mesa.git
11 years agoscons: remove nouveau build
Andreas Boll [Fri, 3 May 2013 09:41:00 +0000 (11:41 +0200)]
scons: remove nouveau build

One build system for linux/unix only drivers should be enough.
Additionally the nouveau target was disabled anyway.

Acked-by: Jose Fonseca <jfonseca@vmware.com>
11 years agoscons: remove radeon build
Andreas Boll [Fri, 3 May 2013 09:33:35 +0000 (11:33 +0200)]
scons: remove radeon build

One build system for linux/unix only drivers should be enough.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=48694

Acked-by: Jose Fonseca <jfonseca@vmware.com>
11 years agor600g: don't emit surface_sync after FLUSH_AND_INV_EVENT
Alex Deucher [Fri, 3 May 2013 13:56:31 +0000 (09:56 -0400)]
r600g: don't emit surface_sync after FLUSH_AND_INV_EVENT

It shouldn't be needed since the FLUSH_AND_INV_EVENT has already
made sure the destination caches are flushed.  Additionally,
we didn't previously emit the surface_sync until this commit:
http://cgit.freedesktop.org/mesa/mesa/commit/?id=e5e4c07e7964a3258ed02b530bcdc24c0650204b
Emitting them together causes hangs in compute on cayman/TN
and hangs in Heaven on evergreen.

Note: this patch is a candidate for the 9.1 branch, but requires:
http://cgit.freedesktop.org/mesa/mesa/commit/?id=156bcca62c9f4e79e78929f72bc085757f36a65a
as well.

Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
11 years agor600g/sb: zero-initialize bytecode structs
Vadim Girlin [Fri, 3 May 2013 12:53:27 +0000 (16:53 +0400)]
r600g/sb: zero-initialize bytecode structs

Signed-off-by: Vadim Girlin <vadimgirlin@gmail.com>
11 years agor600g/sb: fix constant propagation in gvn pass
Vadim Girlin [Fri, 3 May 2013 09:01:31 +0000 (13:01 +0400)]
r600g/sb: fix constant propagation in gvn pass

Fixes the bug that prevented propagation of literals in some cases.

Signed-off-by: Vadim Girlin <vadimgirlin@gmail.com>
11 years agor600g/sb: don't run unnecessary passes
Vadim Girlin [Thu, 2 May 2013 15:36:52 +0000 (19:36 +0400)]
r600g/sb: don't run unnecessary passes

Signed-off-by: Vadim Girlin <vadimgirlin@gmail.com>
11 years agor600g/sb: silence warnings with gcc 4.8
Vadim Girlin [Thu, 2 May 2013 06:36:18 +0000 (10:36 +0400)]
r600g/sb: silence warnings with gcc 4.8

Signed-off-by: Vadim Girlin <vadimgirlin@gmail.com>
11 years agor600g/sb: fix handling of interference sets in post_scheduler
Vadim Girlin [Thu, 2 May 2013 03:53:00 +0000 (07:53 +0400)]
r600g/sb: fix handling of interference sets in post_scheduler

post_scheduler clears interference set for reallocatable values when
the value becomes live first time, and then updates it to take into
account modified order of operations, but this was not handled properly
if the value appears first time as a source in copy operation.

Fixes issues with webgl demo: http://madebyevan.com/webgl-water/

Signed-off-by: Vadim Girlin <vadimgirlin@gmail.com>
11 years agor600g/sb: fix allocation of indirectly addressed input arrays
Vadim Girlin [Thu, 2 May 2013 03:56:32 +0000 (07:56 +0400)]
r600g/sb: fix allocation of indirectly addressed input arrays

Some inputs may be preloaded into predefined GPRs,
so we can't reallocate arrays with such inputs.

Fixes issues with webgl demo: http://oos.moxiecode.com/js_webgl/snake/

Signed-off-by: Vadim Girlin <vadimgirlin@gmail.com>
11 years agor600g/sb: use hex instead of binary constants
Vadim Girlin [Wed, 1 May 2013 17:33:33 +0000 (21:33 +0400)]
r600g/sb: use hex instead of binary constants

This should fix build issues with GCC < 4.3

Signed-off-by: Vadim Girlin <vadimgirlin@gmail.com>
11 years agor600g: use old shader disassembler by default
Vadim Girlin [Fri, 3 May 2013 08:01:20 +0000 (12:01 +0400)]
r600g: use old shader disassembler by default

New disassembler is not completely isolated yet from further processing
in r600g/sb that is not required for printing the dump, so it has higher
probability to fail in case of any unexpected features in the bytecode.

This patch adds "sbdisasm" flag for R600_DEBUG that allows to use new
disassembler in r600g/sb for shader dumps when shader optimization
is not enabled.

If shader optimization is enabled, new disassembler is used by default.

Signed-off-by: Vadim Girlin <vadimgirlin@gmail.com>
11 years agoradeon/uvd: enable interlaced buffers by default
Christian König [Thu, 2 May 2013 14:19:41 +0000 (16:19 +0200)]
radeon/uvd: enable interlaced buffers by default

Kills tilling on UVD buffers, but we currently don't really need that.

Signed-off-by: Christian König <christian.koenig@amd.com>
11 years agovl/idct: fix for commit 7d2f2a0c890b1993532a45c8c392c28950ddc06e
Christian König [Thu, 2 May 2013 14:02:05 +0000 (16:02 +0200)]
vl/idct: fix for commit 7d2f2a0c890b1993532a45c8c392c28950ddc06e

We still need the option for handling 3D textures as well.

Should fix: https://bugs.freedesktop.org/show_bug.cgi?id=64143

Signed-off-by: Christian König <christian.koenig@amd.com>
11 years agovl/buffers: fix typo in function name
Christian König [Thu, 2 May 2013 13:42:24 +0000 (15:42 +0200)]
vl/buffers: fix typo in function name

Signed-off-by: Christian König <christian.koenig@amd.com>
11 years agoradeon/uvd: fix some MPEG4 artifacts
Christian König [Wed, 1 May 2013 12:33:49 +0000 (14:33 +0200)]
radeon/uvd: fix some MPEG4 artifacts

Still not perfect, but a step in the right direction.

Signed-off-by: Christian König <christian.koenig@amd.com>
11 years agodraw: Update for u_assembled_primitive -> u_assembled_prim rename.
José Fonseca [Fri, 3 May 2013 07:35:17 +0000 (08:35 +0100)]
draw: Update for u_assembled_primitive -> u_assembled_prim rename.

Mesa build is too complex to rely on successful builds. On refactorings
it is always a good idea to use git grep to prevent missing cases:

  $ git grep u_assembled_primitive
  src/gallium/auxiliary/draw/draw_pt_fetch_shade_pipeline_llvm.c:      u_assembled_primitive(in_prim);

11 years agost/egl: fix bulid errors on Android 4.2
Chia-I Wu [Thu, 2 May 2013 08:13:28 +0000 (16:13 +0800)]
st/egl: fix bulid errors on Android 4.2

The differences from the previous releases that affect st/egl are

 - logging macros are prefixed with an 'A'
 - dequeueBuffer() and enqueueBuffer() require an additoinal argument for
   fence fd, acquired from libsync

Additionally, include gralloc_drm.h with extern "C".

11 years agoilo: use u_reduced_prims_for_vertices()
Chia-I Wu [Thu, 2 May 2013 06:32:30 +0000 (14:32 +0800)]
ilo: use u_reduced_prims_for_vertices()

We do not need our own prim_count() anymore.

11 years agoutil/prim: add u_reduced_prims_for_vertices()
Chia-I Wu [Thu, 2 May 2013 06:28:12 +0000 (14:28 +0800)]
util/prim: add u_reduced_prims_for_vertices()

The function returns the number of reduced/tessellated primitives for the
given vertex count.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Acked-by: Zack Rusin <zackr@vmware.com>
11 years agoutil/prim: assorted fixes for u_decomposed_prims_for_vertices()
Chia-I Wu [Thu, 2 May 2013 06:12:27 +0000 (14:12 +0800)]
util/prim: assorted fixes for u_decomposed_prims_for_vertices()

Switch to '>=' for comparisons, and it becomes obvious that the comparison for
PIPE_PRIM_QUAD_STRIP was wrong.

Add minimum vertex count check for PIPE_PRIM_LINE_LOOP.  Return 1 for
PIPE_PRIM_POLYGON with 3 vertices.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Acked-by: Zack Rusin <zackr@vmware.com>
11 years agoutil/prim: use vertex count info in u_validate_pipe_prim()
Chia-I Wu [Thu, 2 May 2013 05:49:30 +0000 (13:49 +0800)]
util/prim: use vertex count info in u_validate_pipe_prim()

As a side effect, primitives with adjacency are now correctly validated.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Acked-by: Zack Rusin <zackr@vmware.com>
11 years agoutil/prim: fix the name of the include guard
Chia-I Wu [Thu, 2 May 2013 05:46:21 +0000 (13:46 +0800)]
util/prim: fix the name of the include guard

It should be U_PRIM_H, not U_BLIT_H.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Acked-by: Zack Rusin <zackr@vmware.com>
11 years agodraw: use u_assembled_prim() instead of u_assembled_primitive()
Chia-I Wu [Thu, 2 May 2013 07:49:35 +0000 (15:49 +0800)]
draw: use u_assembled_prim() instead of u_assembled_primitive()

The latter function is also removed as a result of the change.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Acked-by: Zack Rusin <zackr@vmware.com>
11 years agoutil/prim: clean up and add comments
Chia-I Wu [Thu, 2 May 2013 05:29:59 +0000 (13:29 +0800)]
util/prim: clean up and add comments

Move together (or add) functions to decompose/reduce/assemble a primitive,
give them consistent names, and document them.  Add u_prim_vertex_count() so
that the vertex count information can be used elsewhere.

u_assembled_primitive() will be removed in a folow-on commit.

[olv: fix a warning when -Wold-style-declaration is enabled]

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Acked-by: Zack Rusin <zackr@vmware.com>
11 years agoutil/prim: fix primitive trimming for triangles with adjacency
Chia-I Wu [Thu, 2 May 2013 03:25:39 +0000 (11:25 +0800)]
util/prim: fix primitive trimming for triangles with adjacency

Fix for PIPE_PRIM_TRIANGLES_ADJACENCY and PIPE_PRIM_TRIANGLE_STRIP_ADJACENCY.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Acked-by: Zack Rusin <zackr@vmware.com>
11 years agoi965/vs: Add instruction scheduling.
Eric Anholt [Sat, 1 Dec 2012 00:13:34 +0000 (16:13 -0800)]
i965/vs: Add instruction scheduling.

While this is ignorant of dependency control, it's still good for a 0.39%
+/- 0.08% performance improvement on GLBenchmark 2.7 (n=548)

v2: Rewrite as a subclass of the base class for the FS instruction
    scheduler, inheriting the same latency information.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
11 years agoi965: Move most of the FS instruction scheduler code to a general class.
Eric Anholt [Mon, 29 Apr 2013 21:05:33 +0000 (14:05 -0700)]
i965: Move most of the FS instruction scheduler code to a general class.

About half of this is shareable with the VS code.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
11 years agoi965: Pull a couple of FS scheduling functions out to methods.
Eric Anholt [Mon, 29 Apr 2013 23:45:10 +0000 (16:45 -0700)]
i965: Pull a couple of FS scheduling functions out to methods.

These will get virtualized as we add VS scheduling support.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
11 years agoi965: Move FS instruction scheduling to a non-FS-specific file.
Eric Anholt [Mon, 29 Apr 2013 20:59:38 +0000 (13:59 -0700)]
i965: Move FS instruction scheduling to a non-FS-specific file.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
11 years agoi965: Share the register file enum between the two backends.
Eric Anholt [Mon, 29 Apr 2013 23:05:05 +0000 (16:05 -0700)]
i965: Share the register file enum between the two backends.

I need this so I can look at vec4 and fs registers' files from the same
.cpp file without namespaces.  As far as I can tell we never rely on the
particular numerical values of the files, though I thought it sounded like
a good idea when doing the VS (it turns out having 0 be BAD_FILE is nicer).

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
11 years agoi965: Make dump_instructions be a virtual method of the visitor.
Eric Anholt [Mon, 29 Apr 2013 21:21:14 +0000 (14:21 -0700)]
i965: Make dump_instructions be a virtual method of the visitor.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
11 years agoi965/vs: Do round-robin register allocation on gen6+ like we do in the FS.
Eric Anholt [Mon, 29 Apr 2013 18:48:22 +0000 (11:48 -0700)]
i965/vs: Do round-robin register allocation on gen6+ like we do in the FS.

This will free instruction scheduling to make better choices.  No
statistically significant performance difference on GLB2.7 (n=93).

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
11 years agowayland: Make eglQueryBufferWL succeed for width and height requests too
Rob Bradford [Wed, 10 Apr 2013 10:36:24 +0000 (11:36 +0100)]
wayland: Make eglQueryBufferWL succeed for width and height requests too

Following the addition of the EGL_WIDTH and EGL_HEIGHT this function should
return EGL_TRUE for those requested attributes too.

11 years agodraw/gs: don't crash when vs/gs signatures don't match
Zack Rusin [Thu, 2 May 2013 01:09:32 +0000 (21:09 -0400)]
draw/gs: don't crash when vs/gs signatures don't match

instead of crashing just fill zeros at the input slots that don't
match, that's the mandated behavior and it avoids debug asserts.

Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
11 years agotgsi: allow negation of all integer types
Zack Rusin [Sun, 28 Apr 2013 14:50:55 +0000 (10:50 -0400)]
tgsi: allow negation of all integer types

It's valid because we reuse certain arithmetic operations
for both signed and unsigned types (e.g. uadd, umad, which
have a bit unfortunate naming)

Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
11 years agoi965: Fix SNB GPU hangs when a blorp batch is the first thing to execute.
Eric Anholt [Wed, 1 May 2013 23:08:12 +0000 (16:08 -0700)]
i965: Fix SNB GPU hangs when a blorp batch is the first thing to execute.

The GPU apparently goes looking for constants even though there are no
shader stages enabled, and gets stuck because we haven't told it there are
no constants to collect.  If any other user of the 3D pipeline had run
(even the Render accel of the X server!) since power on, then the in-GPU
constant buffers would have been set up with some contents we didn't use,
and we would succeed.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=56416
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Tested-by: Dave Airlie <airlied@redhat.com>
NOTE: This is a candidate for the stable branches.

11 years agor600g: Don't set the dest cache bits on surface sync for R600_CONTEXT_FLUSH_AND_INV
Tom Stellard [Fri, 26 Apr 2013 17:21:06 +0000 (13:21 -0400)]
r600g: Don't set the dest cache bits on surface sync for R600_CONTEXT_FLUSH_AND_INV

We are already emitting a EVENT_TYPE_CACHE_FLUSH_AND_INV_EVENT packet
when this flush flag is set, so flushing the dest caches with a
SURFACE_SYNC should not be necessary.

The motivation for this change is that emitting a SURFACE_SYNC packet with
the CB bits set was causing compute shaders to hang on Cayman.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
11 years agor600g/compute: Fix build error in debug code
Tom Stellard [Fri, 26 Apr 2013 17:21:04 +0000 (13:21 -0400)]
r600g/compute: Fix build error in debug code

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
11 years agoradeon: Fix build with LLVM 3.3
Armin K [Thu, 2 May 2013 10:55:05 +0000 (12:55 +0200)]
radeon: Fix build with LLVM 3.3

Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
11 years agogallivm: Fix build with LLVM 3.3
Armin K [Thu, 2 May 2013 10:55:04 +0000 (12:55 +0200)]
gallivm: Fix build with LLVM 3.3

Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
11 years agomesa: update comments, simplify code in vtxfmt.c
Brian Paul [Thu, 2 May 2013 01:15:33 +0000 (19:15 -0600)]
mesa: update comments, simplify code in vtxfmt.c

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: update GLvertexformat comments
Brian Paul [Thu, 2 May 2013 01:15:33 +0000 (19:15 -0600)]
mesa: update GLvertexformat comments

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: remove GLvertexformat::EvalMesh1(), EvalMesh2()
Brian Paul [Thu, 2 May 2013 01:15:33 +0000 (19:15 -0600)]
mesa: remove GLvertexformat::EvalMesh1(), EvalMesh2()

See previous commit comments.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: remove GLvertexformat::Rectf()
Brian Paul [Thu, 2 May 2013 01:15:33 +0000 (19:15 -0600)]
mesa: remove GLvertexformat::Rectf()

As with the glDraw* functions, this doesn't have to be in GLvertexformat.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: simplify dispatch for glDraw* functions
Brian Paul [Thu, 2 May 2013 01:15:33 +0000 (19:15 -0600)]
mesa: simplify dispatch for glDraw* functions

Remove all the glDraw* functions from the GLvertexformat structure.
The point of that dispatch struct is to handle all the functions which
dispatch differently depending on whether we're inside glBegin/End.
glDraw* are never allowed inside glBegin/End so we can remove those
entries.

This simplifies the code paths and gets rid of quite a bit of code.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agovbo: add new vbo_initialize_exec_dispatch(), vbo_initialize_save_dispatch()
Brian Paul [Thu, 2 May 2013 01:15:33 +0000 (19:15 -0600)]
vbo: add new vbo_initialize_exec_dispatch(), vbo_initialize_save_dispatch()

First step in simplifying the vertex array / glDraw dispatch code.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: remove _MESA_INIT_EVAL_VTXFMT() macro
Brian Paul [Thu, 2 May 2013 01:15:33 +0000 (19:15 -0600)]
mesa: remove _MESA_INIT_EVAL_VTXFMT() macro

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: remove _MESA_INIT_ARRAYELT_VTXFMT() macro
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
mesa: remove _MESA_INIT_ARRAYELT_VTXFMT() macro

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: remove _MESA_INIT_DLIST_VTXFMT() macro
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
mesa: remove _MESA_INIT_DLIST_VTXFMT() macro

Just expand the code.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: change _mesa_inside_dlist_begin_end() to handle PRIM_UNKNOWN
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
mesa: change _mesa_inside_dlist_begin_end() to handle PRIM_UNKNOWN

If the currently compiled primitive state is PRIM_UNKNOWN we should
not return true from _mesa_inside_dlist_begin_end().  This lets us
simplify the calls to that function.

Note, the call to _mesa_inside_dlist_begin_end() in vbo_save_EndList()
should have probably been checking for PRIM_UNKNOWN too, but it wasn't.
So there's no code change change.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: add names of geometry shader prims in gl_enums.py
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
mesa: add names of geometry shader prims in gl_enums.py

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agovbo: fix initial value of ctx->Driver.CurrentSavePrimitive
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
vbo: fix initial value of ctx->Driver.CurrentSavePrimitive

This is set during context creation/initialization.  We know we're
not inside glBegin/glEnd at this point so use PRIM_OUTSIDE_BEGIN_END.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agovbo: fix error detection in vbo_save_playback_vertex_list()
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
vbo: fix error detection in vbo_save_playback_vertex_list()

The old code didn't make sense.  The clause in question did the
same thing as the next else-if clause.  If we're already executing
a glBegin/End pair and we're starting a new primitive, that's an
error.

Fixes more failures in piglit gl-1.0-beginend-coverage test.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: comments, formatting fixes in dlist code
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
mesa: comments, formatting fixes in dlist code

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agovbo: remove redundant vfmt->Begin = _save_Begin assignment
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
vbo: remove redundant vfmt->Begin = _save_Begin assignment

The same assignment appears later in the function.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: don't install glDraw* functions into the BeginEnd dispatch table
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
mesa: don't install glDraw* functions into the BeginEnd dispatch table

Functions like glDrawArrays, glDrawElements, etc. are illegal between
glBegin/glEnd and should generate GL_INVALID_OPERATION.

Fixes several piglit gl-1.0-beginend-coverage failures.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agovbo: fix parameter validation for saving dlist glDraw* functions
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
vbo: fix parameter validation for saving dlist glDraw* functions

The _save_OBE_DrawArrays/Elements/RangeElements() functions are
called when building a display list and we know we're outside
glBegin/End.

We shouldn't call the normal _mesa_validate_DrawArrays/Elements()
functions here because those functions only work properly in immediate
mode or during dlist execution.  At dlist compile time, we can't call
_mesa_update_state(), etc. and examine the current state since it won't
apply when the list is executed later.

Fixes several failures in piglit's gl-1.0-beginend-coverage test.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: add missing error check in _mesa_EndList()
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
mesa: add missing error check in _mesa_EndList()

If we're in GL_COMPILE_AND_EXECUTE mode and inside glBegin, calling
glEndList() should generate an error.

Fixes a failure in piglit's gl-1.0-beginend-coverage test.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: remove unused PRIM_INSIDE_UNKNOWN_PRIM constant
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
mesa: remove unused PRIM_INSIDE_UNKNOWN_PRIM constant

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: simplify save_Begin() error checking
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
mesa: simplify save_Begin() error checking

The old code was hard to understand and not entirely correct.
Note that PRIM_INSIDE_UNKNOWN_PRIM is no longer set anywhere so
we'll be able to remove that next.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: refactor _mesa_valid_prim_mode()
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
mesa: refactor _mesa_valid_prim_mode()

...in terms of new _mesa_is_valid_prim_mode().  We need a mode validater
function that doesn't depend on current state for the display list code.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: fix CurrentSavePrimitive <= GL_POLYGON tests
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
mesa: fix CurrentSavePrimitive <= GL_POLYGON tests

Use the new PRIM_MAX value instead so that new geometry shader primitive
types are accounted for.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: adjust PRIM_x constants for geometry shaders
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
mesa: adjust PRIM_x constants for geometry shaders

These values pertain to display lists, and the new types of geometry
shader primitives can be used in display lists.

And add new PRIM_MAX constant for follow-on changes.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agomesa: fix save_ShadeModel() logic and add new comments
Brian Paul [Thu, 2 May 2013 01:15:32 +0000 (19:15 -0600)]
mesa: fix save_ShadeModel() logic and add new comments

This removes the test for _mesa_inside_dlist_begin_end().
If ctx->Driver.CurrentSavePrimitive==PRIM_UNKNOWN (the initial value),
_mesa_inside_dlist_begin_end() will, confusingly, return TRUE.
So we didn't set the ctx->ListState.Current.ShadeModel value and it
remained in its indeterminate state.

This didn't effect correctness, but it defeated the intended optimization
of dropping redundant glShadeModel() state changes in order to
coalesce sequences of drawing commands.

Verified with new piglit gl-1.0-dlist-shademodel test.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
11 years agogallivm: Fix altivec intrinsics for 8xi16 add/sub
Adam Jackson [Thu, 25 Apr 2013 19:45:38 +0000 (15:45 -0400)]
gallivm: Fix altivec intrinsics for 8xi16 add/sub

Signed-off-by: Adam Jackson <ajax@redhat.com>
11 years agor600/sb: Fix build failure with non-standard libdrm installation prefix
Lauri Kasanen [Wed, 1 May 2013 16:35:07 +0000 (19:35 +0300)]
r600/sb: Fix build failure with non-standard libdrm installation prefix

Just like radeon/uvd, r600/sb fails to find the libdrm includes.

Signed-off-by: Lauri Kasanen <cand@gmx.com>
11 years agoradeon/uvd: Fix build failure with non-standard libdrm installation prefix
Lauri Kasanen [Wed, 1 May 2013 16:26:06 +0000 (19:26 +0300)]
radeon/uvd: Fix build failure with non-standard libdrm installation prefix

Without this patch, radeon_uvd failed to find the libdrm includes:

In file included from radeon_uvd.c:48:
../../winsys/radeon/drm/radeon_winsys.h:44:35: error:
libdrm/radeon_surface.h: No such file or directory

Signed-off-by: Lauri Kasanen <cand@gmx.com>
11 years agomesa: implement glFramebufferTexture
Jordan Justen [Thu, 18 Apr 2013 17:46:12 +0000 (10:46 -0700)]
mesa: implement glFramebufferTexture

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
11 years agomesa: add Layered field to framebuffers
Jordan Justen [Thu, 18 Apr 2013 17:20:05 +0000 (10:20 -0700)]
mesa: add Layered field to framebuffers

When checking framebuffer completeness, we test each attachment.
We verify that all attachments are consistent in terms of layers.

1. They must all be layered, or all non-layered
2. If they are layered, they must match in depth

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
11 years agomesa: add renderbuffer attachment Layered field
Jordan Justen [Thu, 18 Apr 2013 17:08:50 +0000 (10:08 -0700)]
mesa: add renderbuffer attachment Layered field

If glFramebufferTexture is used, then the framebuffer attachment is
layered.

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
11 years agomesa: add renderbuffer Depth field
Jordan Justen [Thu, 18 Apr 2013 17:05:51 +0000 (10:05 -0700)]
mesa: add renderbuffer Depth field

With glFramebufferTexture, a renderbuffer may support
all layers of the texture, so we need the depth of the
renderbuffer to check for consistency which is required
for framebuffer completeness.

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
11 years agomesa: add usage examples to get-pick-list and shortlog scripts
Andreas Boll [Thu, 18 Apr 2013 07:32:39 +0000 (09:32 +0200)]
mesa: add usage examples to get-pick-list and shortlog scripts

NOTE: This is a candidate for the stable branches.

11 years agodocs: add info about bugzilla_mesa.sh script
Andreas Boll [Wed, 17 Apr 2013 16:45:15 +0000 (18:45 +0200)]
docs: add info about bugzilla_mesa.sh script

11 years agomesa: Add a script to generate the list of fixed bugs
Andreas Boll [Wed, 17 Apr 2013 16:14:44 +0000 (18:14 +0200)]
mesa: Add a script to generate the list of fixed bugs

This list appears in the fixed bugs section of the release notes.

v2: Add usage examples

NOTE: This is a candidate for the stable branches.

11 years agoscons: remove IN_DRI_DRIVER
Andreas Boll [Fri, 19 Apr 2013 21:20:01 +0000 (23:20 +0200)]
scons: remove IN_DRI_DRIVER

Not used anymore.

11 years agobuild: remove unused API_DEFINES
Andreas Boll [Wed, 27 Feb 2013 09:10:14 +0000 (10:10 +0100)]
build: remove unused API_DEFINES

Reviewed-by: Matt Turner <mattst88@gmail.com>
11 years agoconfigure: remove IN_DRI_DRIVER
Brian Paul [Tue, 26 Feb 2013 15:49:34 +0000 (08:49 -0700)]
configure: remove IN_DRI_DRIVER

Not used anymore.

v2: Andreas Boll <andreas.boll.dev@gmail.com>
    - split patch into two patches
    - remove more unused code

Reviewed-by: Matt Turner <mattst88@gmail.com>
11 years agoconfigure: remove FEATURE_GL/ES1/ES2
Brian Paul [Tue, 26 Feb 2013 15:49:34 +0000 (08:49 -0700)]
configure: remove FEATURE_GL/ES1/ES2

Not used anymore.

v2: Andreas Boll <andreas.boll.dev@gmail.com>
    - split patch into two patches

Reviewed-by: Matt Turner <mattst88@gmail.com>
11 years agointel: use automake conditionals for defining FEATURE_{ES1,ES2}
Andreas Boll [Fri, 19 Apr 2013 18:41:08 +0000 (20:41 +0200)]
intel: use automake conditionals for defining FEATURE_{ES1,ES2}

Removes the need of API_DEFINES.

Reviewed-by: Matt Turner <mattst88@gmail.com>
11 years agoegl-static: use automake conditionals for defining FEATURE_{GL,ES1,ES2}
Andreas Boll [Wed, 17 Apr 2013 09:16:27 +0000 (11:16 +0200)]
egl-static: use automake conditionals for defining FEATURE_{GL,ES1,ES2}

Removes the need of API_DEFINES.

Reviewed-by: Matt Turner <mattst88@gmail.com>
11 years agointel: remove executable bit from C file
Andreas Boll [Fri, 19 Apr 2013 16:37:52 +0000 (18:37 +0200)]
intel: remove executable bit from C file

Reviewed-by: Matt Turner <mattst88@gmail.com>
11 years agodocs: s/Aprile/April/
Brian Paul [Wed, 1 May 2013 19:17:21 +0000 (13:17 -0600)]
docs: s/Aprile/April/

11 years agodocs: fix 9.1.2 release notes
Andreas Boll [Wed, 1 May 2013 19:00:54 +0000 (21:00 +0200)]
docs: fix 9.1.2 release notes

11 years agovbo: fix possible use-after-free segfault after a VAO is deleted
Marek Olšák [Tue, 23 Apr 2013 23:29:26 +0000 (01:29 +0200)]
vbo: fix possible use-after-free segfault after a VAO is deleted

This like the fifth attempt to fix the issue.

Also with the new "validating" flag, we can set recalculate_inputs to FALSE
earlier in vbo_bind_arrays, because _mesa_update_state won't change it.

NOTE: This is a candidate for the stable branches.

v2: fixed a typo

Reviewed-by: Brian Paul <brianp@vmware.com>
11 years agoi965/vs: Fix textureGrad() with shadow samplers on Haswell.
Kenneth Graunke [Wed, 13 Feb 2013 05:51:17 +0000 (21:51 -0800)]
i965/vs: Fix textureGrad() with shadow samplers on Haswell.

The shadow comparitor needs to be loaded into the Z component of the
last DWord.

Fixes es3conform's shadow_execution_vert and oglconform's
shadow-grad advanced.textureGrad.1D tests on Haswell.

NOTE: This is a candidate for stable branches.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
11 years agoi965: Lower textureGrad() for samplerCubeShadow.
Kenneth Graunke [Wed, 13 Feb 2013 05:51:16 +0000 (21:51 -0800)]
i965: Lower textureGrad() for samplerCubeShadow.

According to the Ivybridge PRM, Volume 4 Part 1, page 130, in the
section for the sample_d message: "The r coordinate contains the faceid,
and the r gradients are ignored by hardware."

This doesn't match GLSL, which provides gradients for all of the
coordinates.  So we would need to do some math to compute the face ID
before using sample_d.  We currently don't have any code to do that.

However, we do have a lowering pass that converts textureGrad to
textureLod, which solves this problem.  Since textureGrad on three
components is sufficiently obscure, it's not a performance path.

For now, only handle samplerCubeShadow; we need tests for samplerCube
and samplerCubeArray.

Fixes es3conform's shadow_comparison_frag test on Haswell.

NOTE: This is a candidate for stable branches.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
11 years agoradeon/uvd: fix quant scan order for mpeg2
Christian König [Tue, 30 Apr 2013 17:38:24 +0000 (19:38 +0200)]
radeon/uvd: fix quant scan order for mpeg2

Signed-off-by: Christian König <christian.koenig@amd.com>
11 years agost/vdpau: fix background handling in the mixer
Christian König [Tue, 30 Apr 2013 12:55:14 +0000 (14:55 +0200)]
st/vdpau: fix background handling in the mixer

Signed-off-by: Christian König <christian.koenig@amd.com>
11 years agovl/buffer: use 2D_ARRAY instead of 3D textures
Christian König [Tue, 30 Apr 2013 12:40:40 +0000 (14:40 +0200)]
vl/buffer: use 2D_ARRAY instead of 3D textures

Signed-off-by: Christian König <christian.koenig@amd.com>
11 years agovl/compositor: cleanup background clearing
Christian König [Mon, 29 Apr 2013 15:43:04 +0000 (17:43 +0200)]
vl/compositor: cleanup background clearing

Add an extra parameter to specify if we should clear the render target.

Signed-off-by: Christian König <christian.koenig@amd.com>
11 years agoswrast: add casts for ImageSlices pointer arithmetic
Brian Paul [Tue, 30 Apr 2013 19:35:23 +0000 (13:35 -0600)]
swrast: add casts for ImageSlices pointer arithmetic

MSVC doesn't like pointer arithmetic with void * so use GLubyte *.

Reviewed-by: Jose Fonseca<jfonseca@vmware.com>
11 years agoilo: fix PIPE_CAP_MAX_STREAM_OUTPUT_BUFFERS
Chia-I Wu [Wed, 1 May 2013 09:40:50 +0000 (17:40 +0800)]
ilo: fix PIPE_CAP_MAX_STREAM_OUTPUT_BUFFERS

On GEN7+, is->dev.has_gen7_sol_reset is required.

11 years agoilo: enable SO support on GEN7
Chia-I Wu [Mon, 29 Apr 2013 00:47:33 +0000 (08:47 +0800)]
ilo: enable SO support on GEN7

11 years agoilo: reset SO write offsets for new SO targets
Chia-I Wu [Sun, 28 Apr 2013 23:26:37 +0000 (07:26 +0800)]
ilo: reset SO write offsets for new SO targets

When the SO targets are changed and no appending is requested, we need to send
SOL_RESET on GEN7+.

11 years agoilo: correctly program SO states for GEN7
Chia-I Wu [Sun, 28 Apr 2013 23:47:05 +0000 (07:47 +0800)]
ilo: correctly program SO states for GEN7

With the commands supported by GPE, we can finally program the states.

11 years agoilo: implement GEN7 SO GPE functions
Chia-I Wu [Sun, 28 Apr 2013 19:27:29 +0000 (03:27 +0800)]
ilo: implement GEN7 SO GPE functions

They were just stubs before.

11 years agoilo: add gen6_pipeline_update_max_svbi()
Chia-I Wu [Wed, 1 May 2013 08:58:10 +0000 (16:58 +0800)]
ilo: add gen6_pipeline_update_max_svbi()

Move max_svbi calculation to a helper function and make it available for other
GENs.

11 years agoilo: expose register indices of OUTs in ilo_shader
Chia-I Wu [Mon, 29 Apr 2013 00:25:27 +0000 (08:25 +0800)]
ilo: expose register indices of OUTs in ilo_shader

pipe_stream_output_info tells us which of OUT[i] needs to be written out.
We need the info to map OUT[i] to VUE offset.

11 years agoilo: allow one-off flags to be specified for CP
Chia-I Wu [Sun, 28 Apr 2013 23:22:00 +0000 (07:22 +0800)]
ilo: allow one-off flags to be specified for CP

It will be used for SOL_RESET on GEN7.

11 years agoilo: fix tiling/size for special-purpose resources
Chia-I Wu [Tue, 30 Apr 2013 07:30:01 +0000 (15:30 +0800)]
ilo: fix tiling/size for special-purpose resources

We do not allocate such resources yet though.