mesa.git
13 years agoglsl: Fix type mismatch when incrementing or decrementing uint.
Paul Berry [Tue, 1 Nov 2011 01:22:48 +0000 (18:22 -0700)]
glsl: Fix type mismatch when incrementing or decrementing uint.

When converting an expression like "++x" to GLSL IR we were failing to
account for the possibility that x might be an unsigned integral type.
As a result the user would receive a bogus error message "Could not
implicitly convert operands to arithmetic operator".

Fixes piglit tests {vs,fs}-{increment,decrement}-uint.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
13 years agogallium/dri: Remove references to vblank.
José Fonseca [Wed, 2 Nov 2011 16:16:50 +0000 (16:16 +0000)]
gallium/dri: Remove references to vblank.

13 years agoFix gallium dri compile and a merge conflict
Kristian Høgsberg [Wed, 2 Nov 2011 16:01:23 +0000 (12:01 -0400)]
Fix gallium dri compile and a merge conflict

13 years agosvga: Tag cursors with the scanout hint
Jakob Bornecrantz [Tue, 1 Nov 2011 04:10:17 +0000 (05:10 +0100)]
svga: Tag cursors with the scanout hint

Signed-off-by: Jakob Bornecrantz <jakob@vmware.com>
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
13 years agodri: Remove unused memops.h
Kristian Høgsberg [Fri, 28 Oct 2011 21:47:53 +0000 (17:47 -0400)]
dri: Remove unused memops.h

13 years agodri: Remove a few unused dri helper functions
Kristian Høgsberg [Fri, 28 Oct 2011 21:44:01 +0000 (17:44 -0400)]
dri: Remove a few unused dri helper functions

13 years agodri: Remove unused dri texmem.c
Kristian Høgsberg [Fri, 28 Oct 2011 21:39:40 +0000 (17:39 -0400)]
dri: Remove unused dri texmem.c

13 years agodri: Move _dri_texformat_* to radeon_texture.c
Kristian Høgsberg [Fri, 28 Oct 2011 21:35:16 +0000 (17:35 -0400)]
dri: Move _dri_texformat_* to radeon_texture.c

They are only used by the r200 driver now.

13 years agor200: Don't use driIsTextureResident helper
Kristian Høgsberg [Fri, 28 Oct 2011 21:04:35 +0000 (17:04 -0400)]
r200: Don't use driIsTextureResident helper

With DRI2, textures are always resident and using the DRI texmem helper
here is broken anyway, since nothing else uses it.

13 years agodri: Drop unused dri renderbuffer helper functions
Kristian Høgsberg [Fri, 28 Oct 2011 20:51:11 +0000 (16:51 -0400)]
dri: Drop unused dri renderbuffer helper functions

13 years agodri: Move driUpdateFramebufferSize to dri_util.c
Kristian Høgsberg [Fri, 28 Oct 2011 20:49:44 +0000 (16:49 -0400)]
dri: Move driUpdateFramebufferSize to dri_util.c

drirenderbuffer.[ch] is going bye-bye.

13 years agodri: Remove unnecessary #defines and #includes
Kristian Høgsberg [Fri, 28 Oct 2011 20:27:12 +0000 (16:27 -0400)]
dri: Remove unnecessary #defines and #includes

13 years agodri: Remove unnecessary timestamp pointer indirection
Kristian Høgsberg [Fri, 28 Oct 2011 20:22:03 +0000 (16:22 -0400)]
dri: Remove unnecessary timestamp pointer indirection

13 years agodri: Remove unused fields from __DriverAPIRec
Kristian Høgsberg [Fri, 28 Oct 2011 20:03:11 +0000 (16:03 -0400)]
dri: Remove unused fields from __DriverAPIRec

13 years agodri: Remove dri2.enabled flag
Kristian Høgsberg [Fri, 28 Oct 2011 19:56:15 +0000 (15:56 -0400)]
dri: Remove dri2.enabled flag

DRI2 is always enabled now.

13 years agodri: Remove DRI1 fields from DRI structs
Kristian Høgsberg [Fri, 28 Oct 2011 19:53:20 +0000 (15:53 -0400)]
dri: Remove DRI1 fields from DRI structs

13 years agodri: Remove remaining DRI1 vblank code
Kristian Høgsberg [Fri, 28 Oct 2011 19:47:41 +0000 (15:47 -0400)]
dri: Remove remaining DRI1 vblank code

13 years agodri: Remove cliprect information from __DRIdrawable
Kristian Høgsberg [Fri, 28 Oct 2011 19:40:01 +0000 (15:40 -0400)]
dri: Remove cliprect information from __DRIdrawable

13 years agodri: Drop unused driCalculateSwapUsage
Kristian Høgsberg [Fri, 28 Oct 2011 19:25:09 +0000 (15:25 -0400)]
dri: Drop unused driCalculateSwapUsage

13 years agodri: Remove unused driIntersectArea
Kristian Høgsberg [Fri, 28 Oct 2011 19:23:39 +0000 (15:23 -0400)]
dri: Remove unused driIntersectArea

13 years agodri: Drop __driUtilUpdateDrawableInfo and helper macros
Kristian Høgsberg [Fri, 28 Oct 2011 19:22:43 +0000 (15:22 -0400)]
dri: Drop __driUtilUpdateDrawableInfo and helper macros

13 years agodri: Remove unused driSwapBuffers
Kristian Høgsberg [Fri, 28 Oct 2011 19:18:22 +0000 (15:18 -0400)]
dri: Remove unused driSwapBuffers

13 years agodri: Drop unused driCreateNewContext
Kristian Høgsberg [Fri, 28 Oct 2011 19:17:20 +0000 (15:17 -0400)]
dri: Drop unused driCreateNewContext

13 years agodri: Fold driCreateNewDrawable into dri2CreateNewDrawable
Kristian Høgsberg [Fri, 28 Oct 2011 19:14:41 +0000 (15:14 -0400)]
dri: Fold driCreateNewDrawable into dri2CreateNewDrawable

13 years agodri: Remove unsused driCreateNewScreen
Kristian Høgsberg [Fri, 28 Oct 2011 19:11:15 +0000 (15:11 -0400)]
dri: Remove unsused driCreateNewScreen

13 years agodri: Remove unused __DRIswapInfoRec
Kristian Høgsberg [Fri, 28 Oct 2011 19:08:24 +0000 (15:08 -0400)]
dri: Remove unused __DRIswapInfoRec

13 years agodri: Remove driMediaStreamCounterExtension
Kristian Høgsberg [Fri, 28 Oct 2011 19:02:42 +0000 (15:02 -0400)]
dri: Remove driMediaStreamCounterExtension

Also DRI1-only.

13 years agodri: Remove driSwapControlExtension
Kristian Høgsberg [Fri, 28 Oct 2011 19:01:20 +0000 (15:01 -0400)]
dri: Remove driSwapControlExtension

DRI1-only as well.

13 years agodri: Drop driCopySubBufferExtension
Kristian Høgsberg [Fri, 28 Oct 2011 18:57:34 +0000 (14:57 -0400)]
dri: Drop driCopySubBufferExtension

Another DRI1-only extension.

13 years agodri: Drop driReadDrawableExtension
Kristian Høgsberg [Fri, 28 Oct 2011 18:55:25 +0000 (14:55 -0400)]
dri: Drop driReadDrawableExtension

All DRI2 drivers support setting a separate read drawable.

13 years agodri: Drop driLegacyExtension
Kristian Høgsberg [Fri, 28 Oct 2011 18:52:28 +0000 (14:52 -0400)]
dri: Drop driLegacyExtension

There are no DRI1 drivers left.

13 years agost/dri/sw: add get_image inline helper similar to put_image
George Sapountzis [Wed, 2 Nov 2011 13:02:10 +0000 (15:02 +0200)]
st/dri/sw: add get_image inline helper similar to put_image

TODO: check if GetImage works with passing the pitch as width, similar to PutImage,
which avoids the extra copy, ala dri_sw_displaytarget_display() in src/gallium/winsys/sw/dri/dri_sw_winsys.c

13 years agost/dri/sw: Fix incorrect cast
Benjamin Franzke [Wed, 2 Nov 2011 14:56:23 +0000 (15:56 +0100)]
st/dri/sw: Fix incorrect cast

Missed one casting error due to rebase for commit
98aa2a8f725e44aec8bd998fe436a134e94f13bb.

13 years agost/dri/sw: Implement texture_from_pixmap
Benjamin Franzke [Fri, 9 Sep 2011 08:12:23 +0000 (10:12 +0200)]
st/dri/sw: Implement texture_from_pixmap

This is a cleanup of commit 02f1b50987c0d24da3dcc36dbb44821c20d0660c.

Update tex buffer using a dri_drawable hook from implemented in sw/drisw.c.
This saves us the duplication of dri_drawable.c.

CC: Stuart Abercrombie <sabercrombie@chromium.org>
CC: Stéphane Marchesin <marcheu@chromium.org>
13 years agoRevert "Duplicate state_tracker/dri/sw/dri_drawable.c"
Benjamin Franzke [Fri, 9 Sep 2011 07:34:59 +0000 (09:34 +0200)]
Revert "Duplicate state_tracker/dri/sw/dri_drawable.c"

This reverts commit 569bde1fa7d03fb7688d0d391b32e61e857ad44e.

CC: Stuart Abercrombie <sabercrombie@chromium.org>
CC: Stéphane Marchesin <marcheu@chromium.org>
13 years agoRevert "state_trackers/dri/sw: Implement texture_from_pixmap."
Benjamin Franzke [Fri, 9 Sep 2011 07:34:50 +0000 (09:34 +0200)]
Revert "state_trackers/dri/sw: Implement texture_from_pixmap."

This reverts commit 02f1b50987c0d24da3dcc36dbb44821c20d0660c.

CC: Stuart Abercrombie <sabercrombie@chromium.org>
CC: Stéphane Marchesin <marcheu@chromium.org>
13 years agor600g: add comment about special shader exports
Alex Deucher [Wed, 2 Nov 2011 13:38:46 +0000 (09:38 -0400)]
r600g: add comment about special shader exports

Certain exports (position, point size, etc.) are treated
specially by the shader and not counted as generic exports.

Note the exports and any relevant related state bits.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
13 years agor600g: take into account PSIZE output for spi mapping
Vadim Girlin [Wed, 2 Nov 2011 07:50:59 +0000 (11:50 +0400)]
r600g: take into account PSIZE output for spi mapping

Skip PSIZE and POSITION when counting VS outputs.
Fixes https://bugs.freedesktop.org/show_bug.cgi?id=42435

Tested without regressions on evergreen.

NOTE: this is a candidate for the 7.11 branch.

Signed-off-by: Vadim Girlin <vadimgirlin@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
13 years agosvga: Update state prototypes to return pipe_error.
José Fonseca [Wed, 2 Nov 2011 10:32:20 +0000 (10:32 +0000)]
svga: Update state prototypes to return pipe_error.

Fixes a bunch of prototype mismatch warnings..

13 years agoBuild fix for -Werror=int-to-pointer-cast -Werror=pointer-to-int-cast
Jeremy Huddleston [Wed, 2 Nov 2011 04:04:34 +0000 (21:04 -0700)]
Build fix for -Werror=int-to-pointer-cast -Werror=pointer-to-int-cast

Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
13 years agointel: fix potential segfault error
Yuanhan Liu [Tue, 1 Nov 2011 09:41:08 +0000 (17:41 +0800)]
intel: fix potential segfault error

Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
13 years agoswrast: Use MapRenderbuffer for glReadPixels(GL_DEPTH_COMPONENT) fast-paths.
Eric Anholt [Tue, 11 Oct 2011 22:08:15 +0000 (15:08 -0700)]
swrast: Use MapRenderbuffer for glReadPixels(GL_DEPTH_COMPONENT) fast-paths.

This fixes issues with the code playing fast and loose with types of
buffers, and as a bonus avoids the wrappers that were previously used
to pull bits out of packed depth/stencil buffers.

Reviewed-by: Brian Paul <brianp@vmware.com>
13 years agomesa: Make unpack_uint_z_row return 32 bits of data.
Eric Anholt [Thu, 13 Oct 2011 16:38:30 +0000 (09:38 -0700)]
mesa: Make unpack_uint_z_row return 32 bits of data.

Some of the return values were u32, some were 24 bits, and z16
returned 16 bits.  The caller would have to do all the work of
interpreting the format all over again.  However, there are no callers
of this function at this point.

Reviewed-by: Brian Paul <brianp@vmware.com>
13 years agonouveau: Add implementation of MapRenderbuffer.
Eric Anholt [Mon, 17 Oct 2011 17:18:30 +0000 (10:18 -0700)]
nouveau: Add implementation of MapRenderbuffer.

Perhaps the easiest implementation, nouveau can directly map buffers
even if tiled, and uses separate surfaces for its texture
renderbuffers so we don't have to worry about that offset.

Reviewed-by: Brian Paul <brianp@vmware.com>
13 years agoradeon: Add implementation of MapRenderbuffer.
Eric Anholt [Mon, 17 Oct 2011 16:50:52 +0000 (09:50 -0700)]
radeon: Add implementation of MapRenderbuffer.

Unlike intel, we do a blit to/from GTT memory in order to
untile/retile the renderbuffer data, since we don't have fence
registers for accessing it.

(There is software tiling code in radeon_tile.c, but it's unused and
doesn't support macro tiling)

Reviewed-by: Brian Paul <brianp@vmware.com>
13 years agointel: Use MapRenderbuffer in spans setup.
Eric Anholt [Tue, 11 Oct 2011 21:46:31 +0000 (14:46 -0700)]
intel: Use MapRenderbuffer in spans setup.

Reviewed-by: Brian Paul <brianp@vmware.com>
13 years agointel: Add an implementation of MapRenderbuffer.
Eric Anholt [Tue, 11 Oct 2011 21:22:45 +0000 (14:22 -0700)]
intel: Add an implementation of MapRenderbuffer.

v2: Add separate stencil S8 W-tile swizzling/deswizzling.  Tested for
    the swizzling case with env INTEL_SEPARATE_STENCIL=1 INTEL_HIZ=1
    ./bin/hiz-depth-stencil-test-fbo-d24-s8
v3: Apply Chad's fix for S8 window system buffers.

Reviewed-by: Chad Versace <chad@chad-versace.us>
13 years agointel: Improve the debug info for renderbuffer allocation.
Eric Anholt [Tue, 1 Nov 2011 00:07:34 +0000 (17:07 -0700)]
intel: Improve the debug info for renderbuffer allocation.

13 years agoAdd MapRenderbuffer implementations for software drivers.
Eric Anholt [Thu, 13 Oct 2011 19:02:43 +0000 (12:02 -0700)]
Add MapRenderbuffer implementations for software drivers.

Mesa core's is generic for things like osmesa.

For swrast_dri.so, we have to do Y flipping.  The front-buffer path
isn't actually tested, though, because both before and after it fails
with a BadMatch in XGetImage.

Reviewed-by: Brian Paul <brianp@vmware.com>
13 years agomesa: Add a driver hook for mapping renderbuffers.
Eric Anholt [Tue, 11 Oct 2011 21:12:50 +0000 (14:12 -0700)]
mesa: Add a driver hook for mapping renderbuffers.

Reviewed-by: Brian Paul <brianp@vmware.com>
13 years agoRevert "intel: Drop the immediate validation of the texture object in TFP."
Eric Anholt [Tue, 1 Nov 2011 16:24:47 +0000 (09:24 -0700)]
Revert "intel: Drop the immediate validation of the texture object in TFP."

This reverts commit abaebcee787eeb8a89bf7a82ed4d1532fcde5e39.

The assertion I made was that "the zero-copy code in validation" would
zero copy.  Of course, I deleted that check back in January because
the two sites that would trigger it (glTexImage() and this one) both
immediately bound their mt to the object, making the other check
pointless.

Removes two extra blits in glx-tfp.  Also fixed the Android home
screen, which wasn't rendering because the extra copy broke the
relationship between the texture and the eglimage.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=42152
Tested-by: Chad Versace <chad@chad-versace.us>
Tested-by: Eugeni Dodonov <eugeni.dodonov@intel.com>
13 years agomesa: make save_EndConditionalRender param list (void) to silence warning
Brian Paul [Tue, 1 Nov 2011 19:34:52 +0000 (13:34 -0600)]
mesa: make save_EndConditionalRender param list (void) to silence warning

13 years agopostprocess: remove const qualifiers on unsigned int parameters
Brian Paul [Tue, 1 Nov 2011 17:34:51 +0000 (11:34 -0600)]
postprocess: remove const qualifiers on unsigned int parameters

to match the prototype in postprocess.h

13 years agoscons: include src/glsl/Makefile.sources
Chia-I Wu [Sun, 30 Oct 2011 11:32:25 +0000 (19:32 +0800)]
scons: include src/glsl/Makefile.sources

Reviewed-by: Brian Paul <brianp@vmware.com>
13 years agoandroid: include src/glsl/Makefile.sources
Chia-I Wu [Sun, 30 Oct 2011 11:32:41 +0000 (19:32 +0800)]
android: include src/glsl/Makefile.sources

This also fixes the build error due to missing link_uniforms.cpp in the source
lists.

Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Chad Versace <chad@chad-versace.us>
[olv: the missing link_uniforms.cpp was added before this patch is committed]

13 years agoglsl: Refactor source lists to Makefile.sources
Chia-I Wu [Sun, 30 Oct 2011 11:28:35 +0000 (19:28 +0800)]
glsl: Refactor source lists to Makefile.sources

With the hope that Android.mk and SConscript can share the file to reduce
future breakage.

Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Chad Versace <chad@chad-versace.us>
13 years agomesa: Fix memory leak in out-of-memory path.
Vinson Lee [Thu, 27 Oct 2011 21:51:08 +0000 (14:51 -0700)]
mesa: Fix memory leak in out-of-memory path.

Fixes Coverity resource leak defect.

13 years agodrisw: Enable SGI_make_current_read for direct contexts
Adam Jackson [Tue, 1 Nov 2011 17:52:48 +0000 (13:52 -0400)]
drisw: Enable SGI_make_current_read for direct contexts

Equivalent to glXMakeContextCurrent, which had better already work since
we advertise GLX 1.3.

Signed-off-by: Adam Jackson <ajax@redhat.com>
13 years agost/xa: Add a scanout flag for surfaces
Thomas Hellstrom [Tue, 1 Nov 2011 16:42:21 +0000 (17:42 +0100)]
st/xa: Add a scanout flag for surfaces

Bump minor to signal availability.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Jakob Bornecrantz <jakob@vmware.com>
13 years agoBuild link_uniforms for android
Eugeni Dodonov [Tue, 1 Nov 2011 17:04:01 +0000 (15:04 -0200)]
Build link_uniforms for android

Otherwise android-x86 build fails.

Signed-off-by: Eugeni Dodonov <eugeni.dodonov@intel.com>
13 years agollvmpipe: Re-advertise support for stencil formats...
José Fonseca [Tue, 1 Nov 2011 12:28:47 +0000 (12:28 +0000)]
llvmpipe: Re-advertise support for stencil formats...

"no fetch_rgba_float for uint formats" => "llvmpipe does not support uint formats"

"stencil == uint" => "llvmpipe does not support stencil"

13 years agollvmpipe: Use -1 instead of ~0 for "no slot".
José Fonseca [Mon, 31 Oct 2011 19:35:55 +0000 (19:35 +0000)]
llvmpipe: Use -1 instead of ~0 for "no slot".

As the value of unsigned ~0 depends on the bit-width.

Fixes fdo 42411.

13 years agodocs: update relnotes-7.12
Marek Olšák [Sat, 29 Oct 2011 12:33:08 +0000 (14:33 +0200)]
docs: update relnotes-7.12

13 years agodri/nouveau: Enable NV_fog_distance on NV10 and NV20 hardware
Nicholas Miell [Tue, 20 Sep 2011 13:20:39 +0000 (15:20 +0200)]
dri/nouveau: Enable NV_fog_distance on NV10 and NV20 hardware

Add support for NV_fog_distance to the NV10 and NV20 drivers.

[ Francisco Jerez: Fix fog coord. signedness for
  GL_EYE_RADIAL_NV/GL_EYE_PLANE on nv20 ]

13 years agost/mesa: Enable NV_fog_distance for Gallium drivers
Nicholas Miell [Tue, 20 Sep 2011 06:17:01 +0000 (23:17 -0700)]
st/mesa: Enable NV_fog_distance for Gallium drivers

The fixed-function generated vertex program is all that's needed for
Gallium drivers.

13 years agomesa: Add NV_fog_distance to the fixed-function vertex program generator
Nicholas Miell [Tue, 20 Sep 2011 06:17:00 +0000 (23:17 -0700)]
mesa: Add NV_fog_distance to the fixed-function vertex program generator

This is all that is needed to implement NV_fog_distance on programmable
hardware.

13 years agomesa: Add the basics for the NV_fog_distance extension
Nicholas Miell [Tue, 20 Sep 2011 06:16:59 +0000 (23:16 -0700)]
mesa: Add the basics for the NV_fog_distance extension

No driver implements it yet.

13 years agoralloc: Use _vscprintf on MinGW.
José Fonseca [Tue, 1 Nov 2011 06:58:29 +0000 (06:58 +0000)]
ralloc: Use _vscprintf on MinGW.

MinGW uses MSVC's runtime DLLs for most of C runtime's functions, and
there has same semantics for vsnprintf.

Not sure how this worked until now -- maybe one of the internal
vsnprintf implementations was taking precedence.

13 years agosvga: Fix potential buffer overflow in rs draw state.
José Fonseca [Mon, 31 Oct 2011 19:26:53 +0000 (19:26 +0000)]
svga: Fix potential buffer overflow in rs draw state.

Reviewed-by: Brian Paul <brianp@vmware.com>
13 years agollvmpipe: Remove unsed variables.
José Fonseca [Mon, 31 Oct 2011 19:36:07 +0000 (19:36 +0000)]
llvmpipe: Remove unsed variables.

13 years agoutil: Add missing initializer.
José Fonseca [Mon, 31 Oct 2011 19:04:08 +0000 (19:04 +0000)]
util: Add missing initializer.

13 years agoi965: Fix flat integral varyings.
Paul Berry [Tue, 25 Oct 2011 00:46:26 +0000 (17:46 -0700)]
i965: Fix flat integral varyings.

Previously, the vertex and fragment shader back-ends assumed that all
varyings were floats.  In GLSL 1.30 this is no longer true--they can
also be of integral types provided that they have an interpolation
qualifier of "flat".

This required two changes in each back-end: assigning the correct type
to the register that holds the varying value during shader execution,
and assigning the correct type to the register that ties the varying
value to the rest of the graphics pipeline (the message register in
the case of VS, and the payload register in the case of FS).

Fixes piglit tests fs-int-interpolation and fs-uint-interpolation.

Acked-by: Kenneth Graunke <kenneth@whitecape.org>
13 years agomesa: Add glsl_type::get_scalar_type() function.
Paul Berry [Tue, 25 Oct 2011 23:24:03 +0000 (16:24 -0700)]
mesa: Add glsl_type::get_scalar_type() function.

This function is similar to get_base_type(), but when called on
arrays, it returns the scalar type composing the array.  For example,
glsl_type(vec4[]) => float_type.

Acked-by: Kenneth Graunke <kenneth@whitecape.org>
13 years agoi965/gen6+: Switch GLSL from ALT to IEEE floating point mode
Paul Berry [Sat, 8 Oct 2011 00:37:32 +0000 (17:37 -0700)]
i965/gen6+: Switch GLSL from ALT to IEEE floating point mode

i965 graphics hardware has two floating point modes: ALT and IEEE.  In
ALT mode, floating-point operations never generate infinities or NaNs,
and MOV instructions translate infinities and NaNs to finite values.
In IEEE mode, infinities and NaNs behave as specified in the IEEE 754
spec.

Previously, we used ALT mode for all vertex and fragment programs,
whether they were GLSL programs or ARB programs.  The GLSL spec is
sufficiently vague about how infs and nans are to be handled that it
was unclear whether this mode was compliant with the GLSL 1.30 spec or
not, and it made it very difficult to test the isinf() and isnan()
functions.

This patch changes i965 GLSL programs to use IEEE floating-point mode,
which is clearly compliant with GLSL 1.30's inf/nan requirements.  In
addition to making the Piglit isinf and isnan tests pass, this paves
the way for future support of the ARB_shader_precision extension.

Unfortunately we still have to use ALT floating-point mode when
executing ARB programs, because those programs require 0^0 == 1, and
i965 hardware generates 0^0 == NaN in IEEE mode.

Fixes piglit tests "isinf-and-isnan fs_fbo", "isinf-and-isnan vs_fbo",
and {fs,vs}-{isinf,isnan}-{vec2,vec3,vec4}.

13 years agoglsl: Add isinf() and isnan() builtins.
Paul Berry [Thu, 6 Oct 2011 17:20:10 +0000 (10:20 -0700)]
glsl: Add isinf() and isnan() builtins.

The implementations are as follows:

isinf(x) = (abs(x) == +infinity)
isnan(x) = (x != x)

Note: the latter formula is not necessarily obvious.  It works because
NaN is the only floating point number that does not equal itself.

Fixes piglit tests "isinf-and-isnan fs_basic" and "isinf-and-isnan
vs_basic".

13 years agoglsl: Add '.ir' extension to builtin IR files
Paul Berry [Thu, 6 Oct 2011 17:31:34 +0000 (10:31 -0700)]
glsl: Add '.ir' extension to builtin IR files

This patch adds the extension '.ir' to all the files in
src/glsl/builtins/ir/, and changes generate_builtins.py so that it no
longer globs on '*' to find the files to build.  This prevents
spurious files (such as EMACS' infamous *~ backup files) from breaking
the build.

13 years agoglsl: Fix copy-paste error in constant_expression_value(ir_binop_nequal)
Paul Berry [Thu, 6 Oct 2011 18:28:42 +0000 (11:28 -0700)]
glsl: Fix copy-paste error in constant_expression_value(ir_binop_nequal)

The implementation of ir_binop_nequal in constant_expression_value()
appears to have been copy-and-pasted from the implementation of
ir_binop_equal, but with all instances of '==' changed to '!='.  This
is correct except for one minor flaw: one of those '==' operators was
in an assertion checking that the types of the two arguments were
equal.  That one needs to stay an '=='.

Fixes piglit tests {fs,vs}-inline-notequal.

13 years agodocs: list GL_ARB_texture_storage in 7.12 release notes
Brian Paul [Mon, 31 Oct 2011 16:52:57 +0000 (10:52 -0600)]
docs: list GL_ARB_texture_storage in 7.12 release notes

13 years agomesa: add note about immutable textures for _mesa_test_texobj_completeness()
Brian Paul [Mon, 31 Oct 2011 16:52:57 +0000 (10:52 -0600)]
mesa: add note about immutable textures for _mesa_test_texobj_completeness()

13 years agoswrast: implement GL_ARB_texture_storage
Brian Paul [Mon, 31 Oct 2011 16:52:57 +0000 (10:52 -0600)]
swrast: implement GL_ARB_texture_storage

13 years agost/mesa: implement GL_ARB_texture_storage
Brian Paul [Mon, 31 Oct 2011 16:52:57 +0000 (10:52 -0600)]
st/mesa: implement GL_ARB_texture_storage

13 years agomesa: plug in dispatch functions for GL_ARB_texture_storage
Brian Paul [Mon, 31 Oct 2011 16:52:57 +0000 (10:52 -0600)]
mesa: plug in dispatch functions for GL_ARB_texture_storage

13 years agomesa: compile new texstore.c file
Brian Paul [Mon, 31 Oct 2011 16:52:56 +0000 (10:52 -0600)]
mesa: compile new texstore.c file

13 years agomesa: init extension entry for GL_ARB_texture_storage
Brian Paul [Mon, 31 Oct 2011 16:52:56 +0000 (10:52 -0600)]
mesa: init extension entry for GL_ARB_texture_storage

13 years agomesa: display list dispatch for GL_ARB_texture_storage
Brian Paul [Mon, 31 Oct 2011 16:52:56 +0000 (10:52 -0600)]
mesa: display list dispatch for GL_ARB_texture_storage

13 years agomesa: new glTexImage error checks for GL_ARB_texture_storage
Brian Paul [Mon, 31 Oct 2011 16:52:56 +0000 (10:52 -0600)]
mesa: new glTexImage error checks for GL_ARB_texture_storage

If the texture memory was allocated with glTexStorage1/2/3D() we can
only change the image data with glTexSubImage calls.

13 years agomesa: glGetTexParameter support for GL_ARB_texture_storage
Brian Paul [Mon, 31 Oct 2011 16:52:56 +0000 (10:52 -0600)]
mesa: glGetTexParameter support for GL_ARB_texture_storage

13 years agomesa: implement new functions for GL_ARB_texture_storage
Brian Paul [Mon, 31 Oct 2011 16:52:56 +0000 (10:52 -0600)]
mesa: implement new functions for GL_ARB_texture_storage

This is the glTexStorage1D/2D/3D() functions.  Basically do error
checking then call the driver hook to actually allocate memory.

13 years agoglapi: regenerated API files for GL_ARB_texture_storage
Brian Paul [Mon, 31 Oct 2011 16:52:56 +0000 (10:52 -0600)]
glapi: regenerated API files for GL_ARB_texture_storage

13 years agoglapi: new API specs for GL_ARB_texture_storage
Brian Paul [Mon, 31 Oct 2011 16:52:56 +0000 (10:52 -0600)]
glapi: new API specs for GL_ARB_texture_storage

13 years agomesa: add new driver hook for AllocTextureStorage()
Brian Paul [Mon, 31 Oct 2011 16:52:56 +0000 (10:52 -0600)]
mesa: add new driver hook for AllocTextureStorage()

13 years agomesa: add new fields for GL_ARB_texture_storage
Brian Paul [Mon, 31 Oct 2011 16:52:56 +0000 (10:52 -0600)]
mesa: add new fields for GL_ARB_texture_storage

13 years agosvga: Flush the hwtnl primitives before writing to buffers.
José Fonseca [Thu, 27 Oct 2011 18:09:25 +0000 (19:09 +0100)]
svga: Flush the hwtnl primitives before writing to buffers.

svga keeps a small queue of similar primitive draws in order to coalesce
them into a single draw primitive command.

But the buffers referred in primitives not yet emitted were being ignored
in the considerations to flush or not the context.

This fixes piglit vbo-map-remap, vbo-subdata-sync, vbo-subdata-zero, and
Seeker.

Based on investigation and patch from Brian Paul.

Reviewed-By: Brian Paul <brianp@vmware.com>
13 years agost/xa: Plug a pipe context leak.
Thomas Hellstrom [Mon, 31 Oct 2011 13:01:39 +0000 (14:01 +0100)]
st/xa: Plug a pipe context leak.

Forgot to destroy the pipe context on xa context destroy.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
13 years agopipebuffer/debug: Fix a recursive mutex lock
Thomas Hellstrom [Mon, 31 Oct 2011 11:14:05 +0000 (12:14 +0100)]
pipebuffer/debug: Fix a recursive mutex lock

pb_debug_manager_dump was trying to take a lock already
held by all callers.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: José Fonseca <jfonseca@vmware.com>
13 years agoradeon/r200: drop remains of r300/r600 support along with old drm 1.x kernel
Dave Airlie [Sat, 29 Oct 2011 07:39:54 +0000 (08:39 +0100)]
radeon/r200: drop remains of r300/r600 support along with old drm 1.x kernel

This drops all the old drmSupports* checks since KMS does them all, and it
also drop R300_CLASS and R600_CLASS.

Signed-off-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
13 years agollvmpipe: fix a crash in non-SSE path
Chia-I Wu [Fri, 23 Sep 2011 03:40:49 +0000 (11:40 +0800)]
llvmpipe: fix a crash in non-SSE path

It is a typo went unnoticed.

13 years agointel: enable GL_OES_draw_texture
Chia-I Wu [Sun, 21 Aug 2011 13:47:39 +0000 (21:47 +0800)]
intel: enable GL_OES_draw_texture

13 years agometa: add _mesa_meta_DrawTex()
Chia-I Wu [Sun, 21 Aug 2011 13:08:25 +0000 (21:08 +0800)]
meta: add _mesa_meta_DrawTex()

It is set to dd->DrawTex.

Reviewed-by: Brian Paul <brianp@vmware.com>
[olv: set dd->DrawTex in _mesa_init_driver_functions]

13 years agor600g: reset the render condition to NULL when rendering unconditionally
Marek Olšák [Sat, 29 Oct 2011 14:19:45 +0000 (16:19 +0200)]
r600g: reset the render condition to NULL when rendering unconditionally