Return true if the context is bound to a thread.
*
* The binding is considered a reference to the context. Drivers should not
* destroy a context when it is bound.
*/
static INLINE EGLBoolean
_eglPutContext(_EGLContext *ctx)
_eglIsContextBound(_EGLContext *ctx)
{
return (ctx) ? _eglPutResource(&ctx->Resource) : EGL_FALSE;
return (ctx->Binding != NULL);
}
/**
* Link a context to its display and return the handle of the link.
* Link a context to a display and return the handle of the link.
* The handle can be passed to client directly.
*/
static INLINE EGLContext
_eglLinkContext(_EGLContext *ctx)
_eglLinkContext(_EGLContext *ctx, _EGLDisplay *dpy)
{
_eglLinkResource(&ctx->Resource, _EGL_RESOURCE_CONTEXT);
_eglLinkResource(&ctx->Resource, _EGL_RESOURCE_CONTEXT, dpy);
return (EGLContext) ctx;
}
@@ -115,4 +120,18 @@ _eglGetContextHandle(_EGLContext *ctx)
}
/**
* Return true if the context is linked to a display.
*
* The link is considered a reference to the context (the display is owning the
* context). Drivers should not destroy a context when it is linked.
*/
static INLINE EGLBoolean
_eglIsContextLinked(_EGLContext *ctx)
{
_EGLResource *res = (_EGLResource *) ctx;
return (res && _eglIsResourceLinked(res));
}
#endif /* EGLCONTEXT_INCLUDED */

View File

@@ -14,7 +14,33 @@
static _EGLThreadInfo dummy_thread = _EGL_THREAD_INFO_INITIALIZER;
#if PTHREADS
#ifdef GLX_USE_TLS
static __thread const _EGLThreadInfo *_egl_TSD
__attribute__ ((tls_model("initial-exec")));
static INLINE void _eglSetTSD(const _EGLThreadInfo *t)
{
_egl_TSD = t;
}
static INLINE _EGLThreadInfo *_eglGetTSD(void)
{
return (_EGLThreadInfo *) _egl_TSD;
}
static INLINE void _eglFiniTSD(void)
{
}
static INLINE EGLBoolean _eglInitTSD(void (*dtor)(_EGLThreadInfo *))
{
/* TODO destroy TSD */
(void) dtor;
(void) _eglFiniTSD;
return EGL_TRUE;
}
#elif PTHREADS
#include <pthread.h>
static _EGL_DECLARE_MUTEX(_egl_TSDMutex);
@@ -22,26 +48,14 @@ static EGLBoolean _egl_TSDInitialized;
static pthread_key_t _egl_TSD;
static void (*_egl_FreeTSD)(_EGLThreadInfo *);
#ifdef GLX_USE_TLS
static __thread const _EGLThreadInfo *_egl_TLS
__attribute__ ((tls_model("initial-exec")));
#endif
static INLINE void _eglSetTSD(const _EGLThreadInfo *t)
{
pthread_setspecific(_egl_TSD, (const void *) t);
#ifdef GLX_USE_TLS
_egl_TLS = t;
#endif
}
static INLINE _EGLThreadInfo *_eglGetTSD(void)
{
#ifdef GLX_USE_TLS
return (_EGLThreadInfo *) _egl_TLS;
#else
return (_EGLThreadInfo *) pthread_getspecific(_egl_TSD);
#endif
}
static INLINE void _eglFiniTSD(void)

View File

@@ -233,53 +233,17 @@ _eglCheckResource(void *res, _EGLResourceType type, _EGLDisplay *dpy)
/**
* Initialize a display resource.
* Link a resource to a display.
*/
void
_eglInitResource(_EGLResource *res, EGLint size, _EGLDisplay *dpy)
_eglLinkResource(_EGLResource *res, _EGLResourceType type, _EGLDisplay *dpy)
{
memset(res, 0, size);
assert(!res->Display || res->Display == dpy);
res->Display = dpy;
res->RefCount = 1;
}
/**
* Increment reference count for the resource.
*/
void
_eglGetResource(_EGLResource *res)
{
assert(res && res->RefCount > 0);
/* hopefully a resource is always manipulated with its display locked */
res->RefCount++;
}
/**
* Decrement reference count for the resource.
*/
EGLBoolean
_eglPutResource(_EGLResource *res)
{
assert(res && res->RefCount > 0);
res->RefCount--;
return (!res->RefCount);
}
/**
* Link a resource to its display.
*/
void
_eglLinkResource(_EGLResource *res, _EGLResourceType type)
{
assert(res->Display);
res->IsLinked = EGL_TRUE;
res->Next = res->Display->ResourceLists[type];
res->Display->ResourceLists[type] = res;
_eglGetResource(res);
res->Next = dpy->ResourceLists[type];
dpy->ResourceLists[type] = res;
}
@@ -306,9 +270,6 @@ _eglUnlinkResource(_EGLResource *res, _EGLResourceType type)
}
res->Next = NULL;
/* do not reset res->Display */
res->IsLinked = EGL_FALSE;
_eglPutResource(res);
/* We always unlink before destroy. The driver still owns a reference */
assert(res->RefCount);
}

View File

@@ -40,7 +40,6 @@ struct _egl_resource
/* which display the resource belongs to */
_EGLDisplay *Display;
EGLBoolean IsLinked;
EGLint RefCount;
/* used to link resources of the same type */
_EGLResource *Next;
@@ -163,19 +162,7 @@ _eglGetDisplayHandle(_EGLDisplay *dpy)
extern void
_eglInitResource(_EGLResource *res, EGLint size, _EGLDisplay *dpy);
PUBLIC void
_eglGetResource(_EGLResource *res);
PUBLIC EGLBoolean
_eglPutResource(_EGLResource *res);
extern void
_eglLinkResource(_EGLResource *res, _EGLResourceType type);
_eglLinkResource(_EGLResource *res, _EGLResourceType type, _EGLDisplay *dpy);
extern void

View File

@@ -9,10 +9,19 @@
#include <stdlib.h>
#include "eglstring.h"
#include "eglconfig.h"
#include "eglcontext.h"
#include "egldefines.h"
#include "egldisplay.h"
#include "egldriver.h"
#include "egllog.h"
#include "eglmisc.h"
#include "eglmode.h"
#include "eglscreen.h"
#include "eglstring.h"
#include "eglsurface.h"
#include "eglimage.h"
#include "eglsync.h"
#include "eglmutex.h"
#if defined(_EGL_OS_UNIX)
@@ -395,62 +404,35 @@ _eglPreloadForEach(const char *search_path,
static const char *
_eglGetSearchPath(void)
{
static char search_path[1024];
static const char *search_path;
#if defined(_EGL_OS_UNIX) || defined(_EGL_OS_WINDOWS)
if (search_path[0] == '\0') {
char *buf = search_path;
size_t len = sizeof(search_path);
EGLBoolean use_env;
char dir_sep;
if (!search_path) {
static char buffer[1024];
const char *p;
int ret;
p = getenv("EGL_DRIVERS_PATH");
#if defined(_EGL_OS_UNIX)
use_env = (geteuid() == getuid() && getegid() == getgid());
dir_sep = '/';
#else
use_env = EGL_TRUE;
dir_sep = '\\';
#endif
if (use_env) {
char *p;
/* extract the dirname from EGL_DRIVER */
p = getenv("EGL_DRIVER");
if (p && strchr(p, dir_sep)) {
ret = _eglsnprintf(buf, len, "%s", p);
if (ret > 0 && ret < len) {
p = strrchr(buf, dir_sep);
*p++ = ':';
len -= p - buf;
buf = p;
}
}
/* append EGL_DRIVERS_PATH */
p = getenv("EGL_DRIVERS_PATH");
if (p) {
ret = _eglsnprintf(buf, len, "%s:", p);
if (ret > 0 && ret < len) {
buf += ret;
len -= ret;
}
}
}
else {
if (p && (geteuid() != getuid() || getegid() != getgid())) {
_eglLog(_EGL_DEBUG,
"ignore EGL_DRIVERS_PATH for setuid/setgid binaries");
p = NULL;
}
#endif /* _EGL_OS_UNIX */
ret = _eglsnprintf(buf, len, "%s", _EGL_DRIVER_SEARCH_DIR);
if (ret < 0 || ret >= len)
search_path[0] = '\0';
_eglLog(_EGL_DEBUG, "EGL search path is %s", search_path);
if (p) {
ret = _eglsnprintf(buffer, sizeof(buffer),
"%s:%s", p, _EGL_DRIVER_SEARCH_DIR);
if (ret > 0 && ret < sizeof(buffer))
search_path = buffer;
}
}
#endif /* defined(_EGL_OS_UNIX) || defined(_EGL_OS_WINDOWS) */
if (!search_path)
search_path = _EGL_DRIVER_SEARCH_DIR;
#else
search_path = "";
#endif
return search_path;
}
@@ -681,6 +663,77 @@ _eglUnloadDrivers(void)
}
/**
* Plug all the available fallback routines into the given driver's
* dispatch table.
*/
void
_eglInitDriverFallbacks(_EGLDriver *drv)
{
/* If a pointer is set to NULL, then the device driver _really_ has
* to implement it.
*/
drv->API.Initialize = NULL;
drv->API.Terminate = NULL;
drv->API.GetConfigs = _eglGetConfigs;
drv->API.ChooseConfig = _eglChooseConfig;
drv->API.GetConfigAttrib = _eglGetConfigAttrib;
drv->API.CreateContext = _eglCreateContext;
drv->API.DestroyContext = _eglDestroyContext;
drv->API.MakeCurrent = _eglMakeCurrent;
drv->API.QueryContext = _eglQueryContext;
drv->API.CreateWindowSurface = _eglCreateWindowSurface;
drv->API.CreatePixmapSurface = _eglCreatePixmapSurface;
drv->API.CreatePbufferSurface = _eglCreatePbufferSurface;
drv->API.DestroySurface = _eglDestroySurface;
drv->API.QuerySurface = _eglQuerySurface;
drv->API.SurfaceAttrib = _eglSurfaceAttrib;
drv->API.BindTexImage = _eglBindTexImage;
drv->API.ReleaseTexImage = _eglReleaseTexImage;
drv->API.SwapInterval = _eglSwapInterval;
drv->API.SwapBuffers = _eglSwapBuffers;
drv->API.CopyBuffers = _eglCopyBuffers;
drv->API.QueryString = _eglQueryString;
drv->API.WaitClient = _eglWaitClient;
drv->API.WaitNative = _eglWaitNative;
#ifdef EGL_MESA_screen_surface
drv->API.ChooseModeMESA = _eglChooseModeMESA;
drv->API.GetModesMESA = _eglGetModesMESA;
drv->API.GetModeAttribMESA = _eglGetModeAttribMESA;
drv->API.GetScreensMESA = _eglGetScreensMESA;
drv->API.CreateScreenSurfaceMESA = _eglCreateScreenSurfaceMESA;
drv->API.ShowScreenSurfaceMESA = _eglShowScreenSurfaceMESA;
drv->API.ScreenPositionMESA = _eglScreenPositionMESA;
drv->API.QueryScreenMESA = _eglQueryScreenMESA;
drv->API.QueryScreenSurfaceMESA = _eglQueryScreenSurfaceMESA;
drv->API.QueryScreenModeMESA = _eglQueryScreenModeMESA;
drv->API.QueryModeStringMESA = _eglQueryModeStringMESA;
#endif /* EGL_MESA_screen_surface */
#ifdef EGL_VERSION_1_2
drv->API.CreatePbufferFromClientBuffer = _eglCreatePbufferFromClientBuffer;
#endif /* EGL_VERSION_1_2 */
#ifdef EGL_KHR_image_base
drv->API.CreateImageKHR = _eglCreateImageKHR;
drv->API.DestroyImageKHR = _eglDestroyImageKHR;
#endif /* EGL_KHR_image_base */
#ifdef EGL_KHR_reusable_sync
drv->API.CreateSyncKHR = _eglCreateSyncKHR;
drv->API.DestroySyncKHR = _eglDestroySyncKHR;
drv->API.ClientWaitSyncKHR = _eglClientWaitSyncKHR;
drv->API.SignalSyncKHR = _eglSignalSyncKHR;
drv->API.GetSyncAttribKHR = _eglGetSyncAttribKHR;
#endif /* EGL_KHR_reusable_sync */
}
/**
* Invoke a callback function on each EGL search path.
*

View File

@@ -80,7 +80,6 @@ extern void
_eglUnloadDrivers(void);
/* defined in eglfallbacks.c */
PUBLIC void
_eglInitDriverFallbacks(_EGLDriver *drv);

View File

@@ -1,99 +0,0 @@
#include <string.h>
#include "egltypedefs.h"
#include "egldriver.h"
#include "eglconfig.h"
#include "eglcontext.h"
#include "eglsurface.h"
#include "eglmisc.h"
#include "eglscreen.h"
#include "eglmode.h"
#include "eglsync.h"
static EGLBoolean
_eglReturnFalse(void)
{
return EGL_FALSE;
}
/**
* Plug all the available fallback routines into the given driver's
* dispatch table.
*/
void
_eglInitDriverFallbacks(_EGLDriver *drv)
{
memset(&drv->API, 0, sizeof(drv->API));
/* the driver has to implement these */
drv->API.Initialize = NULL;
drv->API.Terminate = NULL;
drv->API.GetConfigs = _eglGetConfigs;
drv->API.ChooseConfig = _eglChooseConfig;
drv->API.GetConfigAttrib = _eglGetConfigAttrib;
drv->API.CreateContext = (CreateContext_t) _eglReturnFalse;
drv->API.DestroyContext = (DestroyContext_t) _eglReturnFalse;
drv->API.MakeCurrent = (MakeCurrent_t) _eglReturnFalse;
drv->API.QueryContext = _eglQueryContext;
drv->API.CreateWindowSurface = (CreateWindowSurface_t) _eglReturnFalse;
drv->API.CreatePixmapSurface = (CreatePixmapSurface_t) _eglReturnFalse;
drv->API.CreatePbufferSurface = (CreatePbufferSurface_t) _eglReturnFalse;
drv->API.CreatePbufferFromClientBuffer =
(CreatePbufferFromClientBuffer_t) _eglReturnFalse;
drv->API.DestroySurface = (DestroySurface_t) _eglReturnFalse;
drv->API.QuerySurface = _eglQuerySurface;
drv->API.SurfaceAttrib = _eglSurfaceAttrib;
drv->API.BindTexImage = (BindTexImage_t) _eglReturnFalse;
drv->API.ReleaseTexImage = (ReleaseTexImage_t) _eglReturnFalse;
drv->API.CopyBuffers = (CopyBuffers_t) _eglReturnFalse;
drv->API.SwapBuffers = (SwapBuffers_t) _eglReturnFalse;
drv->API.SwapInterval = _eglSwapInterval;
drv->API.WaitClient = (WaitClient_t) _eglReturnFalse;
drv->API.WaitNative = (WaitNative_t) _eglReturnFalse;
drv->API.GetProcAddress = (GetProcAddress_t) _eglReturnFalse;
drv->API.QueryString = _eglQueryString;
#ifdef EGL_MESA_screen_surface
drv->API.CopyContextMESA = (CopyContextMESA_t) _eglReturnFalse;
drv->API.CreateScreenSurfaceMESA =
(CreateScreenSurfaceMESA_t) _eglReturnFalse;
drv->API.ShowScreenSurfaceMESA = (ShowScreenSurfaceMESA_t) _eglReturnFalse;
drv->API.ChooseModeMESA = _eglChooseModeMESA;
drv->API.GetModesMESA = _eglGetModesMESA;
drv->API.GetModeAttribMESA = _eglGetModeAttribMESA;
drv->API.GetScreensMESA = _eglGetScreensMESA;
drv->API.ScreenPositionMESA = _eglScreenPositionMESA;
drv->API.QueryScreenMESA = _eglQueryScreenMESA;
drv->API.QueryScreenSurfaceMESA = _eglQueryScreenSurfaceMESA;
drv->API.QueryScreenModeMESA = _eglQueryScreenModeMESA;
drv->API.QueryModeStringMESA = _eglQueryModeStringMESA;
#endif /* EGL_MESA_screen_surface */
#ifdef EGL_KHR_image_base
drv->API.CreateImageKHR = NULL;
drv->API.DestroyImageKHR = NULL;
#endif /* EGL_KHR_image_base */
#ifdef EGL_KHR_reusable_sync
drv->API.CreateSyncKHR = NULL;
drv->API.DestroySyncKHR = NULL;
drv->API.ClientWaitSyncKHR = NULL;
drv->API.SignalSyncKHR = NULL;
drv->API.GetSyncAttribKHR = _eglGetSyncAttribKHR;
#endif /* EGL_KHR_reusable_sync */
#ifdef EGL_MESA_drm_image
drv->API.CreateDRMImageMESA = NULL;
drv->API.ExportDRMImageMESA = NULL;
#endif
#ifdef EGL_NOK_swap_region
drv->API.SwapBuffersRegionNOK = NULL;
#endif
}

View File

@@ -2,6 +2,7 @@
#include <string.h>
#include "eglimage.h"
#include "eglcurrent.h"
#include "egllog.h"
@@ -11,57 +12,28 @@
/**
* Parse the list of image attributes and return the proper error code.
*/
EGLint
_eglParseImageAttribList(_EGLImageAttribs *attrs, _EGLDisplay *dpy,
const EGLint *attrib_list)
static EGLint
_eglParseImageAttribList(_EGLImage *img, const EGLint *attrib_list)
{
EGLint i, err = EGL_SUCCESS;
(void) dpy;
memset(attrs, 0, sizeof(attrs));
attrs->ImagePreserved = EGL_FALSE;
attrs->GLTextureLevel = 0;
attrs->GLTextureZOffset = 0;
if (!attrib_list)
return err;
return EGL_SUCCESS;
for (i = 0; attrib_list[i] != EGL_NONE; i++) {
EGLint attr = attrib_list[i++];
EGLint val = attrib_list[i];
switch (attr) {
/* EGL_KHR_image_base */
case EGL_IMAGE_PRESERVED_KHR:
attrs->ImagePreserved = val;
img->Preserved = val;
break;
/* EGL_KHR_gl_image */
case EGL_GL_TEXTURE_LEVEL_KHR:
attrs->GLTextureLevel = val;
img->GLTextureLevel = val;
break;
case EGL_GL_TEXTURE_ZOFFSET_KHR:
attrs->GLTextureZOffset = val;
img->GLTextureZOffset = val;
break;
/* EGL_MESA_drm_image */
case EGL_WIDTH:
attrs->Width = val;
break;
case EGL_HEIGHT:
attrs->Height = val;
break;
case EGL_DRM_BUFFER_FORMAT_MESA:
attrs->DRMBufferFormatMESA = val;
break;
case EGL_DRM_BUFFER_USE_MESA:
attrs->DRMBufferUseMESA = val;
break;
case EGL_DRM_BUFFER_STRIDE_MESA:
attrs->DRMBufferStrideMESA = val;
break;
default:
/* unknown attrs are ignored */
break;
@@ -78,12 +50,41 @@ _eglParseImageAttribList(_EGLImageAttribs *attrs, _EGLDisplay *dpy,
EGLBoolean
_eglInitImage(_EGLImage *img, _EGLDisplay *dpy)
_eglInitImage(_EGLImage *img, _EGLDisplay *dpy, const EGLint *attrib_list)
{
_eglInitResource(&img->Resource, sizeof(*img), dpy);
EGLint err;
memset(img, 0, sizeof(_EGLImage));
img->Resource.Display = dpy;
img->Preserved = EGL_FALSE;
img->GLTextureLevel = 0;
img->GLTextureZOffset = 0;
err = _eglParseImageAttribList(img, attrib_list);
if (err != EGL_SUCCESS)
return _eglError(err, "eglCreateImageKHR");
return EGL_TRUE;
}
_EGLImage *
_eglCreateImageKHR(_EGLDriver *drv, _EGLDisplay *dpy, _EGLContext *ctx,
EGLenum target, EGLClientBuffer buffer,
const EGLint *attr_list)
{
/* driver should override this function */
return NULL;
}
EGLBoolean
_eglDestroyImageKHR(_EGLDriver *drv, _EGLDisplay *dpy, _EGLImage *image)
{
/* driver should override this function */
return EGL_FALSE;
}
#endif /* EGL_KHR_image_base */

View File

@@ -6,23 +6,6 @@
#include "egldisplay.h"
struct _egl_image_attribs
{
/* EGL_KHR_image_base */
EGLBoolean ImagePreserved;
/* EGL_KHR_gl_image */
EGLint GLTextureLevel;
EGLint GLTextureZOffset;
/* EGL_MESA_drm_image */
EGLint Width;
EGLint Height;
EGLint DRMBufferFormatMESA;
EGLint DRMBufferUseMESA;
EGLint DRMBufferStrideMESA;
};
/**
* "Base" class for device driver images.
*/
@@ -30,48 +13,34 @@ struct _egl_image
{
/* An image is a display resource */
_EGLResource Resource;
EGLBoolean Preserved;
EGLint GLTextureLevel;
EGLint GLTextureZOffset;
};
PUBLIC EGLint
_eglParseImageAttribList(_EGLImageAttribs *attrs, _EGLDisplay *dpy,
const EGLint *attrib_list);
PUBLIC EGLBoolean
_eglInitImage(_EGLImage *img, _EGLDisplay *dpy);
_eglInitImage(_EGLImage *img, _EGLDisplay *dpy, const EGLint *attrib_list);
extern _EGLImage *
_eglCreateImageKHR(_EGLDriver *drv, _EGLDisplay *dpy, _EGLContext *ctx,
EGLenum target, EGLClientBuffer buffer, const EGLint *attr_list);
extern EGLBoolean
_eglDestroyImageKHR(_EGLDriver *drv, _EGLDisplay *dpy, _EGLImage *image);
/**
* Increment reference count for the image.
*/
static INLINE _EGLImage *
_eglGetImage(_EGLImage *img)
{
if (img)
_eglGetResource(&img->Resource);
return img;
}
/**
* Decrement reference count for the image.
*/
static INLINE EGLBoolean
_eglPutImage(_EGLImage *img)
{
return (img) ? _eglPutResource(&img->Resource) : EGL_FALSE;
}
/**
* Link an image to its display and return the handle of the link.
* Link an image to a display and return the handle of the link.
* The handle can be passed to client directly.
*/
static INLINE EGLImageKHR
_eglLinkImage(_EGLImage *img)
_eglLinkImage(_EGLImage *img, _EGLDisplay *dpy)
{
_eglLinkResource(&img->Resource, _EGL_RESOURCE_IMAGE);
_eglLinkResource(&img->Resource, _EGL_RESOURCE_IMAGE, dpy);
return (EGLImageKHR) img;
}
@@ -113,4 +82,15 @@ _eglGetImageHandle(_EGLImage *img)
}
/**
* Return true if the image is linked to a display.
*/
static INLINE EGLBoolean
_eglIsImageLinked(_EGLImage *img)
{
_EGLResource *res = (_EGLResource *) img;
return (res && _eglIsResourceLinked(res));
}
#endif /* EGLIMAGE_INCLUDED */

View File

@@ -151,7 +151,6 @@ _eglLog(EGLint level, const char *fmtStr, ...)
{
va_list args;
char msg[MAXSTRING];
int ret;
/* one-time initialization; a little race here is fine */
if (!logging.initialized)
@@ -163,9 +162,7 @@ _eglLog(EGLint level, const char *fmtStr, ...)
if (logging.logger) {
va_start(args, fmtStr);
ret = vsnprintf(msg, MAXSTRING, fmtStr, args);
if (ret < 0 || ret >= MAXSTRING)
strcpy(msg, "<message truncated>");
vsnprintf(msg, MAXSTRING, fmtStr, args);
va_end(args);
logging.logger(level, msg);

View File

@@ -158,3 +158,32 @@ _eglQueryString(_EGLDriver *drv, _EGLDisplay *dpy, EGLint name)
return NULL;
}
}
EGLBoolean
_eglWaitClient(_EGLDriver *drv, _EGLDisplay *dpy, _EGLContext *ctx)
{
/* just a placeholder */
(void) drv;
(void) dpy;
(void) ctx;
return EGL_TRUE;
}
EGLBoolean
_eglWaitNative(_EGLDriver *drv, _EGLDisplay *dpy, EGLint engine)
{
/* just a placeholder */
(void) drv;
(void) dpy;
switch (engine) {
case EGL_CORE_NATIVE_ENGINE:
break;
default:
_eglError(EGL_BAD_PARAMETER, "eglWaitNative(engine)");
return EGL_FALSE;
}
return EGL_TRUE;
}

View File

@@ -37,4 +37,12 @@ extern const char *
_eglQueryString(_EGLDriver *drv, _EGLDisplay *dpy, EGLint name);
extern EGLBoolean
_eglWaitClient(_EGLDriver *drv, _EGLDisplay *dpy, _EGLContext *ctx);
extern EGLBoolean
_eglWaitNative(_EGLDriver *drv, _EGLDisplay *dpy, EGLint engine);
#endif /* EGLMISC_INCLUDED */

View File

@@ -3,9 +3,11 @@
#include <string.h>
#include "egldisplay.h"
#include "egldriver.h"
#include "eglmode.h"
#include "eglcurrent.h"
#include "eglscreen.h"
#include "eglstring.h"
#ifdef EGL_MESA_screen_surface
@@ -29,19 +31,12 @@ _eglLookupMode(EGLModeMESA mode, _EGLDisplay *disp)
/* loop over all screens on the display */
for (scrnum = 0; scrnum < disp->Screens->Size; scrnum++) {
const _EGLScreen *scrn = disp->Screens->Elements[scrnum];
EGLint idx;
/*
* the mode ids of a screen ranges from scrn->Handle to scrn->Handle +
* scrn->NumModes
*/
if (mode >= scrn->Handle &&
mode < scrn->Handle + _EGL_SCREEN_MAX_MODES) {
idx = mode - scrn->Handle;
assert(idx < scrn->NumModes && scrn->Modes[idx].Handle == mode);
return &scrn->Modes[idx];
EGLint i;
/* search list of modes for handle */
for (i = 0; i < scrn->NumModes; i++) {
if (scrn->Modes[i].Handle == mode) {
return scrn->Modes + i;
}
}
}
@@ -49,6 +44,45 @@ _eglLookupMode(EGLModeMESA mode, _EGLDisplay *disp)
}
/**
* Add a new mode with the given attributes (width, height, depth, refreshRate)
* to the given screen.
* Assign a new mode ID/handle to the mode as well.
* \return pointer to the new _EGLMode
*/
_EGLMode *
_eglAddNewMode(_EGLScreen *screen, EGLint width, EGLint height,
EGLint refreshRate, const char *name)
{
EGLint n;
_EGLMode *newModes;
assert(screen);
assert(width > 0);
assert(height > 0);
assert(refreshRate > 0);
n = screen->NumModes;
newModes = (_EGLMode *) realloc(screen->Modes, (n+1) * sizeof(_EGLMode));
if (newModes) {
screen->Modes = newModes;
screen->Modes[n].Handle = n + 1;
screen->Modes[n].Width = width;
screen->Modes[n].Height = height;
screen->Modes[n].RefreshRate = refreshRate;
screen->Modes[n].Optimal = EGL_FALSE;
screen->Modes[n].Interlaced = EGL_FALSE;
screen->Modes[n].Name = _eglstrdup(name);
screen->NumModes++;
return screen->Modes + n;
}
else {
return NULL;
}
}
/**
* Parse the attrib_list to fill in the fields of the given _eglMode
* Return EGL_FALSE if any errors, EGL_TRUE otherwise.

View File

@@ -32,6 +32,11 @@ extern _EGLMode *
_eglLookupMode(EGLModeMESA mode, _EGLDisplay *dpy);
PUBLIC _EGLMode *
_eglAddNewMode(_EGLScreen *screen, EGLint width, EGLint height,
EGLint refreshRate, const char *name);
extern EGLBoolean
_eglChooseModeMESA(_EGLDriver *drv, _EGLDisplay *dpy, _EGLScreen *scrn,
const EGLint *attrib_list, EGLModeMESA *modes,

View File

@@ -18,6 +18,7 @@
#include "egldisplay.h"
#include "eglcurrent.h"
#include "eglmode.h"
#include "eglconfig.h"
#include "eglsurface.h"
#include "eglscreen.h"
#include "eglmutex.h"
@@ -41,8 +42,7 @@ _eglAllocScreenHandle(void)
EGLScreenMESA s;
_eglLockMutex(&_eglNextScreenHandleMutex);
s = _eglNextScreenHandle;
_eglNextScreenHandle += _EGL_SCREEN_MAX_MODES;
s = _eglNextScreenHandle++;
_eglUnlockMutex(&_eglNextScreenHandleMutex);
return s;
@@ -53,54 +53,16 @@ _eglAllocScreenHandle(void)
* Initialize an _EGLScreen object to default values.
*/
void
_eglInitScreen(_EGLScreen *screen, _EGLDisplay *dpy, EGLint num_modes)
_eglInitScreen(_EGLScreen *screen)
{
memset(screen, 0, sizeof(_EGLScreen));
screen->Display = dpy;
screen->NumModes = num_modes;
screen->StepX = 1;
screen->StepY = 1;
if (num_modes > _EGL_SCREEN_MAX_MODES)
num_modes = _EGL_SCREEN_MAX_MODES;
screen->Modes = (_EGLMode *) calloc(num_modes, sizeof(*screen->Modes));
screen->NumModes = (screen->Modes) ? num_modes : 0;
}
/**
* Link a screen to its display and return the handle of the link.
* The handle can be passed to client directly.
*/
EGLScreenMESA
_eglLinkScreen(_EGLScreen *screen)
{
_EGLDisplay *display;
EGLint i;
assert(screen && screen->Display);
display = screen->Display;
if (!display->Screens) {
display->Screens = _eglCreateArray("Screen", 4);
if (!display->Screens)
return (EGLScreenMESA) 0;
}
screen->Handle = _eglAllocScreenHandle();
for (i = 0; i < screen->NumModes; i++)
screen->Modes[i].Handle = screen->Handle + i;
_eglAppendArray(display->Screens, (void *) screen);
return screen->Handle;
}
/**
* Lookup a handle to find the linked config.
* Return NULL if the handle has no corresponding linked config.
* Given a public screen handle, return the internal _EGLScreen object.
*/
_EGLScreen *
_eglLookupScreen(EGLScreenMESA screen, _EGLDisplay *display)
@@ -112,21 +74,39 @@ _eglLookupScreen(EGLScreenMESA screen, _EGLDisplay *display)
for (i = 0; i < display->Screens->Size; i++) {
_EGLScreen *scr = (_EGLScreen *) display->Screens->Elements[i];
if (scr->Handle == screen) {
assert(scr->Display == display);
if (scr->Handle == screen)
return scr;
}
}
return NULL;
}
/**
* Add the given _EGLScreen to the display's list of screens.
*/
void
_eglAddScreen(_EGLDisplay *display, _EGLScreen *screen)
{
assert(display);
assert(screen);
if (!display->Screens) {
display->Screens = _eglCreateArray("Screen", 4);
if (!display->Screens)
return;
}
screen->Handle = _eglAllocScreenHandle();
_eglAppendArray(display->Screens, (void *) screen);
}
static EGLBoolean
_eglFlattenScreen(void *elem, void *buffer)
{
_EGLScreen *scr = (_EGLScreen *) elem;
EGLScreenMESA *handle = (EGLScreenMESA *) buffer;
*handle = _eglGetScreenHandle(scr);
*handle = scr->Handle;
return EGL_TRUE;
}
@@ -142,6 +122,66 @@ _eglGetScreensMESA(_EGLDriver *drv, _EGLDisplay *display, EGLScreenMESA *screens
}
/**
* Drivers should do a proper implementation.
*/
_EGLSurface *
_eglCreateScreenSurfaceMESA(_EGLDriver *drv, _EGLDisplay *dpy, _EGLConfig *conf,
const EGLint *attrib_list)
{
return NULL;
}
/**
* Show the given surface on the named screen.
* If surface is EGL_NO_SURFACE, disable the screen's output.
*
* This is just a placeholder function; drivers will always override
* this with code that _really_ shows the surface.
*/
EGLBoolean
_eglShowScreenSurfaceMESA(_EGLDriver *drv, _EGLDisplay *dpy,
_EGLScreen *scrn, _EGLSurface *surf,
_EGLMode *mode)
{
if (!surf) {
scrn->CurrentSurface = NULL;
}
else {
if (surf->Type != EGL_SCREEN_BIT_MESA) {
_eglError(EGL_BAD_SURFACE, "eglShowSurfaceMESA");
return EGL_FALSE;
}
if (surf->Width < mode->Width || surf->Height < mode->Height) {
_eglError(EGL_BAD_SURFACE,
"eglShowSurfaceMESA(surface smaller than screen size)");
return EGL_FALSE;
}
scrn->CurrentSurface = surf;
scrn->CurrentMode = mode;
}
return EGL_TRUE;
}
/**
* Set a screen's current display mode.
* Note: mode = EGL_NO_MODE is valid (turns off the screen)
*
* This is just a placeholder function; drivers will always override
* this with code that _really_ sets the mode.
*/
EGLBoolean
_eglScreenModeMESA(_EGLDriver *drv, _EGLDisplay *dpy, _EGLScreen *scrn,
_EGLMode *m)
{
scrn->CurrentMode = m;
return EGL_TRUE;
}
/**
* Set a screen's surface origin.
*/
@@ -202,4 +242,33 @@ _eglQueryScreenMESA(_EGLDriver *drv, _EGLDisplay *dpy, _EGLScreen *scrn,
}
/**
* Delete the modes associated with given screen.
*/
void
_eglDestroyScreenModes(_EGLScreen *scrn)
{
EGLint i;
for (i = 0; i < scrn->NumModes; i++) {
if (scrn->Modes[i].Name)
free((char *) scrn->Modes[i].Name); /* cast away const */
}
if (scrn->Modes)
free(scrn->Modes);
scrn->Modes = NULL;
scrn->NumModes = 0;
}
/**
* Default fallback routine - drivers should usually override this.
*/
void
_eglDestroyScreen(_EGLScreen *scrn)
{
_eglDestroyScreenModes(scrn);
free(scrn);
}
#endif /* EGL_MESA_screen_surface */

View File

@@ -8,9 +8,6 @@
#ifdef EGL_MESA_screen_surface
#define _EGL_SCREEN_MAX_MODES 16
/**
* Per-screen information.
* Note that an EGL screen doesn't have a size. A screen may be set to
@@ -22,8 +19,6 @@
*/
struct _egl_screen
{
_EGLDisplay *Display;
EGLScreenMESA Handle; /* The public/opaque handle which names this object */
_EGLMode *CurrentMode;
@@ -38,35 +33,41 @@ struct _egl_screen
PUBLIC void
_eglInitScreen(_EGLScreen *screen, _EGLDisplay *dpy, EGLint num_modes);
PUBLIC EGLScreenMESA
_eglLinkScreen(_EGLScreen *screen);
_eglInitScreen(_EGLScreen *screen);
extern _EGLScreen *
_eglLookupScreen(EGLScreenMESA screen, _EGLDisplay *dpy);
/**
* Return the handle of a linked screen.
*/
static INLINE EGLScreenMESA
_eglGetScreenHandle(_EGLScreen *screen)
{
return (screen) ? screen->Handle : (EGLScreenMESA) 0;
}
PUBLIC void
_eglAddScreen(_EGLDisplay *display, _EGLScreen *screen);
extern EGLBoolean
_eglGetScreensMESA(_EGLDriver *drv, _EGLDisplay *dpy, EGLScreenMESA *screens, EGLint max_screens, EGLint *num_screens);
extern _EGLSurface *
_eglCreateScreenSurfaceMESA(_EGLDriver *drv, _EGLDisplay *dpy, _EGLConfig *conf, const EGLint *attrib_list);
extern EGLBoolean
_eglShowScreenSurfaceMESA(_EGLDriver *drv, _EGLDisplay *dpy, _EGLScreen *scrn, _EGLSurface *surf, _EGLMode *m);
extern EGLBoolean
_eglScreenModeMESA(_EGLDriver *drv, _EGLDisplay *dpy, _EGLScreen *scrn, _EGLMode *m);
extern EGLBoolean
_eglScreenPositionMESA(_EGLDriver *drv, _EGLDisplay *dpy, _EGLScreen *scrn, EGLint x, EGLint y);
extern EGLBoolean
_eglQueryDisplayMESA(_EGLDriver *drv, _EGLDisplay *dpy, EGLint attribute, EGLint *value);
extern EGLBoolean
_eglQueryScreenSurfaceMESA(_EGLDriver *drv, _EGLDisplay *dpy,
_EGLScreen *scrn, _EGLSurface **surface);
@@ -80,6 +81,14 @@ extern EGLBoolean
_eglQueryScreenMESA(_EGLDriver *drv, _EGLDisplay *dpy, _EGLScreen *scrn, EGLint attribute, EGLint *value);
extern void
_eglDestroyScreenModes(_EGLScreen *scrn);
PUBLIC void
_eglDestroyScreen(_EGLScreen *scrn);
#endif /* EGL_MESA_screen_surface */

View File

@@ -17,12 +17,12 @@
static void
_eglClampSwapInterval(_EGLSurface *surf, EGLint interval)
{
EGLint bound = surf->Config->MaxSwapInterval;
EGLint bound = GET_CONFIG_ATTRIB(surf->Config, EGL_MAX_SWAP_INTERVAL);
if (interval >= bound) {
interval = bound;
}
else {
bound = surf->Config->MinSwapInterval;
bound = GET_CONFIG_ATTRIB(surf->Config, EGL_MIN_SWAP_INTERVAL);
if (interval < bound)
interval = bound;
}
@@ -263,13 +263,14 @@ _eglInitSurface(_EGLSurface *surf, _EGLDisplay *dpy, EGLint type,
return EGL_FALSE;
}
if ((conf->SurfaceType & type) == 0) {
if ((GET_CONFIG_ATTRIB(conf, EGL_SURFACE_TYPE) & type) == 0) {
/* The config can't be used to create a surface of this type */
_eglError(EGL_BAD_CONFIG, func);
return EGL_FALSE;
}
_eglInitResource(&surf->Resource, sizeof(*surf), dpy);
memset(surf, 0, sizeof(_EGLSurface));
surf->Resource.Display = dpy;
surf->Type = type;
surf->Config = conf;
@@ -302,6 +303,24 @@ _eglInitSurface(_EGLSurface *surf, _EGLDisplay *dpy, EGLint type,
}
EGLBoolean
_eglSwapBuffers(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surf)
{
/* Drivers have to do the actual buffer swap. */
return EGL_TRUE;
}
EGLBoolean
_eglCopyBuffers(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surf,
EGLNativePixmapType target)
{
/* copy surface to native pixmap */
/* All implementation burdon for this is in the device driver */
return EGL_FALSE;
}
EGLBoolean
_eglQuerySurface(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surface,
EGLint attribute, EGLint *value)
@@ -314,7 +333,7 @@ _eglQuerySurface(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surface,
*value = surface->Height;
break;
case EGL_CONFIG_ID:
*value = surface->Config->ConfigID;
*value = GET_CONFIG_ATTRIB(surface->Config, EGL_CONFIG_ID);
break;
case EGL_LARGEST_PBUFFER:
*value = surface->LargestPbuffer;
@@ -369,6 +388,51 @@ _eglQuerySurface(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surface,
}
/**
* Drivers should do a proper implementation.
*/
_EGLSurface *
_eglCreateWindowSurface(_EGLDriver *drv, _EGLDisplay *dpy, _EGLConfig *conf,
EGLNativeWindowType window, const EGLint *attrib_list)
{
return NULL;
}
/**
* Drivers should do a proper implementation.
*/
_EGLSurface *
_eglCreatePixmapSurface(_EGLDriver *drv, _EGLDisplay *dpy, _EGLConfig *conf,
EGLNativePixmapType pixmap, const EGLint *attrib_list)
{
return NULL;
}
/**
* Drivers should do a proper implementation.
*/
_EGLSurface *
_eglCreatePbufferSurface(_EGLDriver *drv, _EGLDisplay *dpy, _EGLConfig *conf,
const EGLint *attrib_list)
{
return NULL;
}
/**
* Default fallback routine - drivers should usually override this.
*/
EGLBoolean
_eglDestroySurface(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surf)
{
if (!_eglIsSurfaceBound(surf))
free(surf);
return EGL_TRUE;
}
/**
* Default fallback routine - drivers might override this.
*/
@@ -381,7 +445,7 @@ _eglSurfaceAttrib(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surface,
switch (attribute) {
case EGL_MIPMAP_LEVEL:
confval = surface->Config->RenderableType;
confval = GET_CONFIG_ATTRIB(surface->Config, EGL_RENDERABLE_TYPE);
if (!(confval & (EGL_OPENGL_ES_BIT | EGL_OPENGL_ES2_BIT))) {
err = EGL_BAD_PARAMETER;
break;
@@ -393,7 +457,7 @@ _eglSurfaceAttrib(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surface,
case EGL_MULTISAMPLE_RESOLVE_DEFAULT:
break;
case EGL_MULTISAMPLE_RESOLVE_BOX:
confval = surface->Config->SurfaceType;
confval = GET_CONFIG_ATTRIB(surface->Config, EGL_SURFACE_TYPE);
if (!(confval & EGL_MULTISAMPLE_RESOLVE_BOX_BIT))
err = EGL_BAD_MATCH;
break;
@@ -410,7 +474,7 @@ _eglSurfaceAttrib(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surface,
case EGL_BUFFER_DESTROYED:
break;
case EGL_BUFFER_PRESERVED:
confval = surface->Config->SurfaceType;
confval = GET_CONFIG_ATTRIB(surface->Config, EGL_SURFACE_TYPE);
if (!(confval & EGL_SWAP_BEHAVIOR_PRESERVED_BIT))
err = EGL_BAD_MATCH;
break;
@@ -472,6 +536,40 @@ _eglBindTexImage(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surface,
}
EGLBoolean
_eglReleaseTexImage(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surface,
EGLint buffer)
{
/* Just do basic error checking and return success/fail.
* Drivers must implement the real stuff.
*/
if (surface->Type != EGL_PBUFFER_BIT) {
_eglError(EGL_BAD_SURFACE, "eglBindTexImage");
return EGL_FALSE;
}
if (surface->TextureFormat == EGL_NO_TEXTURE) {
_eglError(EGL_BAD_MATCH, "eglBindTexImage");
return EGL_FALSE;
}
if (buffer != EGL_BACK_BUFFER) {
_eglError(EGL_BAD_PARAMETER, "eglReleaseTexImage");
return EGL_FALSE;
}
if (!surface->BoundToTexture) {
_eglError(EGL_BAD_SURFACE, "eglReleaseTexImage");
return EGL_FALSE;
}
surface->BoundToTexture = EGL_FALSE;
return EGL_TRUE;
}
EGLBoolean
_eglSwapInterval(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surf,
EGLint interval)
@@ -479,3 +577,24 @@ _eglSwapInterval(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surf,
_eglClampSwapInterval(surf, interval);
return EGL_TRUE;
}
#ifdef EGL_VERSION_1_2
/**
* Example function - drivers should do a proper implementation.
*/
_EGLSurface *
_eglCreatePbufferFromClientBuffer(_EGLDriver *drv, _EGLDisplay *dpy,
EGLenum buftype, EGLClientBuffer buffer,
_EGLConfig *conf, const EGLint *attrib_list)
{
if (buftype != EGL_OPENVG_IMAGE) {
_eglError(EGL_BAD_PARAMETER, "eglCreatePbufferFromClientBuffer");
return NULL;
}
return NULL;
}
#endif /* EGL_VERSION_1_2 */

View File

@@ -51,10 +51,34 @@ _eglInitSurface(_EGLSurface *surf, _EGLDisplay *dpy, EGLint type,
_EGLConfig *config, const EGLint *attrib_list);
extern EGLBoolean
_eglSwapBuffers(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surf);
extern EGLBoolean
_eglCopyBuffers(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surf, EGLNativePixmapType target);
extern EGLBoolean
_eglQuerySurface(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surf, EGLint attribute, EGLint *value);
extern _EGLSurface *
_eglCreateWindowSurface(_EGLDriver *drv, _EGLDisplay *dpy, _EGLConfig *conf, EGLNativeWindowType window, const EGLint *attrib_list);
extern _EGLSurface *
_eglCreatePixmapSurface(_EGLDriver *drv, _EGLDisplay *dpy, _EGLConfig *conf, EGLNativePixmapType pixmap, const EGLint *attrib_list);
extern _EGLSurface *
_eglCreatePbufferSurface(_EGLDriver *drv, _EGLDisplay *dpy, _EGLConfig *conf, const EGLint *attrib_list);
extern EGLBoolean
_eglDestroySurface(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surf);
extern EGLBoolean
_eglSurfaceAttrib(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surf, EGLint attribute, EGLint value);
@@ -63,40 +87,45 @@ PUBLIC extern EGLBoolean
_eglBindTexImage(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surf, EGLint buffer);
extern EGLBoolean
_eglReleaseTexImage(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surf, EGLint buffer);
extern EGLBoolean
_eglSwapInterval(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSurface *surf, EGLint interval);
/**
* Increment reference count for the surface.
*/
static INLINE _EGLSurface *
_eglGetSurface(_EGLSurface *surf)
{
if (surf)
_eglGetResource(&surf->Resource);
return surf;
}
#ifdef EGL_VERSION_1_2
extern _EGLSurface *
_eglCreatePbufferFromClientBuffer(_EGLDriver *drv, _EGLDisplay *dpy,
EGLenum buftype, EGLClientBuffer buffer,
_EGLConfig *conf, const EGLint *attrib_list);
#endif /* EGL_VERSION_1_2 */
/**
* Decrement reference count for the surface.
* Return true if there is a context bound to the surface.
*
* The binding is considered a reference to the surface. Drivers should not
* destroy a surface when it is bound.
*/
static INLINE EGLBoolean
_eglPutSurface(_EGLSurface *surf)
_eglIsSurfaceBound(_EGLSurface *surf)
{
return (surf) ? _eglPutResource(&surf->Resource) : EGL_FALSE;
return (surf->CurrentContext != NULL);
}
/**
* Link a surface to its display and return the handle of the link.
* Link a surface to a display and return the handle of the link.
* The handle can be passed to client directly.
*/
static INLINE EGLSurface
_eglLinkSurface(_EGLSurface *surf)
_eglLinkSurface(_EGLSurface *surf, _EGLDisplay *dpy)
{
_eglLinkResource(&surf->Resource, _EGL_RESOURCE_SURFACE);
_eglLinkResource(&surf->Resource, _EGL_RESOURCE_SURFACE, dpy);
return (EGLSurface) surf;
}
@@ -138,4 +167,18 @@ _eglGetSurfaceHandle(_EGLSurface *surf)
}
/**
* Return true if the surface is linked to a display.
*
* The link is considered a reference to the surface (the display is owning the
* surface). Drivers should not destroy a surface when it is linked.
*/
static INLINE EGLBoolean
_eglIsSurfaceLinked(_EGLSurface *surf)
{
_EGLResource *res = (_EGLResource *) surf;
return (res && _eglIsResourceLinked(res));
}
#endif /* EGLSURFACE_INCLUDED */

View File

@@ -50,7 +50,10 @@ _eglInitSync(_EGLSync *sync, _EGLDisplay *dpy, EGLenum type,
!(type == EGL_SYNC_FENCE_KHR && dpy->Extensions.KHR_fence_sync))
return _eglError(EGL_BAD_ATTRIBUTE, "eglCreateSyncKHR");
_eglInitResource(&sync->Resource, sizeof(*sync), dpy);
memset(sync, 0, sizeof(*sync));
sync->Resource.Display = dpy;
sync->Type = type;
sync->SyncStatus = EGL_UNSIGNALED_KHR;
sync->SyncCondition = EGL_SYNC_PRIOR_COMMANDS_COMPLETE_KHR;
@@ -63,6 +66,37 @@ _eglInitSync(_EGLSync *sync, _EGLDisplay *dpy, EGLenum type,
}
_EGLSync *
_eglCreateSyncKHR(_EGLDriver *drv, _EGLDisplay *dpy,
EGLenum type, const EGLint *attrib_list)
{
return NULL;
}
EGLBoolean
_eglDestroySyncKHR(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSync *sync)
{
return EGL_TRUE;
}
EGLint
_eglClientWaitSyncKHR(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSync *sync,
EGLint flags, EGLTimeKHR timeout)
{
return EGL_FALSE;
}
EGLBoolean
_eglSignalSyncKHR(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSync *sync,
EGLenum mode)
{
return EGL_FALSE;
}
EGLBoolean
_eglGetSyncAttribKHR(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSync *sync,
EGLint attribute, EGLint *value)

View File

@@ -28,41 +28,38 @@ _eglInitSync(_EGLSync *sync, _EGLDisplay *dpy, EGLenum type,
const EGLint *attrib_list);
extern _EGLSync *
_eglCreateSyncKHR(_EGLDriver *drv, _EGLDisplay *dpy,
EGLenum type, const EGLint *attrib_list);
extern EGLBoolean
_eglDestroySyncKHR(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSync *sync);
extern EGLint
_eglClientWaitSyncKHR(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSync *sync,
EGLint flags, EGLTimeKHR timeout);
extern EGLBoolean
_eglSignalSyncKHR(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSync *sync,
EGLenum mode);
extern EGLBoolean
_eglGetSyncAttribKHR(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSync *sync,
EGLint attribute, EGLint *value);
/**
* Increment reference count for the sync.
*/
static INLINE _EGLSync *
_eglGetSync(_EGLSync *sync)
{
if (sync)
_eglGetResource(&sync->Resource);
return sync;
}
/**
* Decrement reference count for the sync.
*/
static INLINE EGLBoolean
_eglPutSync(_EGLSync *sync)
{
return (sync) ? _eglPutResource(&sync->Resource) : EGL_FALSE;
}
/**
* Link a sync to its display and return the handle of the link.
* Link a sync to a display and return the handle of the link.
* The handle can be passed to client directly.
*/
static INLINE EGLSyncKHR
_eglLinkSync(_EGLSync *sync)
_eglLinkSync(_EGLSync *sync, _EGLDisplay *dpy)
{
_eglLinkResource(&sync->Resource, _EGL_RESOURCE_SYNC);
_eglLinkResource(&sync->Resource, _EGL_RESOURCE_SYNC, dpy);
return (EGLSyncKHR) sync;
}
@@ -103,6 +100,20 @@ _eglGetSyncHandle(_EGLSync *sync)
}
/**
* Return true if the sync is linked to a display.
*
* The link is considered a reference to the sync (the display is owning the
* sync). Drivers should not destroy a sync when it is linked.
*/
static INLINE EGLBoolean
_eglIsSyncLinked(_EGLSync *sync)
{
_EGLResource *res = (_EGLResource *) sync;
return (res && _eglIsResourceLinked(res));
}
#endif /* EGL_KHR_reusable_sync */

View File

@@ -24,8 +24,6 @@ typedef struct _egl_extensions _EGLExtensions;
typedef struct _egl_image _EGLImage;
typedef struct _egl_image_attribs _EGLImageAttribs;
typedef struct _egl_mode _EGLMode;
typedef struct _egl_resource _EGLResource;

View File

@@ -40,7 +40,7 @@ depend: $(C_SOURCES) $(CPP_SOURCES) $(ASM_SOURCES) $(SYMLINKS) $(GENERATED_SOURC
touch depend
$(MKDEP) $(MKDEP_OPTIONS) $(INCLUDES) $(C_SOURCES) $(CPP_SOURCES) $(ASM_SOURCES) $(GENERATED_SOURCES) 2> /dev/null
$(PROGS): % : %.o $(PROGS_DEPS)
$(PROGS): % : %.o
$(LD) $(LDFLAGS) $(filter %.o,$^) -o $@ -Wl,--start-group $(LIBS) -Wl,--end-group
# Emacs tags

View File

@@ -1,135 +1,30 @@
Import('env')
import os
#
# Auxiliary modules
#
Import('*')
SConscript('auxiliary/SConscript')
#
# Drivers
#
SConscript([
'drivers/failover/SConscript',
'drivers/galahad/SConscript',
'drivers/identity/SConscript',
'drivers/llvmpipe/SConscript',
'drivers/rbug/SConscript',
'drivers/softpipe/SConscript',
'drivers/svga/SConscript',
'drivers/trace/SConscript',
])
if not env['msvc']:
# These drivers do not build on MSVC compilers
SConscript([
'drivers/i915/SConscript',
'drivers/i965/SConscript',
'drivers/r300/SConscript',
])
if env['drm']:
# These drivers depend on drm headers
if env['drm_radeon']:
SConscript([
'drivers/r600/SConscript',
])
# XXX: nouveau drivers have a tight dependency on libdrm, so to enable
# we need some version logic before we enable them. Also, ATM there is
# no nouveau target in scons
# if env['drm_nouveau']:
# SConscript([
# 'drivers/nouveau/SConscript',
# 'drivers/nv50/SConscript',
# 'drivers/nvfx/SConscript',
# ])
#
# State trackers
#
for driver in env['drivers']:
SConscript(os.path.join('drivers', driver, 'SConscript'))
# Needed by some state trackers
SConscript('winsys/sw/null/SConscript')
SConscript('state_trackers/python/SConscript')
if env['platform'] != 'embedded':
SConscript('state_trackers/vega/SConscript')
if platform != 'embedded':
SConscript('state_trackers/glx/xlib/SConscript')
SConscript('state_trackers/dri/SConscript')
SConscript('state_trackers/xorg/SConscript')
SConscript('state_trackers/egl/SConscript')
SConscript('state_trackers/vega/SConscript')
if env['x11']:
SConscript('state_trackers/glx/xlib/SConscript')
if env['dri']:
SConscript('state_trackers/dri/SConscript')
if env['dri'] and env['xorg']:
SConscript('state_trackers/xorg/SConscript')
if env['platform'] == 'windows':
SConscript([
'state_trackers/egl/SConscript',
'state_trackers/wgl/SConscript',
])
#
# Winsys
#
if platform == 'windows':
SConscript('state_trackers/wgl/SConscript')
SConscript('winsys/SConscript')
#
# Targets
#
SConscript('targets/SConscript')
SConscript([
'targets/graw-null/SConscript',
])
if env['x11']:
SConscript([
'targets/graw-xlib/SConscript',
'targets/libgl-xlib/SConscript',
])
if env['platform'] == 'windows':
SConscript([
'targets/graw-gdi/SConscript',
'targets/libgl-gdi/SConscript',
#'egl-gdi/SConscript',
])
if env['dri']:
SConscript([
'targets/SConscript.dri',
'targets/dri-swrast/SConscript',
'targets/dri-vmwgfx/SConscript',
#'targets/dri-nouveau/SConscript',
])
if env['drm_intel']:
SConscript([
'targets/dri-i915/SConscript',
'targets/dri-i965/SConscript',
])
if env['drm_radeon']:
SConscript([
'targets/dri-r300/SConscript',
'targets/dri-r600/SConscript',
])
if env['xorg'] and env['drm']:
SConscript([
#'targets/xorg-i915/SConscript',
#'targets/xorg-i965/SConscript',
#'targets/xorg-nouveau/SConscript',
#'targets/xorg-radeon/SConscript',
'targets/xorg-vmwgfx/SConscript',
])
#
# Unit tests & tools
#
if env['platform'] != 'embedded':
SConscript('tests/unit/SConscript')
SConscript('tests/graw/SConscript')
if platform != 'embedded':
SConscript('tests/unit/SConscript')
SConscript('tests/graw/SConscript')

View File

@@ -8,7 +8,6 @@ C_SOURCES = \
cso_cache/cso_context.c \
cso_cache/cso_hash.c \
draw/draw_context.c \
draw/draw_fs.c \
draw/draw_gs.c \
draw/draw_pipe.c \
draw/draw_pipe_aaline.c \
@@ -122,18 +121,17 @@ C_SOURCES = \
util/u_handle_table.c \
util/u_hash.c \
util/u_hash_table.c \
util/u_index_modify.c \
util/u_keymap.c \
util/u_linear.c \
util/u_linkage.c \
util/u_network.c \
util/u_math.c \
util/u_mempool.c \
util/u_mm.c \
util/u_rect.c \
util/u_ringbuffer.c \
util/u_sampler.c \
util/u_simple_shaders.c \
util/u_slab.c \
util/u_snprintf.c \
util/u_staging.c \
util/u_surface.c \
@@ -142,7 +140,8 @@ C_SOURCES = \
util/u_tile.c \
util/u_transfer.c \
util/u_resource.c \
util/u_upload_mgr.c
util/u_upload_mgr.c \
target-helpers/wrap_screen.c
# Disabling until pipe-video branch gets merged in
#vl/vl_bitstream_parser.c \
@@ -154,7 +153,6 @@ C_SOURCES = \
GALLIVM_SOURCES = \
gallivm/lp_bld_arit.c \
gallivm/lp_bld_assert.c \
gallivm/lp_bld_bitarit.c \
gallivm/lp_bld_const.c \
gallivm/lp_bld_conv.c \
gallivm/lp_bld_debug.c \
@@ -170,12 +168,10 @@ GALLIVM_SOURCES = \
gallivm/lp_bld_printf.c \
gallivm/lp_bld_quad.c \
gallivm/lp_bld_sample.c \
gallivm/lp_bld_sample_aos.c \
gallivm/lp_bld_sample_soa.c \
gallivm/lp_bld_struct.c \
gallivm/lp_bld_swizzle.c \
gallivm/lp_bld_tgsi_aos.c \
gallivm/lp_bld_tgsi_info.c \
gallivm/lp_bld_tgsi_soa.c \
gallivm/lp_bld_type.c \
draw/draw_llvm.c \
@@ -185,7 +181,7 @@ GALLIVM_SOURCES = \
draw/draw_pt_fetch_shade_pipeline_llvm.c
GALLIVM_CPP_SOURCES = \
gallivm/lp_bld_misc.cpp
gallivm/lp_bld_misc.cpp
GENERATED_SOURCES = \
indices/u_indices_gen.c \
@@ -203,6 +199,9 @@ CPP_SOURCES += \
endif
LIBRARY_DEFINES += -D__STDC_CONSTANT_MACROS
include ../Makefile.template

View File

@@ -7,6 +7,8 @@ env.Append(CPPPATH = [
'util',
])
env.Tool('udis86')
env.CodeGenerate(
target = 'indices/u_indices_gen.c',
script = 'indices/u_indices_gen.py',
@@ -52,7 +54,6 @@ source = [
'cso_cache/cso_context.c',
'cso_cache/cso_hash.c',
'draw/draw_context.c',
'draw/draw_fs.c',
'draw/draw_gs.c',
'draw/draw_pipe.c',
'draw/draw_pipe_aaline.c',
@@ -169,19 +170,18 @@ source = [
'util/u_handle_table.c',
'util/u_hash.c',
'util/u_hash_table.c',
'util/u_index_modify.c',
'util/u_keymap.c',
'util/u_linear.c',
'util/u_linkage.c',
'util/u_network.c',
'util/u_math.c',
'util/u_mempool.c',
'util/u_mm.c',
'util/u_rect.c',
'util/u_resource.c',
'util/u_ringbuffer.c',
'util/u_sampler.c',
'util/u_simple_shaders.c',
'util/u_slab.c',
'util/u_snprintf.c',
'util/u_staging.c',
'util/u_surface.c',
@@ -196,45 +196,40 @@ source = [
#'vl/vl_compositor.c',
#'vl/vl_csc.c',
#'vl/vl_shader_build.c',
'target-helpers/wrap_screen.c',
]
if env['llvm']:
if env['UDIS86']:
env.Append(CPPDEFINES = [('HAVE_UDIS86', '1')])
source += [
'gallivm/lp_bld_arit.c',
'gallivm/lp_bld_assert.c',
'gallivm/lp_bld_bitarit.c',
'gallivm/lp_bld_const.c',
'gallivm/lp_bld_conv.c',
'gallivm/lp_bld_debug.c',
'gallivm/lp_bld_flow.c',
'gallivm/lp_bld_format_aos.c',
'gallivm/lp_bld_format_soa.c',
'gallivm/lp_bld_format_yuv.c',
'gallivm/lp_bld_gather.c',
'gallivm/lp_bld_init.c',
'gallivm/lp_bld_intr.c',
'gallivm/lp_bld_logic.c',
'gallivm/lp_bld_misc.cpp',
'gallivm/lp_bld_pack.c',
'gallivm/lp_bld_printf.c',
'gallivm/lp_bld_quad.c',
'gallivm/lp_bld_sample.c',
'gallivm/lp_bld_sample_aos.c',
'gallivm/lp_bld_sample_soa.c',
'gallivm/lp_bld_struct.c',
'gallivm/lp_bld_swizzle.c',
'gallivm/lp_bld_tgsi_aos.c',
'gallivm/lp_bld_tgsi_info.c',
'gallivm/lp_bld_tgsi_soa.c',
'gallivm/lp_bld_type.c',
'draw/draw_llvm.c',
'draw/draw_llvm_sample.c',
'draw/draw_llvm_translate.c',
'draw/draw_pt_fetch_shade_pipeline_llvm.c',
'draw/draw_vs_llvm.c'
'gallivm/lp_bld_arit.c',
'gallivm/lp_bld_assert.c',
'gallivm/lp_bld_const.c',
'gallivm/lp_bld_conv.c',
'gallivm/lp_bld_debug.c',
'gallivm/lp_bld_flow.c',
'gallivm/lp_bld_format_aos.c',
'gallivm/lp_bld_format_soa.c',
'gallivm/lp_bld_format_yuv.c',
'gallivm/lp_bld_gather.c',
'gallivm/lp_bld_init.c',
'gallivm/lp_bld_intr.c',
'gallivm/lp_bld_logic.c',
'gallivm/lp_bld_misc.cpp',
'gallivm/lp_bld_pack.c',
'gallivm/lp_bld_printf.c',
'gallivm/lp_bld_quad.c',
'gallivm/lp_bld_sample.c',
'gallivm/lp_bld_sample_soa.c',
'gallivm/lp_bld_struct.c',
'gallivm/lp_bld_swizzle.c',
'gallivm/lp_bld_tgsi_aos.c',
'gallivm/lp_bld_tgsi_soa.c',
'gallivm/lp_bld_type.c',
'draw/draw_llvm.c',
'draw/draw_llvm_sample.c',
'draw/draw_llvm_translate.c',
'draw/draw_pt_fetch_shade_pipeline_llvm.c',
'draw/draw_vs_llvm.c'
]
gallium = env.ConvenienceLibrary(
@@ -242,6 +237,4 @@ gallium = env.ConvenienceLibrary(
source = source,
)
env.Alias('gallium', gallium)
Export('gallium')

View File

@@ -63,32 +63,19 @@ draw_get_option_use_llvm(void)
}
#endif
/**
* Create new draw module context.
*/
struct draw_context *
draw_create(struct pipe_context *pipe)
{
return draw_create_gallivm(pipe, NULL);
}
/**
* Create new draw module context with gallivm state for LLVM JIT.
*/
struct draw_context *
draw_create_gallivm(struct pipe_context *pipe, struct gallivm_state *gallivm)
struct draw_context *draw_create( struct pipe_context *pipe )
{
struct draw_context *draw = CALLOC_STRUCT( draw_context );
if (draw == NULL)
goto fail;
#if HAVE_LLVM
if (draw_get_option_use_llvm() && gallivm) {
draw->llvm = draw_llvm_create(draw, gallivm);
if(draw_get_option_use_llvm())
{
lp_build_init();
assert(lp_build_engine);
draw->engine = lp_build_engine;
draw->llvm = draw_llvm_create(draw);
}
#endif
@@ -104,8 +91,6 @@ fail:
return NULL;
}
boolean draw_init(struct draw_context *draw)
{
/*
@@ -350,7 +335,6 @@ draw_set_mapped_constant_buffer(struct draw_context *draw,
case PIPE_SHADER_VERTEX:
draw->pt.user.vs_constants[slot] = buffer;
draw->pt.user.vs_constants_size[slot] = size;
draw->pt.user.planes = (float (*) [12][4]) &(draw->plane[0]);
draw_vs_set_constants(draw, slot, buffer, size);
break;
case PIPE_SHADER_GEOMETRY:
@@ -429,42 +413,6 @@ draw_set_force_passthrough( struct draw_context *draw, boolean enable )
}
/**
* Allocate an extra vertex/geometry shader vertex attribute.
* This is used by some of the optional draw module stages such
* as wide_point which may need to allocate additional generic/texcoord
* attributes.
*/
int
draw_alloc_extra_vertex_attrib(struct draw_context *draw,
uint semantic_name, uint semantic_index)
{
const int num_outputs = draw_current_shader_outputs(draw);
const int n = draw->extra_shader_outputs.num;
assert(n < Elements(draw->extra_shader_outputs.semantic_name));
draw->extra_shader_outputs.semantic_name[n] = semantic_name;
draw->extra_shader_outputs.semantic_index[n] = semantic_index;
draw->extra_shader_outputs.slot[n] = num_outputs + n;
draw->extra_shader_outputs.num++;
return draw->extra_shader_outputs.slot[n];
}
/**
* Remove all extra vertex attributes that were allocated with
* draw_alloc_extra_vertex_attrib().
*/
void
draw_remove_extra_vertex_attribs(struct draw_context *draw)
{
draw->extra_shader_outputs.num = 0;
}
/**
* Ask the draw module for the location/slot of the given vertex attribute in
* a post-transformed vertex.
@@ -498,12 +446,12 @@ draw_find_shader_output(const struct draw_context *draw,
return i;
}
/* Search the extra vertex attributes */
for (i = 0; i < draw->extra_shader_outputs.num; i++) {
if (draw->extra_shader_outputs.semantic_name[i] == semantic_name &&
draw->extra_shader_outputs.semantic_index[i] == semantic_index) {
return draw->extra_shader_outputs.slot[i];
}
/* XXX there may be more than one extra vertex attrib.
* For example, simulated gl_FragCoord and gl_PointCoord.
*/
if (draw->extra_shader_outputs.semantic_name == semantic_name &&
draw->extra_shader_outputs.semantic_index == semantic_index) {
return draw->extra_shader_outputs.slot;
}
return 0;
@@ -522,18 +470,16 @@ draw_find_shader_output(const struct draw_context *draw,
uint
draw_num_shader_outputs(const struct draw_context *draw)
{
uint count;
uint count = draw->vs.vertex_shader->info.num_outputs;
/* If a geometry shader is present, its outputs go to the
* driver, else the vertex shader's outputs.
*/
if (draw->gs.geometry_shader)
count = draw->gs.geometry_shader->info.num_outputs;
else
count = draw->vs.vertex_shader->info.num_outputs;
count += draw->extra_shader_outputs.num;
if (draw->extra_shader_outputs.slot > 0)
count++;
return count;
}
@@ -725,11 +671,6 @@ draw_set_samplers(struct draw_context *draw,
draw->samplers[i] = NULL;
draw->num_samplers = num;
#ifdef HAVE_LLVM
if (draw->llvm)
draw_llvm_set_sampler_state(draw);
#endif
}
void
@@ -737,9 +678,9 @@ draw_set_mapped_texture(struct draw_context *draw,
unsigned sampler_idx,
uint32_t width, uint32_t height, uint32_t depth,
uint32_t last_level,
uint32_t row_stride[PIPE_MAX_TEXTURE_LEVELS],
uint32_t img_stride[PIPE_MAX_TEXTURE_LEVELS],
const void *data[PIPE_MAX_TEXTURE_LEVELS])
uint32_t row_stride[DRAW_MAX_TEXTURE_LEVELS],
uint32_t img_stride[DRAW_MAX_TEXTURE_LEVELS],
const void *data[DRAW_MAX_TEXTURE_LEVELS])
{
#ifdef HAVE_LLVM
if(draw->llvm)

View File

@@ -46,17 +46,12 @@ struct draw_context;
struct draw_stage;
struct draw_vertex_shader;
struct draw_geometry_shader;
struct draw_fragment_shader;
struct tgsi_sampler;
struct gallivm_state;
#define DRAW_MAX_TEXTURE_LEVELS 13 /* 4K x 4K for now */
struct draw_context *draw_create( struct pipe_context *pipe );
struct draw_context *
draw_create_gallivm(struct pipe_context *pipe, struct gallivm_state *gallivm);
void draw_destroy( struct draw_context *draw );
void draw_flush(struct draw_context *draw);
@@ -124,9 +119,9 @@ draw_set_mapped_texture(struct draw_context *draw,
unsigned sampler_idx,
uint32_t width, uint32_t height, uint32_t depth,
uint32_t last_level,
uint32_t row_stride[PIPE_MAX_TEXTURE_LEVELS],
uint32_t img_stride[PIPE_MAX_TEXTURE_LEVELS],
const void *data[PIPE_MAX_TEXTURE_LEVELS]);
uint32_t row_stride[DRAW_MAX_TEXTURE_LEVELS],
uint32_t img_stride[DRAW_MAX_TEXTURE_LEVELS],
const void *data[DRAW_MAX_TEXTURE_LEVELS]);
/*
@@ -142,17 +137,6 @@ void draw_delete_vertex_shader(struct draw_context *draw,
struct draw_vertex_shader *dvs);
/*
* Fragment shader functions
*/
struct draw_fragment_shader *
draw_create_fragment_shader(struct draw_context *draw,
const struct pipe_shader_state *shader);
void draw_bind_fragment_shader(struct draw_context *draw,
struct draw_fragment_shader *dvs);
void draw_delete_fragment_shader(struct draw_context *draw,
struct draw_fragment_shader *dvs);
/*
* Geometry shader functions
*/

View File

@@ -1,73 +0,0 @@
/**************************************************************************
*
* Copyright 2010 VMware, Inc.
* All Rights Reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sub license, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice (including the
* next paragraph) shall be included in all copies or substantial portions
* of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT.
* IN NO EVENT SHALL VMWARE AND/OR ITS SUPPLIERS BE LIABLE FOR
* ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
**************************************************************************/
#include "pipe/p_shader_tokens.h"
#include "util/u_math.h"
#include "util/u_memory.h"
#include "util/u_prim.h"
#include "tgsi/tgsi_parse.h"
#include "draw_fs.h"
#include "draw_private.h"
#include "draw_context.h"
struct draw_fragment_shader *
draw_create_fragment_shader(struct draw_context *draw,
const struct pipe_shader_state *shader)
{
struct draw_fragment_shader *dfs;
dfs = CALLOC_STRUCT(draw_fragment_shader);
if (dfs) {
dfs->base = *shader;
tgsi_scan_shader(shader->tokens, &dfs->info);
}
return dfs;
}
void
draw_bind_fragment_shader(struct draw_context *draw,
struct draw_fragment_shader *dfs)
{
draw_do_flush(draw, DRAW_FLUSH_STATE_CHANGE);
draw->fs.fragment_shader = dfs;
}
void
draw_delete_fragment_shader(struct draw_context *draw,
struct draw_fragment_shader *dfs)
{
FREE(dfs);
}

View File

@@ -1,42 +0,0 @@
/**************************************************************************
*
* Copyright 2010 VMware, Inc.
* All Rights Reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sub license, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice (including the
* next paragraph) shall be included in all copies or substantial portions
* of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT.
* IN NO EVENT SHALL VMWARE AND/OR ITS SUPPLIERS BE LIABLE FOR
* ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
**************************************************************************/
#ifndef DRAW_FS_H
#define DRAW_FS_H
#include "tgsi/tgsi_scan.h"
struct draw_fragment_shader
{
struct pipe_shader_state base;
struct tgsi_shader_info info;
};
#endif /* DRAW_FS_H */

File diff suppressed because it is too large Load Diff

View File

@@ -41,6 +41,7 @@
#include <llvm-c/Target.h>
#include <llvm-c/ExecutionEngine.h>
#define DRAW_MAX_TEXTURE_LEVELS 13 /* 4K x 4K for now */
struct draw_llvm;
struct llvm_vertex_shader;
@@ -51,13 +52,9 @@ struct draw_jit_texture
uint32_t height;
uint32_t depth;
uint32_t last_level;
uint32_t row_stride[PIPE_MAX_TEXTURE_LEVELS];
uint32_t img_stride[PIPE_MAX_TEXTURE_LEVELS];
const void *data[PIPE_MAX_TEXTURE_LEVELS];
float min_lod;
float max_lod;
float lod_bias;
float border_color[4];
uint32_t row_stride[DRAW_MAX_TEXTURE_LEVELS];
uint32_t img_stride[DRAW_MAX_TEXTURE_LEVELS];
const void *data[DRAW_MAX_TEXTURE_LEVELS];
};
enum {
@@ -68,10 +65,6 @@ enum {
DRAW_JIT_TEXTURE_ROW_STRIDE,
DRAW_JIT_TEXTURE_IMG_STRIDE,
DRAW_JIT_TEXTURE_DATA,
DRAW_JIT_TEXTURE_MIN_LOD,
DRAW_JIT_TEXTURE_MAX_LOD,
DRAW_JIT_TEXTURE_LOD_BIAS,
DRAW_JIT_TEXTURE_BORDER_COLOR,
DRAW_JIT_TEXTURE_NUM_FIELDS /* number of fields above */
};
@@ -96,51 +89,46 @@ struct draw_jit_context
{
const float *vs_constants;
const float *gs_constants;
float (*planes) [12][4];
float *viewport;
struct draw_jit_texture textures[PIPE_MAX_VERTEX_SAMPLERS];
};
#define draw_jit_context_vs_constants(_gallivm, _ptr) \
lp_build_struct_get(_gallivm, _ptr, 0, "vs_constants")
#define draw_jit_context_vs_constants(_builder, _ptr) \
lp_build_struct_get(_builder, _ptr, 0, "vs_constants")
#define draw_jit_context_gs_constants(_gallivm, _ptr) \
lp_build_struct_get(_gallivm, _ptr, 1, "gs_constants")
#define draw_jit_context_gs_constants(_builder, _ptr) \
lp_build_struct_get(_builder, _ptr, 1, "gs_constants")
#define draw_jit_context_planes(_gallivm, _ptr) \
lp_build_struct_get(_gallivm, _ptr, 2, "planes")
#define DRAW_JIT_CTX_TEXTURES 2
#define draw_jit_context_viewport(_gallivm, _ptr) \
lp_build_struct_get(_gallivm, _ptr, 3, "viewport")
#define DRAW_JIT_CTX_TEXTURES 4
#define draw_jit_context_textures(_gallivm, _ptr) \
lp_build_struct_get_ptr(_gallivm, _ptr, DRAW_JIT_CTX_TEXTURES, "textures")
#define draw_jit_header_id(_gallivm, _ptr) \
lp_build_struct_get_ptr(_gallivm, _ptr, 0, "id")
#define draw_jit_header_clip(_gallivm, _ptr) \
lp_build_struct_get_ptr(_gallivm, _ptr, 1, "clip")
#define draw_jit_header_data(_gallivm, _ptr) \
lp_build_struct_get_ptr(_gallivm, _ptr, 2, "data")
#define draw_jit_context_textures(_builder, _ptr) \
lp_build_struct_get_ptr(_builder, _ptr, DRAW_JIT_CTX_TEXTURES, "textures")
#define draw_jit_vbuffer_stride(_gallivm, _ptr) \
lp_build_struct_get(_gallivm, _ptr, 0, "stride")
#define draw_jit_vbuffer_max_index(_gallivm, _ptr) \
lp_build_struct_get(_gallivm, _ptr, 1, "max_index")
#define draw_jit_header_id(_builder, _ptr) \
lp_build_struct_get_ptr(_builder, _ptr, 0, "id")
#define draw_jit_vbuffer_offset(_gallivm, _ptr) \
lp_build_struct_get(_gallivm, _ptr, 2, "buffer_offset")
#define draw_jit_header_clip(_builder, _ptr) \
lp_build_struct_get(_builder, _ptr, 1, "clip")
#define draw_jit_header_data(_builder, _ptr) \
lp_build_struct_get_ptr(_builder, _ptr, 2, "data")
typedef int
#define draw_jit_vbuffer_stride(_builder, _ptr) \
lp_build_struct_get(_builder, _ptr, 0, "stride")
#define draw_jit_vbuffer_max_index(_builder, _ptr) \
lp_build_struct_get(_builder, _ptr, 1, "max_index")
#define draw_jit_vbuffer_offset(_builder, _ptr) \
lp_build_struct_get(_builder, _ptr, 2, "buffer_offset")
typedef void
(*draw_jit_vert_func)(struct draw_jit_context *context,
struct vertex_header *io,
const char *vbuffers[PIPE_MAX_ATTRIBS],
@@ -151,7 +139,7 @@ typedef int
unsigned instance_id);
typedef int
typedef void
(*draw_jit_vert_func_elts)(struct draw_jit_context *context,
struct vertex_header *io,
const char *vbuffers[PIPE_MAX_ATTRIBS],
@@ -163,16 +151,8 @@ typedef int
struct draw_llvm_variant_key
{
unsigned nr_vertex_elements:8;
unsigned nr_samplers:8;
unsigned clip_xy:1;
unsigned clip_z:1;
unsigned clip_user:1;
unsigned clip_halfz:1;
unsigned bypass_viewport:1;
unsigned need_edgeflags:1;
unsigned nr_planes:4;
unsigned pad:6;
unsigned nr_vertex_elements:16;
unsigned nr_samplers:16;
/* Variable number of vertex elements:
*/
@@ -246,19 +226,21 @@ struct draw_llvm {
struct draw_jit_context jit_context;
struct gallivm_state *gallivm;
struct draw_llvm_variant_list_item vs_variants_list;
int nr_variants;
/* LLVM JIT builder types */
LLVMModuleRef module;
LLVMExecutionEngineRef engine;
LLVMModuleProviderRef provider;
LLVMTargetDataRef target;
LLVMPassManagerRef pass;
LLVMTypeRef context_ptr_type;
LLVMTypeRef vertex_header_ptr_type;
LLVMTypeRef buffer_ptr_type;
LLVMTypeRef vb_ptr_type;
LLVMTypeRef vertex_header_ptr_type;
};
static INLINE struct llvm_vertex_shader *
llvm_vertex_shader(struct draw_vertex_shader *vs)
{
@@ -267,7 +249,7 @@ llvm_vertex_shader(struct draw_vertex_shader *vs)
struct draw_llvm *
draw_llvm_create(struct draw_context *draw, struct gallivm_state *gallivm);
draw_llvm_create(struct draw_context *draw);
void
draw_llvm_destroy(struct draw_llvm *llvm);
@@ -284,7 +266,7 @@ struct draw_llvm_variant_key *
draw_llvm_make_variant_key(struct draw_llvm *llvm, char *store);
LLVMValueRef
draw_llvm_translate_from(struct gallivm_state *gallivm,
draw_llvm_translate_from(LLVMBuilderRef builder,
LLVMValueRef vbuffer,
enum pipe_format from_format);
@@ -292,16 +274,13 @@ struct lp_build_sampler_soa *
draw_llvm_sampler_soa_create(const struct lp_sampler_static_state *static_state,
LLVMValueRef context_ptr);
void
draw_llvm_set_sampler_state(struct draw_context *draw);
void
draw_llvm_set_mapped_texture(struct draw_context *draw,
unsigned sampler_idx,
uint32_t width, uint32_t height, uint32_t depth,
uint32_t last_level,
uint32_t row_stride[PIPE_MAX_TEXTURE_LEVELS],
uint32_t img_stride[PIPE_MAX_TEXTURE_LEVELS],
const void *data[PIPE_MAX_TEXTURE_LEVELS]);
uint32_t row_stride[DRAW_MAX_TEXTURE_LEVELS],
uint32_t img_stride[DRAW_MAX_TEXTURE_LEVELS],
const void *data[DRAW_MAX_TEXTURE_LEVELS]);
#endif

View File

@@ -32,7 +32,6 @@
#include "pipe/p_defines.h"
#include "pipe/p_shader_tokens.h"
#include "gallivm/lp_bld_const.h"
#include "gallivm/lp_bld_debug.h"
#include "gallivm/lp_bld_type.h"
#include "gallivm/lp_bld_sample.h"
@@ -85,13 +84,12 @@ struct draw_llvm_sampler_soa
*/
static LLVMValueRef
draw_llvm_texture_member(const struct lp_sampler_dynamic_state *base,
struct gallivm_state *gallivm,
LLVMBuilderRef builder,
unsigned unit,
unsigned member_index,
const char *member_name,
boolean emit_load)
{
LLVMBuilderRef builder = gallivm->builder;
struct draw_llvm_sampler_dynamic_state *state =
(struct draw_llvm_sampler_dynamic_state *)base;
LLVMValueRef indices[4];
@@ -101,13 +99,13 @@ draw_llvm_texture_member(const struct lp_sampler_dynamic_state *base,
debug_assert(unit < PIPE_MAX_VERTEX_SAMPLERS);
/* context[0] */
indices[0] = lp_build_const_int32(gallivm, 0);
indices[0] = LLVMConstInt(LLVMInt32Type(), 0, 0);
/* context[0].textures */
indices[1] = lp_build_const_int32(gallivm, DRAW_JIT_CTX_TEXTURES);
indices[1] = LLVMConstInt(LLVMInt32Type(), DRAW_JIT_CTX_TEXTURES, 0);
/* context[0].textures[unit] */
indices[2] = lp_build_const_int32(gallivm, unit);
indices[2] = LLVMConstInt(LLVMInt32Type(), unit, 0);
/* context[0].textures[unit].member */
indices[3] = lp_build_const_int32(gallivm, member_index);
indices[3] = LLVMConstInt(LLVMInt32Type(), member_index, 0);
ptr = LLVMBuildGEP(builder, state->context_ptr, indices, Elements(indices), "");
@@ -134,10 +132,10 @@ draw_llvm_texture_member(const struct lp_sampler_dynamic_state *base,
#define DRAW_LLVM_TEXTURE_MEMBER(_name, _index, _emit_load) \
static LLVMValueRef \
draw_llvm_texture_##_name( const struct lp_sampler_dynamic_state *base, \
struct gallivm_state *gallivm, \
LLVMBuilderRef builder, \
unsigned unit) \
{ \
return draw_llvm_texture_member(base, gallivm, unit, _index, #_name, _emit_load ); \
return draw_llvm_texture_member(base, builder, unit, _index, #_name, _emit_load ); \
}
@@ -148,10 +146,6 @@ DRAW_LLVM_TEXTURE_MEMBER(last_level, DRAW_JIT_TEXTURE_LAST_LEVEL, TRUE)
DRAW_LLVM_TEXTURE_MEMBER(row_stride, DRAW_JIT_TEXTURE_ROW_STRIDE, FALSE)
DRAW_LLVM_TEXTURE_MEMBER(img_stride, DRAW_JIT_TEXTURE_IMG_STRIDE, FALSE)
DRAW_LLVM_TEXTURE_MEMBER(data_ptr, DRAW_JIT_TEXTURE_DATA, FALSE)
DRAW_LLVM_TEXTURE_MEMBER(min_lod, DRAW_JIT_TEXTURE_MIN_LOD, TRUE)
DRAW_LLVM_TEXTURE_MEMBER(max_lod, DRAW_JIT_TEXTURE_MAX_LOD, TRUE)
DRAW_LLVM_TEXTURE_MEMBER(lod_bias, DRAW_JIT_TEXTURE_LOD_BIAS, TRUE)
DRAW_LLVM_TEXTURE_MEMBER(border_color, DRAW_JIT_TEXTURE_BORDER_COLOR, FALSE)
static void
@@ -167,7 +161,7 @@ draw_llvm_sampler_soa_destroy(struct lp_build_sampler_soa *sampler)
*/
static void
draw_llvm_sampler_soa_emit_fetch_texel(const struct lp_build_sampler_soa *base,
struct gallivm_state *gallivm,
LLVMBuilderRef builder,
struct lp_type type,
unsigned unit,
unsigned num_coords,
@@ -182,7 +176,7 @@ draw_llvm_sampler_soa_emit_fetch_texel(const struct lp_build_sampler_soa *base,
assert(unit < PIPE_MAX_VERTEX_SAMPLERS);
lp_build_sample_soa(gallivm,
lp_build_sample_soa(builder,
&sampler->dynamic_state.static_state[unit],
&sampler->dynamic_state.base,
type,
@@ -213,10 +207,6 @@ draw_llvm_sampler_soa_create(const struct lp_sampler_static_state *static_state,
sampler->dynamic_state.base.row_stride = draw_llvm_texture_row_stride;
sampler->dynamic_state.base.img_stride = draw_llvm_texture_img_stride;
sampler->dynamic_state.base.data_ptr = draw_llvm_texture_data_ptr;
sampler->dynamic_state.base.min_lod = draw_llvm_texture_min_lod;
sampler->dynamic_state.base.max_lod = draw_llvm_texture_max_lod;
sampler->dynamic_state.base.lod_bias = draw_llvm_texture_lod_bias;
sampler->dynamic_state.base.border_color = draw_llvm_texture_border_color;
sampler->dynamic_state.static_state = static_state;
sampler->dynamic_state.context_ptr = context_ptr;

View File

@@ -3,7 +3,7 @@
#include "draw_llvm.h"
#include "gallivm/lp_bld_const.h"
#include "gallivm/lp_bld_arit.h"
#include "gallivm/lp_bld_struct.h"
#include "gallivm/lp_bld_format.h"
#include "gallivm/lp_bld_debug.h"
@@ -17,279 +17,272 @@
#define DRAW_DBG 0
static LLVMValueRef
from_64_float(struct gallivm_state *gallivm, LLVMValueRef val)
from_64_float(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef bc = LLVMBuildBitCast(gallivm->builder, val,
LLVMPointerType(LLVMDoubleTypeInContext(gallivm->context), 0) , "");
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, bc, "");
return LLVMBuildFPTrunc(gallivm->builder, l, LLVMFloatTypeInContext(gallivm->context), "");
LLVMValueRef bc = LLVMBuildBitCast(builder, val,
LLVMPointerType(LLVMDoubleType(), 0) , "");
LLVMValueRef l = LLVMBuildLoad(builder, bc, "");
return LLVMBuildFPTrunc(builder, l, LLVMFloatType(), "");
}
static LLVMValueRef
from_32_float(struct gallivm_state *gallivm, LLVMValueRef val)
from_32_float(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef bc = LLVMBuildBitCast(gallivm->builder, val,
LLVMPointerType(LLVMFloatTypeInContext(gallivm->context), 0) , "");
return LLVMBuildLoad(gallivm->builder, bc, "");
LLVMValueRef bc = LLVMBuildBitCast(builder, val,
LLVMPointerType(LLVMFloatType(), 0) , "");
return LLVMBuildLoad(builder, bc, "");
}
static INLINE LLVMValueRef
from_8_uscaled(struct gallivm_state *gallivm, LLVMValueRef val)
from_8_uscaled(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, val, "");
return LLVMBuildUIToFP(gallivm->builder, l, LLVMFloatTypeInContext(gallivm->context), "");
LLVMValueRef l = LLVMBuildLoad(builder, val, "");
return LLVMBuildUIToFP(builder, l, LLVMFloatType(), "");
}
static INLINE LLVMValueRef
from_16_uscaled(struct gallivm_state *gallivm, LLVMValueRef val)
from_16_uscaled(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef bc = LLVMBuildBitCast(gallivm->builder, val,
LLVMPointerType(LLVMIntTypeInContext(gallivm->context, 16), 0) , "");
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, bc, "");
return LLVMBuildUIToFP(gallivm->builder, l, LLVMFloatTypeInContext(gallivm->context), "");
LLVMValueRef bc = LLVMBuildBitCast(builder, val,
LLVMPointerType(LLVMIntType(16), 0) , "");
LLVMValueRef l = LLVMBuildLoad(builder, bc, "");
return LLVMBuildUIToFP(builder, l, LLVMFloatType(), "");
}
static INLINE LLVMValueRef
from_32_uscaled(struct gallivm_state *gallivm, LLVMValueRef val)
from_32_uscaled(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef bc = LLVMBuildBitCast(gallivm->builder, val,
LLVMPointerType(LLVMIntTypeInContext(gallivm->context, 32), 0) , "");
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, bc, "");
return LLVMBuildUIToFP(gallivm->builder, l, LLVMFloatTypeInContext(gallivm->context), "");
LLVMValueRef bc = LLVMBuildBitCast(builder, val,
LLVMPointerType(LLVMIntType(32), 0) , "");
LLVMValueRef l = LLVMBuildLoad(builder, bc, "");
return LLVMBuildUIToFP(builder, l, LLVMFloatType(), "");
}
static INLINE LLVMValueRef
from_8_sscaled(struct gallivm_state *gallivm, LLVMValueRef val)
from_8_sscaled(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, val, "");
return LLVMBuildSIToFP(gallivm->builder, l, LLVMFloatTypeInContext(gallivm->context), "");
LLVMValueRef l = LLVMBuildLoad(builder, val, "");
return LLVMBuildSIToFP(builder, l, LLVMFloatType(), "");
}
static INLINE LLVMValueRef
from_16_sscaled(struct gallivm_state *gallivm, LLVMValueRef val)
from_16_sscaled(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef bc = LLVMBuildBitCast(gallivm->builder, val,
LLVMPointerType(LLVMIntTypeInContext(gallivm->context, 16), 0) , "");
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, bc, "");
return LLVMBuildSIToFP(gallivm->builder, l, LLVMFloatTypeInContext(gallivm->context), "");
LLVMValueRef bc = LLVMBuildBitCast(builder, val,
LLVMPointerType(LLVMIntType(16), 0) , "");
LLVMValueRef l = LLVMBuildLoad(builder, bc, "");
return LLVMBuildSIToFP(builder, l, LLVMFloatType(), "");
}
static INLINE LLVMValueRef
from_32_sscaled(struct gallivm_state *gallivm, LLVMValueRef val)
from_32_sscaled(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef bc = LLVMBuildBitCast(gallivm->builder, val,
LLVMPointerType(LLVMIntTypeInContext(gallivm->context, 32), 0) , "");
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, bc, "");
return LLVMBuildSIToFP(gallivm->builder, l, LLVMFloatTypeInContext(gallivm->context), "");
LLVMValueRef bc = LLVMBuildBitCast(builder, val,
LLVMPointerType(LLVMIntType(32), 0) , "");
LLVMValueRef l = LLVMBuildLoad(builder, bc, "");
return LLVMBuildSIToFP(builder, l, LLVMFloatType(), "");
}
static INLINE LLVMValueRef
from_8_unorm(struct gallivm_state *gallivm, LLVMValueRef val)
from_8_unorm(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, val, "");
LLVMValueRef uscaled = LLVMBuildUIToFP(gallivm->builder, l, LLVMFloatTypeInContext(gallivm->context), "");
return LLVMBuildFDiv(gallivm->builder, uscaled,
lp_build_const_float(gallivm, 255.), "");
LLVMValueRef l = LLVMBuildLoad(builder, val, "");
LLVMValueRef uscaled = LLVMBuildUIToFP(builder, l, LLVMFloatType(), "");
return LLVMBuildFDiv(builder, uscaled,
LLVMConstReal(LLVMFloatType(), 255.), "");
}
static INLINE LLVMValueRef
from_16_unorm(struct gallivm_state *gallivm, LLVMValueRef val)
from_16_unorm(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef bc = LLVMBuildBitCast(gallivm->builder, val,
LLVMPointerType(LLVMIntTypeInContext(gallivm->context, 16), 0) , "");
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, bc, "");
LLVMValueRef uscaled = LLVMBuildUIToFP(gallivm->builder, l, LLVMFloatTypeInContext(gallivm->context), "");
return LLVMBuildFDiv(gallivm->builder, uscaled,
lp_build_const_float(gallivm, 65535.), "");
LLVMValueRef bc = LLVMBuildBitCast(builder, val,
LLVMPointerType(LLVMIntType(16), 0) , "");
LLVMValueRef l = LLVMBuildLoad(builder, bc, "");
LLVMValueRef uscaled = LLVMBuildUIToFP(builder, l, LLVMFloatType(), "");
return LLVMBuildFDiv(builder, uscaled,
LLVMConstReal(LLVMFloatType(), 65535.), "");
}
static INLINE LLVMValueRef
from_32_unorm(struct gallivm_state *gallivm, LLVMValueRef val)
from_32_unorm(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef bc = LLVMBuildBitCast(gallivm->builder, val,
LLVMPointerType(LLVMIntTypeInContext(gallivm->context, 32), 0) , "");
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, bc, "");
LLVMValueRef uscaled = LLVMBuildUIToFP(gallivm->builder, l, LLVMFloatTypeInContext(gallivm->context), "");
LLVMValueRef bc = LLVMBuildBitCast(builder, val,
LLVMPointerType(LLVMIntType(32), 0) , "");
LLVMValueRef l = LLVMBuildLoad(builder, bc, "");
LLVMValueRef uscaled = LLVMBuildUIToFP(builder, l, LLVMFloatType(), "");
return LLVMBuildFDiv(gallivm->builder, uscaled,
lp_build_const_float(gallivm, 4294967295.), "");
return LLVMBuildFDiv(builder, uscaled,
LLVMConstReal(LLVMFloatType(), 4294967295.), "");
}
static INLINE LLVMValueRef
from_8_snorm(struct gallivm_state *gallivm, LLVMValueRef val)
from_8_snorm(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, val, "");
LLVMValueRef uscaled = LLVMBuildSIToFP(gallivm->builder, l, LLVMFloatTypeInContext(gallivm->context), "");
return LLVMBuildFDiv(gallivm->builder, uscaled,
lp_build_const_float(gallivm, 127.0), "");
LLVMValueRef l = LLVMBuildLoad(builder, val, "");
LLVMValueRef uscaled = LLVMBuildSIToFP(builder, l, LLVMFloatType(), "");
return LLVMBuildFDiv(builder, uscaled,
LLVMConstReal(LLVMFloatType(), 127.0), "");
}
static INLINE LLVMValueRef
from_16_snorm(struct gallivm_state *gallivm, LLVMValueRef val)
from_16_snorm(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef bc = LLVMBuildBitCast(gallivm->builder, val,
LLVMPointerType(LLVMIntTypeInContext(gallivm->context, 16), 0) , "");
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, bc, "");
LLVMValueRef uscaled = LLVMBuildSIToFP(gallivm->builder, l, LLVMFloatTypeInContext(gallivm->context), "");
return LLVMBuildFDiv(gallivm->builder, uscaled,
lp_build_const_float(gallivm, 32767.0f), "");
LLVMValueRef bc = LLVMBuildBitCast(builder, val,
LLVMPointerType(LLVMIntType(16), 0) , "");
LLVMValueRef l = LLVMBuildLoad(builder, bc, "");
LLVMValueRef uscaled = LLVMBuildSIToFP(builder, l, LLVMFloatType(), "");
return LLVMBuildFDiv(builder, uscaled,
LLVMConstReal(LLVMFloatType(), 32767.0f), "");
}
static INLINE LLVMValueRef
from_32_snorm(struct gallivm_state *gallivm, LLVMValueRef val)
from_32_snorm(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef bc = LLVMBuildBitCast(gallivm->builder, val,
LLVMPointerType(LLVMIntTypeInContext(gallivm->context, 32), 0) , "");
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, bc, "");
LLVMValueRef uscaled = LLVMBuildSIToFP(gallivm->builder, l, LLVMFloatTypeInContext(gallivm->context), "");
LLVMValueRef bc = LLVMBuildBitCast(builder, val,
LLVMPointerType(LLVMIntType(32), 0) , "");
LLVMValueRef l = LLVMBuildLoad(builder, bc, "");
LLVMValueRef uscaled = LLVMBuildSIToFP(builder, l, LLVMFloatType(), "");
return LLVMBuildFDiv(gallivm->builder, uscaled,
lp_build_const_float(gallivm, 2147483647.0), "");
return LLVMBuildFDiv(builder, uscaled,
LLVMConstReal(LLVMFloatType(), 2147483647.0), "");
}
static INLINE LLVMValueRef
from_32_fixed(struct gallivm_state *gallivm, LLVMValueRef val)
from_32_fixed(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef bc = LLVMBuildBitCast(gallivm->builder, val,
LLVMPointerType(LLVMIntTypeInContext(gallivm->context, 32), 0) , "");
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, bc, "");
LLVMValueRef uscaled = LLVMBuildSIToFP(gallivm->builder, l, LLVMFloatTypeInContext(gallivm->context), "");
LLVMValueRef bc = LLVMBuildBitCast(builder, val,
LLVMPointerType(LLVMIntType(32), 0) , "");
LLVMValueRef l = LLVMBuildLoad(builder, bc, "");
LLVMValueRef uscaled = LLVMBuildSIToFP(builder, l, LLVMFloatType(), "");
return LLVMBuildFDiv(gallivm->builder, uscaled,
lp_build_const_float(gallivm, 65536.0), "");
return LLVMBuildFDiv(builder, uscaled,
LLVMConstReal(LLVMFloatType(), 65536.0), "");
}
static LLVMValueRef
to_64_float(struct gallivm_state *gallivm, LLVMValueRef fp)
to_64_float(LLVMBuilderRef builder, LLVMValueRef fp)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, fp, "");
return LLVMBuildFPExt(gallivm->builder, l, LLVMDoubleTypeInContext(gallivm->context), "");
LLVMValueRef l = LLVMBuildLoad(builder, fp, "");
return LLVMBuildFPExt(builder, l, LLVMDoubleType(), "");
}
static LLVMValueRef
to_32_float(struct gallivm_state *gallivm, LLVMValueRef fp)
to_32_float(LLVMBuilderRef builder, LLVMValueRef fp)
{
return LLVMBuildLoad(gallivm->builder, fp, "");
return LLVMBuildLoad(builder, fp, "");
}
static INLINE LLVMValueRef
to_8_uscaled(struct gallivm_state *gallivm, LLVMValueRef fp)
to_8_uscaled(LLVMBuilderRef builder, LLVMValueRef fp)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, fp, "");
return LLVMBuildFPToUI(gallivm->builder, l, LLVMIntTypeInContext(gallivm->context, 8), "");
LLVMValueRef l = LLVMBuildLoad(builder, fp, "");
return LLVMBuildFPToUI(builder, l, LLVMIntType(8), "");
}
static INLINE LLVMValueRef
to_16_uscaled(struct gallivm_state *gallivm, LLVMValueRef fp)
to_16_uscaled(LLVMBuilderRef builder, LLVMValueRef fp)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, fp, "");
return LLVMBuildFPToUI(gallivm->builder, l, LLVMIntTypeInContext(gallivm->context, 16), "");
LLVMValueRef l = LLVMBuildLoad(builder, fp, "");
return LLVMBuildFPToUI(builder, l, LLVMIntType(16), "");
}
static INLINE LLVMValueRef
to_32_uscaled(struct gallivm_state *gallivm, LLVMValueRef fp)
to_32_uscaled(LLVMBuilderRef builder, LLVMValueRef fp)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, fp, "");
return LLVMBuildFPToUI(gallivm->builder, l, LLVMIntTypeInContext(gallivm->context, 32), "");
LLVMValueRef l = LLVMBuildLoad(builder, fp, "");
return LLVMBuildFPToUI(builder, l, LLVMIntType(32), "");
}
static INLINE LLVMValueRef
to_8_sscaled(struct gallivm_state *gallivm, LLVMValueRef fp)
to_8_sscaled(LLVMBuilderRef builder, LLVMValueRef fp)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, fp, "");
return LLVMBuildFPToSI(gallivm->builder, l, LLVMIntTypeInContext(gallivm->context, 8), "");
LLVMValueRef l = LLVMBuildLoad(builder, fp, "");
return LLVMBuildFPToSI(builder, l, LLVMIntType(8), "");
}
static INLINE LLVMValueRef
to_16_sscaled(struct gallivm_state *gallivm, LLVMValueRef fp)
to_16_sscaled(LLVMBuilderRef builder, LLVMValueRef fp)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, fp, "");
return LLVMBuildFPToSI(gallivm->builder, l, LLVMIntTypeInContext(gallivm->context, 16), "");
LLVMValueRef l = LLVMBuildLoad(builder, fp, "");
return LLVMBuildFPToSI(builder, l, LLVMIntType(16), "");
}
static INLINE LLVMValueRef
to_32_sscaled(struct gallivm_state *gallivm, LLVMValueRef fp)
to_32_sscaled(LLVMBuilderRef builder, LLVMValueRef fp)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, fp, "");
return LLVMBuildFPToSI(gallivm->builder, l, LLVMIntTypeInContext(gallivm->context, 32), "");
LLVMValueRef l = LLVMBuildLoad(builder, fp, "");
return LLVMBuildFPToSI(builder, l, LLVMIntType(32), "");
}
static INLINE LLVMValueRef
to_8_unorm(struct gallivm_state *gallivm, LLVMValueRef fp)
to_8_unorm(LLVMBuilderRef builder, LLVMValueRef fp)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, fp, "");
LLVMValueRef uscaled = LLVMBuildFPToUI(gallivm->builder, l,
LLVMIntTypeInContext(gallivm->context, 8), "");
return LLVMBuildFMul(gallivm->builder, uscaled,
lp_build_const_float(gallivm, 255.), "");
LLVMValueRef l = LLVMBuildLoad(builder, fp, "");
LLVMValueRef uscaled = LLVMBuildFPToUI(builder, l, LLVMIntType(8), "");
return LLVMBuildFMul(builder, uscaled,
LLVMConstReal(LLVMFloatType(), 255.), "");
}
static INLINE LLVMValueRef
to_16_unorm(struct gallivm_state *gallivm, LLVMValueRef fp)
to_16_unorm(LLVMBuilderRef builder, LLVMValueRef fp)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, fp, "");
LLVMValueRef uscaled = LLVMBuildFPToUI(gallivm->builder, l,
LLVMIntTypeInContext(gallivm->context, 32), "");
return LLVMBuildFMul(gallivm->builder, uscaled,
lp_build_const_float(gallivm, 65535.), "");
LLVMValueRef l = LLVMBuildLoad(builder, fp, "");
LLVMValueRef uscaled = LLVMBuildFPToUI(builder, l, LLVMIntType(32), "");
return LLVMBuildFMul(builder, uscaled,
LLVMConstReal(LLVMFloatType(), 65535.), "");
}
static INLINE LLVMValueRef
to_32_unorm(struct gallivm_state *gallivm, LLVMValueRef fp)
to_32_unorm(LLVMBuilderRef builder, LLVMValueRef fp)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, fp, "");
LLVMValueRef uscaled = LLVMBuildFPToUI(gallivm->builder, l,
LLVMIntTypeInContext(gallivm->context, 32), "");
LLVMValueRef l = LLVMBuildLoad(builder, fp, "");
LLVMValueRef uscaled = LLVMBuildFPToUI(builder, l, LLVMIntType(32), "");
return LLVMBuildFMul(gallivm->builder, uscaled,
lp_build_const_float(gallivm, 4294967295.), "");
return LLVMBuildFMul(builder, uscaled,
LLVMConstReal(LLVMFloatType(), 4294967295.), "");
}
static INLINE LLVMValueRef
to_8_snorm(struct gallivm_state *gallivm, LLVMValueRef val)
to_8_snorm(LLVMBuilderRef builder, LLVMValueRef val)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, val, "");
LLVMValueRef uscaled = LLVMBuildFPToSI(gallivm->builder, l,
LLVMIntTypeInContext(gallivm->context, 8), "");
return LLVMBuildFMul(gallivm->builder, uscaled,
lp_build_const_float(gallivm, 127.0), "");
LLVMValueRef l = LLVMBuildLoad(builder, val, "");
LLVMValueRef uscaled = LLVMBuildFPToSI(builder, l, LLVMIntType(8), "");
return LLVMBuildFMul(builder, uscaled,
LLVMConstReal(LLVMFloatType(), 127.0), "");
}
static INLINE LLVMValueRef
to_16_snorm(struct gallivm_state *gallivm, LLVMValueRef fp)
to_16_snorm(LLVMBuilderRef builder, LLVMValueRef fp)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, fp, "");
LLVMValueRef uscaled = LLVMBuildFPToSI(gallivm->builder, l,
LLVMIntTypeInContext(gallivm->context, 16), "");
return LLVMBuildFMul(gallivm->builder, uscaled,
lp_build_const_float(gallivm, 32767.0f), "");
LLVMValueRef l = LLVMBuildLoad(builder, fp, "");
LLVMValueRef uscaled = LLVMBuildFPToSI(builder, l, LLVMIntType(16), "");
return LLVMBuildFMul(builder, uscaled,
LLVMConstReal(LLVMFloatType(), 32767.0f), "");
}
static INLINE LLVMValueRef
to_32_snorm(struct gallivm_state *gallivm, LLVMValueRef fp)
to_32_snorm(LLVMBuilderRef builder, LLVMValueRef fp)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, fp, "");
LLVMValueRef uscaled = LLVMBuildFPToSI(gallivm->builder, l,
LLVMIntTypeInContext(gallivm->context, 32), "");
LLVMValueRef l = LLVMBuildLoad(builder, fp, "");
LLVMValueRef uscaled = LLVMBuildFPToSI(builder, l, LLVMIntType(32), "");
return LLVMBuildFMul(gallivm->builder, uscaled,
lp_build_const_float(gallivm, 2147483647.0), "");
return LLVMBuildFMul(builder, uscaled,
LLVMConstReal(LLVMFloatType(), 2147483647.0), "");
}
static INLINE LLVMValueRef
to_32_fixed(struct gallivm_state *gallivm, LLVMValueRef fp)
to_32_fixed(LLVMBuilderRef builder, LLVMValueRef fp)
{
LLVMValueRef l = LLVMBuildLoad(gallivm->builder, fp, "");
LLVMValueRef uscaled = LLVMBuildFPToSI(gallivm->builder, l,
LLVMIntTypeInContext(gallivm->context, 32), "");
LLVMValueRef l = LLVMBuildLoad(builder, fp, "");
LLVMValueRef uscaled = LLVMBuildFPToSI(builder, l, LLVMIntType(32), "");
return LLVMBuildFMul(gallivm->builder, uscaled,
lp_build_const_float(gallivm, 65536.0), "");
return LLVMBuildFMul(builder, uscaled,
LLVMConstReal(LLVMFloatType(), 65536.0), "");
}
typedef LLVMValueRef (*from_func)(struct gallivm_state *, LLVMValueRef);
typedef LLVMValueRef (*to_func)(struct gallivm_state *, LLVMValueRef);
typedef LLVMValueRef (*from_func)(LLVMBuilderRef, LLVMValueRef);
typedef LLVMValueRef (*to_func)(LLVMBuilderRef, LLVMValueRef);
/* so that underneath can avoid function calls which are prohibited
* for static initialization we need this conversion */
@@ -302,21 +295,21 @@ enum ll_type {
};
static INLINE LLVMTypeRef
ll_type_to_llvm(struct gallivm_state *gallivm, enum ll_type type)
ll_type_to_llvm(enum ll_type type)
{
switch (type) {
case LL_Double:
return LLVMDoubleTypeInContext(gallivm->context);
return LLVMDoubleType();
case LL_Float:
return LLVMFloatTypeInContext(gallivm->context);
return LLVMFloatType();
case LL_Int32:
return LLVMInt32TypeInContext(gallivm->context);
return LLVMInt32Type();
case LL_Int16:
return LLVMIntTypeInContext(gallivm->context, 16);
return LLVMIntType(16);
case LL_Int8:
return LLVMIntTypeInContext(gallivm->context, 8);
return LLVMIntType(8);
}
return LLVMIntTypeInContext(gallivm->context, 8);
return LLVMIntType(8);
}
static INLINE int
@@ -422,42 +415,42 @@ struct draw_llvm_translate {
static LLVMValueRef
fetch(struct gallivm_state *gallivm,
fetch(LLVMBuilderRef builder,
LLVMValueRef ptr, int val_size, int nr_components,
from_func func)
{
int i;
int offset = 0;
LLVMValueRef res =
LLVMConstNull(LLVMVectorType(LLVMFloatTypeInContext(gallivm->context), 4));
LLVMValueRef res = LLVMConstNull(
LLVMVectorType(LLVMFloatType(), 4));
LLVMValueRef defaults[4];
defaults[0] =
defaults[1] =
defaults[2] = lp_build_const_float(gallivm, 0.0);
defaults[3] = lp_build_const_float(gallivm, 1.0);
defaults[0] = LLVMConstReal(LLVMFloatType(), 0);
defaults[1] = LLVMConstReal(LLVMFloatType(), 0);
defaults[2] = LLVMConstReal(LLVMFloatType(), 0);
defaults[3] = LLVMConstReal(LLVMFloatType(), 1);
for (i = 0; i < nr_components; ++i) {
LLVMValueRef src_index = lp_build_const_int32(gallivm, offset);
LLVMValueRef dst_index = lp_build_const_int32(gallivm, i);
LLVMValueRef src_index = LLVMConstInt(LLVMInt32Type(), offset, 0);
LLVMValueRef dst_index = LLVMConstInt(LLVMInt32Type(), i, 0);
LLVMValueRef src_tmp;
LLVMValueRef component;
src_tmp = LLVMBuildGEP(gallivm->builder, ptr, &src_index, 1, "src_tmp");
src_tmp = LLVMBuildGEP(builder, ptr, &src_index, 1, "src_tmp");
/* convert src_tmp to float */
component = func(gallivm, src_tmp);
component = func(builder, src_tmp);
/* vec.comp = component */
res = LLVMBuildInsertElement(gallivm->builder,
res = LLVMBuildInsertElement(builder,
res,
component,
dst_index, "");
offset += val_size;
}
for (; i < 4; ++i) {
LLVMValueRef dst_index = lp_build_const_int32(gallivm, i);
res = LLVMBuildInsertElement(gallivm->builder,
LLVMValueRef dst_index = LLVMConstInt(LLVMInt32Type(), i, 0);
res = LLVMBuildInsertElement(builder,
res,
defaults[i],
dst_index, "");
@@ -467,7 +460,7 @@ fetch(struct gallivm_state *gallivm,
LLVMValueRef
draw_llvm_translate_from(struct gallivm_state *gallivm,
draw_llvm_translate_from(LLVMBuilderRef builder,
LLVMValueRef vbuffer,
enum pipe_format from_format)
{
@@ -484,7 +477,7 @@ draw_llvm_translate_from(struct gallivm_state *gallivm,
for (i = 0; i < Elements(translates); ++i) {
if (translates[i].format == from_format) {
/*LLVMTypeRef type = ll_type_to_llvm(translates[i].type);*/
return fetch(gallivm,
return fetch(builder,
vbuffer,
ll_type_size(translates[i].type),
translates[i].num_components,
@@ -501,6 +494,6 @@ draw_llvm_translate_from(struct gallivm_state *gallivm,
*/
format_desc = util_format_description(from_format);
zero = LLVMConstNull(LLVMInt32TypeInContext(gallivm->context));
return lp_build_fetch_rgba_aos(gallivm, format_desc, type, vbuffer, zero, zero, zero);
zero = LLVMConstNull(LLVMInt32Type());
return lp_build_fetch_rgba_aos(builder, format_desc, type, vbuffer, zero, zero, zero);
}

View File

@@ -406,7 +406,6 @@ aaline_create_texture(struct aaline_stage *aaline)
texTemp.width0 = 1 << MAX_TEXTURE_LEVEL;
texTemp.height0 = 1 << MAX_TEXTURE_LEVEL;
texTemp.depth0 = 1;
texTemp.array_size = 1;
texTemp.bind = PIPE_BIND_SAMPLER_VIEW;
aaline->texture = screen->resource_create(screen, &texTemp);
@@ -442,10 +441,10 @@ aaline_create_texture(struct aaline_stage *aaline)
/* This texture is new, no need to flush.
*/
transfer = pipe->get_transfer(pipe,
aaline->texture,
level,
PIPE_TRANSFER_WRITE,
&box);
aaline->texture,
u_subresource(0, level),
PIPE_TRANSFER_WRITE,
&box);
data = pipe->transfer_map(pipe, transfer);
if (data == NULL)
@@ -689,9 +688,10 @@ aaline_first_line(struct draw_stage *stage, struct prim_header *header)
aaline->tex_slot = draw_current_shader_outputs(draw);
aaline->pos_slot = draw_current_shader_position_output(draw);;
/* allocate the extra post-transformed vertex attribute */
(void) draw_alloc_extra_vertex_attrib(draw, TGSI_SEMANTIC_GENERIC,
aaline->fs->generic_attrib);
/* advertise the extra post-transformed vertex attribute */
draw->extra_shader_outputs.semantic_name = TGSI_SEMANTIC_GENERIC;
draw->extra_shader_outputs.semantic_index = aaline->fs->generic_attrib;
draw->extra_shader_outputs.slot = aaline->tex_slot;
/* how many samplers? */
/* we'll use sampler/texture[pstip->sampler_unit] for the stipple */
@@ -744,7 +744,7 @@ aaline_flush(struct draw_stage *stage, unsigned flags)
draw->suspend_flushing = FALSE;
draw_remove_extra_vertex_attribs(draw);
draw->extra_shader_outputs.slot = 0;
}

View File

@@ -701,9 +701,9 @@ aapoint_first_point(struct draw_stage *stage, struct prim_header *header)
aapoint->pos_slot = draw_current_shader_position_output(draw);
/* allocate the extra post-transformed vertex attribute */
(void) draw_alloc_extra_vertex_attrib(draw, TGSI_SEMANTIC_GENERIC,
aapoint->fs->generic_attrib);
draw->extra_shader_outputs.semantic_name = TGSI_SEMANTIC_GENERIC;
draw->extra_shader_outputs.semantic_index = aapoint->fs->generic_attrib;
draw->extra_shader_outputs.slot = aapoint->tex_slot;
/* find psize slot in post-transform vertex */
aapoint->psize_slot = -1;
@@ -754,7 +754,7 @@ aapoint_flush(struct draw_stage *stage, unsigned flags)
draw->suspend_flushing = FALSE;
draw_remove_extra_vertex_attribs(draw);
draw->extra_shader_outputs.slot = 0;
}

View File

@@ -393,8 +393,8 @@ pstip_update_texture(struct pstip_stage *pstip)
*/
pipe->flush( pipe, PIPE_FLUSH_TEXTURE_CACHE, NULL );
transfer = pipe_get_transfer(pipe, pstip->texture, 0, 0,
PIPE_TRANSFER_WRITE, 0, 0, 32, 32);
transfer = pipe_get_transfer(pipe, pstip->texture, 0, 0, 0,
PIPE_TRANSFER_WRITE, 0, 0, 32, 32);
data = pipe->transfer_map(pipe, transfer);
/*
@@ -440,7 +440,6 @@ pstip_create_texture(struct pstip_stage *pstip)
texTemp.width0 = 32;
texTemp.height0 = 32;
texTemp.depth0 = 1;
texTemp.array_size = 1;
texTemp.bind = PIPE_BIND_SAMPLER_VIEW;
pstip->texture = screen->resource_create(screen, &texTemp);

View File

@@ -172,7 +172,7 @@ static struct draw_stage *validate_pipeline( struct draw_stage *stage )
wide_lines = (rast->line_width > draw->pipeline.wide_line_threshold
&& !rast->line_smooth);
/* drawing large/sprite points (but not AA points)? */
/* drawing large points? */
if (rast->sprite_coord_enable && draw->pipeline.point_sprite)
wide_points = TRUE;
else if (rast->point_smooth && draw->pipeline.aapoint)
@@ -207,7 +207,7 @@ static struct draw_stage *validate_pipeline( struct draw_stage *stage )
precalc_flat = TRUE;
}
if (wide_points) {
if (wide_points || rast->sprite_coord_enable) {
draw->pipeline.wide_point->next = next;
next = draw->pipeline.wide_point;
}

View File

@@ -57,24 +57,26 @@
#include "util/u_memory.h"
#include "pipe/p_defines.h"
#include "pipe/p_shader_tokens.h"
#include "draw_fs.h"
#include "draw_vs.h"
#include "draw_pipe.h"
struct widepoint_stage {
struct draw_stage stage; /**< base class */
struct draw_stage stage;
float half_point_size;
float xbias;
float ybias;
/** for automatic texcoord generation/replacement */
uint num_texcoord_gen;
uint texcoord_gen_slot[PIPE_MAX_SHADER_OUTPUTS];
uint texcoord_slot[PIPE_MAX_SHADER_OUTPUTS];
uint texcoord_enable[PIPE_MAX_SHADER_OUTPUTS];
uint num_texcoords;
uint texcoord_mode;
int psize_slot;
int point_coord_fs_input; /**< input for pointcoord */
};
@@ -94,20 +96,30 @@ widepoint_stage( struct draw_stage *stage )
static void set_texcoords(const struct widepoint_stage *wide,
struct vertex_header *v, const float tc[4])
{
const struct draw_context *draw = wide->stage.draw;
const struct pipe_rasterizer_state *rast = draw->rasterizer;
const uint texcoord_mode = rast->sprite_coord_mode;
uint i;
for (i = 0; i < wide->num_texcoords; i++) {
if (wide->texcoord_enable[i]) {
uint j = wide->texcoord_slot[i];
v->data[j][0] = tc[0];
if (wide->texcoord_mode == PIPE_SPRITE_COORD_LOWER_LEFT)
v->data[j][1] = 1.0f - tc[1];
else
v->data[j][1] = tc[1];
v->data[j][2] = tc[2];
v->data[j][3] = tc[3];
}
}
for (i = 0; i < wide->num_texcoord_gen; i++) {
const uint slot = wide->texcoord_gen_slot[i];
if (wide->point_coord_fs_input >= 0) {
/* put gl_PointCoord into the extra vertex slot */
uint slot = wide->stage.draw->extra_shader_outputs.slot;
v->data[slot][0] = tc[0];
if (texcoord_mode == PIPE_SPRITE_COORD_LOWER_LEFT)
if (wide->texcoord_mode == PIPE_SPRITE_COORD_LOWER_LEFT)
v->data[slot][1] = 1.0f - tc[1];
else
v->data[slot][1] = tc[1];
v->data[slot][2] = tc[2];
v->data[slot][3] = tc[3];
v->data[slot][2] = 0.0F;
v->data[slot][3] = 1.0F;
}
}
@@ -189,9 +201,18 @@ static void widepoint_point( struct draw_stage *stage,
}
static void
widepoint_first_point(struct draw_stage *stage,
struct prim_header *header)
static int
find_pntc_input_attrib(struct draw_context *draw)
{
/* Scan the fragment program's input decls to find the pointcoord
* attribute. The xy components will store the point coord.
*/
return 0; /* XXX fix this */
}
static void widepoint_first_point( struct draw_stage *stage,
struct prim_header *header )
{
struct widepoint_stage *wide = widepoint_stage(stage);
struct draw_context *draw = stage->draw;
@@ -223,49 +244,31 @@ widepoint_first_point(struct draw_stage *stage,
stage->point = draw_pipe_passthrough_point;
}
draw_remove_extra_vertex_attribs(draw);
if (rast->point_quad_rasterization) {
const struct draw_fragment_shader *fs = draw->fs.fragment_shader;
uint i;
wide->num_texcoord_gen = 0;
/* Loop over fragment shader inputs looking for generic inputs
* for which bit 'k' in sprite_coord_enable is set.
*/
for (i = 0; i < fs->info.num_inputs; i++) {
if (fs->info.input_semantic_name[i] == TGSI_SEMANTIC_GENERIC) {
const int generic_index = fs->info.input_semantic_index[i];
/* Note that sprite_coord enable is a bitfield of
* PIPE_MAX_SHADER_OUTPUTS bits.
*/
if (generic_index < PIPE_MAX_SHADER_OUTPUTS &&
(rast->sprite_coord_enable & (1 << generic_index))) {
/* OK, this generic attribute needs to be replaced with a
* texcoord (see above).
*/
int slot = draw_find_shader_output(draw,
TGSI_SEMANTIC_GENERIC,
generic_index);
if (slot > 0) {
/* there's already a post-vertex shader attribute
* for this fragment shader input attribute.
*/
}
else {
/* need to allocate a new post-vertex shader attribute */
slot = draw_alloc_extra_vertex_attrib(draw,
TGSI_SEMANTIC_GENERIC,
generic_index);
}
/* add this slot to the texcoord-gen list */
wide->texcoord_gen_slot[wide->num_texcoord_gen++] = slot;
}
/* find vertex shader texcoord outputs */
const struct draw_vertex_shader *vs = draw->vs.vertex_shader;
uint i, j = 0;
wide->texcoord_mode = rast->sprite_coord_mode;
for (i = 0; i < vs->info.num_outputs; i++) {
if (vs->info.output_semantic_name[i] == TGSI_SEMANTIC_GENERIC) {
wide->texcoord_slot[j] = i;
wide->texcoord_enable[j] = (rast->sprite_coord_enable >> j) & 1;
j++;
}
}
wide->num_texcoords = j;
/* find fragment shader PointCoord input */
wide->point_coord_fs_input = find_pntc_input_attrib(draw);
/* setup extra vp output (point coord implemented as a texcoord) */
draw->extra_shader_outputs.semantic_name = TGSI_SEMANTIC_GENERIC;
draw->extra_shader_outputs.semantic_index = 0;
draw->extra_shader_outputs.slot = draw_current_shader_outputs(draw);
}
else {
wide->point_coord_fs_input = -1;
draw->extra_shader_outputs.slot = 0;
}
wide->psize_slot = -1;
@@ -292,8 +295,7 @@ static void widepoint_flush( struct draw_stage *stage, unsigned flags )
stage->point = widepoint_first_point;
stage->next->flush( stage->next, flags );
draw_remove_extra_vertex_attribs(draw);
stage->draw->extra_shader_outputs.slot = 0;
/* restore original rasterizer state */
if (draw->rast_handle) {

View File

@@ -169,9 +169,6 @@ struct draw_context
unsigned vs_constants_size[PIPE_MAX_CONSTANT_BUFFERS];
const void *gs_constants[PIPE_MAX_CONSTANT_BUFFERS];
unsigned gs_constants_size[PIPE_MAX_CONSTANT_BUFFERS];
/* pointer to planes */
float (*planes)[12][4];
} user;
boolean test_fse; /* enable FSE even though its not correct (eg for softpipe) */
@@ -253,11 +250,6 @@ struct draw_context
struct tgsi_sampler **samplers;
} gs;
/** Fragment shader state */
struct {
struct draw_fragment_shader *fragment_shader;
} fs;
/** Stream output (vertex feedback) state */
struct {
struct pipe_stream_output_state state;
@@ -274,10 +266,9 @@ struct draw_context
/* If a prim stage introduces new vertex attributes, they'll be stored here
*/
struct {
uint num;
uint semantic_name[10];
uint semantic_index[10];
uint slot[10];
uint semantic_name;
uint semantic_index;
int slot;
} extra_shader_outputs;
unsigned reduced_prim;
@@ -286,6 +277,7 @@ struct draw_context
#ifdef HAVE_LLVM
struct draw_llvm *llvm;
LLVMExecutionEngineRef engine;
#endif
struct pipe_sampler_view *sampler_views[PIPE_MAX_VERTEX_SAMPLERS];
@@ -370,11 +362,6 @@ void draw_gs_destroy( struct draw_context *draw );
uint draw_current_shader_outputs(const struct draw_context *draw);
uint draw_current_shader_position_output(const struct draw_context *draw);
int draw_alloc_extra_vertex_attrib(struct draw_context *draw,
uint semantic_name, uint semantic_index);
void draw_remove_extra_vertex_attribs(struct draw_context *draw);
/*******************************************************************************
* Vertex processing (was passthrough) code:
*/

View File

@@ -287,84 +287,6 @@ draw_print_arrays(struct draw_context *draw, uint prim, int start, uint count)
}
/** Helper code for below */
#define PRIM_RESTART_LOOP(elements) \
do { \
for (i = start; i < end; i++) { \
if (elements[i] == info->restart_index) { \
if (cur_count > 0) { \
/* draw elts up to prev pos */ \
draw_pt_arrays(draw, prim, cur_start, cur_count); \
} \
/* begin new prim at next elt */ \
cur_start = i + 1; \
cur_count = 0; \
} \
else { \
cur_count++; \
} \
} \
if (cur_count > 0) { \
draw_pt_arrays(draw, prim, cur_start, cur_count); \
} \
} while (0)
/**
* For drawing prims with primitive restart enabled.
* Scan for restart indexes and draw the runs of elements/vertices between
* the restarts.
*/
static void
draw_pt_arrays_restart(struct draw_context *draw,
const struct pipe_draw_info *info)
{
const unsigned prim = info->mode;
const unsigned start = info->start;
const unsigned count = info->count;
const unsigned end = start + count;
unsigned i, cur_start, cur_count;
assert(info->primitive_restart);
if (draw->pt.user.elts) {
/* indexed prims (draw_elements) */
cur_start = start;
cur_count = 0;
switch (draw->pt.user.eltSize) {
case 1:
{
const ubyte *elt_ub = (const ubyte *) draw->pt.user.elts;
PRIM_RESTART_LOOP(elt_ub);
}
break;
case 2:
{
const ushort *elt_us = (const ushort *) draw->pt.user.elts;
PRIM_RESTART_LOOP(elt_us);
}
break;
case 4:
{
const uint *elt_ui = (const uint *) draw->pt.user.elts;
PRIM_RESTART_LOOP(elt_ui);
}
break;
default:
assert(0 && "bad eltSize in draw_arrays()");
}
}
else {
/* Non-indexed prims (draw_arrays).
* Primitive restart should have been handled in the state tracker.
*/
draw_pt_arrays(draw, prim, start, count);
}
}
/**
* Non-instanced drawing.
* \sa draw_arrays_instanced
@@ -473,12 +395,6 @@ draw_vbo(struct draw_context *draw,
for (instance = 0; instance < info->instance_count; instance++) {
draw->instance_id = instance + info->start_instance;
if (info->primitive_restart) {
draw_pt_arrays_restart(draw, info);
}
else {
draw_pt_arrays(draw, info->mode, info->start, info->count);
}
draw_pt_arrays(draw, info->mode, info->start, info->count);
}
}

View File

@@ -34,7 +34,6 @@
#include "draw/draw_pt.h"
#include "draw/draw_vs.h"
#include "draw/draw_llvm.h"
#include "gallivm/lp_bld_init.h"
struct llvm_middle_end {
@@ -73,18 +72,19 @@ llvm_middle_end_prepare( struct draw_pt_middle_end *middle,
struct draw_llvm_variant_list_item *li;
unsigned i;
unsigned instance_id_index = ~0;
const unsigned out_prim = (draw->gs.geometry_shader ?
draw->gs.geometry_shader->output_primitive :
in_prim);
unsigned out_prim = (draw->gs.geometry_shader ?
draw->gs.geometry_shader->output_primitive :
in_prim);
/* Add one to num_outputs because the pipeline occasionally tags on
* an additional texcoord, eg for AA lines.
*/
const unsigned nr = MAX2( shader->base.info.num_inputs,
shader->base.info.num_outputs + 1 );
unsigned nr = MAX2( shader->base.info.num_inputs,
shader->base.info.num_outputs + 1 );
/* Scan for instanceID system value.
* XXX but we never use instance_id_index?!
*/
for (i = 0; i < shader->base.info.num_inputs; i++) {
if (shader->base.info.input_semantic_name[i] == TGSI_SEMANTIC_INSTANCEID) {
@@ -133,10 +133,9 @@ llvm_middle_end_prepare( struct draw_pt_middle_end *middle,
key = draw_llvm_make_variant_key(fpme->llvm, store);
/* Search shader's list of variants for the key */
li = first_elem(&shader->variants);
while (!at_end(&shader->variants, li)) {
if (memcmp(&li->base->key, key, shader->variant_key_size) == 0) {
while(!at_end(&shader->variants, li)) {
if(memcmp(&li->base->key, key, shader->variant_key_size) == 0) {
variant = li->base;
break;
}
@@ -144,16 +143,10 @@ llvm_middle_end_prepare( struct draw_pt_middle_end *middle,
}
if (variant) {
/* found the variant, move to head of global list (for LRU) */
move_to_head(&fpme->llvm->vs_variants_list, &variant->list_item_global);
}
else {
/* Need to create new variant */
unsigned i;
/* First check if we've created too many variants. If so, free
* 25% of the LRU to avoid using too much memory.
*/
if (fpme->llvm->nr_variants >= DRAW_MAX_SHADER_VARIANTS) {
/*
* XXX: should we flush here ?
@@ -182,11 +175,6 @@ llvm_middle_end_prepare( struct draw_pt_middle_end *middle,
draw->pt.user.vs_constants[0];
fpme->llvm->jit_context.gs_constants =
draw->pt.user.gs_constants[0];
fpme->llvm->jit_context.planes =
(float (*) [12][4]) draw->pt.user.planes[0];
fpme->llvm->jit_context.viewport =
(float *)draw->viewport.scale;
}
@@ -229,7 +217,6 @@ llvm_pipeline_generic( struct draw_pt_middle_end *middle,
struct draw_vertex_info gs_vert_info;
struct draw_vertex_info *vert_info;
unsigned opt = fpme->opt;
unsigned clipped = 0;
llvm_vert_info.count = fetch_info->count;
llvm_vert_info.vertex_size = fpme->vertex_size;
@@ -243,7 +230,7 @@ llvm_pipeline_generic( struct draw_pt_middle_end *middle,
}
if (fetch_info->linear)
clipped = fpme->current_variant->jit_func( &fpme->llvm->jit_context,
fpme->current_variant->jit_func( &fpme->llvm->jit_context,
llvm_vert_info.verts,
(const char **)draw->pt.user.vbuffer,
fetch_info->start,
@@ -252,7 +239,7 @@ llvm_pipeline_generic( struct draw_pt_middle_end *middle,
draw->pt.vertex_buffer,
draw->instance_id);
else
clipped = fpme->current_variant->jit_func_elts( &fpme->llvm->jit_context,
fpme->current_variant->jit_func_elts( &fpme->llvm->jit_context,
llvm_vert_info.verts,
(const char **)draw->pt.user.vbuffer,
fetch_info->elts,
@@ -279,9 +266,6 @@ llvm_pipeline_generic( struct draw_pt_middle_end *middle,
FREE(vert_info->verts);
vert_info = &gs_vert_info;
prim_info = &gs_prim_info;
clipped = draw_pt_post_vs_run( fpme->post_vs, vert_info );
}
/* stream output needs to be done before clipping */
@@ -289,11 +273,11 @@ llvm_pipeline_generic( struct draw_pt_middle_end *middle,
vert_info,
prim_info );
if (clipped) {
if (draw_pt_post_vs_run( fpme->post_vs, vert_info )) {
opt |= PT_PIPELINE;
}
/* Do we need to run the pipeline? Now will come here if clipped
/* Do we need to run the pipeline?
*/
if (opt & PT_PIPELINE) {
pipeline( fpme,
@@ -429,7 +413,7 @@ draw_pt_fetch_pipeline_or_emit_llvm(struct draw_context *draw)
{
struct llvm_middle_end *fpme = 0;
if (!draw->llvm->gallivm->engine)
if (!draw->engine)
return NULL;
fpme = CALLOC_STRUCT( llvm_middle_end );

View File

@@ -65,7 +65,19 @@ static void
vs_llvm_delete( struct draw_vertex_shader *dvs )
{
struct llvm_vertex_shader *shader = llvm_vertex_shader(dvs);
struct pipe_fence_handle *fence = NULL;
struct draw_llvm_variant_list_item *li;
struct pipe_context *pipe = dvs->draw->pipe;
/*
* XXX: This might be not neccessary at all.
*/
pipe->flush(pipe, 0, &fence);
if (fence) {
pipe->screen->fence_finish(pipe->screen, fence, 0);
pipe->screen->fence_reference(pipe->screen, &fence, NULL);
}
li = first_elem(&shader->variants);
while(!at_end(&shader->variants, li)) {

View File

@@ -55,33 +55,4 @@
#endif
/**
* Redefine these LLVM entrypoints as invalid macros to make sure we
* don't accidentally use them. We need to use the functions which
* take an explicit LLVMContextRef parameter.
*/
#define LLVMInt1Type ILLEGAL_LLVM_FUNCTION
#define LLVMInt8Type ILLEGAL_LLVM_FUNCTION
#define LLVMInt16Type ILLEGAL_LLVM_FUNCTION
#define LLVMInt32Type ILLEGAL_LLVM_FUNCTION
#define LLVMInt64Type ILLEGAL_LLVM_FUNCTION
#define LLVMIntType ILLEGAL_LLVM_FUNCTION
#define LLVMFloatType ILLEGAL_LLVM_FUNCTION
#define LLVMDoubleType ILLEGAL_LLVM_FUNCTION
#define LLVMX86FP80Type ILLEGAL_LLVM_FUNCTION
#define LLVMFP128Type ILLEGAL_LLVM_FUNCTION
#define LLVMPPCFP128Type ILLEGAL_LLVM_FUNCTION
#define LLVMStructType ILLEGAL_LLVM_FUNCTION
#define LLVMVoidType ILLEGAL_LLVM_FUNCTION
#define LLVMLabelType ILLEGAL_LLVM_FUNCTION
#define LLVMOpaqueType ILLEGAL_LLVM_FUNCTION
#define LLVMUnionType ILLEGAL_LLVM_FUNCTION
#define LLVMMDString ILLEGAL_LLVM_FUNCTION
#define LLVMMDNode ILLEGAL_LLVM_FUNCTION
#define LLVMConstString ILLEGAL_LLVM_FUNCTION
#define LLVMConstStruct ILLEGAL_LLVM_FUNCTION
#define LLVMAppendBasicBlock ILLEGAL_LLVM_FUNCTION
#define LLVMInsertBasicBlock ILLEGAL_LLVM_FUNCTION
#define LLVMCreateBuilder ILLEGAL_LLVM_FUNCTION
#endif /* LP_BLD_H */

File diff suppressed because it is too large Load Diff

View File

@@ -171,12 +171,6 @@ LLVMValueRef
lp_build_itrunc(struct lp_build_context *bld,
LLVMValueRef a);
void
lp_build_ifloor_fract(struct lp_build_context *bld,
LLVMValueRef a,
LLVMValueRef *out_ipart,
LLVMValueRef *out_fpart);
LLVMValueRef
lp_build_sqrt(struct lp_build_context *bld,
LLVMValueRef a);
@@ -214,27 +208,10 @@ LLVMValueRef
lp_build_exp2(struct lp_build_context *bld,
LLVMValueRef a);
LLVMValueRef
lp_build_extract_exponent(struct lp_build_context *bld,
LLVMValueRef x,
int bias);
LLVMValueRef
lp_build_extract_mantissa(struct lp_build_context *bld,
LLVMValueRef x);
LLVMValueRef
lp_build_log2(struct lp_build_context *bld,
LLVMValueRef a);
LLVMValueRef
lp_build_fast_log2(struct lp_build_context *bld,
LLVMValueRef a);
LLVMValueRef
lp_build_ilog2(struct lp_build_context *bld,
LLVMValueRef x);
void
lp_build_exp2_approx(struct lp_build_context *bld,
LLVMValueRef x,

View File

@@ -56,21 +56,20 @@ lp_assert(int condition, const char *msg)
* \param msg a string to print if the assertion fails.
*/
LLVMValueRef
lp_build_assert(struct gallivm_state *gallivm,
LLVMValueRef condition,
lp_build_assert(LLVMBuilderRef builder, LLVMValueRef condition,
const char *msg)
{
LLVMBuilderRef builder = gallivm->builder;
LLVMContextRef context = gallivm->context;
LLVMModuleRef module = gallivm->module;
LLVMModuleRef module;
LLVMTypeRef arg_types[2];
LLVMValueRef msg_string, assert_func, params[2], r;
msg_string = lp_build_const_string_variable(module, context,
msg, strlen(msg) + 1);
module = LLVMGetGlobalParent(LLVMGetBasicBlockParent(
LLVMGetInsertBlock(builder)));
arg_types[0] = LLVMInt32TypeInContext(context);
arg_types[1] = LLVMPointerType(LLVMInt8TypeInContext(context), 0);
msg_string = lp_build_const_string_variable(module, msg, strlen(msg) + 1);
arg_types[0] = LLVMInt32Type();
arg_types[1] = LLVMPointerType(LLVMInt8Type(), 0);
/* lookup the lp_assert function */
assert_func = LLVMGetNamedFunction(module, "lp_assert");
@@ -78,12 +77,12 @@ lp_build_assert(struct gallivm_state *gallivm,
/* Create the assertion function if not found */
if (!assert_func) {
LLVMTypeRef func_type =
LLVMFunctionType(LLVMVoidTypeInContext(context), arg_types, 2, 0);
LLVMFunctionType(LLVMVoidType(), arg_types, 2, 0);
assert_func = LLVMAddFunction(module, "lp_assert", func_type);
LLVMSetFunctionCallConv(assert_func, LLVMCCallConv);
LLVMSetLinkage(assert_func, LLVMExternalLinkage);
LLVMAddGlobalMapping(gallivm->engine, assert_func,
LLVMAddGlobalMapping(lp_build_engine, assert_func,
func_to_pointer((func_pointer)lp_assert));
}
assert(assert_func);

View File

@@ -30,12 +30,10 @@
#include "lp_bld.h"
#include "lp_bld_init.h"
LLVMValueRef
lp_build_assert(struct gallivm_state *gallivm,
LLVMValueRef condition,
lp_build_assert(LLVMBuilderRef builder, LLVMValueRef condition,
const char *msg);

View File

@@ -1,192 +0,0 @@
/**************************************************************************
*
* Copyright 2010 VMware, Inc.
* All Rights Reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sub license, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
* THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
* DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
* OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
* USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* The above copyright notice and this permission notice (including the
* next paragraph) shall be included in all copies or substantial portions
* of the Software.
*
**************************************************************************/
#include "util/u_debug.h"
#include "lp_bld_type.h"
#include "lp_bld_debug.h"
#include "lp_bld_const.h"
#include "lp_bld_bitarit.h"
/**
* Return (a | b)
*/
LLVMValueRef
lp_build_or(struct lp_build_context *bld, LLVMValueRef a, LLVMValueRef b)
{
LLVMBuilderRef builder = bld->gallivm->builder;
const struct lp_type type = bld->type;
LLVMValueRef res;
assert(lp_check_value(type, a));
assert(lp_check_value(type, b));
/* can't do bitwise ops on floating-point values */
if (type.floating) {
a = LLVMBuildBitCast(builder, a, bld->int_vec_type, "");
b = LLVMBuildBitCast(builder, b, bld->int_vec_type, "");
}
res = LLVMBuildOr(builder, a, b, "");
if (type.floating) {
res = LLVMBuildBitCast(builder, res, bld->vec_type, "");
}
return res;
}
/**
* Return (a & b)
*/
LLVMValueRef
lp_build_and(struct lp_build_context *bld, LLVMValueRef a, LLVMValueRef b)
{
LLVMBuilderRef builder = bld->gallivm->builder;
const struct lp_type type = bld->type;
LLVMValueRef res;
assert(lp_check_value(type, a));
assert(lp_check_value(type, b));
/* can't do bitwise ops on floating-point values */
if (type.floating) {
a = LLVMBuildBitCast(builder, a, bld->int_vec_type, "");
b = LLVMBuildBitCast(builder, b, bld->int_vec_type, "");
}
res = LLVMBuildAnd(builder, a, b, "");
if (type.floating) {
res = LLVMBuildBitCast(builder, res, bld->vec_type, "");
}
return res;
}
/**
* Return (a & ~b)
*/
LLVMValueRef
lp_build_andnot(struct lp_build_context *bld, LLVMValueRef a, LLVMValueRef b)
{
LLVMBuilderRef builder = bld->gallivm->builder;
const struct lp_type type = bld->type;
LLVMValueRef res;
assert(lp_check_value(type, a));
assert(lp_check_value(type, b));
/* can't do bitwise ops on floating-point values */
if (type.floating) {
a = LLVMBuildBitCast(builder, a, bld->int_vec_type, "");
b = LLVMBuildBitCast(builder, b, bld->int_vec_type, "");
}
res = LLVMBuildNot(builder, b, "");
res = LLVMBuildAnd(builder, a, res, "");
if (type.floating) {
res = LLVMBuildBitCast(builder, res, bld->vec_type, "");
}
return res;
}
/**
* Shift left.
*/
LLVMValueRef
lp_build_shl(struct lp_build_context *bld, LLVMValueRef a, LLVMValueRef b)
{
LLVMBuilderRef builder = bld->gallivm->builder;
const struct lp_type type = bld->type;
LLVMValueRef res;
assert(!type.floating);
assert(lp_check_value(type, a));
assert(lp_check_value(type, b));
res = LLVMBuildShl(builder, a, b, "");
return res;
}
/**
* Shift right.
*/
LLVMValueRef
lp_build_shr(struct lp_build_context *bld, LLVMValueRef a, LLVMValueRef b)
{
LLVMBuilderRef builder = bld->gallivm->builder;
const struct lp_type type = bld->type;
LLVMValueRef res;
assert(!type.floating);
assert(lp_check_value(type, a));
assert(lp_check_value(type, b));
if (type.sign) {
res = LLVMBuildAShr(builder, a, b, "");
} else {
res = LLVMBuildLShr(builder, a, b, "");
}
return res;
}
/**
* Shift left with immediate.
*/
LLVMValueRef
lp_build_shl_imm(struct lp_build_context *bld, LLVMValueRef a, unsigned imm)
{
LLVMValueRef b = lp_build_const_int_vec(bld->gallivm, bld->type, imm);
assert(imm <= bld->type.width);
return lp_build_shl(bld, a, b);
}
/**
* Shift right with immediate.
*/
LLVMValueRef
lp_build_shr_imm(struct lp_build_context *bld, LLVMValueRef a, unsigned imm)
{
LLVMValueRef b = lp_build_const_int_vec(bld->gallivm, bld->type, imm);
assert(imm <= bld->type.width);
return lp_build_shr(bld, a, b);
}

View File

@@ -1,69 +0,0 @@
/**************************************************************************
*
* Copyright 2009 VMware, Inc.
* All Rights Reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sub license, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice (including the
* next paragraph) shall be included in all copies or substantial portions
* of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT.
* IN NO EVENT SHALL VMWARE AND/OR ITS SUPPLIERS BE LIABLE FOR
* ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
**************************************************************************/
/**
* @file
* Helper bitwise arithmetic functions.
*
* @author Jose Fonseca <jfonseca@vmware.com>
*/
#ifndef LP_BLD_BITARIT_H
#define LP_BLD_BITARIT_H
#include "gallivm/lp_bld.h"
struct lp_type;
struct lp_build_context;
LLVMValueRef
lp_build_or(struct lp_build_context *bld, LLVMValueRef a, LLVMValueRef b);
LLVMValueRef
lp_build_and(struct lp_build_context *bld, LLVMValueRef a, LLVMValueRef b);
LLVMValueRef
lp_build_andnot(struct lp_build_context *bld, LLVMValueRef a, LLVMValueRef b);
LLVMValueRef
lp_build_shl(struct lp_build_context *bld, LLVMValueRef a, LLVMValueRef b);
LLVMValueRef
lp_build_shr(struct lp_build_context *bld, LLVMValueRef a, LLVMValueRef b);
LLVMValueRef
lp_build_shl_imm(struct lp_build_context *bld, LLVMValueRef a, unsigned imm);
LLVMValueRef
lp_build_shr_imm(struct lp_build_context *bld, LLVMValueRef a, unsigned imm);
#endif /* !LP_BLD_ARIT_H */

View File

@@ -39,7 +39,6 @@
#include "lp_bld_type.h"
#include "lp_bld_const.h"
#include "lp_bld_init.h"
unsigned
@@ -212,31 +211,31 @@ lp_const_eps(struct lp_type type)
LLVMValueRef
lp_build_undef(struct gallivm_state *gallivm, struct lp_type type)
lp_build_undef(struct lp_type type)
{
LLVMTypeRef vec_type = lp_build_vec_type(gallivm, type);
LLVMTypeRef vec_type = lp_build_vec_type(type);
return LLVMGetUndef(vec_type);
}
LLVMValueRef
lp_build_zero(struct gallivm_state *gallivm, struct lp_type type)
lp_build_zero(struct lp_type type)
{
if (type.length == 1) {
if (type.floating)
return lp_build_const_float(gallivm, 0.0);
return LLVMConstReal(LLVMFloatType(), 0.0);
else
return LLVMConstInt(LLVMIntTypeInContext(gallivm->context, type.width), 0, 0);
return LLVMConstInt(LLVMIntType(type.width), 0, 0);
}
else {
LLVMTypeRef vec_type = lp_build_vec_type(gallivm, type);
LLVMTypeRef vec_type = lp_build_vec_type(type);
return LLVMConstNull(vec_type);
}
}
LLVMValueRef
lp_build_one(struct gallivm_state *gallivm, struct lp_type type)
lp_build_one(struct lp_type type)
{
LLVMTypeRef elem_type;
LLVMValueRef elems[LP_MAX_VECTOR_LENGTH];
@@ -244,7 +243,7 @@ lp_build_one(struct gallivm_state *gallivm, struct lp_type type)
assert(type.length <= LP_MAX_VECTOR_LENGTH);
elem_type = lp_build_elem_type(gallivm, type);
elem_type = lp_build_elem_type(type);
if(type.floating)
elems[0] = LLVMConstReal(elem_type, 1.0);
@@ -284,11 +283,10 @@ lp_build_one(struct gallivm_state *gallivm, struct lp_type type)
* Build constant-valued element from a scalar value.
*/
LLVMValueRef
lp_build_const_elem(struct gallivm_state *gallivm,
struct lp_type type,
lp_build_const_elem(struct lp_type type,
double val)
{
LLVMTypeRef elem_type = lp_build_elem_type(gallivm, type);
LLVMTypeRef elem_type = lp_build_elem_type(type);
LLVMValueRef elem;
if(type.floating) {
@@ -308,15 +306,15 @@ lp_build_const_elem(struct gallivm_state *gallivm,
* Build constant-valued vector from a scalar value.
*/
LLVMValueRef
lp_build_const_vec(struct gallivm_state *gallivm, struct lp_type type,
lp_build_const_vec(struct lp_type type,
double val)
{
if (type.length == 1) {
return lp_build_const_elem(gallivm, type, val);
return lp_build_const_elem(type, val);
} else {
LLVMValueRef elems[LP_MAX_VECTOR_LENGTH];
unsigned i;
elems[0] = lp_build_const_elem(gallivm, type, val);
elems[0] = lp_build_const_elem(type, val);
for(i = 1; i < type.length; ++i)
elems[i] = elems[0];
return LLVMConstVector(elems, type.length);
@@ -325,10 +323,10 @@ lp_build_const_vec(struct gallivm_state *gallivm, struct lp_type type,
LLVMValueRef
lp_build_const_int_vec(struct gallivm_state *gallivm, struct lp_type type,
long long val)
lp_build_const_int_vec(struct lp_type type,
long long val)
{
LLVMTypeRef elem_type = lp_build_int_elem_type(gallivm, type);
LLVMTypeRef elem_type = lp_build_int_elem_type(type);
LLVMValueRef elems[LP_MAX_VECTOR_LENGTH];
unsigned i;
@@ -345,8 +343,7 @@ lp_build_const_int_vec(struct gallivm_state *gallivm, struct lp_type type,
LLVMValueRef
lp_build_const_aos(struct gallivm_state *gallivm,
struct lp_type type,
lp_build_const_aos(struct lp_type type,
double r, double g, double b, double a,
const unsigned char *swizzle)
{
@@ -358,7 +355,7 @@ lp_build_const_aos(struct gallivm_state *gallivm,
assert(type.length % 4 == 0);
assert(type.length <= LP_MAX_VECTOR_LENGTH);
elem_type = lp_build_elem_type(gallivm, type);
elem_type = lp_build_elem_type(type);
if(swizzle == NULL)
swizzle = default_swizzle;
@@ -389,11 +386,10 @@ lp_build_const_aos(struct gallivm_state *gallivm,
* @param mask TGSI_WRITEMASK_xxx
*/
LLVMValueRef
lp_build_const_mask_aos(struct gallivm_state *gallivm,
struct lp_type type,
lp_build_const_mask_aos(struct lp_type type,
unsigned mask)
{
LLVMTypeRef elem_type = LLVMIntTypeInContext(gallivm->context, type.width);
LLVMTypeRef elem_type = LLVMIntType(type.width);
LLVMValueRef masks[LP_MAX_VECTOR_LENGTH];
unsigned i, j;

View File

@@ -39,7 +39,6 @@
#include "pipe/p_compiler.h"
#include "gallivm/lp_bld.h"
#include "gallivm/lp_bld_init.h"
@@ -74,71 +73,44 @@ lp_const_eps(struct lp_type type);
LLVMValueRef
lp_build_undef(struct gallivm_state *gallivm, struct lp_type type);
lp_build_undef(struct lp_type type);
LLVMValueRef
lp_build_zero(struct gallivm_state *gallivm, struct lp_type type);
lp_build_zero(struct lp_type type);
LLVMValueRef
lp_build_one(struct gallivm_state *gallivm, struct lp_type type);
lp_build_one(struct lp_type type);
LLVMValueRef
lp_build_const_elem(struct gallivm_state *gallivm, struct lp_type type,
lp_build_const_elem(struct lp_type type,
double val);
LLVMValueRef
lp_build_const_vec(struct gallivm_state *gallivm, struct lp_type type,
double val);
lp_build_const_vec(struct lp_type type, double val);
LLVMValueRef
lp_build_const_int_vec(struct gallivm_state *gallivm,
struct lp_type type, long long val);
lp_build_const_int_vec(struct lp_type type, long long val);
LLVMValueRef
lp_build_const_aos(struct gallivm_state *gallivm, struct lp_type type,
lp_build_const_aos(struct lp_type type,
double r, double g, double b, double a,
const unsigned char *swizzle);
LLVMValueRef
lp_build_const_mask_aos(struct gallivm_state *gallivm,
struct lp_type type,
lp_build_const_mask_aos(struct lp_type type,
unsigned mask);
static INLINE LLVMValueRef
lp_build_const_int32(struct gallivm_state *gallivm, int i)
lp_build_const_int32(int i)
{
return LLVMConstInt(LLVMInt32TypeInContext(gallivm->context), i, 0);
}
static INLINE LLVMValueRef
lp_build_const_float(struct gallivm_state *gallivm, float x)
{
return LLVMConstReal(LLVMFloatTypeInContext(gallivm->context), x);
}
/** Return constant-valued pointer to int */
static INLINE LLVMValueRef
lp_build_const_int_pointer(struct gallivm_state *gallivm, const void *ptr)
{
LLVMTypeRef int_type;
LLVMValueRef v;
/* int type large enough to hold a pointer */
int_type = LLVMIntTypeInContext(gallivm->context, 8 * sizeof(void *));
v = LLVMConstInt(int_type, (unsigned long long) ptr, 0);
v = LLVMBuildIntToPtr(gallivm->builder, v,
LLVMPointerType(int_type, 0),
"cast int to ptr");
return v;
return LLVMConstInt(LLVMInt32Type(), i, 0);
}

View File

@@ -63,7 +63,6 @@
#include "util/u_debug.h"
#include "util/u_math.h"
#include "util/u_cpu_detect.h"
#include "lp_bld_type.h"
#include "lp_bld_const.h"
@@ -89,118 +88,66 @@
* return { i32, i32, i32, i32 } where each value is in [0, 2^dst_width-1].
*/
LLVMValueRef
lp_build_clamped_float_to_unsigned_norm(struct gallivm_state *gallivm,
lp_build_clamped_float_to_unsigned_norm(LLVMBuilderRef builder,
struct lp_type src_type,
unsigned dst_width,
LLVMValueRef src)
{
LLVMBuilderRef builder = gallivm->builder;
LLVMTypeRef int_vec_type = lp_build_int_vec_type(gallivm, src_type);
LLVMTypeRef int_vec_type = lp_build_int_vec_type(src_type);
LLVMValueRef res;
unsigned mantissa;
unsigned n;
unsigned long long ubound;
unsigned long long mask;
double scale;
double bias;
assert(src_type.floating);
assert(dst_width <= src_type.width);
src_type.sign = FALSE;
mantissa = lp_mantissa(src_type);
if (dst_width <= mantissa) {
/*
* Apply magic coefficients that will make the desired result to appear
* in the lowest significant bits of the mantissa, with correct rounding.
*
* This only works if the destination width fits in the mantissa.
/* We cannot carry more bits than the mantissa */
n = MIN2(mantissa, dst_width);
/* This magic coefficients will make the desired result to appear in the
* lowest significant bits of the mantissa.
*/
ubound = ((unsigned long long)1 << n);
mask = ubound - 1;
scale = (double)mask/ubound;
bias = (double)((unsigned long long)1 << (mantissa - n));
res = LLVMBuildFMul(builder, src, lp_build_const_vec(src_type, scale), "");
res = LLVMBuildFAdd(builder, res, lp_build_const_vec(src_type, bias), "");
res = LLVMBuildBitCast(builder, res, int_vec_type, "");
if(dst_width > n) {
int shift = dst_width - n;
res = LLVMBuildShl(builder, res, lp_build_const_int_vec(src_type, shift), "");
/* TODO: Fill in the empty lower bits for additional precision? */
/* YES: this fixes progs/trivial/tri-z-eq.c.
* Otherwise vertex Z=1.0 values get converted to something like
* 0xfffffb00 and the test for equality with 0xffffffff fails.
*/
unsigned long long ubound;
unsigned long long mask;
double scale;
double bias;
ubound = (1ULL << dst_width);
mask = ubound - 1;
scale = (double)mask/ubound;
bias = (double)(1ULL << (mantissa - dst_width));
res = LLVMBuildFMul(builder, src, lp_build_const_vec(gallivm, src_type, scale), "");
res = LLVMBuildFAdd(builder, res, lp_build_const_vec(gallivm, src_type, bias), "");
res = LLVMBuildBitCast(builder, res, int_vec_type, "");
res = LLVMBuildAnd(builder, res,
lp_build_const_int_vec(gallivm, src_type, mask), "");
}
else if (dst_width == (mantissa + 1)) {
/*
* The destination width matches exactly what can be represented in
* floating point (i.e., mantissa + 1 bits). So do a straight
* multiplication followed by casting. No further rounding is necessary.
*/
double scale;
scale = (double)((1ULL << dst_width) - 1);
res = LLVMBuildFMul(builder, src,
lp_build_const_vec(gallivm, src_type, scale), "");
res = LLVMBuildFPToSI(builder, res, int_vec_type, "");
}
else {
/*
* The destination exceeds what can be represented in the floating point.
* So multiply by the largest power two we get away with, and when
* subtract the most significant bit to rescale to normalized values.
*
* The largest power of two factor we can get away is
* (1 << (src_type.width - 1)), because we need to use signed . In theory it
* should be (1 << (src_type.width - 2)), but IEEE 754 rules states
* INT_MIN should be returned in FPToSI, which is the correct result for
* values near 1.0!
*
* This means we get (src_type.width - 1) correct bits for values near 0.0,
* and (mantissa + 1) correct bits for values near 1.0. Equally or more
* important, we also get exact results for 0.0 and 1.0.
*/
unsigned n = MIN2(src_type.width - 1, dst_width);
double scale = (double)(1ULL << n);
unsigned lshift = dst_width - n;
unsigned rshift = n;
LLVMValueRef lshifted;
LLVMValueRef rshifted;
res = LLVMBuildFMul(builder, src,
lp_build_const_vec(gallivm, src_type, scale), "");
res = LLVMBuildFPToSI(builder, res, int_vec_type, "");
/*
* Align the most significant bit to its final place.
*
* This will cause 1.0 to overflow to 0, but the later adjustment will
* get it right.
*/
if (lshift) {
lshifted = LLVMBuildShl(builder, res,
lp_build_const_int_vec(gallivm, src_type,
lshift), "");
} else {
lshifted = res;
#if 0
{
LLVMValueRef msb;
msb = LLVMBuildLShr(builder, res, lp_build_const_int_vec(src_type, dst_width - 1), "");
msb = LLVMBuildShl(builder, msb, lp_build_const_int_vec(src_type, shift), "");
msb = LLVMBuildSub(builder, msb, lp_build_const_int_vec(src_type, 1), "");
res = LLVMBuildOr(builder, res, msb, "");
}
/*
* Align the most significant bit to the right.
*/
rshifted = LLVMBuildAShr(builder, res,
lp_build_const_int_vec(gallivm, src_type, rshift),
"");
/*
* Subtract the MSB to the LSB, therefore re-scaling from
* (1 << dst_width) to ((1 << dst_width) - 1).
*/
res = LLVMBuildSub(builder, lshifted, rshifted, "");
#elif 0
while(shift > 0) {
res = LLVMBuildOr(builder, res, LLVMBuildLShr(builder, res, lp_build_const_int_vec(src_type, n), ""), "");
shift -= n;
n *= 2;
}
#endif
}
else
res = LLVMBuildAnd(builder, res, lp_build_const_int_vec(src_type, mask), "");
return res;
}
@@ -212,14 +159,13 @@ lp_build_clamped_float_to_unsigned_norm(struct gallivm_state *gallivm,
* return {float, float, float, float} with values in range [0, 1].
*/
LLVMValueRef
lp_build_unsigned_norm_to_float(struct gallivm_state *gallivm,
lp_build_unsigned_norm_to_float(LLVMBuilderRef builder,
unsigned src_width,
struct lp_type dst_type,
LLVMValueRef src)
{
LLVMBuilderRef builder = gallivm->builder;
LLVMTypeRef vec_type = lp_build_vec_type(gallivm, dst_type);
LLVMTypeRef int_vec_type = lp_build_int_vec_type(gallivm, dst_type);
LLVMTypeRef vec_type = lp_build_vec_type(dst_type);
LLVMTypeRef int_vec_type = lp_build_int_vec_type(dst_type);
LLVMValueRef bias_;
LLVMValueRef res;
unsigned mantissa;
@@ -231,17 +177,6 @@ lp_build_unsigned_norm_to_float(struct gallivm_state *gallivm,
assert(dst_type.floating);
/* Special-case int8->float, though most cases could be handled
* this way:
*/
if (src_width == 8) {
scale = 1.0/255.0;
res = LLVMBuildSIToFP(builder, src, vec_type, "");
res = LLVMBuildFMul(builder, res,
lp_build_const_vec(gallivm, dst_type, scale), "");
return res;
}
mantissa = lp_mantissa(dst_type);
n = MIN2(mantissa, src_width);
@@ -255,11 +190,10 @@ lp_build_unsigned_norm_to_float(struct gallivm_state *gallivm,
if(src_width > mantissa) {
int shift = src_width - mantissa;
res = LLVMBuildLShr(builder, res,
lp_build_const_int_vec(gallivm, dst_type, shift), "");
res = LLVMBuildLShr(builder, res, lp_build_const_int_vec(dst_type, shift), "");
}
bias_ = lp_build_const_vec(gallivm, dst_type, bias);
bias_ = lp_build_const_vec(dst_type, bias);
res = LLVMBuildOr(builder,
res,
@@ -268,7 +202,7 @@ lp_build_unsigned_norm_to_float(struct gallivm_state *gallivm,
res = LLVMBuildBitCast(builder, res, vec_type, "");
res = LLVMBuildFSub(builder, res, bias_, "");
res = LLVMBuildFMul(builder, res, lp_build_const_vec(gallivm, dst_type, scale), "");
res = LLVMBuildFMul(builder, res, lp_build_const_vec(dst_type, scale), "");
return res;
}
@@ -281,13 +215,12 @@ lp_build_unsigned_norm_to_float(struct gallivm_state *gallivm,
* to the lp_type union.
*/
void
lp_build_conv(struct gallivm_state *gallivm,
lp_build_conv(LLVMBuilderRef builder,
struct lp_type src_type,
struct lp_type dst_type,
const LLVMValueRef *src, unsigned num_srcs,
LLVMValueRef *dst, unsigned num_dsts)
{
LLVMBuilderRef builder = gallivm->builder;
struct lp_type tmp_type;
LLVMValueRef tmp[LP_MAX_VECTOR_LENGTH];
unsigned num_tmps;
@@ -308,87 +241,6 @@ lp_build_conv(struct gallivm_state *gallivm,
}
num_tmps = num_srcs;
/* Special case 4x4f --> 1x16ub
*/
if (src_type.floating == 1 &&
src_type.fixed == 0 &&
src_type.sign == 1 &&
src_type.norm == 0 &&
src_type.width == 32 &&
src_type.length == 4 &&
dst_type.floating == 0 &&
dst_type.fixed == 0 &&
dst_type.sign == 0 &&
dst_type.norm == 1 &&
dst_type.width == 8 &&
dst_type.length == 16 &&
util_cpu_caps.has_sse2)
{
int i;
for (i = 0; i < num_dsts; i++, src += 4) {
struct lp_type int16_type = dst_type;
struct lp_type int32_type = dst_type;
LLVMValueRef lo, hi;
LLVMValueRef src_int0;
LLVMValueRef src_int1;
LLVMValueRef src_int2;
LLVMValueRef src_int3;
LLVMTypeRef int16_vec_type;
LLVMTypeRef int32_vec_type;
LLVMTypeRef src_vec_type;
LLVMTypeRef dst_vec_type;
LLVMValueRef const_255f;
LLVMValueRef a, b, c, d;
int16_type.width *= 2;
int16_type.length /= 2;
int16_type.sign = 1;
int32_type.width *= 4;
int32_type.length /= 4;
int32_type.sign = 1;
src_vec_type = lp_build_vec_type(gallivm, src_type);
dst_vec_type = lp_build_vec_type(gallivm, dst_type);
int16_vec_type = lp_build_vec_type(gallivm, int16_type);
int32_vec_type = lp_build_vec_type(gallivm, int32_type);
const_255f = lp_build_const_vec(gallivm, src_type, 255.0f);
a = LLVMBuildFMul(builder, src[0], const_255f, "");
b = LLVMBuildFMul(builder, src[1], const_255f, "");
c = LLVMBuildFMul(builder, src[2], const_255f, "");
d = LLVMBuildFMul(builder, src[3], const_255f, "");
{
struct lp_build_context bld;
bld.gallivm = gallivm;
bld.type = src_type;
bld.vec_type = src_vec_type;
bld.int_elem_type = lp_build_elem_type(gallivm, int32_type);
bld.int_vec_type = int32_vec_type;
bld.undef = lp_build_undef(gallivm, src_type);
bld.zero = lp_build_zero(gallivm, src_type);
bld.one = lp_build_one(gallivm, src_type);
src_int0 = lp_build_iround(&bld, a);
src_int1 = lp_build_iround(&bld, b);
src_int2 = lp_build_iround(&bld, c);
src_int3 = lp_build_iround(&bld, d);
}
/* relying on clamping behavior of sse2 intrinsics here */
lo = lp_build_pack2(gallivm, int32_type, int16_type, src_int0, src_int1);
hi = lp_build_pack2(gallivm, int32_type, int16_type, src_int2, src_int3);
dst[i] = lp_build_pack2(gallivm, int16_type, dst_type, lo, hi);
}
return;
}
/*
* Clamp if necessary
*/
@@ -401,13 +253,13 @@ lp_build_conv(struct gallivm_state *gallivm,
double dst_max = lp_const_max(dst_type);
LLVMValueRef thres;
lp_build_context_init(&bld, gallivm, tmp_type);
lp_build_context_init(&bld, builder, tmp_type);
if(src_min < dst_min) {
if(dst_min == 0.0)
thres = bld.zero;
else
thres = lp_build_const_vec(gallivm, src_type, dst_min);
thres = lp_build_const_vec(src_type, dst_min);
for(i = 0; i < num_tmps; ++i)
tmp[i] = lp_build_max(&bld, tmp[i], thres);
}
@@ -416,7 +268,7 @@ lp_build_conv(struct gallivm_state *gallivm,
if(dst_max == 1.0)
thres = bld.one;
else
thres = lp_build_const_vec(gallivm, src_type, dst_max);
thres = lp_build_const_vec(src_type, dst_max);
for(i = 0; i < num_tmps; ++i)
tmp[i] = lp_build_min(&bld, tmp[i], thres);
}
@@ -432,7 +284,7 @@ lp_build_conv(struct gallivm_state *gallivm,
else if(tmp_type.floating) {
if(!dst_type.fixed && !dst_type.sign && dst_type.norm) {
for(i = 0; i < num_tmps; ++i) {
tmp[i] = lp_build_clamped_float_to_unsigned_norm(gallivm,
tmp[i] = lp_build_clamped_float_to_unsigned_norm(builder,
tmp_type,
dst_type.width,
tmp[i]);
@@ -444,14 +296,14 @@ lp_build_conv(struct gallivm_state *gallivm,
LLVMTypeRef tmp_vec_type;
if (dst_scale != 1.0) {
LLVMValueRef scale = lp_build_const_vec(gallivm, tmp_type, dst_scale);
LLVMValueRef scale = lp_build_const_vec(tmp_type, dst_scale);
for(i = 0; i < num_tmps; ++i)
tmp[i] = LLVMBuildFMul(builder, tmp[i], scale, "");
}
/* Use an equally sized integer for intermediate computations */
tmp_type.floating = FALSE;
tmp_vec_type = lp_build_vec_type(gallivm, tmp_type);
tmp_vec_type = lp_build_vec_type(tmp_type);
for(i = 0; i < num_tmps; ++i) {
#if 0
if(dst_type.sign)
@@ -471,8 +323,7 @@ lp_build_conv(struct gallivm_state *gallivm,
/* FIXME: compensate different offsets too */
if(src_shift > dst_shift) {
LLVMValueRef shift = lp_build_const_int_vec(gallivm, tmp_type,
src_shift - dst_shift);
LLVMValueRef shift = lp_build_const_int_vec(tmp_type, src_shift - dst_shift);
for(i = 0; i < num_tmps; ++i)
if(src_type.sign)
tmp[i] = LLVMBuildAShr(builder, tmp[i], shift, "");
@@ -496,7 +347,7 @@ lp_build_conv(struct gallivm_state *gallivm,
new_type.width = dst_type.width;
new_type.length = dst_type.length;
lp_build_resize(gallivm, tmp_type, new_type, tmp, num_srcs, tmp, num_dsts);
lp_build_resize(builder, tmp_type, new_type, tmp, num_srcs, tmp, num_dsts);
tmp_type = new_type;
num_tmps = num_dsts;
@@ -512,7 +363,7 @@ lp_build_conv(struct gallivm_state *gallivm,
else if(!src_type.floating && dst_type.floating) {
if(!src_type.fixed && !src_type.sign && src_type.norm) {
for(i = 0; i < num_tmps; ++i) {
tmp[i] = lp_build_unsigned_norm_to_float(gallivm,
tmp[i] = lp_build_unsigned_norm_to_float(builder,
src_type.width,
dst_type,
tmp[i]);
@@ -526,7 +377,7 @@ lp_build_conv(struct gallivm_state *gallivm,
/* Use an equally sized integer for intermediate computations */
tmp_type.floating = TRUE;
tmp_type.sign = TRUE;
tmp_vec_type = lp_build_vec_type(gallivm, tmp_type);
tmp_vec_type = lp_build_vec_type(tmp_type);
for(i = 0; i < num_tmps; ++i) {
#if 0
if(dst_type.sign)
@@ -540,7 +391,7 @@ lp_build_conv(struct gallivm_state *gallivm,
}
if (src_scale != 1.0) {
LLVMValueRef scale = lp_build_const_vec(gallivm, tmp_type, 1.0/src_scale);
LLVMValueRef scale = lp_build_const_vec(tmp_type, 1.0/src_scale);
for(i = 0; i < num_tmps; ++i)
tmp[i] = LLVMBuildFMul(builder, tmp[i], scale, "");
}
@@ -552,7 +403,7 @@ lp_build_conv(struct gallivm_state *gallivm,
/* FIXME: compensate different offsets too */
if(src_shift < dst_shift) {
LLVMValueRef shift = lp_build_const_int_vec(gallivm, tmp_type, dst_shift - src_shift);
LLVMValueRef shift = lp_build_const_int_vec(tmp_type, dst_shift - src_shift);
for(i = 0; i < num_tmps; ++i)
tmp[i] = LLVMBuildShl(builder, tmp[i], shift, "");
}
@@ -576,7 +427,7 @@ lp_build_conv(struct gallivm_state *gallivm,
* This is basically a very trimmed down version of lp_build_conv.
*/
void
lp_build_conv_mask(struct gallivm_state *gallivm,
lp_build_conv_mask(LLVMBuilderRef builder,
struct lp_type src_type,
struct lp_type dst_type,
const LLVMValueRef *src, unsigned num_srcs,
@@ -610,11 +461,11 @@ lp_build_conv_mask(struct gallivm_state *gallivm,
if(src_type.width > dst_type.width) {
assert(num_dsts == 1);
dst[0] = lp_build_pack(gallivm, src_type, dst_type, TRUE, src, num_srcs);
dst[0] = lp_build_pack(builder, src_type, dst_type, TRUE, src, num_srcs);
}
else if(src_type.width < dst_type.width) {
assert(num_srcs == 1);
lp_build_unpack(gallivm, src_type, dst_type, src[0], dst, num_dsts);
lp_build_unpack(builder, src_type, dst_type, src[0], dst, num_dsts);
}
else {
assert(num_srcs == num_dsts);

View File

@@ -44,27 +44,27 @@ struct lp_type;
LLVMValueRef
lp_build_clamped_float_to_unsigned_norm(struct gallivm_state *gallivm,
lp_build_clamped_float_to_unsigned_norm(LLVMBuilderRef builder,
struct lp_type src_type,
unsigned dst_width,
LLVMValueRef src);
LLVMValueRef
lp_build_unsigned_norm_to_float(struct gallivm_state *gallivm,
lp_build_unsigned_norm_to_float(LLVMBuilderRef builder,
unsigned src_width,
struct lp_type dst_type,
LLVMValueRef src);
void
lp_build_conv(struct gallivm_state *gallivm,
lp_build_conv(LLVMBuilderRef builder,
struct lp_type src_type,
struct lp_type dst_type,
const LLVMValueRef *srcs, unsigned num_srcs,
LLVMValueRef *dsts, unsigned num_dsts);
void
lp_build_conv_mask(struct gallivm_state *gallivm,
lp_build_conv_mask(LLVMBuilderRef builder,
struct lp_type src_type,
struct lp_type dst_type,
const LLVMValueRef *src, unsigned num_srcs,

View File

@@ -57,8 +57,6 @@ lp_disassemble(const void* func)
#ifdef HAVE_UDIS86
ud_t ud_obj;
uint64_t max_jmp_pc;
uint inst_no;
boolean emit_addrs = TRUE, emit_line_nos = FALSE;
ud_init(&ud_obj);
@@ -78,18 +76,13 @@ lp_disassemble(const void* func)
while (ud_disassemble(&ud_obj)) {
if (emit_addrs) {
#ifdef PIPE_ARCH_X86
debug_printf("0x%08lx:\t", (unsigned long)ud_insn_off(&ud_obj));
debug_printf("0x%08lx:\t", (unsigned long)ud_insn_off(&ud_obj));
#endif
#ifdef PIPE_ARCH_X86_64
debug_printf("0x%016llx:\t", (unsigned long long)ud_insn_off(&ud_obj));
debug_printf("0x%016llx:\t", (unsigned long long)ud_insn_off(&ud_obj));
#endif
}
else if (emit_line_nos) {
debug_printf("%6d:\t", inst_no);
inst_no++;
}
#if 0
debug_printf("%-16s ", ud_insn_hex(&ud_obj));
#endif
@@ -122,10 +115,8 @@ lp_disassemble(const void* func)
}
}
if (ud_obj.mnemonic == UD_Iinvalid ||
(ud_insn_off(&ud_obj) >= max_jmp_pc &&
(ud_obj.mnemonic == UD_Iret ||
ud_obj.mnemonic == UD_Ijmp)))
if ((ud_insn_off(&ud_obj) >= max_jmp_pc && ud_obj.mnemonic == UD_Iret) ||
ud_obj.mnemonic == UD_Iinvalid)
break;
}

View File

@@ -36,13 +36,11 @@
#include "util/u_string.h"
#define GALLIVM_DEBUG_TGSI (1 << 0)
#define GALLIVM_DEBUG_IR (1 << 1)
#define GALLIVM_DEBUG_ASM (1 << 2)
#define GALLIVM_DEBUG_NO_OPT (1 << 3)
#define GALLIVM_DEBUG_PERF (1 << 4)
#define GALLIVM_DEBUG_NO_BRILINEAR (1 << 5)
#define GALLIVM_DEBUG_GC (1 << 6)
#define GALLIVM_DEBUG_TGSI 0x1
#define GALLIVM_DEBUG_IR 0x2
#define GALLIVM_DEBUG_ASM 0x4
#define GALLIVM_DEBUG_NO_OPT 0x8
#define GALLIVM_DEBUG_PERF 0x10
#ifdef DEBUG

View File

@@ -34,60 +34,352 @@
#include "util/u_debug.h"
#include "util/u_memory.h"
#include "lp_bld_init.h"
#include "lp_bld_type.h"
#include "lp_bld_flow.h"
#define LP_BUILD_FLOW_MAX_VARIABLES 64
#define LP_BUILD_FLOW_MAX_DEPTH 32
/**
* Insert a new block, right where builder is pointing to.
* Enumeration of all possible flow constructs.
*/
enum lp_build_flow_construct_kind {
LP_BUILD_FLOW_SCOPE,
LP_BUILD_FLOW_SKIP,
LP_BUILD_FLOW_IF
};
/**
* Variable declaration scope.
*/
struct lp_build_flow_scope
{
/** Number of variables declared in this scope */
unsigned num_variables;
};
/**
* Early exit. Useful to skip to the end of a function or block when
* the execution mask becomes zero or when there is an error condition.
*/
struct lp_build_flow_skip
{
/** Block to skip to */
LLVMBasicBlockRef block;
/** Number of variables declared at the beginning */
unsigned num_variables;
LLVMValueRef *phi; /**< array [num_variables] */
};
/**
* if/else/endif.
*/
struct lp_build_flow_if
{
unsigned num_variables;
LLVMValueRef *phi; /**< array [num_variables] */
LLVMValueRef condition;
LLVMBasicBlockRef entry_block, true_block, false_block, merge_block;
};
/**
* Union of all possible flow constructs' data
*/
union lp_build_flow_construct_data
{
struct lp_build_flow_scope scope;
struct lp_build_flow_skip skip;
struct lp_build_flow_if ifthen;
};
/**
* Element of the flow construct stack.
*/
struct lp_build_flow_construct
{
enum lp_build_flow_construct_kind kind;
union lp_build_flow_construct_data data;
};
/**
* All necessary data to generate LLVM control flow constructs.
*
* This is useful important not only for aesthetic reasons, but also for
* performance reasons, as frequently run blocks should be laid out next to
* each other and fall-throughs maximized.
* Besides keeping track of the control flow construct themselves we also
* need to keep track of variables in order to generate SSA Phi values.
*/
struct lp_build_flow_context
{
LLVMBuilderRef builder;
/**
* Control flow stack.
*/
struct lp_build_flow_construct constructs[LP_BUILD_FLOW_MAX_DEPTH];
unsigned num_constructs;
/**
* Variable stack
*/
LLVMValueRef *variables[LP_BUILD_FLOW_MAX_VARIABLES];
unsigned num_variables;
};
struct lp_build_flow_context *
lp_build_flow_create(LLVMBuilderRef builder)
{
struct lp_build_flow_context *flow;
flow = CALLOC_STRUCT(lp_build_flow_context);
if(!flow)
return NULL;
flow->builder = builder;
return flow;
}
void
lp_build_flow_destroy(struct lp_build_flow_context *flow)
{
assert(flow->num_constructs == 0);
assert(flow->num_variables == 0);
FREE(flow);
}
/**
* Begin/push a new flow control construct, such as a loop, skip block
* or variable scope.
*/
static union lp_build_flow_construct_data *
lp_build_flow_push(struct lp_build_flow_context *flow,
enum lp_build_flow_construct_kind kind)
{
assert(flow->num_constructs < LP_BUILD_FLOW_MAX_DEPTH);
if(flow->num_constructs >= LP_BUILD_FLOW_MAX_DEPTH)
return NULL;
flow->constructs[flow->num_constructs].kind = kind;
return &flow->constructs[flow->num_constructs++].data;
}
/**
* Return the current/top flow control construct on the stack.
* \param kind the expected type of the top-most construct
*/
static union lp_build_flow_construct_data *
lp_build_flow_peek(struct lp_build_flow_context *flow,
enum lp_build_flow_construct_kind kind)
{
assert(flow->num_constructs);
if(!flow->num_constructs)
return NULL;
assert(flow->constructs[flow->num_constructs - 1].kind == kind);
if(flow->constructs[flow->num_constructs - 1].kind != kind)
return NULL;
return &flow->constructs[flow->num_constructs - 1].data;
}
/**
* End/pop the current/top flow control construct on the stack.
* \param kind the expected type of the top-most construct
*/
static union lp_build_flow_construct_data *
lp_build_flow_pop(struct lp_build_flow_context *flow,
enum lp_build_flow_construct_kind kind)
{
assert(flow->num_constructs);
if(!flow->num_constructs)
return NULL;
assert(flow->constructs[flow->num_constructs - 1].kind == kind);
if(flow->constructs[flow->num_constructs - 1].kind != kind)
return NULL;
return &flow->constructs[--flow->num_constructs].data;
}
/**
* Begin a variable scope.
*
* See also llvm/lib/Transforms/Scalar/BasicBlockPlacement.cpp.
*
*/
void
lp_build_flow_scope_begin(struct lp_build_flow_context *flow)
{
struct lp_build_flow_scope *scope;
scope = &lp_build_flow_push(flow, LP_BUILD_FLOW_SCOPE)->scope;
if(!scope)
return;
scope->num_variables = 0;
}
/**
* Declare a variable.
*
* A variable is a named entity which can have different LLVMValueRef's at
* different points of the program. This is relevant for control flow because
* when there are multiple branches to a same location we need to replace
* the variable's value with a Phi function as explained in
* http://en.wikipedia.org/wiki/Static_single_assignment_form .
*
* We keep track of variables by keeping around a pointer to where they're
* current.
*
* There are a few cautions to observe:
*
* - Variable's value must not be NULL. If there is no initial value then
* LLVMGetUndef() should be used.
*
* - Variable's value must be kept up-to-date. If the variable is going to be
* modified by a function then a pointer should be passed so that its value
* is accurate. Failure to do this will cause some of the variables'
* transient values to be lost, leading to wrong results.
*
* - A program should be written from top to bottom, by always appending
* instructions to the bottom with a single LLVMBuilderRef. Inserting and/or
* modifying existing statements will most likely lead to wrong results.
*
*/
void
lp_build_flow_scope_declare(struct lp_build_flow_context *flow,
LLVMValueRef *variable)
{
struct lp_build_flow_scope *scope;
scope = &lp_build_flow_peek(flow, LP_BUILD_FLOW_SCOPE)->scope;
if(!scope)
return;
assert(*variable);
if(!*variable)
return;
assert(flow->num_variables < LP_BUILD_FLOW_MAX_VARIABLES);
if(flow->num_variables >= LP_BUILD_FLOW_MAX_VARIABLES)
return;
flow->variables[flow->num_variables++] = variable;
++scope->num_variables;
}
void
lp_build_flow_scope_end(struct lp_build_flow_context *flow)
{
struct lp_build_flow_scope *scope;
scope = &lp_build_flow_pop(flow, LP_BUILD_FLOW_SCOPE)->scope;
if(!scope)
return;
assert(flow->num_variables >= scope->num_variables);
if(flow->num_variables < scope->num_variables) {
flow->num_variables = 0;
return;
}
flow->num_variables -= scope->num_variables;
}
/**
* Note: this function has no dependencies on the flow code and could
* be used elsewhere.
*/
LLVMBasicBlockRef
lp_build_insert_new_block(struct gallivm_state *gallivm, const char *name)
lp_build_insert_new_block(LLVMBuilderRef builder, const char *name)
{
LLVMBasicBlockRef current_block;
LLVMBasicBlockRef next_block;
LLVMBasicBlockRef new_block;
/* get current basic block */
current_block = LLVMGetInsertBlock(gallivm->builder);
current_block = LLVMGetInsertBlock(builder);
/* check if there's another block after this one */
next_block = LLVMGetNextBasicBlock(current_block);
if (next_block) {
/* insert the new block before the next block */
new_block = LLVMInsertBasicBlockInContext(gallivm->context, next_block, name);
new_block = LLVMInsertBasicBlock(next_block, name);
}
else {
/* append new block after current block */
LLVMValueRef function = LLVMGetBasicBlockParent(current_block);
new_block = LLVMAppendBasicBlockInContext(gallivm->context, function, name);
new_block = LLVMAppendBasicBlock(function, name);
}
return new_block;
}
static LLVMBasicBlockRef
lp_build_flow_insert_block(struct lp_build_flow_context *flow)
{
return lp_build_insert_new_block(flow->builder, "");
}
/**
* Begin a "skip" block. Inside this block we can test a condition and
* skip to the end of the block if the condition is false.
*/
void
lp_build_flow_skip_begin(struct lp_build_skip_context *skip,
struct gallivm_state *gallivm)
lp_build_flow_skip_begin(struct lp_build_flow_context *flow)
{
skip->gallivm = gallivm;
struct lp_build_flow_skip *skip;
LLVMBuilderRef builder;
unsigned i;
skip = &lp_build_flow_push(flow, LP_BUILD_FLOW_SKIP)->skip;
if(!skip)
return;
/* create new basic block */
skip->block = lp_build_insert_new_block(gallivm, "skip");
skip->block = lp_build_flow_insert_block(flow);
skip->num_variables = flow->num_variables;
if(!skip->num_variables) {
skip->phi = NULL;
return;
}
/* Allocate a Phi node for each variable in this skip scope */
skip->phi = MALLOC(skip->num_variables * sizeof *skip->phi);
if(!skip->phi) {
skip->num_variables = 0;
return;
}
builder = LLVMCreateBuilder();
LLVMPositionBuilderAtEnd(builder, skip->block);
/* create a Phi node for each variable */
for(i = 0; i < skip->num_variables; ++i)
skip->phi[i] = LLVMBuildPhi(builder, LLVMTypeOf(*flow->variables[i]), "");
LLVMDisposeBuilder(builder);
}
@@ -96,50 +388,83 @@ lp_build_flow_skip_begin(struct lp_build_skip_context *skip,
* skip block if the condition is true.
*/
void
lp_build_flow_skip_cond_break(struct lp_build_skip_context *skip,
lp_build_flow_skip_cond_break(struct lp_build_flow_context *flow,
LLVMValueRef cond)
{
struct lp_build_flow_skip *skip;
LLVMBasicBlockRef current_block;
LLVMBasicBlockRef new_block;
unsigned i;
new_block = lp_build_insert_new_block(skip->gallivm, "");
skip = &lp_build_flow_peek(flow, LP_BUILD_FLOW_SKIP)->skip;
if(!skip)
return;
current_block = LLVMGetInsertBlock(flow->builder);
new_block = lp_build_flow_insert_block(flow);
/* for each variable, update the Phi node with a (variable, block) pair */
for(i = 0; i < skip->num_variables; ++i) {
assert(*flow->variables[i]);
assert(LLVMTypeOf(skip->phi[i]) == LLVMTypeOf(*flow->variables[i]));
LLVMAddIncoming(skip->phi[i], flow->variables[i], &current_block, 1);
}
/* if cond is true, goto skip->block, else goto new_block */
LLVMBuildCondBr(skip->gallivm->builder, cond, skip->block, new_block);
LLVMBuildCondBr(flow->builder, cond, skip->block, new_block);
LLVMPositionBuilderAtEnd(skip->gallivm->builder, new_block);
LLVMPositionBuilderAtEnd(flow->builder, new_block);
}
void
lp_build_flow_skip_end(struct lp_build_skip_context *skip)
lp_build_flow_skip_end(struct lp_build_flow_context *flow)
{
struct lp_build_flow_skip *skip;
LLVMBasicBlockRef current_block;
unsigned i;
skip = &lp_build_flow_pop(flow, LP_BUILD_FLOW_SKIP)->skip;
if(!skip)
return;
current_block = LLVMGetInsertBlock(flow->builder);
/* add (variable, block) tuples to the phi nodes */
for(i = 0; i < skip->num_variables; ++i) {
assert(*flow->variables[i]);
assert(LLVMTypeOf(skip->phi[i]) == LLVMTypeOf(*flow->variables[i]));
LLVMAddIncoming(skip->phi[i], flow->variables[i], &current_block, 1);
*flow->variables[i] = skip->phi[i];
}
/* goto block */
LLVMBuildBr(skip->gallivm->builder, skip->block);
LLVMPositionBuilderAtEnd(skip->gallivm->builder, skip->block);
LLVMBuildBr(flow->builder, skip->block);
LLVMPositionBuilderAtEnd(flow->builder, skip->block);
FREE(skip->phi);
}
/**
* Check if the mask predicate is zero. If so, jump to the end of the block.
*/
void
static void
lp_build_mask_check(struct lp_build_mask_context *mask)
{
LLVMBuilderRef builder = mask->skip.gallivm->builder;
LLVMValueRef value;
LLVMBuilderRef builder = mask->flow->builder;
LLVMValueRef cond;
value = lp_build_mask_value(mask);
/* cond = (mask == 0) */
cond = LLVMBuildICmp(builder,
LLVMIntEQ,
LLVMBuildBitCast(builder, value, mask->reg_type, ""),
LLVMBuildBitCast(builder, mask->value, mask->reg_type, ""),
LLVMConstNull(mask->reg_type),
"");
/* if cond, goto end of block */
lp_build_flow_skip_cond_break(&mask->skip, cond);
lp_build_flow_skip_cond_break(mask->flow, cond);
}
@@ -152,27 +477,21 @@ lp_build_mask_check(struct lp_build_mask_context *mask)
*/
void
lp_build_mask_begin(struct lp_build_mask_context *mask,
struct gallivm_state *gallivm,
struct lp_build_flow_context *flow,
struct lp_type type,
LLVMValueRef value)
{
memset(mask, 0, sizeof *mask);
mask->reg_type = LLVMIntTypeInContext(gallivm->context, type.width * type.length);
mask->var = lp_build_alloca(gallivm,
lp_build_int_vec_type(gallivm, type),
"execution_mask");
mask->flow = flow;
mask->reg_type = LLVMIntType(type.width * type.length);
mask->value = value;
LLVMBuildStore(gallivm->builder, value, mask->var);
lp_build_flow_scope_begin(flow);
lp_build_flow_scope_declare(flow, &mask->value);
lp_build_flow_skip_begin(flow);
lp_build_flow_skip_begin(&mask->skip, gallivm);
}
LLVMValueRef
lp_build_mask_value(struct lp_build_mask_context *mask)
{
return LLVMBuildLoad(mask->skip.gallivm->builder, mask->var, "");
lp_build_mask_check(mask);
}
@@ -185,10 +504,9 @@ void
lp_build_mask_update(struct lp_build_mask_context *mask,
LLVMValueRef value)
{
value = LLVMBuildAnd(mask->skip.gallivm->builder,
lp_build_mask_value(mask),
value, "");
LLVMBuildStore(mask->skip.gallivm->builder, value, mask->var);
mask->value = LLVMBuildAnd( mask->flow->builder, mask->value, value, "");
lp_build_mask_check(mask);
}
@@ -198,42 +516,42 @@ lp_build_mask_update(struct lp_build_mask_context *mask,
LLVMValueRef
lp_build_mask_end(struct lp_build_mask_context *mask)
{
lp_build_flow_skip_end(&mask->skip);
return lp_build_mask_value(mask);
lp_build_flow_skip_end(mask->flow);
lp_build_flow_scope_end(mask->flow);
return mask->value;
}
void
lp_build_loop_begin(struct lp_build_loop_state *state,
struct gallivm_state *gallivm,
LLVMValueRef start)
lp_build_loop_begin(LLVMBuilderRef builder,
LLVMValueRef start,
struct lp_build_loop_state *state)
{
LLVMBuilderRef builder = gallivm->builder;
LLVMBasicBlockRef block = LLVMGetInsertBlock(builder);
LLVMValueRef function = LLVMGetBasicBlockParent(block);
state->block = lp_build_insert_new_block(gallivm, "loop_begin");
state->counter_var = lp_build_alloca(gallivm, LLVMTypeOf(start), "loop_counter");
state->gallivm = gallivm;
LLVMBuildStore(builder, start, state->counter_var);
state->block = LLVMAppendBasicBlock(function, "loop");
LLVMBuildBr(builder, state->block);
LLVMPositionBuilderAtEnd(builder, state->block);
state->counter = LLVMBuildLoad(builder, state->counter_var, "");
state->counter = LLVMBuildPhi(builder, LLVMTypeOf(start), "");
LLVMAddIncoming(state->counter, &start, &block, 1);
}
void
lp_build_loop_end_cond(struct lp_build_loop_state *state,
LLVMValueRef end,
LLVMValueRef step,
LLVMIntPredicate llvm_cond)
lp_build_loop_end(LLVMBuilderRef builder,
LLVMValueRef end,
LLVMValueRef step,
struct lp_build_loop_state *state)
{
LLVMBuilderRef builder = state->gallivm->builder;
LLVMBasicBlockRef block = LLVMGetInsertBlock(builder);
LLVMValueRef function = LLVMGetBasicBlockParent(block);
LLVMValueRef next;
LLVMValueRef cond;
LLVMBasicBlockRef after_block;
@@ -243,26 +561,44 @@ lp_build_loop_end_cond(struct lp_build_loop_state *state,
next = LLVMBuildAdd(builder, state->counter, step, "");
LLVMBuildStore(builder, next, state->counter_var);
cond = LLVMBuildICmp(builder, LLVMIntNE, next, end, "");
cond = LLVMBuildICmp(builder, llvm_cond, next, end, "");
after_block = lp_build_insert_new_block(state->gallivm, "loop_end");
after_block = LLVMAppendBasicBlock(function, "");
LLVMBuildCondBr(builder, cond, after_block, state->block);
LLVMPositionBuilderAtEnd(builder, after_block);
LLVMAddIncoming(state->counter, &next, &block, 1);
state->counter = LLVMBuildLoad(builder, state->counter_var, "");
LLVMPositionBuilderAtEnd(builder, after_block);
}
void
lp_build_loop_end(struct lp_build_loop_state *state,
LLVMValueRef end,
LLVMValueRef step)
lp_build_loop_end_cond(LLVMBuilderRef builder,
LLVMValueRef end,
LLVMValueRef step,
int llvm_cond,
struct lp_build_loop_state *state)
{
lp_build_loop_end_cond(state, end, step, LLVMIntNE);
LLVMBasicBlockRef block = LLVMGetInsertBlock(builder);
LLVMValueRef function = LLVMGetBasicBlockParent(block);
LLVMValueRef next;
LLVMValueRef cond;
LLVMBasicBlockRef after_block;
if (!step)
step = LLVMConstInt(LLVMTypeOf(end), 1, 0);
next = LLVMBuildAdd(builder, state->counter, step, "");
cond = LLVMBuildICmp(builder, llvm_cond, next, end, "");
after_block = LLVMAppendBasicBlock(function, "");
LLVMBuildCondBr(builder, cond, after_block, state->block);
LLVMAddIncoming(state->counter, &next, &block, 1);
LLVMPositionBuilderAtEnd(builder, after_block);
}
@@ -280,16 +616,24 @@ lp_build_loop_end(struct lp_build_loop_state *state,
Is built with:
// x needs an alloca variable
x = lp_build_alloca(builder, type, "x");
LLVMValueRef x = LLVMGetUndef(); // or something else
flow = lp_build_flow_create(builder);
lp_build_if(ctx, builder, cond);
LLVMBuildStore(LLVMBuildAdd(1, 2), x);
lp_build_else(ctx);
LLVMBuildStore(LLVMBuildAdd(2, 3). x);
lp_build_endif(ctx);
lp_build_flow_scope_begin(flow);
// x needs a phi node
lp_build_flow_scope_declare(flow, &x);
lp_build_if(ctx, flow, builder, cond);
x = LLVMAdd(1, 2);
lp_build_else(ctx);
x = LLVMAdd(2, 3);
lp_build_endif(ctx);
lp_build_flow_scope_end(flow);
lp_build_flow_destroy(flow);
*/
@@ -298,28 +642,53 @@ lp_build_loop_end(struct lp_build_loop_state *state,
* Begin an if/else/endif construct.
*/
void
lp_build_if(struct lp_build_if_state *ifthen,
struct gallivm_state *gallivm,
lp_build_if(struct lp_build_if_state *ctx,
struct lp_build_flow_context *flow,
LLVMBuilderRef builder,
LLVMValueRef condition)
{
LLVMBasicBlockRef block = LLVMGetInsertBlock(gallivm->builder);
LLVMBasicBlockRef block = LLVMGetInsertBlock(builder);
struct lp_build_flow_if *ifthen;
unsigned i;
memset(ifthen, 0, sizeof *ifthen);
ifthen->gallivm = gallivm;
memset(ctx, 0, sizeof(*ctx));
ctx->builder = builder;
ctx->flow = flow;
/* push/create new scope */
ifthen = &lp_build_flow_push(flow, LP_BUILD_FLOW_IF)->ifthen;
assert(ifthen);
ifthen->num_variables = flow->num_variables;
ifthen->condition = condition;
ifthen->entry_block = block;
/* create a Phi node for each variable in this flow scope */
ifthen->phi = MALLOC(ifthen->num_variables * sizeof(*ifthen->phi));
if (!ifthen->phi) {
ifthen->num_variables = 0;
return;
}
/* create endif/merge basic block for the phi functions */
ifthen->merge_block = lp_build_insert_new_block(gallivm, "endif-block");
ifthen->merge_block = lp_build_insert_new_block(builder, "endif-block");
LLVMPositionBuilderAtEnd(builder, ifthen->merge_block);
/* create a phi node for each variable */
for (i = 0; i < flow->num_variables; i++) {
ifthen->phi[i] = LLVMBuildPhi(builder, LLVMTypeOf(*flow->variables[i]), "");
/* add add the initial value of the var from the entry block */
if (!LLVMIsUndef(*flow->variables[i]))
LLVMAddIncoming(ifthen->phi[i], flow->variables[i],
&ifthen->entry_block, 1);
}
/* create/insert true_block before merge_block */
ifthen->true_block =
LLVMInsertBasicBlockInContext(gallivm->context,
ifthen->merge_block,
"if-true-block");
ifthen->true_block = LLVMInsertBasicBlock(ifthen->merge_block, "if-true-block");
/* successive code goes into the true block */
LLVMPositionBuilderAtEnd(gallivm->builder, ifthen->true_block);
LLVMPositionBuilderAtEnd(builder, ifthen->true_block);
}
@@ -327,21 +696,27 @@ lp_build_if(struct lp_build_if_state *ifthen,
* Begin else-part of a conditional
*/
void
lp_build_else(struct lp_build_if_state *ifthen)
lp_build_else(struct lp_build_if_state *ctx)
{
LLVMBuilderRef builder = ifthen->gallivm->builder;
struct lp_build_flow_context *flow = ctx->flow;
struct lp_build_flow_if *ifthen;
unsigned i;
/* Append an unconditional Br(anch) instruction on the true_block */
LLVMBuildBr(builder, ifthen->merge_block);
ifthen = &lp_build_flow_peek(flow, LP_BUILD_FLOW_IF)->ifthen;
assert(ifthen);
/* for each variable, update the Phi node with a (variable, block) pair */
LLVMPositionBuilderAtEnd(ctx->builder, ifthen->merge_block);
for (i = 0; i < flow->num_variables; i++) {
assert(*flow->variables[i]);
LLVMAddIncoming(ifthen->phi[i], flow->variables[i], &ifthen->true_block, 1);
}
/* create/insert false_block before the merge block */
ifthen->false_block =
LLVMInsertBasicBlockInContext(ifthen->gallivm->context,
ifthen->merge_block,
"if-false-block");
ifthen->false_block = LLVMInsertBasicBlock(ifthen->merge_block, "if-false-block");
/* successive code goes into the else block */
LLVMPositionBuilderAtEnd(builder, ifthen->false_block);
LLVMPositionBuilderAtEnd(ctx->builder, ifthen->false_block);
}
@@ -349,32 +724,75 @@ lp_build_else(struct lp_build_if_state *ifthen)
* End a conditional.
*/
void
lp_build_endif(struct lp_build_if_state *ifthen)
lp_build_endif(struct lp_build_if_state *ctx)
{
LLVMBuilderRef builder = ifthen->gallivm->builder;
struct lp_build_flow_context *flow = ctx->flow;
struct lp_build_flow_if *ifthen;
LLVMBasicBlockRef curBlock = LLVMGetInsertBlock(ctx->builder);
unsigned i;
ifthen = &lp_build_flow_pop(flow, LP_BUILD_FLOW_IF)->ifthen;
assert(ifthen);
/* Insert branch to the merge block from current block */
LLVMBuildBr(builder, ifthen->merge_block);
LLVMBuildBr(ctx->builder, ifthen->merge_block);
/*
* Now patch in the various branch instructions.
*/
if (ifthen->false_block) {
LLVMPositionBuilderAtEnd(ctx->builder, ifthen->merge_block);
/* for each variable, update the Phi node with a (variable, block) pair */
for (i = 0; i < flow->num_variables; i++) {
assert(*flow->variables[i]);
LLVMAddIncoming(ifthen->phi[i], flow->variables[i], &curBlock, 1);
/* replace the variable ref with the phi function */
*flow->variables[i] = ifthen->phi[i];
}
}
else {
/* no else clause */
LLVMPositionBuilderAtEnd(ctx->builder, ifthen->merge_block);
for (i = 0; i < flow->num_variables; i++) {
assert(*flow->variables[i]);
LLVMAddIncoming(ifthen->phi[i], flow->variables[i], &ifthen->true_block, 1);
/* replace the variable ref with the phi function */
*flow->variables[i] = ifthen->phi[i];
}
}
FREE(ifthen->phi);
/***
*** Now patch in the various branch instructions.
***/
/* Insert the conditional branch instruction at the end of entry_block */
LLVMPositionBuilderAtEnd(builder, ifthen->entry_block);
LLVMPositionBuilderAtEnd(ctx->builder, ifthen->entry_block);
if (ifthen->false_block) {
/* we have an else clause */
LLVMBuildCondBr(builder, ifthen->condition,
LLVMBuildCondBr(ctx->builder, ifthen->condition,
ifthen->true_block, ifthen->false_block);
}
else {
/* no else clause */
LLVMBuildCondBr(builder, ifthen->condition,
LLVMBuildCondBr(ctx->builder, ifthen->condition,
ifthen->true_block, ifthen->merge_block);
}
/* Insert branch from end of true_block to merge_block */
if (ifthen->false_block) {
/* Append an unconditional Br(anch) instruction on the true_block */
LLVMPositionBuilderAtEnd(ctx->builder, ifthen->true_block);
LLVMBuildBr(ctx->builder, ifthen->merge_block);
}
else {
/* No else clause.
* Note that we've already inserted the branch at the end of
* true_block. See the very first LLVMBuildBr() call in this function.
*/
}
/* Resume building code at end of the ifthen->merge_block */
LLVMPositionBuilderAtEnd(builder, ifthen->merge_block);
LLVMPositionBuilderAtEnd(ctx->builder, ifthen->merge_block);
}
@@ -394,16 +812,15 @@ lp_build_endif(struct lp_build_if_state *ifthen)
* - http://www.llvm.org/docs/tutorial/OCamlLangImpl7.html#memory
*/
LLVMValueRef
lp_build_alloca(struct gallivm_state *gallivm,
lp_build_alloca(LLVMBuilderRef builder,
LLVMTypeRef type,
const char *name)
{
LLVMBuilderRef builder = gallivm->builder;
LLVMBasicBlockRef current_block = LLVMGetInsertBlock(builder);
LLVMValueRef function = LLVMGetBasicBlockParent(current_block);
LLVMBasicBlockRef first_block = LLVMGetEntryBasicBlock(function);
LLVMValueRef first_instr = LLVMGetFirstInstruction(first_block);
LLVMBuilderRef first_builder = LLVMCreateBuilderInContext(gallivm->context);
LLVMBuilderRef first_builder = LLVMCreateBuilder();
LLVMValueRef res;
if (first_instr) {
@@ -413,7 +830,6 @@ lp_build_alloca(struct gallivm_state *gallivm,
}
res = LLVMBuildAlloca(first_builder, type, name);
LLVMBuildStore(builder, LLVMConstNull(type), res);
LLVMDisposeBuilder(first_builder);
@@ -436,17 +852,16 @@ lp_build_alloca(struct gallivm_state *gallivm,
* - http://www.llvm.org/docs/tutorial/OCamlLangImpl7.html#memory
*/
LLVMValueRef
lp_build_array_alloca(struct gallivm_state *gallivm,
lp_build_array_alloca(LLVMBuilderRef builder,
LLVMTypeRef type,
LLVMValueRef count,
const char *name)
{
LLVMBuilderRef builder = gallivm->builder;
LLVMBasicBlockRef current_block = LLVMGetInsertBlock(builder);
LLVMValueRef function = LLVMGetBasicBlockParent(current_block);
LLVMBasicBlockRef first_block = LLVMGetEntryBasicBlock(function);
LLVMValueRef first_instr = LLVMGetFirstInstruction(first_block);
LLVMBuilderRef first_builder = LLVMCreateBuilderInContext(gallivm->context);
LLVMBuilderRef first_builder = LLVMCreateBuilder();
LLVMValueRef res;
if (first_instr) {

View File

@@ -41,49 +41,52 @@
struct lp_type;
/**
* Early exit. Useful to skip to the end of a function or block when
* the execution mask becomes zero or when there is an error condition.
*/
struct lp_build_skip_context
{
struct gallivm_state *gallivm;
struct lp_build_flow_context;
/** Block to skip to */
LLVMBasicBlockRef block;
};
struct lp_build_flow_context *
lp_build_flow_create(LLVMBuilderRef builder);
void
lp_build_flow_skip_begin(struct lp_build_skip_context *ctx,
struct gallivm_state *gallivm);
lp_build_flow_destroy(struct lp_build_flow_context *flow);
void
lp_build_flow_skip_cond_break(struct lp_build_skip_context *ctx,
lp_build_flow_scope_begin(struct lp_build_flow_context *flow);
void
lp_build_flow_scope_declare(struct lp_build_flow_context *flow,
LLVMValueRef *variable);
void
lp_build_flow_scope_end(struct lp_build_flow_context *flow);
void
lp_build_flow_skip_begin(struct lp_build_flow_context *flow);
void
lp_build_flow_skip_cond_break(struct lp_build_flow_context *flow,
LLVMValueRef cond);
void
lp_build_flow_skip_end(struct lp_build_skip_context *ctx);
lp_build_flow_skip_end(struct lp_build_flow_context *flow);
struct lp_build_mask_context
{
struct lp_build_skip_context skip;
struct lp_build_flow_context *flow;
LLVMTypeRef reg_type;
LLVMValueRef var;
LLVMValueRef value;
};
void
lp_build_mask_begin(struct lp_build_mask_context *mask,
struct gallivm_state *gallivm,
struct lp_build_flow_context *flow,
struct lp_type type,
LLVMValueRef value);
LLVMValueRef
lp_build_mask_value(struct lp_build_mask_context *mask);
/**
* Bitwise AND the mask with the given value, if a previous mask was set.
*/
@@ -91,9 +94,6 @@ void
lp_build_mask_update(struct lp_build_mask_context *mask,
LLVMValueRef value);
void
lp_build_mask_check(struct lp_build_mask_context *mask);
LLVMValueRef
lp_build_mask_end(struct lp_build_mask_context *mask);
@@ -107,48 +107,44 @@ lp_build_mask_end(struct lp_build_mask_context *mask);
*/
struct lp_build_loop_state
{
LLVMBasicBlockRef block;
LLVMValueRef counter_var;
LLVMValueRef counter;
struct gallivm_state *gallivm;
LLVMBasicBlockRef block;
LLVMValueRef counter;
};
void
lp_build_loop_begin(struct lp_build_loop_state *state,
struct gallivm_state *gallivm,
LLVMValueRef start);
lp_build_loop_begin(LLVMBuilderRef builder,
LLVMValueRef start,
struct lp_build_loop_state *state);
void
lp_build_loop_end(struct lp_build_loop_state *state,
lp_build_loop_end(LLVMBuilderRef builder,
LLVMValueRef end,
LLVMValueRef step);
LLVMValueRef step,
struct lp_build_loop_state *state);
void
lp_build_loop_end_cond(struct lp_build_loop_state *state,
lp_build_loop_end_cond(LLVMBuilderRef builder,
LLVMValueRef end,
LLVMValueRef step,
LLVMIntPredicate cond);
int cond, /* LLVM condition */
struct lp_build_loop_state *state);
/**
* if/else/endif.
*/
struct lp_build_if_state
{
struct gallivm_state *gallivm;
LLVMValueRef condition;
LLVMBasicBlockRef entry_block;
LLVMBasicBlockRef true_block;
LLVMBasicBlockRef false_block;
LLVMBasicBlockRef merge_block;
LLVMBuilderRef builder;
struct lp_build_flow_context *flow;
};
void
lp_build_if(struct lp_build_if_state *ctx,
struct gallivm_state *gallivm,
struct lp_build_flow_context *flow,
LLVMBuilderRef builder,
LLVMValueRef condition);
void
@@ -158,15 +154,15 @@ void
lp_build_endif(struct lp_build_if_state *ctx);
LLVMBasicBlockRef
lp_build_insert_new_block(struct gallivm_state *gallivm, const char *name);
lp_build_insert_new_block(LLVMBuilderRef builder, const char *name);
LLVMValueRef
lp_build_alloca(struct gallivm_state *gallivm,
lp_build_alloca(LLVMBuilderRef builder,
LLVMTypeRef type,
const char *name);
LLVMValueRef
lp_build_array_alloca(struct gallivm_state *gallivm,
lp_build_array_alloca(LLVMBuilderRef builder,
LLVMTypeRef type,
LLVMValueRef count,
const char *name);

Some files were not shown because too many files have changed in this diff Show More